Archive for June, 2006

No Internet

Firstly, a little note to let you all know that my normal dependence on Internet access has been interrupted by a slight hiccup. On Thursday my computer downloaded about 32GB for no apparent reason. We’re only allowed 10GB in a month so I was summarily disconnected and until I come up with an explaination, it seems I will remain that way. A number of other cool things have happened which I want to update you all on but just to let you know if you’re trying to skype me or IM me or whatever, I have to go to town and use internet access at the coffee shops to get a connection.

Comments

A Break

Thank god, the hell of exams is over tomorrow at noon. What I’m doing with my break:

  • Haircut
  • Bar Work
  • Sleeping
  • Board games
  • Interview
  • Bumps
  • Kite Flying
  • Frisbee Throwing
  • Running
  • Sculling
  • Thinking of a Game for the June Event
  • Making a Google Calendar/Mail SyncML connector
  • Making SQUIGGLE a usable toolkit
  • Thinking about making a cheap motion cap system
  • Not drinking until Saturday night
  • Not going to Barcelona :-(
  • Yoga to get some flexibility back, maybe some external martial art as well
  • Spending at least two full days taking photos nonstop.

That should be enough for now. There are still outlying possibilities for a job in Cambridge or London, so I’m holding out hope. At this point if I do have one I probably will postpone going to the US until Septemberish.

Comments (1)

DumbIdeaCast

I subscribed to a new podcast from the Harvard Business Review called the “Ideacast.” In less than four shows they have demonstrated precisely why business literature runs between useless and stupid. I imagine the sort of jackass manager who reads most of the junk published must have it handed to him while he is standing on a window ledge about to jump from his executive suite. The better choice would be to end it all. I hate to make such generalizations but if HBS’s publishing arm puts out cud like this one imagines that literally any stupid idea makes the cut at some publisher somewhere.

Particularly apropos is the current podcast which contains two topics which are the subject of two books and two articles in HBR. The first is “Deliberate Mistakes” which suggests that following some random 5 part formula, business people should be encouraged to make mistakes. The formula boils down to: if the potential upside of the opportunity is greater than the opportunity cost, do it. Pardon me, but did anyone not standing on a window ledge need to be told the basic premise of entrepreneurship? Of profit? Of economics?

Of course, perhaps I should be so harsh: let’s look at some example case studies, Harvard style! First witness a missed opportunity: organic food retailer Whole Foods has picked up the majority of the organic food market because “big, dumb” supermarkets were too timid with stocking organics in their stores. Of course, had they stocked a small selection of organics, they would have been able to squelch Whole Foods before it had gained market share, and would have one less competitor.

Obviously the “big, dumb” chains made a misstep here, 20/20 hindsight says that they could have stopped Whole Foods (which is itself debatable). But is the prescription really to “make a mistake”? Could the stocking of organic foods be considered a mistake? Not really. Obviously there was a market or Whole Foods wouldn’t have existed. Obviously customers of supermarkets are also customers of niche chains like Whole Foods, given that supermarkets have larger customer bases. Stocking organics wouldn’t have been a total mistake, so why didn’t supermarkets do it sooner? Perhaps it is because the supermarket paradigm didn’t include organics at the time and had no reason to start doing so until a market was already established. Organics cannot afford to pay for shelf placement, organics would require space particularly in the crowded and low margin fresh vegetables aisle, organics turn over is neither seasonal nor matches the volume that supermarkets need to make a profit.

However, obviously when a niche market starts opening stores you rethink your strategy and start to analyze your new competitor. But it wasn’t so simple a decision. What was the advantage that Whole Foods had over the incumbents? That it wasn’t an incumbent! That it was specialized! That its model was different, it was serving the needs of a small demographic very well instead of the needs of a large demographic well enough. These mean that just stocking organics doesn’t equate to facing the competition. In fact facing the competition would have been better served by making your own Whole Foods type stores in existing areas and building dedicated sections into new stores and then reintegrating a successful side project back into the main stores (much like what was done with in supermarket video stores and florists in the 80’s, the height of supermarket subdividing).

