Flying Flux

God Bless and Goodbye

2011-05-11T11:00:00.003-04:00

A couple of months ago, I received a thick packet in the mailbox. Inside was a questionnaire, asking me to participate in a scientific study. "Oh?", I wondered curiously. Having had a closer look at the purpose of the study, I dutifully complied, filling out the forms to the best of my ability and promptly mailed back the results. The eagerness to complete this survey was quite personal. Here's the title of the survey:

Lymphoma. Cancer. This survey brought back a flood of memories, things that I scarely want to remember, but things that nevertheless gave me great insight into myself. After returning the questionnaire, I opened my filing drawer and pulled out a folder thick with dated reading materials. Here's a small sampling:

There were brochures, pamphlets, information guides, medication descriptions, fact sheets, phone directories, and yes, even scientific papers from the Journal of Clinical Oncology, something that I requested of my oncologist to provide for me. As an aside, I'm surprised by the rather low numbers in the sample size of these papers. One paper had a study size of 17 patients, and the other, 65 patients.

Amongst all that reading was, of course, that one single piece of paper that started it all. A blocky looking fax titled Surgical Pathology Report. You can see a snippet of the report below.

If you think I've focused the camera too close, well, those were really the only words that stood out when I first read the report. Lymphoma. Positive.

I kept my cool, mostly, until I got home and sat both of my boys on my lap. Neither of them at the time were yet of kindergarten age and they did not know why their father was crying. But that was the first and only time I shed tears for this. Life went on as before, save for occasional trips to the hospital and a couple of longer trips to Toronto for a second opinion. If one were given the chance to choose the type of cancer to get, mine would be it. My course of treatment was to not treat it at all. Wait it out. See if the body can fight it off. With my faith sustaining me and by the grace of God, I sit here today typing out this blog, cancer-free. One very important note: my decision to do nothing is peculiar to my form of cancer and not a general recommendation to other cancer patients.

Despite what I've written so far, this post isn't really about my cancer. It's just background information. To be a father and a husband, this is what is most important to me. Having been diagnosed with cancer didn't change this priority, but it did bring it into sharper focus. I know I need to take care of myself so I can take care of my family, so I can enjoy my time with them. But over the years, that focus slowly became blurry again as the emotional impact of that diagnosis slowly wore off. I consciously knew that I needed to take care of myself, but I also consciously committed myself to more and more projects that took time away from my family. That was the way it was until that questionnaire prompted me to re-evaluate how I was living my life. I was becoming more and more exhausted and it simply wasn't sustainable. I wasn't living up to what I promised myself years ago to take care of myself.

Having that clarity again is refreshing. What I thought I held dear I really didn't. This includes the online persona that I've cultivated for myself these past two and a half years. It includes my involvement with the successful Engineer Blogs. And yes, it even includes the Flying Flux. Once I came to this realization, it wasn't too hard to take the next step.

I am permanently leaving my blogging life behind and moving towards more meaningful things in the real world. I'll be making a similar announcement on Engineer Blogs later today. When politicians say they're stepping down to spend more time with their families, it usually means they're being forced out or were caught in some hanky panky or dirty money scandal. But in my case, I really am stepping down to spend more time with my family. I rather leave on a high note, on my own terms, than to let things slowly deteriorate into disrepair. And I do this with bittersweet feelings. After all, the Flying Flux has been my personal platform these past 2+ years to rant and rave on topics big and small. Where now can I publicly complain about the obtuseness of management, the paranoia of my colleagues, and the idiocy of my government?

But no matter the amount of fun I've had in maintaining the Flux and writing at Engineer Blogs, it pales in comparison to my real achievements in life, namely that I've managed to convince someone to bear my children (twice!) and that I've been able to keep these children alive for nearly a decade.

So thank you readers for your patronage. I shall miss you all. And on that note, it's time for me to take my plough and head back to the farm. God bless, and goodbye.

Heavy Breathing

2011-05-06T09:40:00.001-04:00

If I cared about doing such things, it would be quite easy to assume the ID of Mr. Poker. It's not hard to find out his credit card numbers, expiry dates, passport number, birthday, his wife's birthday, etc. The reason is that Mr. Poker has a high speaking volume. Very high. So much so, when we're on teleconference calls, I take off my headphones when he speaks. I can hear him loud and clear through the cubicle walls. The same is true when he makes personal phone calls, such as buying things over the phone, talking to people at the passport office, etc. His booming phone voice has driven other co-workers not on our team to move their desks to the other side of the office building. Unfortunately for the rest of us, we're stuck.

But Mr. Poker isn't just a loud talker. He's a loud breather. In conference calls, when he isn't speaking, everyone can still hear him pant like a perverted crank caller. And even after I asked him (nicely) to please put himself on mute when he isn't talking, he doesn't do it. I think he thinks the heavy breathing is done by someone else. Oh well. I guess this is what makes life interesting.

Evergreening the Geosphere

2011-05-05T14:55:00.000-04:00

I learned two new managerial buzzwords recently. The first one is an adjective -- evergreen. It means refreshed, never old, never stale, never withered, always new, always green. For example, let's make sure we have an evergreen view of the Gantt chart that doesn't reflect reality. The second is a verb -- to geo-scale. It means hiring, or scaling up the organization, will take into account all geographies in which we operate. To translate this into more concrete termes, it means we're going to hire people in China and India. To the rest of our geographical centres, here's a big f*&k you.

It's disheartening to see this kind of language being employed internally. Engineers are pretty straightforward people. We'd like straightforward statements. All this obfuscation through word-play doesn't mean squat to us. In fact, it comes across as insincere bullsh*t. If you're going to hire in India and China, just say you're going to hire in India and China. To spin it any other way just means you're no better than the guy that forces us to fill out meaningless forms for the sake of process.

Oh wait, never mind. You are the same guy. Carry on.

When Meaningless Forms Affect Your Salary

2011-05-02T23:27:00.007-04:00

Our super duper director, the one whose bright idea is to require engineering to design a zero-bug prototype, has been hard at work implementing new ways of doing things this past year. New ways of documenting. New ways of presenting data. New checklists. New ways to track schedule. It's all part of the zero-bug effort. And how did he come up with all these new ways of doing things? By not asking, even once, the people who were to actually going do these things -- the engineers.

When we mentioned that all these new processes will cost a lot of schedule overhead, he insisted that it would not. In fact, it'd be barely noticeable above what we're already doing today, he tells us. One of those fun new it-will-take-no-time documents is filling in the design checklist. Often, he'd get angry when he'd check our design documents repository to find it sparsely populated with these checklists. He'd threaten us, telling us that he doesn't want this intransigence to show up on our annual reviews, but he's willing to go there. The checklist is filled with useful items, like "Are all input and output labels spelled using all capital letters?" The answer for us is no, as we have been using lowercase for all of our designs for the past 7 years. There are many, many other useful checklist items, like "Has the design been simulated across all possible corner case permutations?" The answer is strictly no, but we do simulate a relevant subset. Else, the design will take ten years to complete.

So basically, the checklist is meaningless. We already have internal checklists and design reviews that are much more useful than this new fancy schmancy crap on which we were never consulted. And the only thing our director does with these checklists is to see that they exist. He doesn't actually look at them. Process for the sake of process. But if your annual review and salary is still important to you, then process is what you shall do.

Cognitive Dissonance #2

2011-04-28T13:49:00.001-04:00

Snapped this pic on vacation last year north of Toronto. Wonder if the fish and chips come with sweet and sour sauce.

Intent vs. Implementation

2011-04-25T13:47:00.001-04:00

When designing anything, there's what you would like to design, and what you ended up implementing. The design intent may be elegant, beautiful, even sublime. But overlook one small issue and the whole thing can fall apart.

During the last several months, I've been embroiled in issue which directly impacts FluxCorp's bottom line -- manufacturing yield. It all boils down to a bug in the design that causes intermittent failure in a small percentage of parts. The bug is quite simple. The design intent was to add two variables, i and j, together, or i+j. But we ended up implementing was i+i. A typo? Yes. A typo that's causing months of headaches? Yes.

My example here may be a bit abstract, so let me give you a real world example. Here are two pictures of a five star hotel bathroom in India. Beautiful bathroom, except for one minor implementation flaw that renders the bathtub mostly usable. Spot the problem?

Scolded by a Child

2011-04-17T23:43:00.003-04:00

This past week, the manager of the digital design group decided to give a good scolding to us analog guys. His frustration is due to the analog design changing on a regular basis, which sometimes require associated changes to the digital logic. He wonders why we analog guys can't do a bit of up-front feasibility study (let's say, 10% of our design time) and then finalize our designs' pin-outs, specs, and footprints up front. After all, the digital guys are able to do it. When Mr. Halal responded that analog is more complex, we got the "yeah, I knew you were going to say that" response. The manager may have known how we were going to respond, but he didn't seem convinced.

In the digital world, functionality is implemented in software-like code call RTL. Thus, when one finds a functional bug in a digital design, changes to a text file fixes the problem. Functionality aside, there are three other major specs that needs to be met -- power, area, and clock speed. Three variables. That's it. I'm not saying it's trivial work; it isn't. But let's go through an example of what an analog designer needs to worry about when designing an amplifier:

- frequency response
- gain
- phase margin
- gain margin
- biasing
- headroom
- DC offset
- slew rate
- power
- input impedance
- output impedance
- compensation capacitance
- area
- matching
- common mode
- common mode feedback stability
- common mode rejection
- power supply rejection
- gain compression
- intermodulation
- parasitic coupling
- ground loop
- package inductance
- electromigration
- over-voltage

I'm sure there are others I haven't thought of. To top it all off, changing one spec can ripple through and affect ten other specs. Sometimes, the inability to meet one spec means one must throw the design away and start over. And that's just an amplifier. There are many, many other analog circuits that go into a complete system. That's why analog design remains an art created by manual labour while digital circuits are synthesized, created by a software tool using RTL code as its input. By the time enough feasibility has been done on the analog design, the design is pretty much more than half-way finished.

I don't expect the digital manager to fully grasp the complexities of analog design. But at the very least, he can drop his condescension thinking that he knows it all. It's like being scolded by a child who's telling you that rocket science is easy because he just made a rocket yesterday out of paper-mâché.

I'm not big on EE class warfare. But in light of his most unwelcoming comments, perhaps I should just forward him this lovely poster (grabbed from The Amp Hour):

Losing Trust

2011-04-11T15:35:00.000-04:00

One of the new ways of doing things here at FluxCorp is to start tracking bugs we find in verification, no matter how large or small, so (a) the issue doesn't slip through the cracks, (b) management gets to see how hard we're working, and (c) the rate of bug finds is a loose indication of how robust the design is. We're also employing a new web-based bug tracker, something that I personally setup for this very purposed.

But old habits die hard and old perceptions are hard to change. Recently, a bug was entered into the system due to a typo in someone's design. Instead of typing something like "out = in + 5", the line ended up being "out = out + 5". The typo caused the entire design to fail during simulation. The engineer doing the verification (me) logged the problem and indicated to the designer that the problem was a typo. The response was quick. The bug was fixed in minutes and I got an e-mail from the bug tracking software, saying the problem's status has been changed from "open" to "fixed" along with the designer notes. The note said that logging every tiny little bug into the bug tracking system is f%^king bullsh*t.

It's easy to point fingers with a self-righteous sigh, blaming this colleague for being unhelpful and uncooperative. But I do understand where he's coming from. Through the grapevine, I've heard that the digital team has been burnt in the past by logging all of their bugs. Some upper management idiot had used the list to blame the team for too many bugs in their designs. Ever since, the team has been wary of maintaining such lists, or at most, having a list of serious issues shared amongst only a limited handful of people.

Often, management do not see the long term effects of their actions. Once trust has been violated, it is extremely difficult to gain it back again, even many many years later.

A Matter of Scale x3

2011-04-05T14:40:00.000-04:00

Over at Engineer Blogs, I've put up a series of posts discussing the challenges of designing circuits in the nano-world. The posts so far are:

Last week over at EB, we've also been discussing networking as it relates to our careers. I reminisced about how a friend help me land my current job at FluxCorp. So have a look. And leave a comment while you're at it.

Cognitive Dissonance #1

2011-04-01T14:25:00.000-04:00

Spend money? Save money? Here's a partial screen capture from CNN's website a little while back.

And no, this isn't an April Fool's joke.

Politics and Simulation

2011-03-31T15:43:00.006-04:00

It's federal election time in Canada - 5 glorious weeks of campaigning. Of the four parties currently represented in parliament, only two parties have a realistic chance of forming a government - the Conservative Party and the Liberal Party.

I'm glad we now have electronic voting. Here's a screen capture from my circuit simulator.

Our national public broadcaster, the CBC, also has an online election app where you can answer a series of questions on national policy and it will tell you where you are on the political spectrum, and what party you are closest to based on each party's respective policy statements. Here's the graphic of my outcome (I'm the check-mark, if you're wondering where I am politically).

CON = Conservative Party
LIB = Liberal Party
NDP = New Democratic Party
GR = Green Party
BQ = Bloc Quebecois (a Quebec separatist party)

It tells me I'm closer to the Liberals, but just by a hair. I'm pretty equidistant between the two frontrunners. The oval is the standard deviation of my answers. It appears that I am quite centrist, with slightly right leanings. But if this were the United States, our Conservative Party would be considered socialists, the Liberal Party as communists, and the rest of the parties as wacko-commie-fascist-Islamist-Maoist-Marxists-Nazis.

Of course, this is all based on what parties say. What they do is an entirely different matter.

Borscht

2011-03-28T11:03:00.000-04:00

When my manager came to visit us in February, he had a chat with our new Ukrainian colleague, Mr. Borscht, who is not part of our design team. Later on, my manager said that Mr. Borscht, who is in his early 60s, appears to be very proud of the fact that he has a "young Canadian wife". This is true. Mr. Borscht often mentions how much younger his wife is and how his new child is only half a year younger than his grandchild. And it isn't just Mr. Borscht who is playing Family 2.0. Mr. Poker, who's turning 60 this year, also has two "generations" of children.

When I first started to blog about the new Ukrainian colleague, I dubbed him Mr. Borscht. So what did he bring to work for lunch the other day? Borscht! And how does borscht relate to his young Canadian wife? Apparently, Mr. Borscht claims that he was able to "hook" his wife through his cooking of the "best borscht in the world". After she had one taste, he claims, their future nuptial was virtually guaranteed.

I guess the way to a woman's heart is borscht?!

In honour of Mr. Borscht's success with borscht, here's a Ukrainian borscht recipe. According to Mr. Borscht, be very liberal with the garlic. That's the secret.

Ingredients

* 1 (16 ounce) package pork sausage
* 3 medium beets, peeled and shredded
* 3 carrots, peeled and shredded
* 3 medium baking potatoes, peeled and cubed
* 1 tablespoon vegetable oil
* 1 medium onion, chopped
* 1 (6 ounce) can tomato paste
* 3/4 cup water
* 1/2 medium head cabbage, cored and shredded
* 1 (8 ounce) can diced tomatoes, drained
* 3 cloves garlic, minced
* salt and pepper to taste
* 1 teaspoon white sugar, or to taste
* 1/2 cup sour cream, for topping
* 1 tablespoon chopped fresh parsley for garnish

Directions

1. Crumble the sausage into a skillet over medium-high heat. Cook and stir until no longer pink. Remove from the heat and set aside.
2. Fill a large pot halfway with water(about 2 quarts), and bring to a boil. Add the sausage, and cover the pot. Return to a boil. Add the beets, and cook until they have lost their color. Add the carrots and potatoes, and cook until tender, about 15 minutes. Add the cabbage, and the can of diced tomatoes.
3. Heat the oil in a skillet over medium heat. Add the onion, and cook until tender. Stir in the tomato paste and water until well blended. Transfer to the pot. Add the raw garlic to the soup, cover and turn off the heat. Let stand for 5 minutes. Taste, and season with salt, pepper and sugar.
4. Ladle into serving bowls, and garnish with sour cream and fresh parsley.

Come One, Come All

2011-03-24T10:32:00.000-04:00

The provincial engineering regulatory body in Ontario, PEO, proposed last year to drop a long-standing requirement for the licensing of professional engineers. The requirement was that all PEs be either Canadian citizens or landed immigrants. The purpose of this change is to allow internationally trained engineers a smoother path integrating into the Ontario workforce. That recommendation was accepted by the provincial government and was recently passed into law.

Under the old rules, foreign engineers had to apply for immigration to Canada first before they can apply for licensure. Licensing takes a long time, which means these new immigrants were sitting at home unemployed. The new rules allows potential immigrants to obtain licensing first, meaning they can look for jobs the day they arrive, or perhaps even before.

Will foreign licensing reduce the number of engineering immigrants that end up driving taxis? That remains to be seen. But hopefully, advanced licensing will give potential engineering immigrants an earlier look at the job market here to see if it's really worth pulling up their roots to come to Canada.

Six Years, and Beyond?

2011-03-18T23:23:00.000-04:00

This month, I passed my sixth year mark at FluxCorp. This has been the most productive six years of my engineering career. In the previous three jobs that I've had, I've created numerous IC products. None were a success. But in the time that I've been here, I've been involved in two major product releases, each release having four variants. Each of those variants are now in production, with millions of parts per week flowing out the fabs at TSMC. Right now, I'm involved in the R&D phase of my third major product. If you own a PC of any kind purchased in the last three years, it's likely one my designs is living inside.

