Bookmark Aggregation

This is hardly new, but since I’ve often observed the need for better information aggregation tools I figured I’d give a plug. is essentially an online bookmark (or favorites, in IE-speak) repository. It allows you to post sites to your own personal collection of links. This is great for those who frequently access the internet from multiple locations and different browsers (i.e. from work and home) as it is always accessible on the web. But the really powerful thing about is that everyone’s bookmarks are public and easily viewable, and there are all sorts of ways to aggregate and correllate bookmarks. They like to call the system a social bookmarks manager.

The system uses a tagging scheme (or flat hierarchy, if you prefer) to organize links. In the context of a system like, tagging essentially means that for each bookmark you add, you choose a number of labels or categories (tags) which are used to organize your bookmarks so you can find them later. Again, since is a public system, you can see what other people are posting to the same tags. This becomes a good way to keep up on a particular topic (for example, CSS, the economy, movies, tacos or cheese). Jon Udell speculates that posted links would follow a power law distribution, where a few individuals really stand out as the most reliable contributors of valuable links for a given topic. Unfortunately, isn’t particularly great at sorting that out yet (though you may be able to notice such patterns emerging if you really keep up on a topic and who is posting what, which can be somewhat daunting for popular tags like CSS, but perhaps not so for something more obscure like unicode). Udell also notes how useful tagging is when trying to organize something that you think will be useful in the future.

Tagging is a concept whose time has come, and despite its drawbacks, I have a feeling that 10 years from now, we’re all going to look back and wonder how the heck we accomplished anything before something like tagging rolled around. certainly isn’t the only site using tagging (Flickr has tagged photos, Technorati uses tags for blog posts, and there are several other sites). Of course, the concept does have its problems; namely, how do you know which tags to use? For instance, one of the more popular general subjects on is blogs and blogging, but what tags should be used? Blog, Blogging, Blogs, Weblog, Weblogs, blogosphere and so on… Luckily is getting better and better at this – their “experimental post” works wonders because it is actually able to recommend tags you should use based on what tags other people have used.

The system is actually quite simple and easy to use, but there’s not much in the way of documentation. Check out this blog post or John Udell’s screencast for some quick tutorials on how to get started. I’ve been playing around with it more and more, and it’s proving very useful on multiple levels (organizing links I come across as well as finding new links in the first place!). If you’re interested, you can check out my bookmarks. Some other interesting functionality:

  • Every page you view on has an RSS feed, so you can subscribe to feeds you like and read them along with your favorite news sites, blogs, &c.
  • One interesting thing you can do with tags is to create a continually updated set of links directed at one specific person. For instance, let’s say I’m always finding links that I think my brother would enjoy. I can bookmark them with the tag “attn: goober” and send him the link, which will always be updated with the latest links I’ve sent him (and he could subscribe to the RSS for that page too).
  • shows the pages that are being bookmarked most frequently – a good way to keep up with the leading edge. You can also add a tag to see only popular items for that tag. For example, to keep up with the most popular links about blogs, you could try
  • There’s a lot of integration with Mozilla/Firefox, which is one reason for the service’s popularity.
  • There also appears to be a lot of development that leverages data for other uses or in other applications.
  • picks your nose for you! Ok, er, it doesn’t actually do that (and um, even if it did, would anyone use that feature?), but it does lots of other things too. Go sign up and check it out.

Again, it’s a very useful site once you figure out what you’re doing, and I have a few ideas that might show up on the blog (eventually). It should be particularly useful when I attempt to do something like this or this again. The system is far from perfect, and it’s difficult to tell where some of the driving concepts are really going, but it certainly seems like there’s something interesting and very useful going on here.

The important thing about is not that it was designed to create the perfect information resource, but rather an efficient system of collaboration. It’s a systemic improvement; as such, the improvement in information output is an emergent property of internet use. Syndication, aggregation, and filtering on the internet still need to improve considerably, but this seems like a step in the right direction.

4 thoughts on “Bookmark Aggregation”

  1. There’s something strangely satisfying with all this tagging. Maybe it’s some weird obsessive-compulsive organisational thing.

    Anyway, this blogger, Michiel, makes a point here that users could really benefit from an agent that would handle gathering information/search results based on tags that knows what different tags to use that should be grouped together either because of slight variations in the terminology (such as “weblog,” “weblogs,” “blog,” “blogs,” “blogging,” and on and on and on) or language (such as “pants,” “pantelones,” etc.), which sounds like it would be really helpful. Obviously those of us who think of this sort of thing would think to look for interesting weblogs not only under the “weblogs” tag but a number of variants as well, although that requires patience we (or, at least, I) don’t always have.

    For a while, with my own bookmarks, I was throwing in whatever tags I could possibly think of, including somewhat redundant ones (like “reznor” and “trentreznor”) but that gets a little old. The new recommended tag thing is nifty.

  2. Yes agents seem to be a good match to something like I also liked his idea bout collecting all the disparate tags from other sites that use them into one large database. That would be pretty neat. What’s your name? I’ll “subscribe” if you’re interesting enough:P

Comments are closed.