Then, add interlinks to trick Google's citation analysis model. Plaster those word-salad pages with ads, and voila - free cash flow!
Of course, we didn't do it. But even as we developed this idea, the room crackled with a kind of dark, excited dread. We weren't any smarter than many other rooms full of people who were engaged in exercises just like this one. The difference was, we *loved* the web. The idea of someone deliberately poisoning it this way churned our stomachs.
11/
That's where the evil part came in. We were pretty sure we could extract a list of the 100,000 most commonly searched terms from Google, and then we could use our web-crawler to capture the top 100 results for each. We could feed these to our Bayesian machine-learning tool to create statistical models of the semantic structure of these results, and then we could generate thousands of pages of word-salad for each of those keywords that matched those statistical models.
10/
In particular, we were really interested in a new, Linux-based search tool that used citation analysis - a close cousin to our own collaborative filter, harnessing latent clues about relevance implicit in the web's structure - to produce the best search results the web had ever seen. Like us, this company had no idea how to make money, so we were watching it very carefully. That company was called "Google."
9/
It's one of those evil ideas that I keep coming back to. Sometimes, during these money-making brainstorm sessions, we'd decompose the technology we were working on into its component parts to see if any subset of them might make money ("Be the first person to not do something no one has ever not done before" - B. Eno).
We had a (by contemporary standards, primitive) machine-learning system; we had a web crawler; and we had a keen sense of how the early web worked.
8/
(Not quite - some of the programmers who worked on the project have rebooted it, which is very cool!)
But back in the Opencola days, we three partners would have these regular meetings where we'd brainstorm ways that we could make money off of this extremely cool, but frankly very noncommercial idea. As with any good brainstorming session, there were "no bad ideas," so sometimes we would veer off into fanciful territory, or even very *evil* territory.
7/
As with Napster and its successors, you could also talk to the people whose collections enriched your own, allowing you to connect with people who shared even your most esoteric interests.
Opencola didn't make it. Our VCs got greedy when Microsoft offered to buy us and tried to grab all the equity away from the founders. I quit and went to EFF, and my partners got very good jobs at Microsoft, and the company was bought for its tax-credits by Opentext, and that was that.
6/
The way it was gonna work was that you'd have a folder on your desktop and you could put things in it that you liked and the system would crawl other users' folders, and the open web, and copy things into your folder that it found that seemed related to the stuff you liked. You could refine the system's sensibilities by thumbs-up/thumbs-downing the suggestions, and it would refine its conception of your preferences over time.
5/
Back in the late 1990s, I co-founded a dotcom company, Opencola. It was a "free/open, peer-to-peer search and recommendation system." The big idea was that we could combine early machine learning technology with Napster-style P2P file sharing and a web-crawler to help you find things that would interest you.
4/
Now that I am Old, I find myself dwelling on key junctures from my life. It's not nostalgia ("Nostalgia is a toxic impulse" - J. Hodgman) - rather, it's an attempt to figure out how I got here ("My god! What have I done?" - D. Byrne), and also, how the *world* got this way.
There's one incident I return to a lot, a moment that didn't feel momentous at the time, but which, on reflection, seems to have a lot to say about *this* moment - both for me, and for the world we live in.
3/
Instead of marveling that a book that felt to me like it was ten seconds old was a childhood favorite of this full-grown person, I was free to experience the intense gratification of knowing I'd helped this person find their way, and intense gratitude that they'd told me about it (including you, Sean - it was nice to meet you last night at Drawn and Quarterly in Montreal!).
2/