...more recent posts
Right now you are part of the internet. Your computer is connected to my computer, and by extension to every other computer on the internet. But aside from the ability to click buttons and fill in various HTML form elements, your browser software basically only receives information. This is the web as we know it. But soon, or so they keep saying, we will move to the next phase of the internet. On this new frontier your computer will be both a receiver and a broadcaster (a client and a server.) This movement is already pressaged by the likes of napster, gnutella, SETI@home, and mojonation, etc... These are very different services, operating in different manners, and employing different protocols. But all have been lumped together (sometimes without too much sense) under the title P2P. That means Peer to Peer, and it stands in contrast to the more traditional server to client model. Or more generally P2P means a flat hierarchy with content information flowing bi-directionally (communication), as opposed to a hierarchy, with most content information flowing in one direction (consumption). A telephone call is like a P2P application; watching a television program is like a traditional client-server model. You can talk back to your T.V., but it's not listening. Guess which model Big Business wants?
Anyway, the web is seriously abuzz about Sun's new framework for designing P2P applications, JXTA. Wes Felter has the quick first look technical overview: So what is JXTA? Dave Winer has a bunch of links including a package of O'Reilly pages (this one being the overview,) the register's rather negative take, and Sun's own press release.
JXTA is at least one whole level more fundamental than anything I ever deal with, but I understand what it is trying to provide. I'll be very happy to employ some of the things that others might be able to make with JXTA. Or with some other set of P2P building tools. A lot of people are in sync on this issue, now it's just a matter of getting on with the slightly more political fight, as the big gorillas (Microsoft, Sun, etc.) fight it out with each other (.net vs. JXTA) and with those weirdo (;-) independent developers who by providing fun applications play a crucial role in a particular framework's acceptance, but who keep insisting that these fundamental layers of the web should
Right on. Douglas Rushkoff just solved (at least temporarily) one of the things that bugs me most about the web: The New York Times. I don't necessarily trust them, but I do like this paper. Most days I read the front section and the business section at my coffee shop. But they make it difficult for others to link to them, thus circumventing all the best results of putting their articles on the web in the first place. Anyway, Rushkoff points out in this article that the WAP version of the New York Times (that's Wireless Access Protocol, the standard devised for fitting the web onto the small display screens of cell phones and wireless PDAs) can be accessed by any browser at avantgo. You can get the Times here, and unlike some other back doors around the Times mandatory sign in, I'm not sure they'll be able to close this one. Or, at least, here's hoping.
15,000 scientists can't be wrong. Is there a scientific journal boycott brewing? (/. story)
Apple acquires Focal Point Systems, creators of Film Logic. Finally. Final Cut Pro can now seriously compete with high end video editing solutions.
And canon announces a 3d lens for the XL1 (although the Canon site says this is a proof of concept and no price or ship dates have been set, macnn says it will go on sale in October for $8500.)
David McCusker takes a break from talking about stuff I can't understand, and introduces us to his mysticism.
Dan Gillmor writes some more about weblogs (second item.)
Astronomy picture of the day. Beautiful shot of the Jovian moon IO against a background of Jupiter itself.
Peep. Ping. Coming up for air. I think I've got the translation script working. This script walks the tree, starting at a given page of the old system, and translates all the posts into the new system. The hard part is that each post is itself the top node of a smaller tree which represents the downward cascading comments which may or may not be under each post. And these comments cascade down to an arbitrary and theoretically infinite depth. So that means lots of recursive functions (a loop of code that invokes another iteration of itself from inside the execution of that same loop.) This sort of thing is hard for me to grasp. And it can get out of control. (Remember way back in the day when I basically crashed that csoft server with some faulty perl code? That was a recursive function that never came out of its infundibular self invoking spiral.) There still might be some subtle bug in it, but I can't find one. This is the hardest thing I've ever written. Very interesting process. I wonder if people who are really trained work this same way. For me it's like having a conversation with the computer. Eventually I'm going to write something longer about this interesting phenomenon. I guess it's a dialectic. In any case, it lets me make things beyond myself. I don't know exactly how this script works, but I'm pretty sure it does. I figured it out, like I said, by having a conversation with the machine. Both of us (the machine and I, as it were) understand the script, but neither of us alone can understand the whole thing. Or something like that. Cool stuff.
So if I'm right, and it is working, it won't be too long now. First of May seems possible. Or, hey, the third is a Thrusday. That's a nice date.