Williams or Downs?

Are you more like the kid with WMS, or the kid with DNS? The difference between these two forms of mental retardation is absolutely fascinating, and in my opinion represents two extremes of a dichotomy that is useful for categorizing all people — not just those classified as mentally retarded.

WMS people and DNS people have roughly the same IQ level, but their abilities are very different. The most obvious is that DNS kids are socially and verballyimpaired, while WMS kids give the appearance of social and linguistic savvy that belies their mental retardation. WMS kids improvise freely with language and music, using words and creating patterns that the normal high-functioning person would never think of. On the other hand, WMS suffer from fractional attentiveness and seem to always miss the “big picture” or overall pattern. WMS are locally (and fractionally)attentive but globally blind, while DNS are exact opposites — recognizing the overall pattern but missing important details that any normalperson would easily see.

So the question is, which is most like you? Are you the kind of person who instinctively, inexplicably can match an overall structure to its label when others are still struggling to name it? Or are you the kind who bounces along from cognitive connection to cognitive connection, never caring too much about whether the label or concept is exactly right, so long as itfits right?


Speaking of connections, I was listening to some new Frank Black music last night, and finally decided to research my theory that Frank has been secretly stealing lyrics from Leonard Cohen all these years (and not just the Pixies cover of “I Can’t Forget”). Google found a recent interview with Black where he compares himself to Cohen, but he gives away no secrets. The article is pretty interesting, though. Both Cohen and Black produce lyrics and music that emulate the best of WMS.

While reading, I also noticed that the article uses “Frank”, “Chet”, and “chum” in the same few paragraphs. Those three words immediatelytransport youto the world of “The Hardy Boys”. Do you think it was accidental that the author used all three words in the same article? Hardly! A quick Google search demonstrates that this is not the sort of juxtaposition anyone makes accidentally.

Blah Blah, XML

Anyone want to take bets on how soon “The Reg” starts using Word or InfoPathto publish articles to their blog?


Among other things, we tried Duke’s at Alki Beach for the first time this weekend. We would definitely go back, and recommend for good food at a good price. The chowder was great, steamed clams were OK, and etouffe was pretty good. They have all sorts of mixed drinks they call “Kazi” (as in Kamakazi) which are good beach drinks. And Duke’s doesn’t skimp on the alcohol content of their beach drinks like some places do.


This is life as a cog in a BigCo wheel:

1. This week I got a “ship-it” prize for helping to ship Windows Server 2003. I am not sure what I did to help ship Windows Server 2003, but I snatched it and ran before they had a chance to notice anything wrong.

2. Yesterday was “take your son or daughter to work day”. My daughter made shapes with my Mega-Magz and disrupted an important conference call.

3. Ciam mentions the Canadian Geese. For some reason I am reminded of my tenure at Ameritech (now SBC), and the yearly entertainment when the geese declared open season on office ladies and would chase the women across the parking lot.

4. All of the cafeterias and starbucks on campus allow us to use our smart-card embedded badges to pay for food. This week they are running a special where every dollar you deposit to your badge is matched by a 10% deposit from MSFT. So I deposited $120 this week, feeling clever about the 4 free lattes that I am getting.

5.The cafes on campus do the typical “buy 10, get one free” punch cards for lattes. I finally caved and started a card of my own yesterday.

6.Yesterday was the Windows Server2003 launch on campus. They decorated our cafeteria, and handed out free Krispy Kremes (a box per person).

New World

Michael Winser’s blog first introduced me to the perspective that we are experiencing “a disease that spreads like the common cold but kills 4 percent of its victims”.

This puts me in an interesting dilemma regarding a previously planned trip to Hunan this fall. Skipping or delaying the trip is highly undesirable, but since it’s a somewhat extended period, I’m estimating my chances of avoiding SARS while on the trip would be about 50%. Ihave to ask myself, isa two percent chance of dying worth it?

This is why the dilemma is so interesting. We gamble with our lives all the time. Air travel in the U.S. carries a 1 in 65 million chance of death. And life itself carries a 100% chance of dying eventually. But to put things in perspective, both of the recent military campaigns in Afghanistan and Iraq have so far had far less than a2% casualty rate for American soldiers. By this measure, I would be insane to travel to Hunan, even if it is for a once-in-a-lifetime event.

