I returned from BarCamp in New York last night. Overall, it was more fun than educational, but that’s only because it was so fun.

Some geeks in Palo Alto started BarCamp last summer as a response to O’Reilly’s invite-only Foo Camp. Then came BarCamp in Amsterdam, and then BarCamp in New York. About 70 people showed up for 36 hours of geeking out in the 9th floor office of a t-shirt company. I only knew three of the people there, though as a bonus, Noah T. Winer (yes, THE Noah T. Winer) showed up on Saturday night unannounced. In general, it was very refreshing to be able to make ridiculously technical jokes and still get a few laughs.

Mike Goelzer and I presented our idea for making the entire web as unreliable as Wikipedia. It went over reasonably well.

There are a lot of pictures from the event on Flickr under the tag BarCamp or BarCampNYC. In most of the pictures that include me, I look like an idiot, but there are a couple where I look relatively normal.

Points that I want to make without taking the time to embed in standard prose:

1. It was fun to see stuff about us pop up on Flickr, people’s blogs, and other websites throughout the weekend.

2. It’s interesting to meet people whose blogs/websites I’ve read before (notably Chris Messina, Tara Hunt, and Jesse Chan-Norris).

3. New York was depressing. I’d been avoiding going there for 33 years, and it doesn’t appear that I was missing much. Not going to San Francisco– that would be missing something. I realize that experiencing New York from the 9th floor of an office building is ridiculous, but at least I can rule out the possibility that it’s Nirvana.

Many thanks to Amit Gupta, Mike Goelzer, Nick Gray and everyone else who helped organize the event. You’re all invited to BarCampBoston some day, unless I move to Kuujjuaq, Nunavik.

Stage 2 of my trip to New York City for BarCampNYC begins tomorrow morning around 7:30. Stage 1 was last night when I drove down to Boston from Wiscasset, Maine, where I work. I spent the day debugging a DNS problem and preparing for the Wikr presentation at BarCamp.

Wikr is a Firefox extension that Mike Goelzer and I have been working on. (Yes, it’s a stupid name, but at least it’s short.) The idea is to establish a means for synchronizing web improvements across trusted peer groups.

The “means,” in this case, is Mike’s Rails server and an RSS + SSE feed.

“Web improvements” is a contentious phrase– what improves the web for me doesn’t necessarily improve it for everyone else. The closest implementation I’ve seen of this idea is Greasemonkey scripts– scripts that allow the user to filter, augment, combine, and tweak web pages into something they prefer to the original. What Wikr is trying to do is to allow people to subscribe feeds of improvements from people or organizations that they trust. Myself, I’d love to have a feed of Google maps links embedded in any news story I read, so I can see where places mentioned are, like this imaginary weather report: “It was 10 degrees below zero in Kuujjuaq, Nunavik today.” Someone else might want a feed of spelling corrections, or a feed of bloggers’ posts about political articles they read.

“Trusted peer groups” means any group of people who trust each other and have the organization to get a feed together. Obviously, a lot of this depends on how easy it is to generate and host feeds. The Platypus extension for Firefox is a GUI for creating Greasemonkey scripts– in the end, Wikr might turn into a repackaged combination of Platypus and Greasemonkey with a little Javascript gluing it in between.

The authentication model is still half-baked. Each group will need to decide who is allowed read-only access to the feed of improvements and who is allowed bidirectional synchronization (the SSE in RSS + SSE). We haven’t developed anything beyond that principle yet.

As a demonstration for BarCampNYC, we have Mike’s server set up with an RSS + SSE feed. We also have an extension that synchronizes the pool of Greasemonkey scripts on Mike’s server to a local cache of scripts used by Firefox.

We’re also hoping to have a website and mailing list set up by the time BarCampNYC ends, so interested parties can follow our progress or join the fun.

Daniel Terdiman has posted an interesting interview with Daniel Brandt about his efforts to locate Brian Chase, the man who inserted false information into John Seigenthaler’s biography on Wikipedia. I thought that a remark Daniel Brandt made in the comments attached to the interview was interesting. Talking about tracking down the Chase, he said: ” Yes, I got lucky because there was a server on that address. I have never claimed that I was anything but lucky. Non-technical journalists frequently see the Internet infrastructure as some sort of black box, and ‘cyber-sleuth’ makes better copy than ‘I got lucky.’ That’s not my fault.”

That’s an example of a journalist spreading misinformation about him. Brandt has no recourse, and he doesn’t care. He doesn’t care because it’s a minor distortion, not libelous, but it’s strange that he offers that defense for himself but agrees with John Seigenthaler’s righteous fury. Couldn’t Seigenthaler also say “That’s not my fault”?

I think taking Wikipedia to task for permitting libel despite their best efforts is silly. Wikipedia is about as dangerous as me writing “Special Lecture at Harvard, 7 pm, December 22nd. Professor Henderson will discuss John Seigenthaler involvement in planning the Kennedy assassination,” on a piece of paper and posting it in Harvard Square. The levels of traffic and exposure are about the same. It’s true that Wikipedia in the aggregate gets lots of traffic, but not Seigenthaler’s page. If Seigenthaler’s page were a high traffic page, the error would get caught sooner. The police are not fact-checking flyers in Harvard Square, so why police Wikipedia? If the Harvard Square example is ridiculous, then pick any other forum or mailing list on the internet.

Since I was in 3rd grade, people have been telling me “Don’t believe everything you read!” and “Don’t believe everything you see on TV!” Those are educational cliches. When Nature looks at Wikipedia vs. Britannica statistically, Wikipedia does pretty well. So what’s the problem?

Note: Terdiman’s article bears the ZDNet tag “open source.” Why? Is Daniel Terdiman tagging his own articles about Wikipedia with the tag “open source,” even though he wrote an article last week arguing that the term “open source” was inappropriate for Wikipedia?

I’ve been thinking about the SSE extension and how it might be combined with Greasemonkey to fix a lot of shyste. Here’s an example: you’re reading a webpage, and you see a spelling error. You highlight the word, right click on some icon in the status bar, and select “spellcheck” or something like that. The point is that somehow, a Greasemonkey script records your correction, and whenever you visit that URL, it applies your correction.

Then, on a periodic basis, your browser contacts a website (THIS IS WHERE RAILS WOULD BE USED!) and updates an RSS + SSE feed of your corrections to the web (er, “the living web”). These feeds get aggregated, so as I’m browsing, I have a cache of Greasemonkey scripts shared and maintained in concert with people that I trust through RSS + SSE. The spelling example is minor– it could be expanded to be website commentary, adblocking (ooh! controversial!), or whatever. In fact, it has the potential to make the entire web as unreliable as Wikipedia!

I suspect that there could be scalability problems with this idea. There could also be stupidity problems.