A few more examples in brief: in the 70’s Ma Bell “tested” not requiring certain bad credit risks to put down a security deposit before they could obtain phone service. The test resulted not in hundreds of thousands in bad debt but thousands of new and responsible customers. A company who normally did not answer RFPs that were not directly in its current lines of business decided that it should instead answer all RFPs and build out infrastructure if required to satisfy and RFP. The result? More money and new lines of business!

These sorts of case studies are like hand-picked urban legends worthy of late night infomercials. “I quit my job and decided to start my own business. Now I make a million a year and only work when I want!” It can happen and clearly the opportunity cost (a few year’s salary) is less than the potential reward (a million per year) but clearly it is not a good idea to randomly quit jobs and start businesses.

Further just calling something just a “mistake” implies that you don’t really understand why it’s a mistake. For if it were simply a mistake it should be absolutely clear why it was the wrong thing to do, not just in hindsight but revisiting the situation without knowledge of how things played out. There must be something endemic about the behavior other than the loss of profit. Not every lost opportunity is a mistake. Otherwise the definition of success would be a completely random path through a minefield of potential disasters.

Suggesting that people make “deliberate mistakes” is suggesting that they deliberately not strategize and do things that seem wrong at the time in hopes they will turn out right. If I’m not mistaken that is a popular definition of insanity. Even the successful examples of mistakes are employing a strategy here but by simplifying it into “seems like the wrong idea, but they did it, and they won” or “seems like the wrong idea, so they didn’t do it, and they lost” you suggest a strategy tantamount to “git er dun”, while, as a researcher and academic yourself, you sidestep the question of what the real factors differentiating the winners and losers. Of course when you select case studies for those two outcomes, “did it and won” or “didn’t and lost” what other prescription can you expect than “just do it.”

That is really the problem with case studies. With 20/20 hindsight you can choose any story and make it a “case study” which turns into a sort of fable in which some arbitrary moral is asserted based on your induction, not deduction. Without taking a statistical approach and analyzing some real data you are never going to see the forest for the trees.

I shouldn’t go on but I will… the second topic is “middlescense”, which apparently is the behavior of employees midcareer. First of all there is the premise of a crisis when the Baby Boomers retire and there is no one to replace them. Talk about self groping. I would agree that boomers have entrenched themselves at the top and as they get closer to retirement they’re going to have to dig themselves out, but to suggest there is a crisis that won’t solve itself is to suggest that the boomers are asleep at the wheel. They’ll either train their replacements or their company will fail, it’s clear how lines of succession are drawn in companies and people move up. Logically today’s upper management are tomorrow’s executives. To suggest the problem is more the one of succession is to presume that boomers are somehow better or more capable than “kids these days”: the classic self-indulgence of old codgers.

Second is that it turns out that people are not solely motivated by more money and don’t always want to climb the corporate ladder! That’s news? At least it is true. So how do we solve this “middlesensce” problem? It sounds strangely similar to the old midlife crisis doesn’t it? Glad you asked! We give them more flexibility, let them transfer, and take on new assignments and… if they want it we demote them! How do you tell the difference between the guy who wants to be demoted and the guy who wants to new assignment and transfer? You LISTEN to them!!! So the solution to midcareer apathy is to LISTEN TO YOUR EMPLOYEES! Newsflash Harvard: the solution to nearly every employee morale problem is to listen to your employees and give them what the want. If they ask for a new assignment and you feel they’re capable, it makes no difference if they’re 20 or 50, give them the assignment. Or let them become apathetic or quit.

So unlike the previous topic which suggested that we abandon all logic and do whatever, this random Joe (or rather in this case Jo) wants us to do the obvious thing and pay attention to and satisfy our employees’ needs. My barber gives better advice then Harvard. In fact, perhaps he should publish a book, “Shave your way to Success” or maybe “A Cut Above” where he can outline well thought out theories on how driving a nicer car than you can afford not only attracts nicer women than you can afford, but it results in employee demoralization and lower profit.

Comments (1)

Rock Paper Scissors

It seems that Rock Paper Scissors is finally gaining the legal grounding it needs to become the premier dispute resolution tool available to attorneys. Of course, I decided not to become a lawyer. I would have been the freakin Johnnie Cochran for the new millenium. I can’t even imagine what the trial of the century would have looked like.

Baaaack to work.

