Interviewer: Is studying computer science the best way to prepare to be a programmer?

Gates: No, the best way to prepare is to write programs, and to study great programs that other people have written. In my case, I went to the garbage cans at the Computer Science Center and I fished out listings of their operating systems.

You’ve got to be willing to read other people’s code, and then write your own, then have other people review your code. You’ve got to want to be in this incredible feedback loop where you get the world-class people to tell you what you’re doing wrong.

This wonderful snippet was revealed by Dan Bricklin and found in the old book Programmers at Work. Spread the news.

I’m fascinated by the universe of systems that capture a little value from a very large number of people and then aggregate that into something of high value. The list of such systems is huge, just for example:

• Open Source Projects.
• Amazon reviews.
• The original oxford english dictionary.
• The internet movie database.
• CMU’s clever picture categorization game.
• This trick for fooling a Turing test.
• Slashdot
• Blogger, Live Journal, Friendster, Orkut
• Netnews, Yahoo Groups
• Google
• the web
• this semi-porn site
• This toy.
• Society
• del.icio.us
• Everything 2
• Wikipedia
• Sheep Market

Here’s a fun little idea I had today along these lines. There is a general classification problem about dialogs. For example say you wanted a classifier for angry dialogs. How could you get such a thing?

Well you could create a Bayesian filter and let a large population of volunteers train it to do the classification. This is the pattern above of a large number of contributors summing up into an aggregated thing of value. In this case the thing of value is a classifier that can recognize a class of text or dialog.

For example here we have a guy who hand built a system to entrap child molesters in online forums by posing as a child. I suspect it would be easy to build a classifier that could monitor dialogs in such chat rooms that did the same thing, and it wouldn’t be hard to find the transcripts to train it with.

It would be very interesting to try this on the messages in a mailing list creating a set of classifiers for various kinds of speech acts: baiting, constructive, helpful, query, answer, discussion, debate, argument, etc. etc. I bet that the spooks have such systems but I wonder how they built them.

I keep trying to find a good name for this class of systems. Brain Farming? Talent Scraping? Help Hoarding?

I have to put up with a lot of talk about of how to create organizations that generate innovation. Smoke bellowing factories of innovation fed by train loads of recently clearcut ideas.

So I liked these two quotes that crossed my screen yesterday.

“Great ideas, it has been said, come into the world as gently as doves. Perhaps then, if we listen attentively, we shall hear amid the uproar of empires and nations, a faint flutter of wings, a gentle stirring of life and hope. Some will say that this hope lies in a nation; others in a man. I believe rather that it is awakened, revived, nourished by millions of solitary individuals whose deeds and works every day negate frontiers and the crudest implications of history. As a result, there shines forth fleetingly the ever-threatened truth that each and every man, on the foundation of his own sufferings and joys, builds for all.” ~ Albert Camus

and this one too, which though from an entirely different context strikes me as one of the things that makes open source work.

“No missed deadlines, only missed opportunities.”

Well, gotta run.

Here is an article that I find quite amusing about the free revealing question. Putting on their rational-man everything is a market thinking cap these researchers tackled the privacy problem. They ran a little experiment to see at what price points people would be willing to reveal

For example here’s a little table showing what percentage of their respondents demanded more than $100.00 to reveal the information mentioned:

The article has this great title: “Privacy and Deviance.” That pleasing title arises out their not too surprising discovery that if the respondent percieves the information he is being asked to reveal as being highy deviant compaired to the median he will demand a higher price; particularly if the value would imply a social negative. Overall it’s an fun bit of research. I just love that little be about revealing the spousal sallary.

The image on the right is a fragment from an ad that  O’Reilly  runs.

One of the reasons you want smart friends is that they can put a bee in your bonnet that  gets  stuck there for a long time.

Eric von Hippel has done that to me at least twice now. First asking how Open Source solves the coordination problem, and second asking why Open Source developers “freely reveal” (their innovations).

Of course what the bees do once they are in my head is entirely out of the control of the friends that stick them in there.

So I’ve begun making a list that attempts to accumulate why people do and don’t reveal. Why does the young woman reveal her belly button? Why does the coworker say “How’s it going?” It’s a very amusing topic.

Here are just a few reasons why people don’t reveal.

• Fear of  embarrassment, or reputation damage. “Familiarity  breeds  contempt.”
• Fear of gaining a reputation as having loose lips.
• Fear of seeming weak or desperate as you reveal and those around you don’t.
• Fear of seeming  naive  and child like exhibiting  unwarranted  trust in others.
• Fear of appear to be a gossip, and hence a member of a social clique of similar people.
• Fear that you’ll hurt others
• Fear that you might reveal what is a valuable secret to another in your one of your communities and hence be shunned by that community.
• Fear of loosing the trade secret advantages the knowledge enables.
• Confusion that ideas are like physical property – if you give away an apple you can’t eat it.
• Cost of translating the ideas into something your audience will understand.
• Cost of finding an audience, i.e. distribution costs
• Fear that revealing will create a “relationship” you have to maintain over time.

