Archive of UserLand's first discussion group, started October 5, 1998.

Re: Sample code and utilities?

Author:Dave Winer
Posted:3/29/1999; 6:07:00 AM
Topic:Sample code and utilities?
Msg #:4636 (In response to 4634)
Prev/Next:4635 / 4637

Having framed the discussion, let me be the first to respond.

I think we have what we need in the Discussion Group software and the Search Engine that are both integral parts of Frontier 6. Add to that, the Subscription feature in Frontier 6, it should be very easy to set up a mirrored arrangement, so that no individual controls the stuff that's available to Frontier users. This isn't an issue for me, but it may be an issue for others, it's been a long time since we've had a power struggle, I trust people to distribute code as it was submitted, and have no problems if someone else operates the server. I can't think of a single person in the Frontier community that I wouldn't trust if they volunteered to be the archivist.

But we've seen that knowledgable Frontier people have day jobs, and are getting more busy, and have less time to invest. When someone wants to contribute an elegant bit of code I think it should be as easy as possible. As one of those busy people I want it to be very easy and automatic on the authorship side. I want most of my energy to go into making the code really good and having good docs. I want the server to take care of letting everyone know it's here, and help people find it six months from now when I've forgotten that I wrote it and so has everyone else (except Seth).

Anyway, with the half-done editorial suite, it's relatively easy to get script bits attached to a DG message. With a few hours of work I could get it to be braindead simple. I think it's worth doing this work. The solution would require XML-RPC and would work not just for Frontier, but for any coding environment that supports XML-RPC. But that's a diversion, my interest is in getting the act together for this community of developers.

The DG is almost ideal because it's indexed by the search engine. However, we could also produce an XMLization of the contents of the samples site, so that the contents of the samples site could be mirrored, again, since it would be XML-RPC-based, there is no limit based on platform, the same content could be stored in PHP on Linux or FileMaker on the Mac or Access on Windows, you get the idea. Or we could probably figure out a connection to Perl's distribution system and Python's.

The key point is that the connection between discuss.userland.com and samples.userland.com is very thin. It's just an attribute that makes something appear in the samples site. And from there doing an XMLization is just one script, probably located here:

http://samples.userland.com/samples.xml

Or something like that. As always the technology isn't the problem, it's the will to solve the problem.

There are responses to this message:

This page was archived on 6/13/2001; 4:49:07 PM.

© Copyright 1998-2001 UserLand Software, Inc.