A bit late to this panel, got caught talking outside in the lobby. Niall Kennedy is up on the stage, he is currently talking about wikipedia. He is now explaining web feeds as alternates. They are highly structured, discovery by Google, Yahoo, and MSN, and others. the user-facing in al modern browsers (he shows examples). He quickly goes through how to create a feed. The base vocabularies for RSS, include ATPM, IETF RFC or RSS 2.0. Extended vocabs; dubmin core meta data, comments, photo, audio, video, opensearch, creative commons, geo, item pricing, weather report, etc. You do not want to create a confusion, you need to use a structure and format that is standard. Popular parsing libs include python's universal feed parser, php's magpie, java's rome, perl's xml:rss and more. After you publish, you want to make sure to ping the engines. Always check your feeds for errors, validate it with feedvalidator.org. Also make sure to "claim your feed" in Technorati, in Google Sitemap, and in Yahoo! Site Explorer. Make sure to subscribe to your feed, populate the subscription search index, tag, add notes, view states and more. Watch out for "masked links" such as with feedburner, they now enable 302s. Watch out for false updates, such as Topix.net. Avoid reinvention, digg did this, and they made a digg:category name space. He then runs through some final things really quickly, nothing that crazy because most blog software apps handle this stuff automatically.
Rick Klau from FeedBurner. He gives the "who we are slide" you can learn about them at feedburner.com, they do rock, I use them. He shows a slop of feeds they manage, and you see the slope rocketing up. He explains that feeds are now not just about blog consumption but it is growing into podcasts, video blogs, print, online, news, retail, e-commerce, and web services. He explains the new IE7 and Firefox 2, then the social services like Digg, and Google Reader's shared posts, he shows TechMeme, FeedDemon and more. He shows of a new service named edgeio, a distributed classified service. He shows off sphere (I dislike that service, but maybe they changed since I reviewed it months ago). Full text vs. partial, the more content in the feed, the more these services can give back to you. Links matter. He then talks about "FeedFlare" which enables you to add functionality to your feeds, by leveraging 3rd part APIs. Examples include a digg this link, where it shows digged stories.
Owen Byrne the co-founder of Digg.com. The idea was user driven news. Went live on Dec. 1, 2004. They grew about 20% a month for the first few months. Paris Hilton hack on Feb. 2005 was a huge article that got digged and got them a ton of awareness. The server was crawling, load factor was at 100%, it took him 20 minutes just to login via SSH. Since then it took off. July 2005 they got some seed money and went from one server to three, they hired a designer, they added some cool features, added diggspy - that was version two. In June 2006, version 3.0 was released, they hired a DBA, finance people, got an office. It is a democratic process, no editors. Users are extremely vocal and motivated. They vote down spam, control home page, top stories are what are important to users. Passionate users - really passionate. They have over a 1/2M register users, 4,000+ stories submitted stories daily... They want to be a slashdot killer, and they sorta have done that, he said. There is a process for multiple developers, technical and management issues, avoid premature optimization, cache, cache and more cache. Hardware is cheap, but downtime is not. Lots of servers - spares, monitoring, testing and developing. They introduced a Digg API.
Chris Tolles from Topix.net. The team behind Topix created the ODP. He explains how it works, how many feeds they look at, etc. You can interact with the news on the Topix site by geographic location, http://www.topix.net/forum/geo. Topix provides fresh content for your site, local and topical content. Fresh content from your site, pull instead of push, rss feeds to increase distribution of content.