So, when we gamble with our lives, what are acceptable odds and payoffs? What activities would be worth a 2% chance of dying, and why? What happens to the decision-making process as the odds get worse? For example, if you had a 50% chance of dying, but an 85% chance of saving two family members from certain death, would you take the gamble?

The other interesting issue with SARS in particular is that we might not have a choice ultimately. Hope is diminishing quickly that we’ll be able to contain the spread of the disease, and many affected areas are still having trouble identifying or accurately reporting all cases. As far as I can tell, SARS is here to stay, and it will continue to spread. Will we be able to develop a vaccine or treatment before every person on earth experiences the disease? If we can develop a vaccine or treatment, will we be able to do so before you get infected? We are now in a race against time to do a number of things simultaneously:

  • Find a vaccine or treatment
  • Find ways to slow the infection rate of the disease
  • Find ways to reduce the mortality rate of the disease

All of these variables interact in a formula that directly relates to your chances of dying from the disease. If all 3 had zero percent success over the next five years, you would have a 4% chance of dying in the next five years. That is the baseline against which we hope to improve. What kinds of costs we are willing to pay to achieve improvements in these three variables will pose some more dilemmas.

Suppose hypotheticallythat we have only a 1 in10 chance of developing a vaccine or treatment within five years. These odds are still pretty crummy, so the emphasis shifts to slowing the disease and reducing its fatality rate. Some improvements we will likely get for free. For example,a better understanding of exactly how the disease is transmitted could result in significant slowing of the infection rate.

However, other improvements could come at very high cost. We would slow the spread of the disease within the U.S.to a crawl by banning all international flight and forced quarantining of anyone with a cold. This would have devastating impact on the economy, though. And if no cure were in sight, society would essentially be trading a significant reduction in civil liberties for a much lower mortality rate.

And once we’ve put SARS behind us, will this be the last or the worst to come?

I’m sure that medical ethicists have crisp mental frameworks for considering these tradeoffs with regards to epidemics, but the advent of SARS makesthings so much more personal. I would like to understand these issues better.

Radioactive Hiring

So I leavetown for a week with no Internet access, and when I come back Scoble has been assimilated! Scoble is hinting that another well-known blogger is going to announce soon. If it’s who I think it is,all I can say is we’re definitely getting an infusion of talent and energythese recentmonths.We search for people who are really passionate about technology and the softwareecosystem; the kind of people who would be working on the stuff one way or another even if they didn’t have the job.

I have a little theory about how this works: You can’t really buy passion-driven people. You have to pay for them, of course, but they have to choose you. Passion-driven people are attracted by many things, but one of the biggest attractions is the desire to work with other passion-drivenpeople who can challenge and appreciate them. As an organization grows, the challenge is to keep the bar high enough that critical mass can be maintained and new people will still continue to be attracted. It’s likerunning a nuclear reactor — if your fuel rods aren’t pure enough, you won’t be able to get/keep the reaction going; but if you enrich and purify thematerial, the reaction is self-perpetuating.

Theneutron bombtheory of hiring is not really appropriate for most hiring situations, but is fun for usethinking about MSFT. For example, what happens if someone uses a lead plate to separate the radioactive material? It’s also one of the reasons I so strongly reject the criticisms of those who claim that MSFT people are simplythralls to some intricate corporate machinations. MSFT isjust a bunch of radioactive people colliding with one another and causing explosions. Particles colliding randomly can generate tremendous energy.

Sure, there is some degree of coordination taking place within all of the chaos, but it’s a huge mistake to think that MSFT strength comes from intricately plotted orchestration of the employees. The managers aren’t sufficiently complicated and the employee’s aren’t sufficiently pliable. Instead, every part of the system is generating energy internally; and the outcome of any effort is as much the result of the magnitude (or lack of)of the eventual explosion as it is a result of any pre-planned strategy.

Vacation Report

Back from a week in SoCal. Five days at theme parks and only one Segway sighting. We visited family, ate lots of great Persian food, and even shopped at Fry’s for a card keyreader. Fry’s has everything except smart card readers, apparently. I didn’t believe the sales guy at first, but after sneaking some time on a vacant sales terminal, I was able to verify for myself. Although that was a disappointment, I would still shop there at least once a week if they had them in Seattle; the place is even cooler than Scoble says.


