Weblog

Jun 22, 03:57 pm: Remixing the web

I’ve argued before that the web is at 3.0. That 1.0 is simply getting GUI. And that 2.0 is the relative ease with which we can put a variety of content on the web. This first was true of text (and that is, in large part, the real revolution of blogging) and photos. It’s becoming increasingly true of other kinds of media—video and audio files.

Web 3.0, I’ll be happy to argue, is about the ability of users to remix all of that information in ways that suit them. When that remixing happens easily—as easily as I’m publishing this entry—we’ll be at web 3.0.

I believe remixing is already happening with tagging and syndication. The easiest example of this is, I think, del.icio.us. Ed Batista wrote about this in his entry [Ed: I can’t find your entry. Help!]. Del.icio.us, with a simple if uninteresting interface, allows users to create a stream of information from across websites and powered by people they’ll never meet. Other sites use tagging, but, for my money, no other site better illustrates the ability to remix on the fly than del.icio.us.

Why is this so important? It unhinges the web. It no longer makes the website the central repository and instead makes containers for ideas—tags basically—the central repository for the web. It is going to, I believe, remake the way we find and store information on the web.

When I’m searching for specific content on the web—the hours for my library, the year in which George Washington was born, an overview of CompuMentor—Yahoo!, Google and Wikipedia can be my prime sources. But when I want to find out ideas—and their relation to people and a variety of websites—well, then maybe I’m after discovery and aggregation rather than search.

Maybe I’m looking for a way to add use tagging as a stand in for authority. Maybe I want to know more about the users who are tagging something so I get some way to rank their authority/interest/inquisitiveness on a given topic. I really don’t know.

I do know that I’ll want to be able to republish—in the stream/syndicate fashion that RSS allows for—to either myself or a group. I know that I’ll want to see duplicates (that gives me an idea of value) but I’ll want to know how many times it was duplicated, and I know that I’ll want suggestions for related tag words.

This remixing started on the ubiquitious links pages—some of them taking you in a circle and never adding anything new—that were on the early web, the blogrolls that are on blogs, and even blogs themselves. It’s a way of saying—here, look at this, I think you’ll find it interesting. The significant difference is, with the addition of user created metadata, the ability to move across sites, to add and substract pieces of information, to search, display, share and stream based on the metadata makes it all, well, easier and tremendously more exciting.

Where is this remixing going? Greasemonkey. It doesn’t just allow for the remixing of information but the remixing of functionality. What happens when that gets easy?

Bonus link: Practice or why I keep writing about the same things over and over

Marnie Webb  | 

Comments made

  1. Ed Batista

    Hey Marnie,
    I’ve talked about Greasemonkey (http://www.edbatista.com/2005/06/do_it_yourself.html) and linked to Beth Kanter’s del.icio.us documentation (http://www.edbatista.com/2005/06/delicious_docum.html) lately, but I’m not sure if those are the posts you’re thinking of.

    Great post—very thought-provoking. Definitely inspired me to think more seriously about social bookmarking and see if I can get as much out of it as I do out of tagging.

    Even if my answer to that particular question’s different than yours, I agree with the larger ideas that 1) tagging is changing the nature of discoverability on the web (not to mention changing our perspective on all top-down information classification schemes), and 2) we need a trustworthiness or relevance algorithm, a la PageRank, to make tagging more meaningful.
    Jun 22, 06:47 pm

  2. marnie webb

    Thanks, Ed. It’s neither of those posts. I’ll see if I can find it.

    I think that tagging can help to provide the kind of trustworthiness you’re talking about. One of the tricks, as I see it, is figuring out how to do that without creating power law problems.
    Jun 23, 05:31 pm

  3. Ed Batista

    Could you explain “power law problems” further? I would assume that more taggers and more tags = better analysis = better results. An essentially self-policing intellectual marketplace, like Wikipedia, or the blogosphere in general, for that matter (political differences notwithstanding). I’d also assume that idiosyncratic tags will die out, and that we’ll reach a balance between intelligible stasis (i.e. common meanings we all understand) and receptivity to innovation (i.e. if a really good new tag comes along, it’ll reach a tipping point and be adopted.)

    But perhaps that’s not what you’re talking about at all.
    Jun 23, 07:23 pm

  4. marnie webb

    Ed,

    First a couple of links:

    * Weblogs and power laws
    * Power laws and priorities

    I agree that tagging behavior can add a quality control element to tags. Certainly, it’s a strong recommendations. Right now, I don’t know who’s del.icio.us account is most popular and the owners can be obscured which means my choice to subscribe to someone’s inbox has to do with my view into their selections—and little else.

    Over time though, as reputation systems and measures become a part of the inbox, that will change. What if there is a way for me to see the most popular del.icio.us posters (as measured by the inbox subscriptions of other users)? That’ll be helpful to me and, if their interests match mine, I’ll subscribe. The result? New and other posters to del.icio.us have a harder time getting attention because they aren’t already popular.

    That’s not necessarily bad. I do worry about what it means to diversity. Diversity of opinion, background etc. Early adopters probably share characteristics.

    It’s the oppposite, I think, of the long tail.
    Jun 24, 12:50 pm

  5. Ed Batista

    Sorry, I wasn’t clear. I understand power laws and their inevitable impact on (or, more accurately, reflection of) our online behavior. But I’m not making the connection between establishing trustworthiness and “creating power law problems.”

    Perhaps this is a topic we need to discuss further over beers ;-)
    Jun 24, 04:26 pm

  6. marnie webb

    Ah, it just seems that many of mechanisms that talk about trustworthiness also talk about popularity (credibility is often a bridge between the two). If popularity becomes a stand-in for trust—or promotes certain items so that they become the same thing—that’s where I worry about power laws.

    But, more conversations; beer (or bourbon). Sounds good.
    Jun 24, 08:41 pm

  7. Ed Batista

    Alles ist klar. I don’t think I fully share your concerns, but I agree that the interrelationship of trustworthiness, credibility and popularity (I’d throw relevance in there, too, for good measure) is a rich topic. Lots of grist for the mill. Beer, bourbon, bring it on.
    Jun 24, 10:29 pm

  8. Marshall

    Good discussion. (found your blog again via interview on Beth’s blog) There really are a wealth of possibilities with social bookmarking/tagging tools. I syndicate with RSS Digest my Furl catagories for Podcasts Downloaded, RSS and Search in the sidebar of my blog. And I always try to follow a blog post about, say wikis, with a link to my Furl archive catagory wiki. Right along with technorati tags, etc.

    One thing I wish I understood was why, oh why, so many people are using Delicious when in my mind Furl is so much better. The automatic recomendations, the easy long clippings, the rate this archive item function. Export to MLA, APA or other format of bibliography…but more than anything: a cached copy of the page from the moment you saved it! Plus the interface is so much more friendly than delicious, which I’m afraid non-nerds are going to be too intimidated by. I know there’s a huge community of people creating delicious add-ons, etc, but I still prefer Furl so far.
    Jun 29, 12:08 am

  9. marnie webb

    Marshall,

    Thanks for your comments. I agree tha possibilities in these new technologies are vast.

    I haven’t used Furl in a while so things may have changed but my preference for del.icio.us has to do with the ease with which I can tag surf, the simplicity of the interface (I like the bare bones thing) and the ability to look at users who have tagged items. As I said, the last time I used furl it wasn’t as easy to do all of those things.

    The biblio export—nice feature.

    I’m working, with others, on a social bookmarking tool analysis to try and describe what’s offered by each of the tools and which situations are appropriate for their use. Maybe, with the work there, we can get at a better answer to your question.
    Jun 29, 03:58 pm