Not surprisingly, working at FluxCorp has been professionally satisfying. But satisfaction doesn't come from just achieving meaningful results at work. That's really only half the story. The other half is working with really, really great people at our local office. I've written about my international cast of co-workers before. Not only are they experienced in the field, they are also generous with their time, patience, and knowledge. As a junior member (age-wise) of the team, I was made to feel at ease from day one.

Of course, I'd be remiss if I didn't mention all the free food we get at FluxCorp. The latest addition to the pantry include chocolate-caramel ice-cream cones. Mmm...satisfaction!

With successful products under my belt, I've built a reputation for myself at FluxCorp. Coupled with my colleagues and side-perks, it seems inconceivable that I'd consider leaving this land of bliss for something else. But as my recent meeting with a recruiter shows, I'm not ruling out the idea altogether. Perhaps six years is getting a bit too comfortable, a bit too predictable, and a bit too familiar. Perhaps as I approach that seven year itch, I may finally take the plunge and look to fulfill my ex-pat dreams in earnest. Perhaps.

T-Shirt for Your Slides

2011-03-11T00:06:00.000-05:00

Two years ago, I happened upon a blog titled Positive Influence (unfortunately, the blog is now stale). The author is Jerry Pounds, a veteran management consultant. In my short interactions with Jerry, he was kind enough to send me his unpublished book called Praise for Profit: How Rewards and Incentives Are Demotivating America's Workforce. I didn't know what to expect from the book. After all, he was a management consultant and I was an engineer. We are natural enemies. But after reading just the preface, I was hooked. It is so refreshing to see the demoralizing management practices that we all instinctively know about put into such eloquent and powerful words.

The reason I bring up this book, almost two years after reading it, is a recent Dilbert-like e-mail that Mr. Halal and I received from our director. The director wants us to implement a way of doing things that both Mr. Halal and I think are huge time-wasters, filled with unnecessary overhead, especially given that we are already understaffed. When your designs don't work, do you spend 80% of your time making up slides tracking the most minute and insignificant detail, elongating bars on a Gantt chart, and creating executive overviews that provide no real insight, or should you really be spending that 80% working on the damn design. Sensing this resistance, the director tried to compare our project to the space program. We must leave no stone unturned, just like the space program, in ensuring we get every last detail correct. Perhaps someone forgot to tell him about Challenger and Columbia. The poor analogy aside (for I would love a billion dollar R&D budget), he tried to coax us from another angle -- bribery. In the e-mail, he states that he is willing to be "very supportive of doing something special" in order to help us overcome our inertia and engage in this "behavior shift", like "getting the team some T-shirts or something." I nearly fell off my chair laughing after I read this, especially when Mr. Halal suggested passing out key-chains instead (FluxCorp recently handed a corporate key-chain to everyone single fluxee worldwide...for what purpose, I don't know.)

Hilarity of the offer aside, both for its cheapness and crassness, the e-mail reminded me of Jerry's book again. In the preface of Praise for Profit, Jerry writes:

Is it necessary in today's business environment to publish a book that warns against manipulative motivational tactics? [...]

Are there any companies that still implement performance poker or Employee of the Month/Quarter/Year? Are managers and supervisors still standing up at company functions and singling out one or two individuals as top contributors, thereby creating envy, competition, and resentment among coworkers?

Are some companies still using stars, praise, bribes, coffee mugs, T-shirts, tickets, special parking spaces, and plaques as gimmicks to "pump up" performance instead of adapting their culture, policies, practices, values, and ethics to support the needs of their employees?

[...]

Unfortunately, the answer to all of the above is yes. [...] The infamous stick that was once the classic symbol of intimidating management tactics has been replaced at the end of the 20th century by the super-carrot. If one carrot will induce the donkey to move, why not hid a carrot behind every improvement, every good idea, and every contribution? [...] The merchandisers of motivation have managed to create the commonly held belief that people will not perform without the presence of a conspicuous payoff. As a nation, we have fully embraced the notion that praise, incentives, and rewards are the driving force without which our whole performance edifice would tumble down.

How true.

The irony in all of this is if the T-shirt scheme actually worked, our product probably wouldn't. The team would be spending a majority of their time making up slides rather than making sure of their designs' performance.

So what would you rather be -- insulted or ineffectual?

Workcation

2011-03-08T07:20:00.007-05:00

Fortune came out with its list of top 100 companies to work for in the US. FluxCorp didn't make the list, but I'm not complaining. Between the free food, decent gym, and generous pension plan (for high tech) for a small satellite office of 15 people, I'm quite happy. However, it does make me wonder about all the perks at these top 100 companies.

Looking at what's on offer at some of the best of the best -- free on-site restaurants, wine bars, compressed work weeks, cheap child-care -- it appears going to work is really just a small step away from a resort vacation. Drop off your kids. Head down to the company restaurant. Grab a latte and sit down with your laptop. Write a few blog posts for your favourite engineering blog. Play a bit of squash at the company gym. Relax in the sauna. Eat lunch. Drink wine. Take a nap. Read some e-mails. Pick up your kids and go home. Sounds great.

We've Only Just Begun

2011-03-05T21:08:00.002-05:00

Exactly a month ago, I contemplated the possibility of leaving FluxCorp in a post titled The End?. I was feeling restless. With my goal of landing a cushy ex-pat job seemingly impossible in this economy, I was willing to consider a change just for the sake of change. The recruiter was just another greasy salesman, telling me all sorts of lies he thinks I want to hear despite acknowledging half way through our converstaion he didn't really know much about what the job entailed beside the piece of paper in front of him. Still, I wondered, is the grass really greener on the other side? I tried to do a bit of due diligence and e-mailed someone I know at the hiring company. "Let's do lunch; it's been a long time." It certainly has, because that fellow has since moved to Toronto. Oh well. I guess it's too soon to say good-bye; after all, FluxCorp, we've only just begun.

A Year that Sucketh

2011-02-28T11:18:00.002-05:00

Last post, I wrote about my then-impending annual ritual of humiliation, the performance review. Here's how the ritual works. Each fluxee is sequestered in a room, blinds drawn, facing his or her manager with a sheet of paper in front of them. On that sheet is their annual ranking on a 5-level scale:

1 - you suck
2 - you suck less
3 - you ok
4 - you da man
5 - (upper management, where everything glitters like gold)

I don't qualify for level 5 and the average score company-wide for non-executives is 2. For two previous consecutive years, I was 'da man', which resulted in a promotion last year. For those two 'da man' reviews, I received a cumulative raise of 4%, which includes the raise due to the promotion. This year, I was told that I suck (less) -- a score of 2. It wasn't unexpected. The past year was all about debugging the Flying Flux. High profile bugs. Bugs that made customers angry. Bugs that became high profile. Bugs that made management wonder why things can't be designed with zero bugs. Expectations were also higher since my promotion. And with all that negativity, I still ended up with a 3% raise this year. So let's see -- 2 years of being a top performer gets me 4% in total. One year of lesser suckatage gets me 3%. Looks like I'm going to kick back and relax in 2011. Time to cruise for the rest of my career.

From One American Overlord to Another

2011-02-23T01:18:00.000-05:00

There's been a lot of instability with the jobs of overlords in northern Africa and the Middle East recently. However, the jobs of our American overlords at FluxCorp remain strong and stable. One visited us last week -- our director. Today, another one will be visiting our alien outpost -- our manager. This visit is supposed to be part of an annual ritual whereby we are subjected to the last and final step in our yearly performance review, the personal feedback from management. His previous visit was two years ago. He was unable to make it last year due to family reasons, that's why he's going to make up for it by staying a whole week. Woohoo! An extra week of meetings. I can't wait. In fact, I'm preparing my sucking up techniques as I type. *pucker*

Women in Engineering, Part IV

2011-02-22T16:12:00.000-05:00

I originally had a big long post on women in engineering in what would have been part four of this series. But instead, I've retitled it Women, Engineering, and Perception and posted it on my favourite engineering blog (and it should be yours too).

You can find previous postings on this series here: Part I, Part II, Part III.

Rethinking Monday. Finished Thinking Thursday.

2011-02-18T23:09:00.000-05:00

As I wrote two days ago, my directory came to visit us this past Wednesday. Bright and early yesterday morning, he and I had an hour long chat, 1-on-1. Besides the various queries I had on the future of our design team, our product roadmap, and corproate initiatives, I also made what I consider to be a definitive statement. It may yet turn out to be the turning point in my career, but that's still too early to say. In short, I indicated that I'd like to be considered for a manager position should one open up in the future.

Just this past Monday, I was still rethinking whether I really wanted to be a manager. Yesterday, I was done thinking. But even while I was saying this to my director, I was thinking about something else in the back of my mind. The hedge, if you will. The backup plan. The job opportunity presented by a recruiter two Mondays ago. I'm still unsure what to do with that one. Interview just to see what's there? My only concern is that they'll offer me $$$ I can't refuse, but will feel like crap taking it because I'll be leaving a great job with great colleagues all for some dirty corporate dollars. Ugh...I feel dirty already.

Right now, my feeling is that I'll most likely stay at FluxCorp. I may still yet go for the interview, but since I know a couple of people that work there, I'll need to do a bit of espionage first. In the meantime, I'm going to kick back and enjoy reading some of my fellow bloggers over at my favourite engineering blog. This week's theme is all about engineering education.

The Quiet Revolution

2011-02-16T14:24:00.001-05:00

My American director (my manager's manager) is here today as part of his Satellite-Office-World-Tour. He heads off to India tomorrow. Yesterday, I wrote a post on our favourite engineering blog (EngineerBlogs.org), complaining again about the zero-bug prototype policy at FluxCorp. It's titled Prototype vs. Simulation. During today's morning group discussion with the director, he reiterated the wish of FluxCorp to have perfect prototypes but at the same time acknowledged we are both short on time and staff. Kind of like me telling you to hit a hole-in-one every time, but acknowledging that you have no arms.

Acknowledgment is a good first step. But what is he going to do about it? So far, he tells me that the zero-bug policy has gotten widespread revolt from the engineers across the entire business division. But the policy remains in place. Somehow, it sounds like that other guy in Egypt that vowed to stay in place despite widespread disapproval from the masses. I think the policy is going to be meet the same fate -- hanging on officially for now, but will no doubt soon be toppled.

I may be premature, but I'm going to be half-glass-full today. Hurray for common sense!!

Thank You Mike Fisher. All the Best in Nashville.

2011-02-15T10:06:00.002-05:00

Three posts ago, I blamed Carrie Underwood for the departure of a popular local NHL hockey player, Mike Fisher. That was unfair and a knee jerk reaction on my part. Fisher himself said that he did not request to be traded from Ottawa to Nashville and I have no reason not to believe him. He and Underwood even recently completed the building of a new home in the west end of the city. And after all, players changing teams in professional sports is just business as usual, especially with Canadian hockey players that marry blonde American women (Gretzky, anyone?). It appears the Senators wanted to trade Fisher anyway and did him a favour by putting him in Nashville.

My lament isn't that we've lost a hockey player. Fisher is a decent player, but better players have come and gone from the Senators without much notice. My lament is for the upstanding citizen that the community is losing. Someone who came to Ottawa at 19 and is leaving at 30. Someone who by all accounts is humble, generous, and gracious. And someone whose popularity stems mostly from what his does off-ice, rather than his skills on-ice. In the world of over-sized personalities in pro sports, this is uncommon indeed. A devout Christian, Fisher relies on his faith for guidance and he certainly lives up to it. If there's scripture that describes Fisher well, it would be from Matthew 6:1-2 :

Be careful not to practice your righteousness in front of others to be seen by them. If you do, you will have no reward from your Father in heaven. So when you give to the needy, do not announce it with trumpets, as the hypocrites do in the synagogues and on the streets, to be honored by others.

This is something that Fisher practices. Requests for his time in support of charities are given without fanfare. Unannounced visits to the children's hospital are done without formal announcements. And the single event that cemented his status in the community is shown in the photo above. He's holding Elgin-Alexander Fraser, a three year old boy days away from dying due to a rare form of cancer. Elgin's last wish was to see a live Ottawa Senators game. The Senators made it happen and more; Fisher later visited the boy in his home on his own time, with the public only finding out about it much later when Elgin passed away and the boy's father thanked Fisher publicly for the joy he brought to his son in his remaining days.

This is just a single example, amongst many, of his generosity. And this is what Ottawans will miss most with his departure. Thanks Mike Fisher, for your decade long community service to Ottawa. We wish you and your family all the best in Nashville.

Ottawa Citizen editorial: Missing Mike. A good player, a better person; City already feeling the loss of the quiet man and his big heart.

Rethinking The Path to Management

2011-02-14T15:05:00.000-05:00

The proverbial shit has hit the proverbial fan lately at FluxCorp. The FlyingFlux that we delivered to one of our customers has a 0.3% defect rate. According to them, that's way too high. One of their own large customers has returned products to them. This makes it a very, very bad situation. Our customer told us they have now shut down the production line with millions of products on hold. They won't restart production until we figure out a fix. We did figure out a fix only to find out they lied to us. They continued their production line with an experimental firmware change. When we indicated that the experimental change was not recommended and that our new recommendation completely and totally takes care of the problem, they then asked us to quantify the risk of having used the experimental version. And they want an answer in 4 hours. I don't think I can even provide an answer within 4 days.

Defect rate is one of those things that is very difficult to predict analytically, even less with a bug that is only half understood. Their executive VP has talked to our executive VP, who has talked to our VP, who has talked to our director, who has talked to the manager interfacing with the customer, who then talks to me. Luckily, I'm just a lowly engineer who's giving his technical opinion to the problem. I don't actually need to craft half-truths to tell the customers so they feel better about us an about their products.

But the story is completely different for the manager facing the customer. He gets berated every day from the disgruntled customer, sits in meetings until midnight, gets pressure from the higher up to solve the problem quickly, all with his hands tied as he's not allowed to reveal too much information. Just look at the inset graphic -- over 1800 unread e-mails in his Inbox (don't ask me how I managed to capture his desktop). And all he does all day is make up Powerpoint slides. Every time I get the urge to jump into management, something pulls me back. I think this is one of those moments again.

The Path to Engineering (Management?)

2011-02-13T23:29:00.003-05:00

Recently, the authors at Engineer Blogs, including yours truly, discussed how each of us ended up in the field of engineering. Some took a circuitous route, like Cherish and FrauTech. Chris Gammell was inspired by physics. Paul Clarke started as a baby engineer. As for me, I didn't even know what engineering was as I approached the end of high school.

Not matter how each of us ended up in our chosen engineering specialties, or where we got our engineering degrees, I think it's safe to say that all of us are happy with our choices. But the grass always looks greener on the other side. FrauTech recently blogged about whether moving into engineering management is "worth it". Last Monday, I met a recruiter to discuss an open local position. The position on offer isn't a management position, but it is a formal technical lead position as opposed to the informal leadership role that I currently play. So where will all this lead... I'm still looking for a $ign.

Damn You Carrie Underwood!

2011-02-10T18:04:00.001-05:00

One of the most popular hockey players in our local NHL team, the Ottawa Senators, is Mike Fisher. Last year, he married Carrie Underwood, who as a country signer, is based in Nashville. With the Senators playing poorly this year and wanting to rebuild by unloading some heavy salaries, Fisher was traded to Nashville's NHL team, the Predators. Obviously, the move to Nashville was not a coincidence.

Fisher wasn't just a hard-working player, he's also a community role model. I'm disappointed. All I can say is, "DAMN YOU CARRIE UNDERWOOD!". There, I feel better.

All Your Border Are Belongs to Us

2011-02-08T01:43:00.002-05:00

Stephen Harper, our intrepid Prime Minister of this northern nation, dared to venture south into the land of tea parties with no tea to talk to their embattled Muslim radical-Christian socialist death-panel Kenyan president Barack Obama. The little sojourn took place at the White House and the major topic on the agenda was border security. Sure, Americans fear terrorists will stream through the porous border with Canada, the longest undefended border in the world. But Canadians fear all the undocumented Mexican workers living in the US walking to the Canadian border to fill out asylum claims. Thus, the two countries have decided to band together. My border is your border kind of cooperation. More information will be shared to nab undesirables while trying to keep the border as open as possible as to not impede the largest international flow of goods in the world -- more than a billion dollars worth of trade crosses the Canada-US border each day.

The details of the agreement are a bit sketchy right now, leading opposition leaders to claim that the Prime Minister is selling out our Canadian sovereignty on border control in order to appease the restless giant to the south. Bah...just a bunch of predictable noise makers. How I long for those days when I could cross into the US with just my library card. OK, so it wasn't that easy, but I did flash my citizenship card with my 12 year old picture on it and that allowed me enter the US and re-enter Canada. This was all pre-9/11 of course. While I certainly don't expect things to go back to the way they were a decade ago, I hope this new understanding will, at the very least, make those border agents less of a power-tripping SOB than they usually are. I mean, if they search my car one more goddamn time...

The End?

