stating the obvious archives | about

The Next Usenet
Mar 02, 2001 :: Michael Sippey

Note -- this is a highly caveated essay in progress. I'm so new to RSS that I'm not even publishing one for theobvious, yet. But this came to me in a burst of energy, and wanted to get it out there for comment / contributions / revisions.

I've been reading with interest the ongoing discussion blasting the Googlification of Deja. While we were only occasional Deja users, we found the tool incredibly useful -- they managed to make finding the haystacked needle remarkably easy. And while we can appreciate that there was probably some well-reasoned economic argument on the part of Google for doing what they did, as an end user we think that they've blown it. Period. They've alienated the Deja customer base, and created a whole bunch of ill will. And frankly we're left wondering just what, exactly, they thought they were buying, if not the UI.

But we see the death of Deja as an opportunity to create the next usenet. Not a literal replacement for Usenet, but a platform that becomes the next place for distributed content creation, publishing, commentary and syndication. The death of Deja (which we predict (ed. note -- my God, Sippey, haven't you learned never to predict anything?) will accelerate the slide of the Usenet system into obscurity) could be the push the web publishing industry needs to morph from a single-site mentality into something approaching the semantic web.

The way I see it, we already have about half of the solution implemented. First, there seems to be an almost endless supply of content creators. Webloggers, personal home page publishers, professional journalists, community participants are creating millions of words of new content. Second, we have the protocol: simple HTTP, and addressable URLs. Third, we have (at least the beginnings of) a common data format in RSS.

We're missing the other half, which consists of two elements -- a common semantic space (i.e. "categories of content"), and the client- and server-side tools to easily create and distribute content. The tools will come, and will come in multiple colors and flavors. Whether it's Blogger, or Radio Userland, or Microsoft Word, or EMACS, or a server-side tool provided by Geocities, there will be tools for reading, writing, commenting on, and publishing, RSS-based content (ed. note -- there you go again).

But I think we could steal a lesson from Usenet in the categorization and distribution of content. The alt.* hierarchy of Usenet is a free-for-all -- if you have access to a news server, you can create a new group in the hierarchy fairly easily. And while it may seem that alt.* has spiraled out of control over the years, it's actually fairly tame all things considered. Two things have helped keep it in check -- the fact that it's a hierarchy (which makes it relatively easy to determine if a group for your topic already exists), and the distribution mechanism for Usenet.

Usenet's an early peer-to-peer architecture, and is one that the next could likely leverage. Today, each Usenet server selectively subscribes to the newsgroups it wants to receive from its neighbor. Imagine a series of webservers that exchange RSS feeds in a similar way. Since RSS 1.0 is extensible via XML namespaces, it would be easy to add one or more categorization elements to each and every item posted, in addition to a categorization for a whole channel. Additionally, RSS could be extended to describe types of publishers, so custom syndication servers could have their own rulesets to enable them to redistribute RSS feeds matching particular channels, created by particular publishers, or classes of publishers. Thus, theobvious could become a syndicator of RSS feeds of indie tech pundit types, while Google becomes the syndicator of record for everything it could get its hands on.

And because the original authors (and the tools they use) control what goes into the RSS, it's not as if every piece of content ever created gets distributed across the entire web. The feeds that flourish will be those that balance inclusion of content in the feed, vs. at the original source.

An additional set of elements could state a set of rules around syndication of that particular item, or micropayment rules for voluntary payment (see Matt Haughey's post on this topic), or even contain a key to enable authorized syndicators to decrypt and published encrypted items.

Usenet is dead. Long live usenet.



Other pieces about online publishing: