Into the Wild Spaces

When I first bought a laptop and started getting on line, I gravitated almost immediately to LinuxToday, with a regular stream of news and commentary on a subject that had captivated me. I learned tons from the articles and the comments below them. But several revamps led to more advertising and fewer interesting articles, and the comments trickled off almost immediately: I'm not sure what the impact was on revenue, but it killed the community. I spent the next few years at Slashdot, and for the same reason: over time I got to recognize who the intelligent posters were, and the community was a decent one, with a lot of technical ability debating on a site whose interface allowed for good comments to stay more visible ("float to the top"). I continued to learn a lot: there's nothing better than someone recommending a tool or an approach to a problem you've had yourself. Then, in February of this year, Slashdot's corporate overlords decided to redo the interface, destroying most of what made the site good for discussion. The users complained about the reduced functionality beta interface, the Corp told them/us to stuff it and made clear Slashdot's future was the new beta interface, and a hell of a lot of us decided to ditch Slashdot entirely for greener pastures. We were the "Slashcotters," heading up a movement called, affectionately, "Fuck Beta." A sizeable number of us SoylentNews. And another guy began building his own site, with similar functionality but a modern codebase of PHP and CSS: Pipedot.org. The community was fractured, but new communities grew.

Immediately, differences emerged: the Usenet crowd tended to be older - members of the original Usenet generation, probably, or at least old enough to remember and appreciate that culture. SoylentNews had a far larger number of members but no clear vision of how they'd organize themselves. And Pipedot had fewer members but the relative convenience of autocracy: that is, since the code was the work of one man, he had full power to shape the site as he wished, and he quickly and efficiently made the site gorgeous and functional, surpassing the original feature set of Slashdot in just a few weeks by adding things like Twitter integration, Borda and Approval polls, and integrated thumbnail pictures instead of representative story icons. But in the meantime, itchy Slashcotters flocked first to SoylentNews where they'd have more say and felt more empowered, despite the uglier and clunkier interface and the relative emotional churn of changes in leadership, competing and diverging visions for the site's future.

I wound up in all those places at once. I like and participate in SoylentNews, but became a volunteer editor and a very primary contributer to Pipedot, which I like quite a bit more (the fact that it works so well on a small screen might be part of the appeal). And I was one of the more frequent contributors on comp.misc too, as our communities grew. We broke communities, and we made communities. What have we learned?

Well, first, that you need community. SoylentNews is the more fun place to be, despite the lousy software and the clunky interface: there are more people expressing their viewpoint, and there's more to learn (although frankly, not much. Seems a lot of smart people stayed behind at Slashdot. That makes Slashdot, despite the abominable beta interface, still the best place to read and learn). OSNews, where I was also once a contributor until I got tired of my articles being rejected and my suggestions being ignored, is increasingly the voice of one single editor rather than a publication site for the articles of many. It's still working but I sense it's in decline, and that's too bad because it is or was or might still become a great site for reading about news about operating systems.

But we've also learned that software and interfaces can affect community. LinuxToday banished its community when it chose advertisement over threaded comments, turning the site into no more than an RSS feed of news stories, each bearing one single blurb of text. Slashdot has probably done the same, although the damage will take some more time to fully be felt, like a slowly metastasizing necrosis of the liver.

Thirdly, I think it's clear that the Dictator's Handbook applies to online communities too: sometimes design or decision by committee is unwieldy and ultimately self-defeating. Pipedot grew from an idea to a full-fledged site in the hands of a single, talented developer with a vision, while the fully democratic SoylentNews almost immediately suffered acrimony and dissent over competing visions in the hands of a changing and self-appointed management team that struggled to incorporate the viewpoints of its users and the important balance between democracy and technocracy.

Lastly, and this point is perhaps the subtext for the entire conversation: the original weakness of the WWW model remains concentration of expenses in the site owner. On Usenet, costs are distributed and as information flows among news service providers, users simply choose the NSP they want; and some of them are free. But to run a website takes time and money, and it's not distributed. Site owners have to pay for their domain name, and server and bandwidth costs. That means recouping costs either through advertising or membership fees or donations or some other way. Sites that can't generate that revenue, or whose owners decide to stop fronting the costs, disappear. Then we have to go form a community somewhere else.