2011-02-05T01:44:00.002-05:00

I got a LinkedIn email recently, where a recruiter wanted to link to me in order to introduce me to a great opportunity. I normally delete these e-mails immediately, but I don't know what it was that morning. I accepted the invitation and subsequently, I found myself reading a job description from a local company. So far, the job description fits pretty well with what I'm doing already and the recruiter tells me that the pay is great. As much as I like FluxCorp and love my eclectic bunch of colleagues (and the free food! oh, the free food!), I'm nevertheless a bit restless. If I cannot be a wealthy ex-pat, maybe I'll be satisfied with a change of local scenery. But nothing is set in stone. The opportunity has a pretty high barrier to cross before I would consider moving. Still, I did agree to meet with the recruiter on Monday morning.

Is this the end of Fluxor working at FluxCorp? Good thing my latest post on Engineer Blogs, fresh off the press, is all about preparing for a job interview. Perhaps it's advice that I'll need to consult myself on sooner rather than later.

Moral Suasion

2011-02-04T01:35:00.001-05:00

In 2009, FluxCorp instituted a mandatory vacation policy whereby employees were forced to take their entire annual leave on specific company-determined days. While such shenanigans are OK in the free-market world of the US, here in socialist Canuckistan, where labour laws are a little tighter, this sort of random pronouncements are illegal. So we banded together here in our little satellite outpost and organized a little (tiny, really) revolt, and we won (kind of). I documented this tiny little episode back in an April 2009 post titled V-A-C-A-T-I-O-N Part Trois. This policy, not unexpectedly, was rather unpopular amongst Fluxees worldwide; thus, it was revoked in 2010.

However, a bad policy never rests on its laurels. It's making a comeback in 2011 disguised by upper management as friendly advice, or even polite suggestion. Management has asked us to please use some of our vacation days in Q1 of this year to lighten the vacation backlog as to make the finances look prettier on paper. Of course, using your vacation days is purely "optional" (*wink* *wink*). And oh, any vacation days taken should not impact the progress of any project. Thus, we are nudged (elbowed, rather) to use our vacations, but our progress cannot slip, meaning that we must put in overtime to make up for the days that we took off. Sounds like a great plan to lighten the finances.

It's also a bit of déjà vu. When one of my former employers were nearing bankruptcy, it cut the pay of all its engineers across the board by 10%, telling them that the official work week has been shrunk by 10% -- Fridays were now officially half-days. In the same e-mail, with no sense of irony, it states that engineers are expected to continue to work unpaid overtime in order for the keep the project moving without delay. Half-day Fridays, yeah right. Why doesn't management just come out and say to everyone that there's going to be a 10% pay cut and leave it at that? Why play these stupid obviously-transparent word games of a shortened work week? Don't they know this whole charade make them look like patronizing dickwads (or wankers, for our British readers)? Good thing I had left the company by that time, but just in the nick of time.

But a better analogy is when this same company asked us to buy shares into the failing enterprise. Of course, buying shares is purely "optional". We even had to sign a waiver saying that we bought these shares "without coercion" (alarm bells, anyone?). The "optional" part, as we found out soon afterwards, means "optional if you don't care about getting fired". A large round of layoffs soon followed this optional share purchase. Those who didn't exercise their "optional" right to purchase worthless shares bore the brunt of those layoffs. I bought shares -- $5000 worth. That was the recommended amount. I looked at it at the time as a pay cut and I was right. Two years later, I got a bankruptcy notice saying that my $5000 in shares can now be redeemed for $68.22.

So if you'll excuse me, I've got to go and take some optional vacations. Don't worry though. I'll keep blogging here and at Engineer Blogs. After all, progress cannot be delayed just because of some pesky vacation days.

Hippity Hop

2011-02-03T02:08:00.003-05:00

A New Face

2011-02-02T04:25:00.001-05:00

For the past two years, my online pictoral self has been a bar magnet with protruding flux lines. But with my new involvement with Engineer Blogs, I think the time is ripe to put a face to the hitherto faceless Fluxor. So here you go, the picture to the right is my new online persona. A face with concentric flux lines radiating out of my head. And a pretty handsome fellow if I do say so myself.

With my new look in place, it's time for you head on down to Engineer Blogs to see what's new there. I reposted one of my very early posts called It's a Digital World..., but with a few updates to reflect current realities. Hope you'll take a look.

Statistics Porn

2011-02-01T09:59:00.000-05:00

Back when the Canadian government did away with our country's mandatory census and proclaimed that a voluntary census makes no difference to the quality of data, I lamented about the ignorance of our elected officials. In a subsequent post, I lamented further about the lack of numeracy in our general population.

Math, as a school subject, is daunting for some. Educators have sought methods to make the subject more relevant and more interesting to kids. I think a similar marketing campaign is needed for our adults as well. Present data in a way that is both visually appealing and socially relevant. Statistics porn, if you will.

Here's a wonderful example of what I'm talking about, courtesy of the BBC and Dr. Hans Rosling:

The top comment on YouTube for this video even says "lol why did this movie turn me on so much?". Porn indeed.

And if you watched the video, you'll know why I'm interested in getting an ex-pat position in China, specifically Shanghai, two and a half years after first visiting this vibrant city on a business trip.

Pitter Patter of Little Feet

2011-01-28T18:46:00.002-05:00

There's been a baby fest recently at FluxCorp amongst my colleagues. Last month, I wrote about a new co-worker joining our midst. I surmised initially that he was Russian due to his name, but it turns out he's Ukranian. I haven't given him a moniker yet, so let's do that now and label him Mr. Borscht. We all discovered soon after Mr. Borscht's arrival last month, for an avid talker is he, that a fourth child of his came into the world a few months earlier. When everyone returned to work in the new year, we were all informed that during the Christmas break, Mr. Borscht welcomed another baby; this time, his first grandson. It was also during the break that Mr. Halal's wife gave birth to their second child.

And me too. Last week, I welcomed triplets. In my acquarium. I'm so excited!

Transnational Geekery

2011-01-26T15:03:00.000-05:00

Tuesday nights are nights where I go and visit the swimming facility at one of the local universities. Last night, after the car had come to a stop in the parking lot, the headbeams lit up the car in front -- a Mercedes C240.

Nice car, I thought. Then I looked closer at the license plate.

The license plate itself is pure geekery -- MY RFID -- and if you'll notice, it's transnational in nature. It's an Ontario license plate, but the plate holder is from the The Wharton School at U-Penn. I surmise this car may belong to an engineer turned MBA that has done pretty well for himself. He may even be a professor at the university.

This lead me to think of another wonderful license plate that Chris Gammell alerted me to:

A DeLorean. Awesome.

Mr. Smith

2011-01-25T01:15:00.000-05:00

Today's topic of discussion by yours truly over at Engineer Blogs is the Smith chart (inset), simultaneously loved by those working in RF and hated by everyone else.

So if you're intrigued by plotting on a circle rather than a grid, then head over to the latest post titled S Stands for Smith.

If you're not intrigued, well, head over to Engineer Blogs anyway. There's a post on Plumber Envy and another one on research into making greener circuit boards. Or if you're more into marketing, there's even a post on Social Media and the Engineer.

Who Turns on Their Stuff at -40 Degrees?

2011-01-23T22:03:00.000-05:00

Upon leaving a restaurant tonight, I couldn't help but notice, as a baby surely notices a wet diaper, that my nostrils were starting to freeze as I traversed the few parking spaces necessary to reach my temporary moving shelter -- my car. Turning on the radio, the weatherman noted that it was currently -32 degrees Celsius (-26 deg. F). It hadn't been this cold in Ottawa for quite a few number of years. As I was driving home, I couldn't help but think about the cold weather and how it reminded me of -40 degrees. Minus 40 degrees is the only temperature that is the same in either Celsius of Fahrenheit. It's also the lower end of the temperature range that we typically design our integrated circuits to work under. On the high end, it's typically 125 deg. C., well above the boiling point of water.

It's not unusual, when designing circuits, to run into a problem whereby circuits that work at one temperature extreme fails at another. By adjusting circuit parameters to make it work at the failing extreme, one often finds that the original passing extreme is no longer passing. It's a frustrating problem because one must now go and design something much more complicated in order to cancel out the temperature effects. And it's a problem that has caused many an engineer to ask "who turns on their stuff at -40 degrees anyway?"

Not many. But on a day such as today, if someone left their iPhone in the car, and after dinner, decided to make a phone call using the same really cold phone, they still expect it to work. And if it doesn't, it'll be the corporation that suffers the bad internet review once the annoyed customer leaves the device on for long enough that the heat from the dissipated battery power makes 3G work on the iPhone once more.

What else needs to work at -40 degrees? All the electronics in my car, of course. Tonight, more than any other night, I thank those automotive electronics engineers who designed their circuits down to -40. (And a thanks to those mechies too!)

Office Politics 101: Instant Karma & Nepotism

2011-01-18T23:06:00.001-05:00

In my recent posts, I took a job candidate (Mr. Flop) to task for his inability to answer some basic questions (e.g. his past designs; how transistors work), even though his résumé clearly implied this should be common knowledge to him. While Mr. Flop may not be the worst candidate I've ever interviewed, in terms of a performance-to-expectations ratio, he probably ranks as the lowest ever for me.

Yes, I may be a little uncharitable here, both in the paragraph above and in my previous posts on this issue. Some may even call me smug, or that I'm full of myself. And those "some" may be correct. As John Lennon's song goes, Instant Karma's gonna get you.

Days after this horrendous interview, our team received an e-mail from the No. 3 in command at FluxCorp. Number three! He e-mailed us to say that he is aware that we had recently interviewed Mr. Flop. Futhermore, he highly and personally recommends Mr. Flop as an analog IC designer. I'm not sure how No. 3 can even make such a recommendation since he's not an IC designer himself. And I wondered how in the world did No. 3 even hear about this interview down in the trenches. Needless to say, this had become a delicate situation, especially with regards to our own livelihoods.

It's obvious there's some sort of close connection between No. 3 and Mr. Flop. A nephew? An illegitimate son? A May-September love affair? Who knows?! Let's just say my esteemed colleague, Mr. Halal, handled the situation with great care to everyone's satisfaction.

What would you do? Would you choose self-preservation? Sticking to your principles? Or think happy thoughts until the problem goes away?

Transistor Interview Questions

2011-01-16T23:58:00.000-05:00

Last week, I wrote about an interview with a disappointing candidate that was applying for a analog IC design position with FluxCorp. Although he wasn't able to answer many questions regarding his own past designs (or alleged designs), I still thought it necessary to give him a chance by moving to more basic questions, like the transistor. In my opinion, these questions are not difficult. If one bothers to prepare for an interview like this by flipping through some old text books, these questions should be a breeze. Unfortunately, the candidate (let's call him Mr. Flop) in question last week failed horribly, which illustrates a few things. First, he doesn't know how to prepare for an interview. Second, he has not internalized undergrad material in his brain so that he can claim to be qualified to be an analog IC designer. Third, he grossly overstated his abilities on his resume. And I'm already being quite charitable here.

The questions I'm going to outline below are ones which I've encountered over the years both being interviewed as well as interviewing otheres. So here goes.

Continue reading...

Engineers of the World, Unite!

2011-01-13T17:47:00.001-05:00

In my last post, I promised that the next post (this one) would be about some of the basic questions about transistors that an unfortunate interview candidate vying for an analog IC design job wasn't able to answer. But I am preempting that post with this public service announcement.

Engineers of the World, Unite!

I, along with other like minded engineers, have started new collective and we have organized ourselves around a new blog called:

Engineer Blogs (http://engineerblogs.org/)

Yes, there are only four of us writing for the collective at the moment, including yours truly, but that's why I am urging all of you to join us, as readers or as potential writers. And yes, I agree that the collective has a certain Marxist ring to it, that's why Cherish over at Faraday's Cage is where you put Schroedinger's Cat prefers to call it a blogging organization, or a blorg. Unfortunately, the Urban Dictionary defines blorg as "Nauseated or feeling malaise. Combined derivative of blah and urgh." Not exactly a good description for our blog, unless that nauseated feeling is a side-effect from the hyper-euphoric estacy you get from reading our divinely inspired words.

In fact, my first contribution to the collective has been posted. IC Designer’s Guide to IEEE Conferences. Have a gander.

On Paper

2011-01-10T13:59:00.000-05:00

During the dot-com boom, there were a lot of rich people on paper. But when the bubble burst, these folks were back to their regular middle-class selves. Similarly, there are a lot of people I've interviewed over the years that look great on paper, but the bubble they've created for themselves almost always** burst when meeting them in real life.

** I say almost always because if you're a woman of a certain visual quality and you come to the interview dressed a certain way, being exposed for your padded resume may not matter. Case in point: The Siren.

This past week, I interviewed what appeared to be a very promising candidate for a contract position. He graduated from the top private high school in town, has a master's EE degree doing phase-locked loops, and has five years of experience doing analog IC work. In previous posts, I wondered where all the locally born engineers are as every analog designer that I work with in my local office are immigrants. But now, here's what appears to be a locally bred specimen, with seemingly great credentials. When I greeted him at the front door on the day of the interview, he looked vaguely familiar, like an ex-colleague but with much less hair. It then came flashing back that we had worked at the same company before when he was still a student (same company as The Siren).

So, we went to our big corporate conference room, pulled open a few shutters and sat down. We exchanged a bit of small talk before I scanned his resume in a bit more detail and asked him to follow me to the white board located up front. I asked him to draw a few things for me, some of the circuits he's designed. He had trouble. Some circuits he was only able to draw partially, some he wasn't able to draw at all. Red flag! But OK, give the guy a chance. Let's dial this interview down a notch and ask him about some basic questions. I moved on to the transistor.

The transistor is to analog designers what the brick is to masons. If you're an aspiring analog designer and you can't answer a few simple (what I consider simple) questions about transistors, it's going to be hard for me to hire you. And if on top of it, you can't even recreate some of your very own circuits from memory, either from your previous job and your very own Master's thesis, then I question your sanity in seeking an analog IC design position in the first place. After an hour interviewing -- him hemming and hawing, me giving leads and hints -- it was clear to me that his sanity was indeed in question.

Needless to say, he was a big disappointment. It's like getting all excited after having watched a great movie trailer only to find out that the trailer was the best part about the movie.

In my next post, I'll discuss in a bit more detail just what basic transistor questions he was unable to answer and a few other circuit basics he got wrong. If you're a new or almost-new grad thinking about a career in analog IC design, I suggest you play close attention. You will be asked those questions in an interview.

Women in Engineering, Part III (and films!)

2011-01-07T00:11:00.008-05:00

Cherish over at Faraday's Cage is where you put Schroedinger's Cat introduced me to a Hitler parody about academic peer reviews. The clip was interesting enough that afterwards, I went and watched the actual movie, called Downfall (2004). It's always nice to see Germans in WWII movies actually speaking German. It was also a good film to boot. I like foreign language films; it feels exotic, and that in itself makes the movie more interesting to me than it may otherwise if I actually understood the language.

At our satellite office here, I can always count on my colleague Mr. Kosher, a Polish-Israeli-Canadian Jew/Catholic, to recommend a good foreign language film. Early last year, I was engrossed for over a month watching a Polish TV show called Dom (meaning House/Home), a 25 episode mini-series shot over 20 years that depicted WWII Poland and the post-war communist era up to 1980, when the independent trade union Solidarity was allowed to form. The protagonist in the series was a mechanical engineer. A rarity in entertainment. You can watch the whole thing on youtube.

Recently, Mr. Kosher introduced me to a South Korean movie called My Dear Desperado (2010). The entire movie is also available on youtube. This film's numerous turns in themes coupled with a bit of wackiness/comedy/romance was truly enjoyable. It's also a great date movie. Up to now, you might be thinking that this post is all about foreign language films rather than women in engineering. But hold on, I'm getting there. The protagonist in My Dear Desperado is also an engineer, a young woman with a master's degree in IT who left her small seaside town to go into Seoul to work for a big company. Like many of us who's worked in high tech, her company went bankrupt and she fell on hard times. The rest of the movie is about her struggles to find another job.

What's eye opening in this movie is the South Korean interviewing process for entry level engineers. Four people are interviewed simultaneously in front of a panel of judges. In the following clip, everyone has to introduce themselves in English. Have a look, it's quite interesting. Skip to 1:35 for the interview part.

Our main character, Ms. Han, finds it difficult to land a job because of her small town university degree and her meagre three months of work experience. When she finally gets a new interview, she is asked some very pertinent questions, like if she has a boyfriend and when was the last time she dated. The interviewers then ask her to sing and dance. Desperate for a real job, she complies, only to realize half way into the song that the interviewers are simply out to embarrass her for their own amusement. Skip to 6:15 for the interview.

Later on, she took the initiative and went right into the HR office of a company to submit her resume. Although the head of personelle told her they only hire experienced people, another employee subsequently told her that she has a good chance of getting a new entry level job. But only if she first accompanies him to the classiest motel in town. Skip to 2:45 for the scene (ends at 5:35).

Ultimately, after much debacle, she ends up in an interview with a company that appears to know the right questions to ask its candidates. And she gets the job. Skip to 3:00 for the scene.

I won't comment on the harrassment she's received at her job interviews. I don't think for a moment this passes for normal behaviour in South Korea. What I do find surprising is that a woman engineer is being portrayed at all, and as the main character no less. What I liked even further is how the movie deals with her engineering knowledge. It's simply one aspect of her character, but does not solely define her like many scientist/engineer caricatures one finds in TV and film, both male and female. It's a well balanced portrayal of a female engineer where her engineering side is just there, not overdone, but not glossed over either. She has money problems; she has family problems. She has romance problems; she has work problems. Sometimes she's resilient; sometimes she's vulnerable. And oh, by the way, she's a pretty good engineer too. A rare portrayal in the entertainment world. Well done.

