Everyday, web surfers all over the world are building a new meta-web, a fluid tangle of interconnected tags that that sits on top of the web we already know and love. These tags magically arrange themselves into a tagweb that is every bit as dynamic, as sloppy, and perhaps as powerful as the HTML hyperlinks we use today.
Surfers create these tags to categorize their bookmarks on a web site called del.icio.us. For example, you might decide to bookmark and categorize this article with tags like metaweb, nivi, and del.icio.us. Or, perhaps: bonehead.
This tagweb, this tangle, is a generalization of the hierarchies we typically use to organize information. Click on this picture of a small part of the tagweb to see the rich interconnections between a few dozen tags.
 |
| Click image to enlarge – Source: Hippasus |
This all sounds pretty simple. And it is. But it seems like the implications of these little tags are not so little. This series of articles will allow us to think about the impact of tags. It took me too long to start taking tags seriously. Don’t make the same mistake — read this series of articles carefully if you don’t know tags.
del.icio.us: The killer app of tagging?
del.icio.us is a social bookmark manager. It lets you easily add the sites you like to your personal collection of bookmarks. And it lets you easily categorize those sites with tags. Since del.icio.us is web-based, you can share your bookmarks between your machines. And, most importantly, you can share your bookmarks with others.
So what does this all mean?
Bookmark. I can bookmark a page such as my article, Greasemonkey will blow up business models, and access my bookmarks on the web. Yes, del.icio.us looks like hell — don’t let this slow you down. They are working on it.
Tag. I can easily categorize my bookmarks with any tags I like such as greasemonkey, nivi, and web. Click any of these tag to see every bookmark — from everybody on del.icio.us — with that tag. Nobody tells me what words I can and can’t use for tags.
Socialize. I can see everyone who has bookmarked my article and read their annotations. I can look at someone’s personal collections of bookmarks by clicking on their name. For example, let’s look at Factory Joe’s bookmarks. You can see he bookmarked my article on 2005-05-11 and tagged it with browsers, business, extension, and a slew of other tags that he decided would be helpful to him.
Subscribe. I can subscribe to a feed of someone’s bookmarks such as a feed of my bookmarks. Or I can subscribe to a feed for a tag such as a feed of articles tagged with greasemonkey. I can even subscribe to feeds that combine users and tags such as a feed of articles by me about greasemonkey.
Zeitgeist. A so-called tag cloud gives me a high-level view of what people are bookmarking.
Watch Jon Udell’s del.icio.us screencast if any of this didn’t make sense. Watch it even if it all made sense — at the very least, he has better examples. And read the instructions if you want to learn more about del.icio.us.
I personally use del.icio.us to power the daily links feed on this site.
del.icio.us heralds a social web
Wouldn’t it be great if I could talk to whoever is also looking at the web page I’m on right now? If I could annotate that page and share my notes with him automagically? If I could see other pages he had found on the same topic? I believe I once heard Tim Berners-Lee lament that the web (http) is not intrinsically social.
Tim, say hello to del.icio.us.
del.icio.us lets me see who visited a web page in the past, present, and future.
del.icio.us lets me read their thoughts (annotations) regarding that page. For example, see the annotations for this page. If you mash the annotations up with greasemonkey, you can read them without even going to del.icio.us.
del.icio.us filters out a lot of social noise by only tracking people who care enough about a page to bookmark it.
The social features of del.icio.us are rudimentary. But between greasemonkey and del.icio.us, we have the fundamentals for a social web.
See also: Fred Wilson, flickr tags, Technorati tags
Del.icio.us is a very useful way to see how other people have tagged the same web site, or other web sites that have been tagged with the same tag.
There has been a bit of talk about ‘tag spam’ where people use inappropriate tags to get exposure for their site. While social tags are useful in categorizing links under certain key words, it’s still hard to get a measure of “appropriateness” or “relevancy” for that key word. It does help to see how many other people have bookmarked that link, though. Some sites are trying to combine rankings with tagging to provide more relevancy and solve the spam problem, like squik.com. It will be interesting to see what happens as these sites get more mainstream.
It’s important to highlight the fact that this is a social bookmarking site. The intent is to discover, tag, and share. Here are some things that triggered in my mind when I saw your post.
For the system to work appropriately, users should be able to search and browse tags easily. del.icio.us solves both these problems, but the user interface is not sufficient to meet the needs of the average user. I have to admit that even with all my magic foo I was still thrown off when I couldn’t find a search field.
Tag inversion (concatenating strings to make an obfuscated tag) seems to be the latest trend. Is it for privacy? Is it for selective communication? Or do people really tag a web site like Amazon with greatestWebsiteEverWithASuperDuperStore? Should del.icio.us keep a list of unique tags, or tags that are exclusive to a single user?
For new startups, what are the long-term effects of tagging? Can tagging be rolled into every social network, e-commerce site, etc.?
Be sure to also check out Consumerpedia.org – it uses a unique hierarchical tagging system…
You should be interested by Guten Tag. Enjoy !
I have not studied Greasemonkey yet, but found your site via del.icio.us. (abbv: del)
I feel about del the way u feel about Greasemonkey.
First a word about me: I am not a programmer/engineer, but I love the power and significance of the Internet. My primary viewpoint is the nongeek use of the web and tools built by, for, with and on the net.
I collaborate with Rick at theinternetco.net to create quickcard.biz. And Rick is my UGC (ultimate geek connections) and partner in crime at quickcard.biz. OK on to my main point for the day:
I’m gradually morphing my cognitive skills toward embracing the “tagging” “continuum” and away from the “filing/subfiling/folder” model.
I think I first unconsciously grokked the power of tagging years ago when using hypercard and macintosh computers.
But now I’m consciously using it and embracing it thru the use of gmail and del.
What I’d like to find is some utility that would allow me to tag all the docs that i create in Word, Excel and all my other apps. That is, never put any file that I create into a folder again, just tag it and go to my tag menu when i’m looking for something.
Google desktop search has some of those capacities, but no ongoing creator of tags list (as per gmail and del).
Rick is working on adding something along this lines to quickcard.biz, which shd allow people to access a taglist within their quickcard sites, but I’m looking for something that will allow me to have an automatic tag creator for all the files I make in all the apps. that I use.
phew.
We have systems that allows trivial indexed searching. And the world is going crazy adding on capabilities like automatic tagging, semantic “hints”, etc.
All of this is great, but it only addresses web published content. Blogging is one resolution. If all content was in either public or private blogs, the tools out there now would blow a persons hair green.
But 95% of the information I care about is still not web published. I want all the awesome semantic processing to be applied against every piece of digital information that i’ve ever looked at. I use many different systems (laptop, desktop, pda, library terminal, etc) and some of them have spotty internet service. But I should still be able to have all the content archived and processed and made available. And I should be able to mark my information as completely securely private or completely public and a variety of levels between with trivial flagging.
you just pointed people at piggy-bank on your Daily Links. By the same people:
http://simile.mit.edu/hayloft/index.html
A million years ago I worked on the haystack project. It’s a distributed information management system with automatic syntax mining, that is currently focusing on email. The end goal is to obviate the file system.
The semantic web/world is here, and ready to save us time and create connections, but without good integrated, centrally managed solutions, who has time for it?