Related directly to my last post is the current state of the tagosphere. The relative stasis in this space is very frustrating to me, because I believe the user base will continue to grow, but without movement towards a way to access this information consistently and somewhat predictably, most of the valuable network effects remain pent up and isolated, of little use for advancing socially influenced search, or other crossover applications from person to person, app to app, and group to group.
Thomas Vander Wal makes good points, including one I make about portability and interoperability of tags. He notes other discussion that is worth reading from David Weinberger and Matt Mower. Mower, in a must-read follow up, reaches way back in time to Northern Voice 2005 where Stephen Downes took a stand against tagging (I’m not being dismissive– Downes discusses many other interesting things, of which tagging is but one part, and I recommend checking it out). I suspect Stephen is right that tagging creates its own power laws and its own derivation of meaning, but I don’t think that negates their powerful effect and usefulness, it merely limits them. No reason to discard baby and bath water, and Downes does not since he uses tags– correctly with the REL attribute– in OLDaily, among other places.
The conversations seem to more about the lack of this thing you perhaps label “tagosphere” - e.g the tagging capability that extends across multiple tools. I’d agree and even appreciate the observations originally made that on a technical front, tagging has not made big great leaps. Tagging via technorati is a crap-shot at best.
I’m less interested in tags across all the tools, and see little problem or dismal-ness in that tags perform well and provide valuable service at the specific tool level.
I don’t see it as a binary. Tags do perform well and are valuable at the level of specific tools– but I would also love to see the ability for that information to cross platforms, particularly for powering social search, which I think would be a fantastic advance for all of us. Localized tags is tasty, being able to speak and connect across apps would be that much tastier.
Even in the localized space, advances have been dismally slow. With all the information that they have in some of the systems… only flickr and library thing have, to my mind, really started doing some interesting things with it.
Chris,
You know I’ve been of two minds about tagging. But of late I am leaning back toward the pro-tag side—assuming there is a way to implement shared tags, author tags and user tags. They could all be idiosyncratic, and a tag cloud seems a nice way to show which tags are highly valued. I don’t know how close we are to having such an app.
Part of what got me back on the wagon is recent participation in a bbs and the notion of being on topic or “drifting.” I also got to thinking about cross-posting in that environment. Communication by blog, such as this, is much less silo-like in structure. What I got to imagining was a system wherein a post would go up once, but could be infinitely tagged. If I choose to read the “folksonomy” tag it would show up in that context, if I choose to read the “comments by chris” tag it would show up in that context. Maybe this is all obvious at your level of the game. But it would be cool, I think, to develop some kind of reader that could not only give me your site, but give me your tags and allow me to paste my own. The third item, which I suppose technorati offers to some extent, would be a distributed tag db of shared user tags.
Ill formed and rambling, maybe I should have had my morning coffee first. Sorry if it’s all old ground, but if it bears review I’d love to follow along.
Peace,
Robert
I don’t really see a good argument *against* tagging. It seems to work, is convenient, and people are generally willing to do it. But the argument for tagging remains largely based on the localized effect within a particular site. I would like to see those tags– in all their idiosyncratic glory– brought together. Your idea is an interesting one, yet another that could exist if someone were willing to create it.
But that’s really the direction of my post– these future systems are by no means a given (which may well be one of the more negative aspects of the “good enough” approach that makes tagging itself so incredibly useful!).
Leave a Reply