Fidelity, Bravery, Integrity

2011-01-05T00:48:00.001-05:00

Got an interesting piece of mail recently:

It's certainly not everyday that one gets a letter from the FBI. Yes, I know my circuits are supposed to be bug-free, but bugs aren't a federal crime. Or is it? And I'm in Canada; aren't those guys out of their jurisdiction? And where's Agent Scully? Can she handle my case?

Oh, and the post's title? That's the FBI motto.

Hiring Local Foreigners

2011-01-04T07:23:00.007-05:00

Jobs, jobs, jobs. World leaders should be concentrating on job creation, as everyone appears to agree. FluxCorp has been hiring, but mostly in China and India. In North America, it's been a real small trickle. Still, we've added a few to our numbers. In my pre-Christmas post on A New Friend, we hired an aging Ukranian-born (not Russian-born, as I had originally suspected) analog designer locally in Ottawa. By next week, we'll also be adding another warm body to the Ottawa team -- a Chinese-born contractor. And this time, he works for my team! Yeah!

Both of these guys have fairly strong accents. Meetings are going to be fun! And both of these guys are experienced engineers. According to the anonymous commenter in my post on A New Friend, s/he thinks that North American "engineers with a passion for analog design (and in my case testing) all quickly found out that there is no longer such a thing as an entry-level analog job; and so we all got forced into other lines of work/volunteer projects."

I agree. It's tough for new electrical engineers of any stripe these days. But in early 2010, we did hire someone right out of school. In my post on Women in Engineering, Part II, I mentioned that I had interviewed two female and one male candidate for a junior analog test engineer position located in northeastern United States. We didn't hire any of them. I did interview one more person afterwards, a young woman nearing the completion of her Master's from UCLA, and she was by far the best of the bunch. We gave her an offer, but she was unable to start for several more months. The reason? She's a foreign student from India, so FluxCorp had to jump through the necessary hoops to get her an H-1B visa.

For quite a while now, there's been a lot of complaints about the H-1B work visa and how it drives down American wages by allowing companies to easily import cheap engineering labour. IEEE-USA is a particularly loud voice in its opposition of how H-1B visas are being used. I certainly won't wade into this whole debate, partly because I'm not American and partly because if I ever move to the US to work, it won't be on an H1B visa. I'll just mention that in this particular case, we weren't out to save money. We simply hired the best person for the job. And we hired locally; it's just that she, herself, isn't local.

Sorry if this post is meandering between outsourcing, job hunting, experience levels, women in engineering, place of birth, and work visas. But I do have a point somwhere in here. And it all goes back to the same question that I asked just before Christmas. Where are the locally born engineers? Or am I living in a parallel universe where all engineers come from alien lands?

Buffalo, New York, USA Canada

2011-01-03T23:16:00.003-05:00

The World Junior (under 20) Championship in hockey is going on right now. Tonight, it was Canada vs. USA for the semi-final match. The picture above is a scene from that very match showing the Canadians celebrating after a goal, the American goalie out of focus in front, and a homogenous bloc of Canadian fans cheering behind the glass. You'd think with a scene like this, the game is being played on Canadian soil. But not so. In fact, Canadians have invaded Buffalo, New York to cheer on their team in the HSBC Arena, home of the NHL's Buffalo Sabres.

Last year, the US beat Canada in Saskatoon, on Canadian soil, to win the championship. This time, the Canadians got revenge. Good game, boys. Bring on the Russians, eh!

Work, 2011 Edition

2011-01-02T22:44:00.002-05:00

The Christmas break was nice. It was filled with food, family, and lots of socializing. My original plan was to take a trip to Niagara Falls, but canceled those plans at the last minute in order to relax at home. But holidays aren't meant for relaxation, even when I do stay at home. Instead, I completed a couple of minor projects around the house, bumped into an old friend at an ice rink which led to more socializing, and stayed up way past my bedtime on many a nights. Now that the holidays have wound down, I'm ready to relax. That's why I'm heading to work tomorrow. My American colleagues are still on holidays on Jan 3. This means I'll be able to do a bit of light housekeeping chores (work-wise) without the constant interruptions and distracting meetings of a regular work day. It'll be a nice relaxing time.

The rest of 2011 is shaping up to be an uncertain year. Having been promoted last year due to two straight consecutive years of excellent reviews, I can't say that I've repeated that performance for a third year in a row. 2010 was filled with bug fixes, more bug fixes, and even more bug fixes. Management hate bugs, so much so that they've mandated a zero-bug policy in our circuit designs -- perfect prototypes is what they expect. I've ranted on about this before, but 2011 is shaping up to be the year of "follow-the-process". What management cares dearly about, from my vantage point, is a bunch of new formal processes that creates busy work, but not useful work. And it's nowhere close to creating a zero-bug design.

It reminds me of the collective farming practices in Stalinist USSR and the Great Leap Forward period of China. This policy was unequivocally a major catastrophe for both nations, resulting in famines that cost the lives of millions. But while millions were dying, the nation's management team were nevertheless telling its workers that everything is going just swell. I mean, just look at the size of these melons (Chinese propaganda poster from the 1950's):

So while we engineers on the ground can see the utter wastefulness of these new processes thought up by our management comrades, we are nevertheless told that things will be not only be swell, in fact, things are going to be greaaat!

So for 2011, I'll be busy filling in check-lists, preparing forms, writing up formal declarations, and doing all the dog-and-pony show presentations that keep management happy, but nothing to actually make my circuits work any better. But they'll only be happy for a short while, because in the long run, when they've seen how utterly they have failed, they'll be in for a long period of reflection. And when they've finished reflecting, I predict they'll blame the engineers for doing such a poor job. Because if their collective wisdom tells them that zero-bug prototypes in circuits of our complexity can be the norm rather than the extreme exception, then surely, their collective wisdom will tell them that nothing that they ever do is wrong.

Christmas Chain Reaction

2010-12-26T13:44:00.000-05:00

Santa was pretty good to me. I got a HD camcorder. Small and light compared to my old analog video recorder which uses 8mm tape. And it is what I actually wanted for Christmas (I also got a high end vacuum cleaner, but it's not here yet). The downside to upgrading the camcorder is that it is going to force me to upgrade a whole bunch of other things. The software that comes with the camcorder to view and edit videos has a minimum recommendation for CPU speed and memory, which is beyond what I currently have. It, of course, also recommends an HDTV. But I still have a big tube TV, perfectly functional, perfectly fine. So if I'm interested in getting the full potential out of this new toy, I must also now upgrade my computer. The HD video files are quite large too. This means I have to upgrade the size of my backup disk drive. Of course, I must upgrade the aforementioned TV to an HDTV. Which means I must naturally upgrade my digital cable receiver to an HDTV version. A flat screen TV sitting on a TV stand made for a big tube TV also looks pretty stupid. So I have to upgrade my furniture. HDTV requires different cabling as well. So all my current cables that go to and from various audio-visual devices must be upgraded as well. And by the time I finally upgrade everything to work with the camcorder, the camcorder will be obsolete.

The Fruits of Technology

2010-12-25T13:00:00.008-05:00

Was Santa nice to you? Did you get your favourite high tech toy?

Even if you ended up with a lump of coal, fret not; have a gander at the video, perhaps a little smile shall come upon your face.

Another White Christmas

2010-12-24T02:08:00.000-05:00

Bing Crosby would approve, as it's a sure bet that Ottawa will be having a white Christmas this year. A reasonable amount of snow on the ground now with temperatures that won't be exceeding the freezing point any time soon.

Family will be visiting from the US of A. I've purchased a turkey. All that's needed now is some frantic last minute shopping. And of course, some Christmas music to get me into the spirit of things. Here's one of the best singing one of the best:

And let me give a tribute to the one that started it all:

Happy Christmas to All and to All a Good Night!

A New Friend

2010-12-21T04:12:00.001-05:00

Ever since the firing of the Psycho Colleague back in September, our analog circuit design team has been down a person. We've been asking for a replacement ever since and my director has been promising it to us ever since.

So what happened last week? Well, there's now a new employee at FluxCorp Canada. He's an analog circuit designer hired to help with the design of the Flying Flux. Just not our Flying Flux. In fact, he was hired by another Flux-building team based out of the western reaches of the American mainland. As I later found out, my director had misled us. We won't be getting a replacement after all. And to rub salt into our wounds, a highly qualified analog designer here in Ottawa was hired into FluxCorp ... to not work for our team. Thanks!

We gave the new guy the most friendly welcome on his first day. We were simply delighted to have another person diminish our weekly supply of free food and drinks. He's Russian. Or Ukranian. Whatever. As they say, if you've met one Slav... or is it just me that says that. Anyhow, looking up at my hiring-of-minorities scoreboard, here are the current standings of all my Ottawa analog circuit designing colleagues*:

By Ethnicity:
Slavs: 2
Chinese: 2
Muslims: 1

By Geographic Origin:
Europe: 2
East Asia: 2
Middle East: 1

By Sex:
Gents: 5
Ladies: 0

By Age:
56-65: 2
46-55: 2
36-45: 1

* Not including yours truly, as I would never subject myself to such silly workplace classifications.

Our new friend has shifted the balance of power. The Slav-bloc has doubled in strength and the average age just got skewed higher. Basically, our office of analog designers are a bunch of old foreign guys.

I keep wondering, where are the locally born engineers? Perhaps I'll address that in a future post.

Another Nail in Nortel's Coffin

2010-12-17T23:07:00.000-05:00

Nortel, the once proud Canadian telecom giant, has already divested itself of every single one of its business units as part of its bankruptcy process. Today, it has officially sold off one of the most splendid corporate campuses in town to the federal government. Built large enough to house more than 15,000 high tech employees, it will, over the course of the next few years, be transformed into the new home for a large chunk of the Department of National Defence. The only remaining piece of Nortel is its portfolio of more than 4,000 patents. When that is eventually sold, Nortel shall be no more.

The Engineer as a Social Animal

2010-12-17T15:17:00.001-05:00

The stereotypical engineer is introverted and socially awkward. I think I fit that stereotype. So it was with a bit of trepidation when I decided to accept an invitation to a cocktail party, which took place last night. Just me, and six other couples. I had to break out of my engineering shell and engage in conversation that went beyond your favourite Unix shell or triple integrating the surface of a sphere equidistant from a point electrical charge in order to figure out the electric flux per unit area. So I talked about what I actually know very little about -- hockey. I hope I dropped enough buzzwords to make me look credible. And I made small talk about other people's children -- theirs, of course, not just any random children.

Ottawa being the most government of government towns in the land, it was no surprise that half the people at the party drew their salaries from tax payers, be they a government bureaucrat, a traffic cop, or a public school teacher. Even if they were in private industry, their employers depended on government contracts for survival. That must be nice. Because as a high tech worker involved in analog chip design, having a stable job with a guaranteed pension is as close to reality as having female colleagues. And if case you didn't already know, I have no female colleagues. Nor do I anticipate that I will have any. Ever.

One interesting that happened at the party was that I found out that one of the ladies there attended the same high school as me and we even graduated in the same year. Yet, I have no recollection of her at all. I promised to look her up in the yearbook when I got home, but as soon as I left the party, I could no longer remember her name. Horrible. I'm simply horrible with names. Ergo, I'm still an engineer, happier socializing with machines rather than people.

Paranoia and Insubordination

2010-12-14T23:56:00.001-05:00

When the Psycho Colleague was fired, I thought my days of dealing with difficult people were behind me. But it appears I'm now stuck with Mr. Paranoia. Mr. P resides in a south westernly state of the United States and he is very scared of losing his job. Very, very scared. I wrote about this fear in one of my earlier posts on the Paranoid Redundancy Syndrome™. It would appear Mr. P has a severe case of this disease.

It's so severe that he tries to "look good" at every possible turn. Luckily, he doesn't do it by bashing others or stabbing people in the back. But he does it by overcompensating for his fears, such as e-mailing any and all tiny accomplishments to our manager. Recently, he decided he wanted to appear efficient. I'd be happy if he really was efficient, but apparently, he was just going for appearances.

Today, Mr. P was supposed to hand off a set of circuit netlists to another colleague by checking in the files into a repository. I, as the project lead, instructed him to please hold off for 30 minutes as I was about to attend a meeting. The reason for holding off, I explained, was so we can run a new script that I wrote that does this netlist hand-off automatically and adds a few more bells and whistles which makes the circuit easier to debug. It also avoids what up-to-now has been a rather tedious and error-prone manual method of handing off the design. In fact, he knew about this script for a couple of weeks and this would have been the first opportunity to use this script in a real-life scenario. So what does he do? Five minutes before I were to call him, he sends me an e-mail saying he had delivered the netlist already, using the old way of doing things.

I was pissed. After my meeting, I had to take a few deep breathes to collect myself before asking him, as calmly as I could, why he did not do as I had instructed. He replied that he wanted to appear to be on top of things; thus, he went ahead and did things on his own so that he can be "finished" before I called him. He couldn't wait for another 5 bloody minutes! We lost the chance to test the script and the delivery didn't include the bells and whistles that the other colleague was expecting. So he didn't really finish and instead, ended up appearing like an idiot. I'm not sure what sort of reverse logic that go through people's minds when they've been inflicted with Paranoid Redundancy Syndrome™, but apparently, doing everything to hasten your corporate demise while thinking the very opposite thing is one of the newly identified symptoms.

Sincerity

2010-12-13T18:18:00.001-05:00

In order to be a good leader, in order for your flock follow you to the ends of the end earth, it is important to have good relationships. Underpinning all good relationships is sincerity; if you learn how to fake that, you'll have the world eating out of your hands. This is a lesson yet to be learned by my director. I guess he's high up enough in the corporate hierarchy that trivial things such as good relationships with his lowest underlings is deemed unimportant.

It's that time of the season where bells are starting to ring ting tingling, which means it's time for those pesky year-end annual reviews. The usual deadline for completing the self-review portion of this annual rite of flagellation is New Year's Eve. But my director wants it by December 17 because he, the thoughtful soul that he is, wants us to "be able to spend the holiday season with our families without needing to worry about filling out the review forms."

Well, how considerate of him; such a caring message just oozes with sincerity. I love how he likes to plan our holiday season for us. Yeah, right. It seems like at least one person would like to get the review process over with before Christmas. Perhaps he thinks we really are just sheep, willingly lapping up whatever milk of rhetoric that may spill from his keyboard, regardless of whatever foul stench may emanate. The thing is, he could accomplish just as much by being more straightforward. He would also earn kudo sincerity points along the way. Simply explain to everyone that you want it by Dec 17 so you can read all the reviews before Christmas as you'll be away/predisposed/drinking-heavily/gambling over the holidays. That would go over much better than pretending we're children. Because if we're to be treated like children, I expect a two hour afternoon nap to be built into the workday.

Life's Little Victories

2010-12-09T00:04:00.000-05:00

Life is often dreary. That's why it's important to celebrate life's little victories, no matter how small. Tonight, I had such a victory. First, let me take you back to a different time. All the way back, back, back, back, to last month. It was back then that I decided I had had enough of all the junk in my garage and needed to purchase a backyard shed. Winter was approaching and I wanted to move my car into the garage so the snow removal guy can come and, well, remove snow. So I did just that. I bought a shed. Spent $500 on it. Then another $50 in extra stuff to help me level the very non-level corner of my backyard where the shed eventually ended up. Hours of labour leveling. Another few hours putting up the damn shed -- these things never fit quite right. After I had moved all the junk out of the garage and into the shed, I thought it'd be a good idea to put a lock on it. A simple combination lock would do. In fact, I had one in the house and I knew where it was (which is a rarity). The problem was, I hadn't used this lock in probably a deacde and I no longer remember the combination.

A more reasonable person, one that had a reasonable income, would have just tossed the old lock away and spent the $5 to get a new one. This minor problem could have been solved in minutes by jumping into the car and driving two minutes to the local hardware store. But as I said, that's what a more reasonable person would do. But I'm a cheapskate; an obstinate one at that. Because after spending more than $550 and many hours putting up the shed, I just couldn't bring myself to spend another $5 to get the job done. Instead, I scoured the 'net on how to figure out the lock's combination. There are many videos out there, mostly for the Master brand of locks; but alas, mine was not a Master brand. There are other videos on how to hack the lock without the combination, but I wasn't trying to steal someone's stuff. I just wanted to use my own lock. And so after much reading and watching videos and doing this and that, I finally figured out the combination. It took me half of the evening. And I saved $5. But as one of those life's little victories, it sure was satisfying.

Mute

2010-12-07T15:57:00.000-05:00

I use my laptop for conference calls. I have software on my laptop that emulates my office phone with all of the same functionality. My laptop even rings whenever someone calls my office number, as long as I have an Internet connection. So when a meeting needs to happen, I simply start the software and put on my headphones. It works great except when my computer is backing up its files and slows down every other piece of software, the virtual telephone included. And it was during that very moment of backing up files when I moved my mouse cursor to the mute button on the virtual phone and clicked it. The button turned red on my monitor. What I didn't notice was that the button wasn't flashing, which it normally does when on mute. I then started to have my snack -- really crunchy Ruffles potato chips. My colleague in the cubicle adjacent to mine was talking at the time.

