Archive for October, 2004

Autodiscovery Hall of Shame

Friday, October 22nd, 2004

After visibly enduring coporeal pain while trying to autodiscover the feeds produced by all of the major blogging systems, Brad Neuberg just put together this Autodiscovery Hall of Shame (or in some cases accolades).

The basic idea of autodiscovery is to provide a well-defined way for services to locate a blog’s syndication feed (and other semantic resources associated with that site.)

Who really cares if a blog publishing system supports autodiscovery? Mostly, the users (and developers) of news aggregation software care. They use various newsreaders to find and subscribe to news and blog feeds. That way, it is possible to view updates from hundreds of favored sites all in one place as the stories are published. Soon, most blog readers won’t be actually visiting blog sites to read them, but rather aggregating the posts in syndicated form made possible with Atom or RSS.

For example, lets suppose you want to subscribe to my blog. You might know the base URI of the blog, and you can assume that somewhere on my site, there ought to be an RSS or Atom syndication of my posts. Your news aggregator (if it can autodiscover) can look in the metadata of any page on my site and find an HTML <link> tag that tells it exactly where my Atom feed is.

So, what happens if your blogging software doesn’t support autodiscovery? In that case, anyone who wants to subscribe to your feed will need to come up with the full URL for the appropriate syndication file. You could let them hunt or guess, but why leave technical hurdles between your message and your audience? If your blogging software doesn’t support autodiscovery of your feed (or worse yet, doesn’t create a feed at all) you will be missing out on all the readers who wanted to subscribe to your blog but couldn’t do so easily.

Hope is not lost for those who are using something listed in the hall of shame. If your blogging software didn’t set up autodiscovery in your feed for you, just do it yourself. First, find out the full URL of your atom or RSS feed. For my blog, it is: http://www.eakes.org/blog/atom.xml

Then all you need to do is add a line like this in the <head> section of your blogs template:

<link rel="alternate"
type="application/atom+xml"
title="Atom"
href="http://www.eakes.org/blog/atom.xml" />

After that, most readers can subscribe to your feed simply by providing the base URI for the blog, e.g.: www.eakes.org/blog/

Tribe supports FOAFnet via import and export FOAF

Monday, October 11th, 2004

The doll house pleasures of modeling one’s own social network were just enough to get me to set one up back when friendster was catching on. After that, I simply can’t be asked to set up the same doll house again with 18 different social network providers.
But now there is a new motivation to do it at least one more time. Tribe just became the first commercial social networking provider to implement the FOAFnet specification. This minimal specification is designed to allow portability of that carefully-crafted social network information we have been repeatedly feeding into each service.

How does the portability work? Supporting FOAFnet boils down to a social network service allowing the import and export of your relationships with other users as FOAF files. You can take your FOAF file to or from any other FOAFnet enabled service. By extracting your relationship information from a proprietary system, FOAF empowers users to manage their own social relationships.

Its still a bit of a chicken and egg problem, but at least now there is one chicken. You can try it in tribe. Build up a little network, then click on the export FOAF button near your picture. The XML shown is your FOAF file for tribe, which you can save with your browser. Marc Canter gives a good example with his own FOAF file.

What’s the difference between FOAF and FOAFnet? FOAF in its full form is bulky, still evolving, and has not yet been been adopted commercially. FOAFnet however, is a consortium of software vendors that culled out a subset of the much larger FOAF specification. That smaller frozen specification facilitates the immediate commercial uptake of social network portability. Fiona Romeo’s social network ideal is now much closer to reality.

Rojo announced at Web 2.0 conference

Friday, October 8th, 2004

Rojo, the project I have been intently working on for the past few months, is now released as an invite-only beta trial. You can request an invite. Chris Alden, CEO of Rojo Networks, just announced it at the Web 2.0 conference here in San Francisco.

I am excited to be part of the Rojo team working to make relevant information more accessible to the reader. In fact, Rojo is hiring–so if you’ve got the skills to pay the bills, make sure you contact Rojo.

Laszlo is Open Source

Wednesday, October 6th, 2004

I am excited to hear that Laszlo has been released under an open source license:

The Laszlo platform is free and licensed under the Common Public License, version 1.0. Commercial and non-commercial users are encouraged to use, extend and modify the Laszlo platform to suit their needs within the terms of the license.

Laszlo uses Flash as a rendering engine but does not use the Flash object model. Theoretically, some other completely free rendering engine could be used.

Before this there hasn’t really been an open source, rich internet application platform. Well, except for the ol’ DHTML work-horse that has been persistently trying to fill in some of the rich internet app vacuum.