Comments (1)

Asking for it

It appears that my life will be a living hell of cramming and rowing for the next two weeks. Tomorrow exams start. I’m as ready as I’ll be. That is to say not very ready. I think one of the most significant problems facing me this year is that none of my courses have any coherence at all. I am taking math, physics, and a menagerie of computer science subjects, like nearly every other computer science student. Of course you will say, in the US, you’d be taking several subjects, what’s the complaint you whiny bastard? The fact is that in the US, I would not be sitting a two three-hour exam with five topics for computer science, meaning that I need to study for breadth, not depth, and then follow that with a three hour physics exam and two (!) three hour math exams.

The computer science topics are java, ML, operating systems, and algorithms for the first exam and number theory, set theory, digital electronics, professional ethics, regular languages, computing perspectives. These topics are massively diverse, and not only that but I am basically expected to display my knowledge of the entire subject which spanned several weeks of lectures and supervisions work based on my answer to a single question which usually will take less than 30 minutes and be evaluated on a 20 point scale. And that is nearly all that counts. Rather I should say, that is all that differentiates me from anyone else, because every other assignment is completed by everyone and only checked for completeness.

So it feels like a total crap shoot. I do not like the idea of an exam based system and it is not likely that Cambridge will soon change to a continuously assessed system, just because the tripos (the name of the tests as well as the overall course itself) is so famous. It is, in fact, the very definition of educational assessment and perhaps one of the first organized tests administered to students.

I would not accuse Cambridge of being stuck in the past, much of the world is still in this exams based system, many European and Asian countries have national tests which are required for graduated and determine university placement. Many of them are longer, and cover more years of material. And while I don’t like it, I should be reasonably expected to know the things I learned at the beginning of the year if I have actually learned them.

What is perhaps broken about the Cambridge system is the format of the tripos itself. A single question asking for very specific detail on a very specific part of a very large course is ludicrous. It can seem as if the tripos is written up as a challenge from the brightest minds to the would-be claimants of the throne. “If you’re so smart, tell me the average amortised time cost of operations on a splay tree if we consider only splays and not comparisons.” What isn’t considered, of course, is that splay trees were perhaps a thirty minute consideration in a sixteen hour course, and the actual behavior of comparisons versus splays took less than one minute. Is this the lecturer’s fault? Should students be expected to hang on every word and supervisors be expected to psychically know what the important lecture points were?

Further the best testing aids we are given are actually old tripos questions. You might think that having a test bank is valuable until you attempt to do them. Because there is really only one question per subject and each question is on such a small part of the course, you can generally guarantee that you will not see the same question again, so while you can’t assume that you shouldn’t learn that material, it’s certainly not a good gauge of what you can expect on this test.

Beyond that, particularly in computer science (and far less so in static subjects like natural sciences) the courses change drastically every year and the history of a given topic only goes back three years or so reliable. Compound that with the fact that supervisors regularly assign these tripos questions as supervision work and you can generally say that the tripos questions aren’t worth much at all.

And unlike standardized tests, there is no review guide you can buy at Border’s which will provide you some sort of structure for your study. You are left out on your own to fare for yourself. You write revision notes, you read lecture slides, you copy questions, you read random books that are probably not helpful, whatever you can think to get you closer to the day of the exam. People at Wolfson do everything from utter denial to walking around the gardens reading index cards out loud (and the guy reading the index cards aloud is cooler than the people in denial by far).

Speaking of which, the college itself grinds to a halt socially. You never realize this is nerdsville until about May 1st. There hasn’t been anything exciting that I’ve been privy to since the Eurovision Song Contest, a ridiculous Europism that will have to wait until next year to get a blog entry. At the same time the town comes alive, tourists begin wobbling back and forth with a particular penchant for finding my front tire while I am attempting to use a bridge for it’s proper purpose: crossing the river, while they attempt to use it as a backdrop for a rather lame and typical photograph (”Here’s me at the top of another stupid bridge getting run over”). Fortunately few of them have found the rowing area.

Ok, just ranting now and it’s bed time. Today starts the worst days of my life. Immediately after exams finish (and You Shall Know) are the May Bumps. Though I haven’t directly mentioned it, yes I am rowing and hopefully we will do our college proud in the Mays as well. I have just gotten my one-piece unitard which admits me as a full-fledged member of the boaty club, photos are being withheld until I get ripped.

Comments (6)