Monday, January 16, 2006

Semantic Tagging Vs. Search ( and CiteULike)

Over on the sidebar, you'll see three links, two of which are new:

Clips blog @ Bloglines
My CiteULike

"Clips blog" speaks for itself: just raw links to things that seem interesting in my RSS feeds. (Suffice it so say, that after a brief dalliance with Google Reader, I'm still using Bloglines.) is a service that many readers will likely be familiar with, though probably not everyone. It's also a kind of 'clippings' mechanism, though the idea is you store your favorites with tagged (keyword) references. Some people (Kerim) even use it as a way of housing an extended blogroll.

It might seem redundant to have semantic tagging when you can basically find anything you can think of with simple searches in Google or Yahoo. But seems to be most surprising when you're trying to find things that relate to what you're interested in, but that you wouldn't necessarily know to search for. The people who seem to have adopted it first are techies -- many software people seem to use to keep track of links to "helpful hints to debugging in DOT.NET" or "quick Linux hacks" (see Popular). But there's no reason why people interested in literature, art, and politics, can't also be using the service to connect to each other, and to find out about things via keywords.

Incidentally, the process of posting to becomes pretty easy with their handy Firefox extension. Once you've installed the extension, it just takes one click to add a post.

I'm still not 100% sure whether tags will really be a big-time revolution -- "the next phase of the internet" -- or simply a service that appeals to hard-core users (bloggers and the like). is certainly growing fast, and especially so since they were acquired by Yahoo!, but their interface isn't exactly fun.

The title of this post might be somewhat misleading, since obviously I'm not suggesting that semantic tagging services like or Flickr are somehow going to make conventional search engines obsolete. At most, the two forms of accessing resources on the web will be complementary to each other. But there is an additional level of interactivity involved with tagging -- you leave a trail across the net (if you want to), which either you yourself or someone else could possibly make use of later.

Finally, a word on CiteULike. This is an academic-oriented service, that allows you to bookmark articles you've read online, including those that are behind university subscription databases -- such as Project Muse, JSTOR, EBSCOhost, etc.

I've just started playing around with CiteULike, and I'm not quite sure yet whether I'll be using it regularly. It might be a useful tool to get into a better discipline with regards to actually reading journal articles (instead of just doing searches when I'm working on an article or a book chapter). Like, CiteULike allows you to share links to articles and view other people's articles, making it potentially easier to keep up with other people who might be doing research that relates to yours. And the "notes" function allows you to read people's thoughts on the articles they're reading. As with, this service will only be of great value as a relational index if a large number of people in a given field are actually using the service regularly.

Again, I get the feeling that people in the sciences are using this extensively, while humanities folks have barely touched it.

Any thoughts on these services? And what do you think about Tagging vs. Search? Are there other tagging services you would recommend?


IndianPad said...

I love delicious as I find things there that I would never find on Google. Also, other people have done the searching and the number of bookmarks easily tells me how many people think its important. Service like digg, and its clones like ;), are not tagging or searching services but are gaining popularity extremely quickly.

Linked from IndianPad

1:43 PM  
Suresh said...

I use and citeulike; the first is handy because I can subscribe to RSS feeds for specific tag combinations and get all kinds of interesting links.

With citeulike, i can do something similar. I have RSS subscriptions for tags on specific topics and have found many interesting papers this way. I've also found people whose names pop up frequently, and thus have found others with common interests.

I use citeulike to maintain a paper database when i'm doing a lit search; it's extra handy because when I'm writing a paper I can export the biblio information directly into my latex document as a bibtex file.

3:52 PM  
Rik said...

Apart from finding what others have already found, delicious is a very good thing for storing bookmarks for yourself. Also, it makes easy to share a particular set with anyone - just send the link to the tag - I found it very useful for sharing links with people on a project.

And when you are looking for something to read, the 'popular' page always has something interesting. A search for 'interesting' would not give anything on google, and the new stuff there may not be in any search database for quite some time. Ok..this sounds like an ad for delicious, so I'll stop. yes, the interface looks horrible.

1:45 PM  

Post a Comment

<< Home