I bit into a chip. The colleague immediately asked if everyone was still there. He then went on. I bit into another chip. The colleague asked the same question again. What a coincidence, I thought. Then when it happened a third time, I started to think that it may be my fault. And indeed it was. It was then I noticed my mute button didn't quite mute the phone due to the lag in software. That crunch must have sounded like a lot of static on the line.

Oops. Foiled again by gluttony. I blame it on all those corporate perks we get.

Peer Review

2010-12-01T23:00:00.008-05:00

A Novel technique to measure Intensity for lasers

Abstract

In laser and system characterization to get better Energy Efficiency (EE) say 0.34 has become increasingly difficult for system designers and circuit designers. There are numerous techniques published and proposed for measuring the intensity number. Here is the method proposed which is different from traditional techniques by reporting various intensity parameters for the laser devices.
...
...
...

As so it was with those words that I sat back in my office chair and started to read a paper written by a colleague in FluxCorp India. I've slightly modified the topic of the paper in order to protect the author's identity (and to continue to mystify you all with my anonymity). The underlined words were substituted by yours truly; everything else is verbatim, including punctuation and capitalization.

FluxCorp's design office in India is big. It's so big that there's enough critical mass there to hold a fully internal India-office-only technical conference. As with any conference, a call for paper submissions were made, papers were submitted, and referees sought out for those papers. I was chosen to be a referee for the paper above as the topic is directly related to my work.

There are some technical papers that are just gems, but most end up being either too dense to be understood or too unrelated to be applicable. Reading the paper above was another experience altogether, starting with optimistic curiosity, quickly transitioning to confusion, and finally ending in comedic farce.

What piqued my interest initially, the optimistic curiosity that I just mentioned, was the title of the paper. A "Novel" technique, eh? Great! I wondered that technique could be, and was quite eager to dig into it. After the first sentence in the abstract, however, confusion arose. First, the sentence is ungrammatical. Then there's that 'say' in the sentence. I pushed on, telling myself that I shouldn't be so hung up on language. The abstract goes on to promise a method that is "different from traditional techniques". Again, I'm back to being optimistically curious, albeit with slightly dimmed enthusiasm.

Unfortunately, the quality of the paper never improved from that first sentence. There were continued difficulties with writing style and grammar. Acronymns were used without expansion or explanation. Concepts were described without clarity. And then a table filled with numbers simply appeared near the end. No explanation of what those numbers mean. No explanation of a test setup. No explanation of what was being tested. Just a big table of numbers coming out of nowhere.

So what about that "Novel" technique for measuring laser intensity? Most of the paper simply talked about the concept of intensity, which was described in horribly vague and misleading language. One can do better simply by copying the introduction from Wikipedia. Then when it finally got around to the actual measuring of laser intensity, the technique was described in quite a verbose fashion; however, I think I can summarize it in a single sentence. Here we go: One can measure the intensity of a laser by...wait for it...measuring the intensity of the laser.

Brilliant! The author receives full marks from Fluxor in Tautology 101.

Conservation Does Pay!

2010-11-27T23:31:00.000-05:00

At the urging of government stimulus money, I replaced a few old items in the house near the end of 2009. Three 12L toilets: swapped out for new 6L toilets. Hot water tank, which runs on natural gas: replaced with a brand spanking new one (and then got screwed by the tank distributor...but the story has a happy ending). Basement forced air furnace, which also runs on natural gas: replaced with a new high efficiency one. All the old items were original items in the house, which is more than 20 years old. Three toilets added up to about $600. No charge for the new water tank since it's a rental. And the furnace was about $3000, labour included.

I use an open source personal finance software called GnuCash to track the monetary details of my life. So it was easy to pull up the numbers to see how the first ten months of 2008 and 2009 (with all the old stuff) compare with the first ten months of 2010.

- Natural gas (Jan-Oct 2008): $1558.09
- Natural gas (Jan-Oct 2009): $1686.56
- Natural gas (Jan-Oct 2010): $ 964.97

- Water (Jan-Oct 2008): $908.76
- Water (Jan-Oct 2009): $845.09
- Water (Jan-Oct 2010): $355.80

Even forgoing the government dollars that I received, these investments will clearly pay for themselves in the long run. The toilets will do so by the end of the year while the furnace and hot water tank will do so in another four years at most. With the stimulus money already in my pocket, I'll probably break even in one more year.

The next thing to do is to update some of the electricity hogs in my house. The fridge is more than a decade old and so's the dishwasher. But I think the most power hungry item is the clothes dryer. It's close to forty years old (left behind by the previous owner) and the manufacturer no longer exists. It's the same manufacturer that made the stove that I had kept for so long before it burst into flames and died by electromigration.

This was a real eye opener. If I didn't keep such detailed records, I'd be hard pressed to believe that such energy saving investments are worthwhile. And if the government didn't twist my arm by giving me money, I'd never had the chance to make such a comparison.

So what are you doing to save money on water and energy?

China the Quick

2010-11-23T05:53:00.000-05:00

China now owns the fastest commercial supercomputer in the world according to the latest list (Nov. 2010) of the fastest 500 supercomputers in the world. The top dog, named Tianhe-1A, isn't just 50% faster than the runner-up, it also uses 43% less power. It also looks like a bank of vending machines (see inset).

What's surprising isn't that China made it to the top of the list. What's surprising is that in 2001, China wasn't even on the list; yet now, it owns spots #1 and #3. It certainly won't hold on to those spots forever, but it does give everyone else an indication of just how serious China is when it comes to their local R&D efforts.

Meanwhile, I'm still awaiting my cushy appointment to China. And waiting, and waiting, and waiting...

Patent Money

2010-11-22T00:45:00.000-05:00

I just noticed an extra Can$300 on my last paycheque. It's a bonus from the mothership for a successful patent submission. I guess that's another line I can add to the CV whenever I get laid off. If you take that $300 and divide it by the number of hours I spent working on writing it up and subsequently editing the nonsensical legal version of it, it's surely lower than minmum wage. And I don't even own the damn patent.

It used to be that patent bonuses were US$1000 upon submission and another US$1000 upon approval by the patent office. But those were the good old days of three years ago. Since then, FluxCorp has scaled back patent bonuses to US$600. Given the near historic lows of the US dollar coupled with the various deductions and taxes that the corporation witholds before releasing the funds, all I end up with is Can$300. Hardly an incentive, which means the only motivation I have left for submitting future patents is intrinsic motivation. I think I left my intrinsic motivation in the same bag that I put my enthusiasm, vigour, and youth. Naïveté and svelte waistline are probably in there as well. That bag got crushed years ago along the jade laden Highway of the Working Man.

Happy Together

2010-11-17T12:48:00.000-05:00

My managers and colleagues love me and I love them too. In fact, we love each other so much we can scarcely go for more than a few hours without hearing each others' voices. So we schedule meetings and meetings and meetings and meetings. Sometimes, we love each other so much we schedule an orgy of overlapping meetings. Just look at my Outlook calendar for this week:

I feel so loved. There is a downside to all this love. I find that I don't seem to get any work done. To solve this problem, I will call a meeting to solicit ideas from my wise and loving co-workers. In fact, we are going to continue having these meetings everyday, until I find out why no work is getting done.

Boring Research

2010-11-15T18:19:00.001-05:00

Imagine spending five years of your life involved in the detailed study of the semiconductor industry, only to come up with nothing! Nothing! But that nothingness is a welcome result. An epidemiological study on the cancer mortality rates of those that work in semiconductor fabrication shows that "work in the US semiconductor industry, including semiconductor wafer fabrication in cleanrooms, was not associated with increased cancer mortality overall or mortality from any specific form of cancer."

The results are so boring, it's not something that's likely to make even the inner inner headlines of any news article in the mainstream media. Media outlets like to publish dramatic stories, and not dying from your work is hardly dramatic. Yet, it's important research because it shows that the industry, at least in the United States, has adequate safety measures to protect its workers. It also serves as an baseline of statistics for future studies as well as studies elsewhere in the world.

As a follow up, I'd like to see an in-depth study on the psychological effect of bunny suit colours worn by workers in the cleanroom. When I visited the largest fab in the world in Taiwan, the regular workers wore white and management wore green. As a visiting customer, I was forced to wear hot pink. I think they were employing psychological warfare to try to put me in my place. I survived, but not before losing my dignity.

Arc de Chinois

2010-11-14T00:44:00.003-05:00

The archway (below) at Ottawa's Little Italy opened to great fanfare several years ago, announcing to the city's residents the importance of this ethnic neighbourhood.

Not to be outdone, another archway officially opened last month just a few blocks away from the Italian one. Ottawa's Chinatown now has a really big and fancy new archway. Apparently, it's one of the largest ones outside of China. For years, the business community in Chinatown has been trying to raise funds to create a Sun Yat-Sen memorial park/square, something that is not uncommon in Chinatowns across North America. When they couldn't raise the funding privately, the business leaders changed tact and tried to push the idea of a classical Chinese archway. That idea didn't go anywhere either, that is until the economy went sour. All of a sudden, stimulus money was everywhere, and what better way to buy the ethnic vote than to throw a million or two at the archway project. Add in a few feel good words about international collaboration (Beijing provided the workers free of charge) and voila, a new gateway is born.

The thing is, it's all about style over substance. The really good Chinatowns around North America -- New York, San Francisco, Toronto, Vancouver -- either don't have arches or have really small ones. Then there are the lesser Chinatowns boasting big gateways but not much else. Ottawa's Chinatown is a case in point. Big gate, uh... that's it.

Here's Philly's Chinatown gate (I've been to Philly's Chinatown, it ain't great):

Here's Washington, D.C.'s very small Chinatown with its very big gate:

If you look carefully, Washington's and Ottawa's archways look very similar in size and design. No wonder, both are sister cities of Beijing.

11.11.11

2010-11-11T11:00:00.000-05:00

At the 11th hour of the 11th day of the 11th month, we remember.

In Flanders Fields

By John McCrae, Lieutenant Colonel of the
Canadian Expeditionary Force, WWI

In Flanders fields the poppies blow
Between the crosses, row on row,
That mark our place; and in the sky
The larks, still bravely singing, fly
Scarce heard amid the guns below.

We are the Dead. Short days ago
We lived, felt dawn, saw sunset glow,
Loved and were loved, and now we lie,
In Flanders fields.

Take up our quarrel with the foe:
To you from failing hands we throw
The torch; be yours to hold it high.
If ye break faith with us who die
We shall not sleep, though poppies grow
In Flanders fields.

Early Adopters

2010-11-09T22:51:00.002-05:00

Early adopters of technology is a small but important group of people. They are the beta testers that help to improve the product for the masses. One of the technologies that is still in its infancy is the human mimicking robot. Chris and Dave derided such robots on their online radio show The Amp Hour in one of their earlier episodes (esp. Dave, I think). However, I think these humanoid machines have indeed come a long way. Here's a sample video of some of the very life-like moments from a recent tech expo in Japan. Let me warn you, it's a bit creepy, kinda like Chucky from Child's Play kind of creepy. But the reason it's creepy is because the robot's movement are so lifelike in so many respects. It's a real testament to the engineering team.

There's a few remarkable things about this singing and dancing robot. First, its height and weight is the same as an average Japanese young adult female -- 5 feet 2 inches and 95 pounds(!). Second, the singing wasn't a pre-recorded soundtrack, but the robot's very own voice. Talk about creepy.

For a comparison, here's the original song sung by a real Japanese young adult female. The song title in Japanese is 出逢った頃のように, of which I have no idea what it means. By the way, it doesn't mean "Every Little Thing". That's the name of the band.

The inventors of HRP-4C think that the road to acceptance of robots by the general public is best done by having the entertainment industry become early adopters. And what sector of the entertainment industry has always been at the forefront of technology? (Hint: rhymes with 'corn' and starts with the letter 'p').

I'll end off this post with a trivia question. Believe it or not, the first line of the song in the videos above is actually in English. It's repeated a twice more throughout the song. Listen carefully. What are those English lyrics?

Numeracy

2010-11-08T15:08:00.006-05:00

Numeracy is one of those basic skills that schools are supposed to teach you. You know, the three R's -- reading, writing, and arithmetic. Quite ironic really that it's called the three R's. I guess spelling isn't high on the list of basic skills. After all, RWA can be re-arranged to make a real word, like WAR. What's wrong with telling kids that they should learn basic WAR in school. However, this is a post on numeracy, so let me get to that.

Two weeks ago, Prof. Ramanathan got out of the wrong side of bed and wrote this op-ed piece in the Washington Post. For background, Prof. Ramanathan is a professor emeritus of mathematics, statistics and computer science at the University of Illinois at Chicago. His basic premise is this:

Unlike literature, history, politics and music, math has little relevance to everyday life. That courses such as "Quantitative Reasoning" improve critical thinking is an unsubstantiated myth. All the mathematics one needs in real life can be learned in early years without much fuss. Most adults have no contact with math at work, nor do they curl up with an algebra book for relaxation.

It's a rather unfortunate stance in my opinion. I agree that math has little relevance in everyday life. But the operative word here is 'everyday'. Elections are not an everyday event. Neither is investing your money nor buying a house. Nor is researching the efficacy of a particular medical treatment. Yet, these are all important activities that many of us partake during our lifetimes and activities which require a decent amount of numeracy to make the correct decision. Just because something isn't used in everyday life doesn't make it insignificant. I think life insurance is one of those very things. Another is buying a house you cannot afford and then having a balloon interest rate crop up on your humongous mortgage after the initial six-month introductory 0.1% rate runs out.

After all, isn't the whole housing crisis bubble due in part to home owners that had very questionable numeracy? And while buying an unaffordable home may not be an everyday event, for those that made that mistake, it is certainly having an everyday impact on their lives right now.

If the population had better numeracy, they wouldn't be so ho-hum about the recent change of the long-form census in Canada from being mandatory to being voluntary (yes, one more rant in about this topic). One pundit, John Robson, on the local radio station here said, when commenting on the census, that pollsters seem to be able to conduct their polls just fine, why do we need government to run a mandatory census? It seems Robson would do well with a dose of numeracy himself. He doesn't realize pollsters weigh their surveys based on the demographics of the people that answer their questions. And where do pollsters get their demographic numbers? That's right -- the census.

It's ironic that Prof. Ramanathan thinks that "courses such as 'Quantitative Reasoning' improve critical thinking is an unsubstantiated myth." I disagree. Yet, in this particular instance on this particular topic, it is ironically true for the good professor himself.

You know what innumeracy is like? It's like censoring The Count from Sesame Street and then ending up with a complete different interpretation. For your viewing pleasure, this is what I'm talking about:

Uncensored version here.

Fluxor 1, Evil Corporation 0

2010-11-03T23:22:00.000-04:00

I've been locked in a year long battle with a big British evil corporation over the rental contract of my basement hot water tank. I mean, what's up with those Brits? Either it's spewing crude oil with into the Gulf of Mexico or being stubborn with a disgruntled consumer (yes, they're about the same in my eyes). If we didn't share the same queen, I wouldn't hesitate to invade them with our canoe navy. Anyhow, I've been quite patient with their customer non-service department and their continued broken promises. They'll have my issue escalated, they tell me. They'll call me right back, within 24 hours. They'll be providing me a refund soon. Well, none of those things happened until I got sick of phoning them; instead, I wrote a letter threatening to sue them in small claims court. The result? Two weeks later, I got a call saying they will be refunding me all of the amount that I am claiming within 2 weeks. Today, I received that refund.

Lemme tell you, don't mess with the Flux. Or I'll wait and wait and wait and if I feel like it, I'll write a letter! Scared yet?

Perils of Voting Technology

2010-11-02T14:27:00.001-04:00

Chris Gammell asked me in my previous post on voting technology to squeeze in one more post on the topic. Given today's is reckoning day for the Democrats in the Unites States, I thought it'd be a good time to oblige Chris on his request. Unfortunately, I don't have anything particular interesting to say about this topic that I haven't already talked about previously, which is that I'm not particularly convinced that the advantages of electronic voting outweighs the negatives.

If a disgruntled voter is upset with a paper ballot, the worst he can do is to ruin his own ballot. If a disgruntled voter is upset with a voting machine, well, here's another example of one of those negatives:

The smashed voting machine example comes from Allentown, Pennsylvania, a place that I've actually blogged about back in June. Sure, you may think that example is on the extreme side on things; then how about this problem in Virginia:

Voting should be simple. My personal feeling is that if you need to be trained to use a machine to vote, it's too complicated.

Scary Halloween Costume

2010-11-01T12:04:00.000-04:00

For Halloween, I donned a really scary costume for the kids -- I dressed up as an electrical engineer. The kids weren't scared, but they should be, lest they think about being an EE when they grow up.

Needless to say, the effects of the near global meltdown continues to be felt. Employers are getting pickier, thinking high unemployment numbers mean they can afford to be selective. The problem with this thinking is that workers with deep and relevant experience are still in short supply. The result is that open positions go unfilled and excellent-but-not-quite-perfect unemployed engineers become increasingly frustrated. This harms both the interests of employers and job seekers. What's worse for the future of high tech engineering is that some are recommending job seekers move to greener pastures -- Asia, specifically China, India, and South Korea. As this IEEE article shows, there's "no shortage of jobs for EEs and IT professionals in the Asia-Pacific Region".

That's just bloody great. I was angling for a job in China years ago, I don't need more competition.