Duncan talks about the benefits of certification, in response to a developer’s skeptical comments. I don’t have a ton of sympathy for people who feel that they don’t need certification. Admittedly,aperson can be certified and yet be a terrible developer. And there are developers I know who could pass the tests in their sleep, and who don’t need the paper to prove their skills. But do the tests have value?

I always wonder about the people who downplay “paper MCSEs” and claim that the knowledge covered in the tests is trivial. Why do you think they don’t ever take the tests? Last I knew, a test was $100 and a few hours of time — could it be that the expense is too much? I don’t think so.

I used to be one of those nabobs. Being the typical developer, I had a high opinion of my own abilities, and a confidence that people who leaned on the crutch of certification could never approach my own considerable prowess. However, being a man of some intellectual honesty, I took it upon myself to prove the triviality of these tests. I eventually got MCSE+I, MCSD, MCDBA, MCT and many other certifications. Here is what I learned about certification:

  • Some tests can be passed by nothing more than rote memorization and cramming. But most require real ability in addition to the study. If you are the kind of developer I was, you wouldn’t necessarily think more highly of a co-worker who passed the tests, but you sure as heck would think differently of a co-worker who couldn’t pass after a few tries.
  • Passing the tests forces you to round out your strengths. You might know everything there is to know about VLDB design and tuning, but absolutely nothing about merge replication. If you are so smart about VLDB, you’ll have no problem picking up what you need to know about merge replication, and you’ll never regret having shored up the weak spots in your expertise.
  • Even the parts of the tests that succumb to memorization and cramming are often useful — genuinely useful — at unexpected points in the future.You mayalways be able to do a good job and awe your peerswithout these random extraneous facts stuffed in your head, but it sure is nice when one of those worthless tidbits comes to the fore and points you to a more efficient route to getting a desired result. A couple of hours of efficiency every now and then really adds up.
  • More is better. The more certs you get, the more value you extract from the certs you already have. This isn’tjustcertifications — in general, all learning is synergistic — but certs force you to learn with some breadth and consistency, and this promotes synergy.
  • Different is better. Learning things outside your area of expertise is good for the brain. Drive a few yards in someone else’s rut. Again, this is about learning in general, and not just certs. But a cert demonstrates a certain level of competence; especially when granted to someone who does not consider the area to be their domain of expertise. If I were king of all techies, I would require all devs to get MCSE first and all sys admins to get MCAD first. Achievement in the non-core areas would help them break through to new levels in their core areas, without fail.

P.S. Alert readers will have notice that I have MCSD and not MCAD. I keep telling myself thatI don’t need to upgrade to MCAD since I know the material so well already. Intellectual honesty is always under attack.

Google News Ethics

News.com is finally picking up on the story about Google publishing press releases. Now, it’s true that you can’t believe everything you read in a press release, but everyone knows that. And press releases are no less believable than half of the stuff these self-righteous journalists regularly write. Rather than insult our intelligence with this false dichotomy and its implications that there is a class of “respectable journalists” who are somehow defacto trustworthy, I would prefer to see journalists actually behaving in a trustworthy manner. Gaining credibility takes more than just some stupid article about “you can’t trust press releases, so you need to trust me”.

Billy Goats Gruff and Babel

[Singapore Straits Times] According to well-placed sources in Beijing, Chinese leaders do not know whether to laugh or cry at the outside world’s belief that they have leverage over Pyongyang.

This is really clever; point the finger at Russia on the same day that China shoots down any U.N. Security Council action against North Korea. Mr. Fong should be ashamed.


Now that we have troops in the city of Babel, and a shining new age of enlightenment en route to the land of Shinar, how long until things are stable enough for us to hold the digital memory conferences there? What a beatiful location this would be for the new library of the human genome. As we continue to chase the dream of a new world’sglobal brotherhood and human liberty, what better tribute could we have — what better affirmation of hope in the face of swirling chaos than the shining new “library of human knowledge”? A Library of Congress for all of the world’s people. Make it tall.

Slave of the Corporation

Awww yeah! When 14 year-old Emily can play protest ruckus on XBox while her 17 year-old sister goes out to meet boys and fight globalization, you know the protest scene is going mainstream. The game trailer is great; they’ve got a Che impersonator doing voice and the rhetoric is strangely similar to the rhetoric being broadcast by coalition forces to the Iraqi people right now: “You are all slaves; rise up and fight for your freedom!”

State of Emergency: A shadowy corporation has seized control of your country. The people have taken to the streets in protest. Fight for your freedom.