Infrequently Noted

Alex Russell on browsers, standards, and the process of progress.

Huh.

Note to self: -khtml-box-flex plus XUL <box> box elements plus CSS expressions in IE equals a cross-browser XUL box implementation via netWindows.

missing the barbecue

Back from Texas. Yesterday was more of the same exciting and engaging stuff. The plane ride back was full of thoughts about onion-style privacy networks that use standard socket interfaces and the implications of not understanding trust in building social networks.

I'm sure more will percolate out of my brain in the next couple of days before I forget all the interesting people I met and the neat conversations I had.

yesterday

So yesterday was good. Jeff Veen from Adaptive Path gave an excellent presentation on user interaction design and the process he uses to address the expectation gulf. Other interesting panels included Howard Rheingold's fairly realistic talk on the future of swarms and smart networks. He described an information space with several properties that I have been noodling on for some time now. His point was that the current network is starting to see some real command-and-control structures being built in and that it's up to us (the geeks) to expand the information space beyond central control structures at a rate that makes control structures irrelevant. OTOH, this all depends on the viability of an open platform, which is now in doubt.

After a lunch of barbecue (mmm! barbecue!) I had lunch and a good discussion with Adam Keys. Also wound up meeting Wes Felter and had a good time hanging out with Simon Willison and company.

Anyway, yesterday was great. I'm already not wanting to leave today.

Lunch

So I'm having lunch on 6th street at a place named "Jazz" which plays nothing but the blues. Not that I mind it. I've got a shiner (draught!) on the table and an open AP in close proximity. It's really hard to beat this.

Last night I got a chance to see my old roomate from school who is working at AMD these days. After stuffing ourselves on amazing barbecue (SF might have food, but it doesn't have this), we spent some of the evening on 6th street, a couple of blocks up from the hotel I'm staying at for the conference. Like a lot of things here, it hasn't changed much from when I lived here, and that's generally a good thing.

I also got a chance to stop Sam Ruby in the hall of the hotel and pitch my idea of signing Atom posts. He seemed generally receptive, even giving that I'm something of a nobody. He raised several good points in relation to it (what about metadata?, lots of interesting attacks there). I'm going to have to work up a set of proposals before I do what he suggested and post to the Atom syntax list. I'm sure I'll have more on that later.

I think my blackened shrimp is almost here, and I've got a shiner that needs tending. More later.

Sat keynote/opening

So I'm waiting for the beginning of today's opening keynote by Brenda Laurel. So far I've gotten into an argument with cellular industry reps and government regulators, and had lunch with a wireless industry columnist. It's been a good start to the conference.

Basically, I asked the question "so why are you colluding to keep your development and app deployment platforms closed? Can't you just make money transiting data?" Seems the cell industry guys have this funny notion that if they "open up" their networks to multi-billion dollar companies to provide another outlet for recycled "content", then they've effected a sea-change in the way business is done. What I think really got my goat was all of the bogus discussion about how competitive the market for cellular services is, yet when pressured about what minor steps the carriers can take right now to open up new markets on existing networks, there's an automatic "oh, that's 5 years away" response. I say that's bogus (and I think I offended them by saying so somewhat frankly). The tools are there NOW. MIDP 2.0 is being deployed ubiquitously, most new devices have some sort of HTTP-capable client, IP stacks, and many have processors fast enough to do crypto and/or media of some form. The ONLY thing missing from this equation that would allow independent (read: open source) app developers to start exploiting this platform is a willingness on the part of the carriers to get the hell out of the way of the impending network effects.

Given their reactions to my points today, I don't think I'll be writing very many J2ME apps for distribution any time soon, which is a shame, because as the carriers keep trying to point out, these phones can do some really neat stuff. They just want to be the censors, not just the toll takers.