Snow, San Diego, and Sanity

2010-10-31T01:57:00.001-04:00

While Jon Stewart was having his Rally to Restore Sanity on the National Mall in Washington, D.C., we here in the northern capital have been getting a little bit of insanity ourselves. What happened? It snowed. Sure, you may be thinking, "so it f^&*king snowed in Canada. That's news?" And you would be right for thinking so. Still, it's just so damn early. I guess there's only one conclusion: global warming is over.

Neighbourhood scene (10/30/10)

I love snow. It's pretty. I love how it glistens in the sunlight. And I love to make snow angels, because it numbs the senses so I don't remember how I have to pay to have it cleared from the driveway. How I have to brush it off my car. How I need to buy winter tires. How it increases my heating bill. How it clogs traffic. And how it creates icicles that threatens to impale you whenever you walk under an overhang.

Four years ago, a recruiter from Qualcomm called to ask me if I was interested in a mostly research job located at headquarters in San Diego. The research group wouldn't be under a strict deadline to produce products, but would instead brainstorm about next-next-generation, 10-20 years down-the-line type of ideas and then try to prove those concepts out by protyping them on silicon. It truly was a big-R little-D R&D job and it sounded great! We talked several times and it progressed as far as a chat with the hiring manager. At the end, they wanted to fly me down to San Diego for a face-to-face interview. It was at that point I balked. San Diego -- scientifically proven to have great weather 98.272% of the time. Yet, I never made it to San Diego, not even for the interview. Looks like Jon Stewart's call to restore sanity came four years too late for me. I chose to stay in Ottawa. Snowy, snowy Ottawa.

Training in Sexual Harassment

2010-10-28T11:15:00.000-04:00

Actually, the title should read "Training in the Prevention of Sexual Harassment", but it was too long to fit in the title line. This week, FluxCorp asked all its Fluxees to be trained on what is and isn't sexual harassment. At regular intervals during the training, we were asked to consider whether a scenario constituted sexual harassment. Let me tell you, those questions were pretty tough. Like this first one:

Joe tells Mary she'll get a raise if she sleeps with him. Is this sexual harassment?

Or this one:

Leah is fondled by a drunk co-worker at a social event. Is this sexual harassment?

If employees can't answer these questions without training, I think it's time to let them go.

Here's a quiz for all of you. Is the picture below sexual harassment?

Millions and Millions of Little Black Squares

2010-10-27T00:42:00.000-04:00

When I first arrived at FluxCorp, I was assigned to fix a broken oscillator design. The chip that the oscillator design went into has now been in production for over a year. At the height of production, we were making 1 million parts per week. Yet, it wasn't all that satisfying despite the fact that it was the first product I had ever worked on that actually made it into commercial production and can be considered a commercial success. That dulled sense of achievement was due to having only played a minor role in the entire project.

The next project was different. Not only was I involved right at the beginning and took on a decent amount of responsiblity, it was to be the first ever Flying Flux designed by our team. Five years and many versions and bug fixes later, we can finally say that the design is a commercial success. Sometimes, that's how long it takes to go from absolutely nothing to design to satisfying our customer's commercialization requirements. And satisfied they were. We just got our first ever large order for our first ever Flying Flux -- nearly 6 million parts ordered, to be delivered over the next four months.

Now that feels satisfying.

Time to round up the workers and start cranking out those little black squares.

Superstar!

2010-10-26T09:30:00.000-04:00

My director just sent me an e-mail to congratulate me on being featured on the main page of our internal website. By the time I read the e-mail and opened up my browser, I was no longer there (on the website). I guess fame really is transient. I didn't even get my 15 minutes. More like 15 seconds.

Electronic Voting

2010-10-26T00:41:00.000-04:00

Yesterday, I mused about the value, or lack thereof, of machine voting versus the traditional pen and paper method. What I didn't mention is that although provincial and federal elections in Canada continue to be strictly pen and paper, many muncipalities have gone toward the Scantron (fill in the bubble) voting method.

So how has this experiment worked out in Ontario? A news story just posted a couple of hours ago indicated that in Arnprior, a town on the outskirts of Ottawa, voting will be extended for 24 hours because their electronic voting system broke down. The returning officer for Arnprior said that "almost all municipal units in Ontario (using electronic voting) are facing this problem."

I love technology.

Democracy and Technology

2010-10-25T11:43:00.001-04:00

I exercised my democratic right today by voting in Ottawa's municipal elections. Despite advances in technology, Canada by and large still votes with the pen and paper method. No hanging chads. No computers to crash. No printer paper to jam. Easily re-countable.

Many parts of the world have gone toward machine assisted voting. The United States has a particular long history with voting technology, starting with its first official mechanical lever voting machine in the 1892. Yet, voting machines are not without problems. Anyone remember Bush vs. Gore and those pesky hanging chads?

The major advantages of voting machines are speed in counting ballots and overall lower costs. I'm not convinced on those lower costs given all the technology that's required to make a modern voting system secure. As for faster counting, really, the world can't wait for an extra day or two?

Am I a neo-luddite?

Lies, Damned Lies, and Statistics

2010-10-21T19:49:00.002-04:00

Today is UN's first ever World Statistics Day. According to the National Post, Statistics Canada has long had a strong international reputation, and in the early 1990s, the agency earned the top spot in The Economist’s "good statistics guide." Thus, it was no surprise that the Industry Minister Tony Clement wrote a memo today to praise the organization, "It is my pleasure, as the minister responsible for Statistics Canada, to congratulate the men and women of Statistics Canada on building what is universally recognized as a world-leading national statistical office."

Of course, this is the very organization he fought so hard to tarnish by first replacing the mandatory census with a voluntary one, and then going on to proclaim that based on the opinion of Stats Can, the voluntary census is just as accurate. This caused the head of Stats Can to resign, saying he cannot abide but such ministerial bullshit (I'm paraphrasing here) and he'd rather resign and clear the air than to sit by and have his organization tarnished by two-faced politicians.

You may think this is an isolated instance in mathematical incompetence. Unfortunately, at around the same time that the mandatory census was scrapped, an announcement was made to put billions into new prisons because of an increase in unreported crime. Yes, we're going to imprison criminals that we don't even know exist. And we're going to spend billions to do so.

It's been three months since I first ranted on about our mathematically challenged government, but it still makes me whince every time I hear about the census. And imprisoning imaginary criminals.

Lies, damned lies by a politician is not surprising. But this time, there are no cherry picked statistics to hide behind. Just bold faced lies.

Aerospace

2010-10-18T15:55:00.004-04:00

When I was still a skinny little boy nearing the end of high school, I dreamed of being either an architect or creating spacecrafts. Along the way, I found out being an architect requires artistic skills, so I dropped that idea. But creating rockets, fighter jets, or space shuttles was still high on my list. I never did go into aerospace engineering. My dad, a mechanical engineer himself, persuaded me to go into electrical engineering, where he thought the jobs were more plentiful. He also mentioned that aircrafts required electrical systems, so I can still get into that industry if I cared to. I never regretted my decision to studying EE. In fact, I did have a slight brush with the aerospace industry. Halfway through my undergrad studies, I worked for four months at Canadian Marconi Company (since acquired by Esterline). Back then, colour LCDs screen were very new and the plan was to put these new screens in to display artificial horizons (inset) in jet fighters in place of the then-current cathode-ray-tubes. However, LCDs with their very sharp pixels are pretty harsh on the eyes and didn't mimic the experience of using CRTs. Thus, I was assigned to implement a bit of image processing -- removing aliasing artifacts -- in order to pretty up the on-screen display. It was a job in embedded programming using a TI DSP board and lots of firmware coding to implement a 2-D guassian anti-aliasing filter. It worked well enough and I had a good time doing it. It also beat serving burgers and fries for four months (paid better too!).

As I mentioned recently, I've been having the blahs at work. My procrastination had led me to surf a job website dedicated to Indian nationals even though I'm not Indian. Given the recent news of the Canadian government buying a boatload of jet fighters from Lockheed Martin, I've also headed over to browse what's available in the aerospace/defense industry. In Montreal, there's Bombardier (planes/trains) and CAE (flight simulators). In the US, there's a few I've browsed through -- Boeing, General Atomics, Esterline, Raytheon. Turns out most electrical jobs require a good grounding in electro-mechanical interfaces. Sounds interesting, but not enticing enough to start from being a junior engineer again.

Although I ultimately did not choose a career in aerospace, every once so often, I still wonder about the romantic notions of those aircrafts I never got the chance to work on. Day dreaming, back to those skinny days of sixteen going on seventeen. Or was it seventeen going on eighteen?

(Skip the first 30 seconds of this video to get to the real clip.)

Goodbye. Hello.

2010-10-13T17:50:00.006-04:00

Yesterday, I mentioned that the Psycho Colleague (PC) has been fired. It's the first time I've ever seen someone actually getting fired as opposed to just getting laid off. And despite much consternation heaped upon me by PC over the years, despite his firing being entirely warranted and solely due to his own doing, I'm not entirely unsympathetic to his situation. I did give him a call at home to give him my best wishes for the future.

But now the team is a man down and a replacement is required. We're already short on staff and this doesn't help. Many management and self-help gurus erroneously believe that the Chinese word for crisis is the fusion of the characters for danger and opportunity. Yes, there's certainly danger here, but there's also wonderful opportunity. As the saying goes, out with the old and in with the new. Goodbye to one. Hello to another. And what better way to boost the importance of our satellite office than to say hello to the new hire here in Ottawa rather than in a place far, far away, i.e. where my manager sits in the foothills of the Rocky Mountains.

For purely self-preservation, I must now modify my role to become part time local recruiter, part time saboteur of efforts to hire elsewhere.

So how will I actually become a saboteur? Easy, I will start using the new résumé reading method proposed by Chris Gammell, which is to read the résumé from the bottom up to see if the prospective candidate has hobbies related to our line of work. For those not interested in working in Ottawa, I will accuse them of having no such hobbies, thus implying they are lazy and unpassionate drones working only for the love of money. Using good for evil, you say? Blame Chris Gammell, he's the one that proposed this method. Not fair, you say? Well, you must have missed my post where at FluxCorp, what's good for the goose isn't necessarily good for the gander. Not ethical, you say? You must have missed my post where Calvin (of Calvin and Hobbes) expounds on his ethical theory. The end justifies the means, but only for me.

Saying Goodbye

2010-10-12T16:39:00.005-04:00

... to the Psycho Colleague (PC). Those that have followed this blog for a while know the various antics that PC has foisted upon yours truly. He has truly earned the moniker that I have bestowed upon him. Yet, it was with surprising ambivalence that I found myself when I learned of PC's departure from FluxCorp.

I was home sick when I first learned about this news through e-mail. Mr. Poker asked me whether this was my doing and my reply to my fellow colleagues is that they all owe me lunch. Horrible joke given someone just lost his job, but I couldn't resist. As I poked around a bit more, I discovered PC was actually fired rather than laid off, which is how problem employees typically meet their fate. Despite his workplace antics, I didn't think he deserved to be turfed out with 2 weeks notice and nothing more. I think his 20+ years of service is at least worth more respect than he has been given.

Let it be known that I think it was a good move for the company. Yet, it's a bit scary how FluxCorp went about implementing this decision. I wish I knew who in the higher-ups pulled the trigger so I know who not to piss off.

(You may be wondering why the video below has Chinese subtitles. My original link to YouTube was taken down due to copyright complaints by 20th Century Fox. But fear not, a bit of googling turned up the same clip on a Chinese video site. The Chinese can't be bothered by such nuisances as copyright law.)

The Goose and The Gander

2010-10-10T21:27:00.001-04:00

At FluxCorp, what's good for the goose isn't good for the gander. Recently, I ranted on about management's new metrics that expect engineering to create perfect prototypes, each time and every time. At the same time these new metrics came out, management also told us that they too, have set metrics for themselves. Management must estimate to within 30% the amount of resources used in any particular project.

Someone in my group was bold enough to ask this question -- why are engineers given a 100% perfection target while management give themselves a 30% leeway? What did our director give as an answer? Because typically, management have never been able to get closer than 30%. That's the answer. No requirements for improving themselves even marginally, yet, a requirement for others to be perfect. What can I say. Leadership by example. This is a situation made for Dilbert -- life imitating art imitating life.

Later on, I asked the same question of our director through e-mail regarding 0% error for us and 30% error for them. I was told that in order to be successful, engineering must be perfect every time. Oh boy, if FluxCorp's business plan is to bet the farm on engineers creating perfect prototypes time after time after time, I think it's time to look for a new job. Well, maybe not. I think I'll wait for my severance first.

Too Ethical

2010-10-07T13:53:00.001-04:00

I am a licensed Professional Engineer in the province of Ontario. The P.Eng. title is a legally protected designation, as is the word "engineer". Like the title Medical Doctor, it's illegal to pass yourself off as an engineer. However, the P.Eng. title hasn't been of any benefit in my career. I don't even mention it on my business card (that is, if I ever get around to ordering business cards). All I get out of being a P.Eng. is the privledge to be charged an annual licensing fee and to receive quarterly magazines focusing mainly on civil and enviornmental engineering topics.

One of the requirements in getting licensed is to pass an exam on tort law and ethics. It's quite simply really. A bit of common sense and intense studying for a week is all that's required to pass that section. In fact, I bought the law and ethics text book from a local university book store six days before this exam, wrote the test, and then returned the book the day after, as the book store has a 7 day refund policy. Is that ethical? Irony at its best.

I had asked one of my undergrad classmates if he was planning to apply for licensure. He said no. Professional Engineers Ontario, the provincial licensing body, requires all licensed engineers to abide by its code of ethics. My classmate didn't want to be bound by those rules. I always found that reason to be quite amusing. Was he planning to be unethical? Did he want to have free reign to be unethical if the situation warrants in the future?

Well, that classmate eventually left engineering to go into the business side of things. And if his success at that is any indication, it's a good thing he decided to leave his ethics at the door.

Metrics Redux

2010-10-06T18:30:00.000-04:00

Just before Christmas last year, my director came up with metrics to measure the effectiveness of our group. I had a good rant on it in this post. One of my major complaints is that the metrics proposed do not make any sense. One of the metrics is that 95% of our designs per year will have first pass success (meaning zero bugs). We do about 1 design per year. There are others in the group doing smaller projects, but we certainly don't have 20 projects per year to measure down to 5% accuracy.

Recently, our director came back to us with a new set of metrics:

- 100% first pass success rate
- 95% success rate when integrating our product within our customers' products

As you'll note, he did remove the 95% from first pass success metric. He moved it instead to integrating our project within the customers' products. Of course, we don't have 20 customer projects per year. But I'll give him a 'D-' for trying.

But let's go back to the 100% first pass success metric. I'm not sure if our director enjoys setting us up for failure, but based on this metric, I think he rather does. Unfortunately, I just don't think he understands the complexity of the Flying Flux. As I mentioned above, it takes a year to design one, and we're still pressed for time. We're constantly pushing the bounds of technology, working in the smallest nanometer technology available, creating new types of architecutres, new types of analog circuits, complex digital-analog interactions, and equally complex embedded microcontroller firmware. That's why there's something called a prototype. That's what a first pass design is supposed to be. Something that performs pretty well and have most of the important functionality and performance, but there might be a few problems here and there that makes it not production worthy. To say that the prototype of a new cutting edge design will be perfect is to laugh at the whole concept of prototyping.

Another problem I have with the new set of metrics is how projects are treated. A small project is viewed the same as a larger project. Our director, before inheriting the design group for the Flying Flux, only had people that did small projects. An clock circuit here. A sensor circuit there. You can add all the small projects together under his purview and it won't have half the complexity of the Flying Flux. Yet, when calculating his metrics, the Flying Flux is treated as a single unit, like all other small projects. Someone did ask why he didn't weigh the projects according to simplicity. He said he wanted to keep things simple. I'm all for simplicity, but if simplicity=stupid, do you still go with simplicity?

Sometimes, I wonder where along the line management loses sight on reality and instead believe in their own delusions.

What is Work?

2010-09-29T16:55:00.001-04:00

As the cliché goes, do what you love and you'll never work a day. According to Christine Scivicque at Forbes, this is just bad career advice. In her blog on the topic, she lambastes this cliché as:

... a blatant, hurtful lie that far too many people fall for. And they end up feeling like something is wrong with them, when really something is wrong with the idea they’ve been sold.

I, too, was sold on this when I was younger. But the fact is, it's difficult for me to make a decent living from procrastination. Or watching TV. Or sleeping. Is something wrong with me?

She goes on to write:

Work is called work because it’s not play. Once you depend on something to put food on your table, it becomes something different. It’s no longer “that thing you do for fun,” it’s “that thing you have to do for survival.”

That doesn’t mean you won’t end up enjoying or maybe even loving the work you do. But it will also be work. You probably won’t mistake it for anything else.

How true. Still, I think the chasm between Ms. Scivicque and the old cliché is one of semantics. I'm sure the cliché is referring to "work" as really crappy work. The type where you dread Mondays and thank God for Fridays. Like cleaning toilets. Or working at McDonald's. Of course, tell that to McDonald's. They think everyone loves to serve burgers.

Actual hiring pamphlet from a local McDonald's:

Getting Things Done

2010-09-28T23:14:00.001-04:00

After blogging about blogging at work, I've decided to buckle down at work and just get things done. It's ironic that I have been having trouble getting things done as I try to follow closely that very popular organizational concept by David Allen called Getting Things Done.