After a while it’s not clear why anybody would reveal anything.  This helps to explain people’s  bewilderment when they encounter free revealing on the net, for example blogging or open source.  The blogging case is particularly curious since the audience you reach is large so the chance of one of those risks kicking off is higher.

But, as my son pointed out, all these reasons have their inverse that can be framed as a positive.

Clay Shirky writes:

(I remember, a decade or so ago, someone asking on a newsgroup  “How do I post to all the soc. groups at once? soc.* doesn’t seem to work – surely I don’t have to enter all the group names in _by hand!”)

What a delightful example of something I’ve been thinking about recently about how inside of community membranes, where trust exists, queries over the community become more approprate.

The rest of Clay’s posting, about how open enables increased innovation while the capture of that innovation is an orthagonal issue is the tip of a big iceberg!

I learned in Cornes & Sandlers very academic text on public and club goods that there are two interesting edge cases in managing the boundary of a club that can be characterized as how the membrane filters it’s inputs – does it take the maximum or the minimum over the inputs?

Consider the example of trying to find the cure for cancer, a clear example of innovation. In that scenario we want a lot of people searching for the cure, but all we need is for one person to find that cure. In this case the club searching for the cure wants to be maximially open, it wants to filter out the maximal or best cure.

Now consider the example of the club that trying to maintain the leeve along a river. Each property owner along the river contributes to the effort; but the quality of the effort overall is decided by the minimum contribution. In this case (a safety example?) the club’s quality is defined by the minimum over the contributions.

That insight was one of the reasons I became very attracted to the club good cartoon of open source. Open source projects need both. Consider Apache’s HTTPD. It’s like the levee in that any flaw creates a security hole that allows hackers to pour in. It’s like a cure for cancer in that we never know where on the planet some web master might create the next delightful and useful innovation. So the club boundary around the server project has to manage to both maximize the innovative contributions and minimize the bogus ones.

ApacheCon is like being a field of locusts. You can almost hear the swarm of
chewing away at the future. There is no buisness model, no project plan, no marketing campaign. Just a hordes and hordes of dudes (though interestingly this year there were substantially more women than last) going – “Hum, I wonder what might happen if I hacked this thing together with that thing… cool, that makes me smile, that was neat.” over and over again. These are the people that take pens apart, that put light bulbs inside of other things just because they can.

Doc Searls gave a very nice talk at ApacheCon. Doc obviously has spent much of his life helping firms figure out how to tell their story so that it will go down the gullet of the media machine reasonably smoothly. Like everybody else he’s noticing that bottleneck like so many others is suffering a firestorm of disintermediation.

So on the one hand you could see him seeking a way to frame the story of what is going on here (what is the story of this open source thing) into the frameworks that his craft knows will make it an appetizing dish to place out on the buffet table where that the passing journalists will heap a helping onto their plate. One part hero. One problem, preferably dress in a war metaphore. A respoution – presumably open source. While on the hand noticing that that question is becoming increasingly uninteresting.

On the otherhand you could see he was in the midst of the puzzling out what would rise from those ashes. It was this puzzling that was delightful to observe because he is a very smart guy and he knows all the current memes with a degree of insight that’s rare. So it was just a joy to watch him reach into his bag of memes and pull one out and turn it over in his hands, say something wise about how it does or doesn’t fit what he was observing in the conference and then set it down, respectfully, and pull out another one.

For example he drew out the blogging meme. He had a wonderful slide where screen snapshots of dozens of blogs written by folks in his audience (a nice touch that) piled up on the screen. A pretty way to show the disintermediation of the traditional media. A compeling way to make clear that even the question “what is the story” is just possibly the wrong question when it’s clear that hundreds of voices are now all speaking at once telling what ever random story they might happen to want to tell.

I guess the blogs reflect that some of the locust’s are mumbling outloud as they chew.

He touched on the DYI IT meme. The end-to-end meme and it’s friend disintermediation and the many volunteers standing by offering to become the intermediary. He touched on the IP enclosure movement meme. etc. etc.

A very nice talk and in a way a very ironic one because Open Source has an undercurrent of shifting power from marketing to engineering, from the PR department alway back to the technologists in R&D. No. Not to R&D; further toward the technologists in the field.

So Doc’s presence, the PR marketing maven, in this crowd has a curious irony to it. It’s as if just as they were loading the Golgafrincham’s B-ship a few of the wisest and cleverest passengers paused and wandered back into the crowd. Thinking as they went: “My, this needs some more thought; something odd going on here.” I’m glad he came.