Blog less. Work more. Sometimes, there's no choice but to eat your vegetables.

Fly Me to the Moon

2010-09-22T13:55:00.001-04:00

To my Chinese colleagues sitting next to my cubes and to my Chinese colleagues in China, may they have a wonderful Mid-Autumn Festival, where the tradition is to admire the full moon (the festival always falls on a full moon) and eat mooncakes (inset).

It's obvious that the Chinese population in Ottawa is rising. Mooncakes are no longer limited to just ethnic stores such as the mega Asian supermarket that opened last year. Regular grocery stores and Costco now carry them as well.

The Chinese legend related to this festival is about some gal taking a potion of eternal life and then flying off to live in the moon. In that spirit, I think the following song is quite appropriate. And c'mon, it's Doris Day. Who doesn't like Doris Day?

Differentiation in Undergraduate Education

2010-09-21T00:05:00.006-04:00

For the longest time now, my impression is that the choice of undergraduate school matters little as far as actual learning of material goes. Yes, one school's reputation may get your foot in the door for a job interview and the contacts you make may serve you well later in life. But in terms of the actual academic content, I didn't think it mattered much. My rationale is that undergraduate material is information that is well established, thus, widely available. In fact, many universities use the same text book in teaching their material. So where's the differentiation?

Let me just say this is not a strongly held viewpoint. In fact, this assumption is due for a review after listening to The Amp Hour's seventh podcast. Hosts Chris and Dave, at around 41:00 minutes in, discussed the complicated nature of software used in drawing the physical layout of integrated circuits (chips). Dave noted the very dear price tag on this type of software, which can indeed costs hundreds of thousands of dollars. Chris then followed up by asking how anyone gets started in the field of chip layout. After all, universities can hardly to afford such software.

I can only surmise from Chris's and Dave's comments that neither of them learned to do any chip layout in university. Frankly, I don't know the educational background of Chris and Dave. Perhaps they had the opportunity to do chip layout at school but simply passed up on those courses. This is the very opposite of my undergraduate experience. The very first chip I laid out was a bipolar transistor and resistor array with two available layers of metal for routing. The transistors and resistors are fixed in locations, but the metals and vias is entirely up to you to draw. Back then, we used Cadence software at school. Today, I still used Cadence software at work. The project was simple enough -- design a simple circuit and then wire it up with two layers of metal. We were given a few types of circuits to choose from. My lab partner and I chose the multivibrator. The chip was fabricated and returned to us in about four weeks, in which then we were responsible for testing in it in the lab. It worked! It crystallized all that abstract stuff I was learning on holes and electrons, current gain and threshold voltage. All that knowledge presented itself as a tangible, physical product that does what you designed it to do in a virtual environment. That was a great moment.

Another course project had us custom designing a digital circuit from CMOS transistors. We had to layout the circuit by drawing every single feature of every single transistor. We started with a blank screen and started putting down rectangles that represented transistor gates, drains, sources, n-wells, vias, wires, and so on. We had to run DRC (design rule check) to make sure wires weren't too close, vias are square, etc. and we had to run LVS (layout versus schematic) to ensure the circuit represented by the layout equaled the design intent in a schematic diagram drawn elsewhere.

From what I can remember, the custom CMOS layout that we did sucked. But that's OK, because the purpose of the course wasn't to turn us into layout designers, but to provide hands-on experience in using industry design tools in creating integrated circuits. I got a taste of what integrated circuit design was all about and ended up pursuing grad school on this very topic.

In contrast to Chris's comment that universities cannot afford such tools, this doesn't appear to be the case. In fact, industry-strength software are often heavily discounted to educational institutions. It's not all altruistic, as the software makers no doubt hope that the students who learn on their software will mostly likely be the purchasers of their software once in industry, where the real money is made. Still, the software is not free. And I wonder if some of the smaller universities, where engineering is not a major focus, whether they will be willing to spend the money at all, albeit discounted, in getting such software, or have the right mixture of professors to deploy such software to the students.

This all goes to challenging my original assumption that undergraduate schools matter little. Perhaps a better statement is to say that the differentiation of undergraduate programs amongst large, fairly well funded schools, is small.

When Transistors Get Married

2010-09-20T01:30:00.000-04:00

In my last post on transistors, the PMOS and NMOS transistors were introduced, representing the yin and yang of electrical devices. As we know, opposites attract, and when a PMOS and a NMOS really really love one another, a miracle is made, and a logic gate is born.

So why are logic gates important? Because most of the world's electronics are created with logic gates. This post introduces two of the most basic logic gates -- the inverter and the NAND gate. Conceptually speaking, only these two gates are required to build all other digital circuitry out there in the world. All other digital gates can be mathematically derived from just these two.

So, let's take a look at how the NMOS and PMOS are married to create an inverter. I'll be using the water pipe analogy one last time. After this post, I'll be abadoning this simplified analogy.

As I wrote in Transistors: Yin and Yang, the NMOS is akin to a valve in which water flows when the valve is pushed in while the PMOS is aking to a valve in which water stops when the valve is pushed in. If we connect the two valves together with the PMOS is on top and mechanically join the valves together, we get the following.

What we care about is the water pressure in between the two valves. As can be seen from the drawing above, if the valve is left out, the water pressure is maximum and equals the pressure at the top end of the valve since water has no where to flow. If the valve is pushed in, all water flows out and water pressure goes to zero.

Putting it simply:

no pressure on valve => max pressure in the center
max pressure on valve => no pressure in the center

Thus, the pressure "inverts" going from the valve control to the center of the pipe. Replacing the water pipes with transistor symbols, an inverter schematic looks like this:

max voltage on transistor gates => min voltage at transistor drains
min voltage on transistor gates => max voltage at transistor drains

The inverter symbol is drawn like a triangle with a bubble at the end indicating an inversion. The supply voltage and ground is typically left out of the drawing and is implicitly assumed. "Max voltage" is a 1 while "min voltage" is a 0. Thus, we end up with the following table.

As for the NAND gate, the diagram below shows how it's constructed with transistors, its customary symbol, and the input to output table (a.k.a. the truth table):

By marrying one PMOS and one NMOS transistor, an inverter is born. So what do you call putting two PMOS and two NMOS transistors together to create a NAND gate? This is left as an exercise to the reader.

Blogging at Work

2010-09-19T00:31:00.001-04:00

I've been blogging a lot at work recently. In fact, the really long and detailed Part 2 on building a happiness machine was written entirely at work, including the creation of all the plots and schematics. You'd think there's nothing to do at work, but on the contrary. Work is busy. But work is also, well, just work.

To put it briefly, I'm bored. I'm restless. I'm procrastinating.

My productivity is low and I'm riding on the good will that's been afforded me since my promotion earlier this year. That good will won't be lasting too long if I continue like this. One of my recent procrastination activites is to search through various online job sites while I'm at work. Is IT eavesdropping? Who cares. I've even gone to browse a website dedicated to Indians living in the US that are willing to consider moving back to India to work in their ever-growing hi-tech industry. Just to be clear, I'm neither Indian (Asian or American) nor do I live in the US.

Am I burned out? I was working some pretty long hours for quite a many months right up to about the end of June. Am I suffering from vacation withdrawal symptom? A big long road trip to the east coast and another week-long trip to Mexico sure was a nice way to spend the summer. Could it be the job itself? I have been doing this analog circuit design stuff for more than a decade now. Maybe I'm just sick of it. Or maybe the recent promotion has left me doing a lot of really boring stuff, much of it related to the process and methodology of designing circuits rather than actually designing circuits. This means I'm tracking progress, ensuring good design methodology, tracking bugs, following up on team members' action items, and creating PowerPoint dog and pony shows for the customers.

Whatever it is, it better passes soon.

All Stars in the Capital

2010-09-17T11:34:00.000-04:00

There hasn't been much hockey happiness in Ottawa ever since the Senators made it to the Stanley Cup finals in 2007. And lost. But at least a little bit of hockey madness is guaranteed to return to town in 2012, regardless of how the Sens perform, as the '12 NHL All Star game has been awarded to Ottawa.

I can see it now. Three days of traffic gridlock as fans descend upon the west side of the city, which is where I both live and work. I can't wait.

Office Politics 101: Begging

2010-09-16T13:53:00.005-04:00

This is a continuing series on office politics.

---

Begging for your job. Literally. It's a risky strategy, but desperate times call for deperate measures.

Two winters ago, when the US housing crisis exploded worldwide, every job in the world felt insecure, including the high tech sector in high flying Shanghai. At that time, the Flying Flux was midway through its design phases and we were short on people who can do verification on the digital portion of the design. This involved writing code to verify the proper behaviour of what I would consider fairly non-complex state machines. After all, even an analog guy like me could've done the job. So I sent out a request for help and we got a loaner from our Shanghai office -- a Ph.D. holder in EE with 5 years of experience. Great, I thought, we'll be finished in no time.

How wrong I was. As with the vast, vast majority of ASIC and FPGA designers, we use Verilog, a hardware description language, to design digital circuits. It turned out our assigned loaner (let's call him Dr. Clueless) has never seen Verilog. Crap. During our first phone conference, our US-based lead digital designer spent three hours talking to Dr. Clueless on what essentially was a crash course in Verilog. I hung up after 15 minutes in.

No big deal, I thought, the guy has a Ph.D. in digital systems. Surely this Verilog stuff should be easy. Two months later, the lead digital designer was still spending more time teaching and guiding him along than we were saving in trying to get useful work out of him. Typically, things went like this:

1. Monday evening, Eastern US (Tuesday morning in Shanghai), Dr. Clueless presents work that was done over the weekend. The lead digital designer pores through every single detail and points out where mistakes were made and then goes on to explain why these are mistakes.

2. Thursday evening, Eastern US (Friday morning, Shanghai), Dr. Clueless presents work that was done since Monday. The lead digital designer pores through every single detail and points out where mistakes were made and then goes on to explain why these are mistakes.

3. Loop back to 1.

Dr. Clueless joined the team in October. Two months later, as the year was nearing its end, I had to start writing performance reviews for about 12 people to send upwards toward management. Dr. Clueless knew he was doing poorly. What's worse was that a junior fellow, another loaner based out of the same Shanghai office, was doing exceptional work on a different digital section that's an order of magnitude more difficult.

Right before the Christmas break and right at the end of another one of those regular "show me what you did" meetings, Dr. Clueless made one of the most daring political moves I've ever seen. He begged for his job. He told me that he "need an excellent review" because he "can't be without a job" since he has to support his wife and child. He "really needs it to look good."

What a shrewd move. Mere days before Christmas, am I to become Ebenezer Scrooge and condemn little Tiny Tim to an uncharitable death? Am I to hand down a five year prison judgement on Jean Valjean for stealing a loaf of bread for his sister's starving family? But of course. I shall be both judge and executioner! Bwahahaha... Actually, I relayed to him that he shall be treated fairly. No more, no less.

His request wasn't a sudden emotional outburst of despair. It was clearly a calculated, premeditated move. I can certainly understand from his point of view how begging, while being entirely unbecoming, undignified, and unprofessional, is actually the least riskly of all options. First, FluxCorp pays very well in Shanghai. He probably makes 5x of an average white collar worker. Second, there was a major global recession and even China was feeling the pinch; thus, poor performance may very well lead to extended unemployment. Third, he has a family to support. Fourth, he was probably betting that even if his begging didn't work, it wouldn't end up hurting him. Lots of upside if it worked, low risk of downside if it didn't.

On that fourth point he was absolutely correct. I could have derailed his career by noting this behaviour in his review, but I omitted it completely from the official record. The lead digital designer, when asked informally by the higher-ups through e-mail about Clueless's performance, embellished the "accomplishments" in his reply. As the lead digital designer later told me, he wasn't "about to throw his family onto the streets". Thus, the calculated gamble did indeed paid off.

Although Dr. Clueless didn't get all of the upside he was looking for, it didn't materially cost him anything to have asked. I can't say though if it cost him his self-respect. That will be something for the good doctor himself to figure out.

Perhaps all this capitalism is just too hard. Surely, FluxCorp is nothing but the Man keeping the slave labour of well paid engineers. It's time to do what Valjean did -- get angry, sing a show tune, and follow Liberty into violent rebellion.

www.flyingflux.com

2010-09-16T03:10:00.003-04:00

Being a cheapskate is not easy. Sometimes, I just can't help but indulge myself. Yes, I coughed up the US$10 and bought my own domain name. Dave and Chris over at The Amp Hour tried for weeks searching for a .com domain for their radio show, but wouldn't you know it, the Flying Flux had a .com readily available.

So thanks, you readers, all 7 of you, including mom, dad, bro, and little Timmy, for being so dedicated to my rants and raves.

These posts shall now live on forever under the Flying Flux banner.

Modeling: Easy Peasy Lemon Squeezy

2010-09-15T10:00:00.002-04:00

How apropos for xkcd to comment on the easiness of mathematical modeling during my on-going attempt at modeling happiness experienced during a vacation.

Here's today's xkcd (click on the comic to go to xkcd and mouseover to see the rest of the joke):

Easy peasy lemon squeezy.

Of course, the flip side to easy is: "difficult, difficult ... lemon, difficult."

Modeling Happiness, Part 2: Vacation length

2010-09-13T17:12:00.005-04:00

In part 1 of this series, I stated that the goal is to create a happiness machine can be used to predict, monitor, and store your personal vacation happiness. A secondary goal, which I forgot to mention, is to create this machine using as much analog circuitry as possible as analog circuit design is what I do for my livelihood.

Two equations were developed in the last post. I'll reproduce them below as a quick refresher:

[1] h=n*c*(0.5*w + 0.35*f + 0.15*s)

[2] avg(h) = h/n = c*(0.5*w + 0.35*f + 0.15*s)

where,

cost = c;
food = f;
service = s;
weather = w;
number of days = n;
happiness = h.

Variables c, f, s, and w are real numbers bounded between 0 and 1 inclusive and are measurements of how happy I am with each of those individual factors with 0 being completely unhappy and 1 being completely happy. Variable n is a whole number (non-negative integer). All variables are assumed to be orthogonal.

One important assumption I've made is that all variables are orthogonal. This is in all likelihood not true, yet, I'm going to claim that this is good enough as an engineering assumption. It's an important assumption because orthogonality allows me to model each variable independently and then put them together to create a final model rather than trying to model all of the intricate interdependencies amongst all the variables. In other words, assuming orthogonality makes my life simpler.

In this post, I shall be creating a more realistic model of how the length of vacation impacts vacation happiness. Looking back at equation [1], we see that the entire equation is multiplied by variable 'n' and if we leave it at that, assuming all other variable to be the same, every day on the vacation provides equal happiness as every other day. However, this is typically not true, and certainly not true for me. As Cherish points out in the comment section of part 1, after about a week of vacation, her happiness diminishes as she becomes antsy to get home. Similarly for me, there is a point in time in which I start getting sick of living in a hotel room, although my tolerance for being away from home is higher than Cherish's having spent two weeks in Shanghai on business and absolutely loving it.

Before we begin creating a new model, I'm going to redefine variable 'n' from being a whole number indicating the number of days on a vacation to a non-negative real number representing days on a vacation or parts thereof. With that little detail out of the way, let's get to modeling happiness versus days on vacation. This new model will end in a new happiness versus days function, which I'll term hn(n) and I'll subsitute hn(n) into equation [1] to come up with a refined model.

Originally, I had planned to model this decrease in happiness over time as an exponential decay because it's easy to do with a resistor and capacitor. Basically, this can be easily model by equating it to a 1V starting voltage on a simple parallel RC circuit that decays over time. The values of R and C are chosen to model the amount of decay that suits a particular individual. For example, a exponentially decaying equation can be generally written as: hn(n)=k*e^(-L*n). With a starting voltage of 1, k=1, n=time, and L=1/(R*C), so that hn(n)=e^(-n/(R*C)).

However, this isn't how I typically experience the happiness of a vacation as this means the largest drop in happiness is at the beginning. This 'time happiness gradient' (dhn/dn) is most negative at n=0 and then reduces in magnitude (although not in sign) as time progresses. This just isn't reality for most people. More typically, the amount of happiness remains relatively constant for the first few days of vacation and then starts to drop off rather sharply as Cherish had mentioned.

Thus, it's wiser to model the time happiness function as a hyperbolic tangent function, typically written as tanh(x), reproduced below:

To have it mimic happiness, we can transform this plot by flipping it along the y-axis, adding an offset, and normalizing happiness to be bounded between 0 and 1, and then adding a scaling factor to control the drop-off. Voila, my own personal happiness equation versus time:

[3] hn(n)=0.5*tanh(0.15*(14-n))+0.512.

As can be seen from the plot above, happiness for me starts to drop off quickly after about 7 days with the drop-off tapering off after about three weeks.

But how can we model this with analog circuitry?

One way is to use an amplifier in its open loop configuration. Amplifiers will 'compress' when a large enough signal is applied. Thay is to say when the input signal is small, the output signal is A times (A being the amplifier gain) the input signal. In mathematical terms:

vout=A*vin; for small vin

But when vin gets sufficiently large, the small signal gain of the amplifier goes down. Thus, for larger signals,

vout < A*vin

This is called gain compression, where gain is reduced as input strength increases. Gain compression can be desirable, such as in audio amplifiers where compression creates that 'warm' sound or in the case where you're building a limiting amplifier. But often, gain compression is highly undesirable and creates third harmonics that creates all sorts of havoc with system performance (e.g. third-order intermodulation in RF systems).

However, in building our happiness machine, gain compression is actually a useful amplifier characteristic and we'll be using this gain compression to model equation [3]. The reason is that many amplifiers naturally exhibit a gain compression curve that is very close to a tanh(x) function. Thus, we can exploit the amplifier gain to model the time happiness model.

If we fix the positive input of the amplifier to a constant voltage, let's say half way of the supply voltage, and then we slowly charge up the negative input of the amplifier by having a constant current go into a capacitor, we can reproduce the shape of the plot of equation [3]. Here's a schematic diagram of a quick circuit that I threw together. I've decided to use ideal current sources, resistors, and capacitors here to save time, but the transistors are real-life models.

Here's what the voltages 'vm' (green) and 'vout' (purple) looks like over time:

The node 'vm' rises linearly as expected while 'vout' looks very much like a tanh function. Zoomed in a bit, we get this:

There you go. The output, vout, looks just like equation [3]. To customize the behaviour of the output, choose your own input current, capacitor value, and amplifier gain to create your own personal curve.

With a new happiness vs. time equation (hn(n)=0.5*tanh(0.15*(14-n))+0.512), we can now substitute 'n' in equation [1] with the cumulative sum of hn(n) from 0 to N days. Putting it mathematically, equation [1] is now transformed into:

[4]

Got any ideas on this model or its implementation? I'd love to hear your ideas.

Modeling Happiness, Part 1: An algebraic model

2010-09-11T21:50:00.007-04:00

You want a vacation. In fact, you've already decided on a tropical all-inclusive resort. But there's so much to choose from and there's so much that determines how happy you'll be with the vacation -- cost, food, service, weather, number of days. How can one make an informed decision?

Often, months after you're back from a vacation and someone asks you how it went, a faded memory means that it's hard to give the person an accurate and precise answer.

Luckily, there is a solution to both problems. You can build a machine -- a happiness machine that can predict how happy you'll be before you go on a vacation, monitor real time your happiness during the vacation, and recall for you how happy you were even months later when the vacation is but a distant memory.

I will show you how this machine can be made, but building such a happiness machine is too large a project for one post. So let's take this step by step. Let's start!

For my first step, I'm going to create a mathematical foundation for this machine from which I can build upon and refine. I'm going to make a reasonable assumption and take the five factors mentioned earlier -- cost, food, service, weather, number of days -- as the sole five factors in determining how happy I'll be with my vacation.

OK, time for that simple equation on vacation happiness. Your personal equation may be different from mine, but I think mine is good enough as an illustration. A few variable definitions before I begin:

cost = c;
food = f;
service = s;
weather = w;
number of days = n;
happiness = h.

Variables c, f, s, and w are real numbers bounded between 0 and 1 inclusive and are measurements of how happy I am with each of those individual factors with 0 being completely unhappy and 1 being completely happy. Variable n is a whole number (non-negative integer). All variables are assumed to be orthogonal.

So, here's my personal vacation happiness equation:

[1] h=n*c*(0.5*w + 0.35*f + 0.15*s)

As you can see, happiness is cumulative in this equation based on the number of days vacationing. Assuming all other variables to be constant, having a 5 day vacation will give provide me five times the happiness versus a one day vacation, since I got to experience that daily happiness five times. We can also re-write this vacation to give an average daily happiness:

[2] avg(h) = h/n = c*(0.5*w + 0.35*f + 0.15*s)

That's it for today, two simple equations. Note that avg(h) is necessarily bounded between 0 and 1. But really, how satisfying is that? Simple equations like this just aren't machine-worthy. But don't worry, I shall be relying on equations [1] and [2] in future posts on which to base my happiness machine.

C'mon. Let's head to Part 2.

Transistors: Yin and Yang

2010-09-07T23:27:00.000-04:00

Polar opposites, it's all around us. And in the world of electricity, this is no different. There are male and female plugs, male and female engineers (although the latter one is rare), protons and electrons, anode and cathode, north and south poles, and with transistors, there are also two flavours that are polar opposites of teach other. The yin and yang of transistors, if you will.

In this third installment on transistors and the continuing series on electrical devices, I describe how these two opposite types of transistors work. But first, let's start off with the philosophical underpinnings of it all. Wikipedia writes:

In Chinese philosophy, the concept of yin yang is used to describe how polar or seemingly contrary forces are interconnected and interdependent in the natural world, and how they give rise to each other in turn.

In transistorland, there's the P-type transistor and the N-type transistor. Both operate with seemingly contrary forces, yet are nevertheless interconnected and interdependent in today's microelectronic world. And if you believe the transistor was created based on Chinese philosophy, you may be interested in the millions of dollars I have stashed away in a Nigerian bank (shhh...send me an e-mail with your bank information).

First, a little refresher from the last installment on transistors. I made the analogy that the transistor acts like a push-button valve where a force is required to allow water to flow through the valve. Here's is the picture again.

The picture above shows that water can only flow by applying a force (pushing in) on the valve. This is akin to the N-type transistor where positive voltage applied to the gate terminal allows current to flow between the other two terminals, the source and drain.

The yang to the N-type's yin is the P-type transistor. Here's the water pipe equivalent, where applying pressure on the valve restricts water flow:

Opposite to the N-type transistor, the P-type transistor restricts current flow between the source and drain terminals when the voltage is high and allows current flow when the voltage is low.

When creating a circuit schematic, there are many ways that a transistor may be drawn. Below, I show two common ways of drawing N-type and P-type transistors. There's what I personally term the bubble method, where there's a bubble in the P-type transistor,

and then there's the arrow method, where arrow indicates the direction of positive current flow and is place on the source terminal:

These particular types of transistors are called MOSFET transistors. There are others such as the BJT, JFET, and MESFET, but the MOSFET is by far the most manufactured transistor in the world. Hence, I've limited my discussion to these types alone. In the workplace, the N-type and P-type MOSFETs are typically referred to as NMOS (EN-Moss) and PMOS (PEE-Moss) transistors, where 'MOS' is short for 'MOSFET', which in turn is short for 'metal-oxide-semiconductor field effect transistor'. Taken together, they're called CMOS (SEE-Moss) transistors, where the 'C' standards for 'complementary', i.e yin and yang.

If you understand how CMOS transistors work, you have the fundamental basis on which to understand how integrated digital circuits are created. Today, digital circuitry inside chips contain only CMOS transistors. No capacitors. No resistors. Only transistors. And how to unleash the interconnected yin yang of transistors to make digital circuitry is what I'll be talking about next in this series on electrical circuits.

Ramble Rambo

2010-09-06T01:48:00.000-04:00

Imagine this scenario. You see your neighbour washing his car, so you make a friendly comment, "almost finished I see." Your neighbour then responds, "yeah, I was going to wash it three and a half hours earlier, but Mary Sue called and said she'll be bringing Buffy to the barbeque, then my mom came over and the donuts got cold, so I picked up the dog and went back to into the grocery store. Coconuts are up 12.4% from last week and the hemroid cream is on sale for 20%. So that's why I'm still not done washing the car." At this point, you nod, walk away, and think your neighbour is bats*#t crazy.

Mr. Poker talks just like that. He is our resident sexist, borderline gambling addict, and conspiracy theorist colleague who also happens to be the most senior engineer on the team. At FluxCorp, we're an equal opportunity employer, meaning we're willing to hire anyone, loonies included. I do get along quite well witih Mr. Poker. A funny curmudgeon he is. Back to the story. We have a team located in northeastern United States dedicated to testing the Flying Flux. In our regular weekly meeting, one of the test engineers asked Mr. Poker a simple question about a particular setting on the chip. The answer should have been quick. One number, maybe two could have been offered. Instead, we get a long winded treatment of how he went about doing the design, why the settings need to be changed, and the origin of those changes. It would have been bearable if it all made sense, but he kept leaving out details in between thoughts. It's not a logical progression from A-B-C-D. Instead, his train of thought goes from A-E-G-H-I-Z. The irony is that the circuit he was talking about originally came from me, and even then, I only managed to catch about half of what he was saying. I could only imagine the utter horror the rest of the team had to go through listening to him go on and on and on about stuff that is totally meaningless.

Once he gets started, he doesn't stop. He could be off on a tangent and he doesn't stop. People give him hints to stop, and he doesn't stop. The pain caused by his non-stop machine gun rambling has me thinking about a new nickname for him - the Ramble Rambo.

So is there a Beatles song that mimics Mr. Poker's untamed stream of conciousness ramblings? Sure there is. The Beatles were good, but they also recorded a bunch of crazy wacko stuff. Here's Revolution 9 from The White Album.

Ignoramus

2010-09-04T02:28:00.010-04:00

Blogs are meant for public consumption. In the case of the Flying Flux, the public wasn't consuming all that much. Thus, taking the advice of another blogger, I decided to do a little self-promotion on reddit.com in the ECE and electronics categories. Soon, I noticed a dramatic pickup in traffic to the Flux. With increased readership comes increased criticism, none more so than a comment made by tty2 at reddit.com. Here's what s/he had to say about my series on electrical devices:

Honestly, I think these articles are very dangerous to young electrical engineers. Someone who doesn't fully understand the concepts is giving vague and inaccurate descriptions intended for people with no knowledge. It's like intellectual sabotage.

W.O.W.

First, let's give credit where credit is due. A big thanks to tty2 for thinking I'm so "very dangerous". I think chicks like that.

Second, regarding the rest of tty2's comment, well, I'm just really pissed off. I really am. In fact, I may even end up launching a lawsuit. As you can see, tty2 informs me that I am "someone who doesn't fully understand the concepts" of basic electronic components.

Five years of undergraduate studies and work programs followed by that hazing ritual called graduate school, and only now do I find out that my schools weren't even able to teach me the most basic of basics! I mean, wtf! I'm so pissed! Thanks again, tty2, for letting me know the real truth. I'm packing my bags man, because tomorrow, I'm heading down to my two alma maters and demanding my money back. What a total scam! I mean, I could've gotten one of those "life experience" Ph.D.s I see advertised online for $39.99. But these two so-called "universities" were able to keep up the scam for so many years, I just can't believe I fell for it all. Seriously though, these two "universities" sure put some serious effort into their scam. They had to hire fake professors to talk about fake topics to allow me to get paid with fake money doing fake internship work. In "grad school", I must have been assigned a fake supervisor to tell me to do some fake research which I ended up publishing in a fake conference run by some fake organization called the IEEE. Oh yeah, remind me to contact the US Patent Office to tell them to nullify all of my patents because surely, since I don't know what I'm doing, those patents couldn't possibly be useful.

Now that I think about it, it all makes sense. Coming out of undergrad, I felt like I didn't really know all that much about electrical engineering. No wonder! At the time, I had asked "professors" at my first "school" what I should do after I "graduated" since I felt I knew so little. Of course, they suggested that I go do "grad school", just so they can scam me out of even more money. Gosh, I feel like such an ignoramus that I needed to look the word up.

I'm so distraught. I feel like, I don't know. I feel like I've just been forced to watch Highlander II all over again. Once FluxCorp finds out about this, I'll surely be fired. I do want to go back to school and study the real stuff that tty2 was talking about. But without a job, how would I get the money? Luckily, I heard of this new online university that just opened up for $9.95 per month. Thanks Glenn Beck, for making Beck University so affordable. I'm pretty sure this isn't a scam this time around.

---

** Just to be clear, this post is sarcastic from top to bottom. I couldn't care less about any negativity thrown my way, although it does make for good blogging fodder.

Capacitors: A Bucket Full of Charge

2010-09-02T10:29:00.002-04:00

REPOST. This was originally posted on June 10, 2009.

---

Before continuing with the series on transistors and having already described the resistor, I'd like to touch upon one more fundamental electrical device, the capacitor. The typical drawing of a capacitor on a circuit diagram is shown below. The symbol represents two metal plates that are parallel to each other with wired terminals jutting out from each plate.

Continuing with the water pipe analogy of electricity from previous posts, if the transistor is a valve and the resistor is a narrow pipe, surely the best fitting description for the capacitor is a bucket. A bucket stores water. A capacitor stores electrons.

The diagram below shows a bucket being filled with water. How exciting. The important thing to note is that given a constant flow of water into the bucket, the water pressure at the bottom of the bucket increases linearly with time.

Similarly, given a constant flow of electrical current into a capacitor, the voltage difference across the terminals increases linearly with time.

A bucket full of charge. This electrical devices stuff is easy, eh.

Movies and High Tech

2010-08-31T02:44:00.002-04:00

This past weekend, I finally saw The Transformers (2007). In it, I find out many major human advancements in technology were due to reverse engineering of the Decepticon leader, Megatron, which at that point in the movie was cryogenically frozen under Hoover Dam. One of those reverse engineered advancements mentioned was the integrated circuit, a.k.a. the chip.

Arthur C. Clarke wrote that "any sufficiently advanced technology is indistinguishable from magic." I guess the integrated circuit is magical enough that the movie writers thought it worthy of being supplied to us from aliens. Even though I work in the field, I still find it amazing that those little non-descript black squares are capable of so much.

That little moment in The Transformers reminded me of another movie scene showing a supposed "chip" that is responsible for monitoring the entire global atmosphere. It was a horrible, sad, sad movie and I'm even sadder to say that I paid to see it in the theatre. That movie is Highlander II (1991), one of the worst ever sequels to anything anywhere.

Early on in the movie, the protagonist's eventual love interest breaks into an evil corporation's main evil machine room and removes a sensor that is monitoring the global atmosphere. Here's a screen capture from the movie:

Remember, this is supposed to be the future with really futurey future stuff. Here's what the sensor contains:

Six bloody resistors and two capacitors! That's the future?! I guess if they can make an atmospheric sensor from eight discrete passive components, it must be futuristic indeed. I do distinctly remember chuckling to myself when that scene flashed across the screen. The movie only got worse.

There are countless sad and funny portrayal of technology in film. What's your favourite?

Resistors: Transistor's Older Simpleton Cousin

2010-08-30T09:39:00.000-04:00

REPOST. This was originally posted on May 26, 2009.

---

In the continuing series on electrical devices, I present to you the resistor, typically drawn on a circuit diagram as a jagged lines shown below:

Using the water pipe analogy, a resistor can be thought of as a narrow pipe:

As water molecules on the left side of the constricted pipe crowd together to try to fit through the constriction, pressure increases. As water molecules leave the constriction on the right side, there's more space for the water molecules and thus, pressure decreases.

Similarly for a resistors, an electrical current flowing through a material that impedes the flow of electrons will have a higher voltage on the side where current is entering the resistor and lower voltage on the side where current is exiting the resistor.

Resistors can be considered the transistor's older simpleton cousin. Older because the resistor has been around for much longer than the transistor. Simpleton because a resistor is both simple to understand and simple mathematically. Consider the voltage to current relationship of a resistor based on Ohm's Law:

v = i * R

Where v is the voltage difference across the resistor, i is the current through the resistor, and R is the resistance, a measure of how much the resistor impedes electron flow.

Now consider the voltage to current relationship of a field-effect transistor:

Whew! That's a lot of variables.

There are other cousins in this family of electronic devices. I shall touch upon one more -- the capacitor -- before returning to the continuing story on transistors.

Transistors: Turn Me On Half Way

2010-08-28T01:25:00.000-04:00

REPOST. This was originally posted on July 13, 2009. This is part two of my qualitative description on transistors. Your feedback is welcome.

---

In the continuing series on electrical devices and following the first post on Transistors: Turn Me On or Not At All, I will now discuss how to only be turned on half way. Not you, the transistor that is.

Remember that the simple view of a transistor using the water pipe analogy is that it is a valve. You turn it on, current flows. You turn it off, current stops. However, life is really never that simple. We all know that in order to turn a valve on, like unscrewing the cap off your favourite non-alcoholic beer, a bit of force is necessary before it will turn at all, after which it becomes quite easy to twist. A transistor is similar in that in order to turn it on just a little bit, an initial force is necessary, but once the transistor twist cap is off, any additional force will cause lots of current to flow. This initial force, applied on to the gate terminal, is called "threshold voltage" in electrical parlance.

Once the threshold voltage has been applied, anything additional voltage beyond that will increase the current through the transistor, from the drain to source terminal, until a practical limit is reached. Thus, the transistor doesn't have to just be on or off. It can be somewhere in between. By varying the gate voltage applied to the transistor, the current through the device also varies correspondingly.

Instead of thinking about the transistor as a twist valve, let's re-do the valve design into something more like a push-button valve shown below. This is a more accurate analogy due to the addition of the spring. If the force is released on the push button valve, the spring will force the plunger to plug the opening preventing further water flow. Similary, lowering the voltage applied to the transistor below the threshold value will prevent electrical current from flowing through the transistor.

And here is where the digital model and the analog model of the transistor diverge. The digital model only concerns itself with the transistor being fully on or fully off -- simple and easy. The analog model concerns itself with the transistor being kind of on, kind of off, and anywhere in between -- more complex and difficut.

So voila, thus ends the basic lessons on the transistor. In the next episode on electrical devices, I will talk about the offspring from the marriage of a turned-on transistor and a reticent resistor. It electifying! *groan*