Why del.icio.us Sucks
posted by brian at 07:51 AM
I sent an extremely condensed version (one sentence long) of this post to a friend recently, unprompted, for no reason at all. I wasn't even sure he had used del.icio.us before, and he sure had never talked to me about it. So what drove me to despise this service so much that I just had to share my feelings about it with him? Truthfully, I'm not even that passionate about this particular opinion of mine. It's unlikely that he even cares what I think about del.icio.us. It's even more unlikely that you care what I think about del.icio.us. And this is exactly the reason, among others, for my dissatisfaction with it—I propose that:
If someone is viewing a web page, clearly they find it interesting—if not, why are they viewing it? And since they're interested in it, there's a good chance that they will post it if they use del.icio.us. And since 99% of the web is complete garbage, so is 99% of what people post to del.icio.us.
"But," you quickly respond, "people obviously don't bookmark every page they come across! Browser bookmarks usually only contain resources people read frequently or intend to read in the near future."
Well, not anymore! The 'personal bookmarks' aspect of the service may be useful, but that is clearly not the main reason for many people using del.icio.us—that ability has existed forever. The feature people take interest in is the ability to share links with the masses. Judging from the majority of stuff posted on there, it is obvious that people are just sharing any old crap they come across. Ultimately, my thought process went like this:
- Hey, this seems kinda cool, and a lot of people are using it.
- Hmm, there sure is a lot of stuff on here.
- After browsing through a few pages for each tag that technically interests me, I still haven't found any cool bookmarks!
- What a waste of time.
Even so, this doesn't make the service inherently bad. Since most of the links posted are new, the service also serves as a live, categorized index of the web. Wow! Who needs web directories from sites like Google and Yahoo! when users are voluntarily "organizing the world's information" right now, and at every hour of the day? Well, unfortunately, the ability to take advantage of this aspect of the service is extremely limited by the default interface. Resources like taga.licio.us offer features that should be available directly on the main site. They shouldn't require me to install a PHP script on my own server.
Maybe the site's creator just doesn't have the server resources to support so many users doing advanced queries. I think it's more likely that he just made the service as simple as possible in order to let the users decide what it should transform into. That's not a bad design idea, but as of now I'm still not impressed with what I see.
Chances are, if there's a topic out there that interests you, there's already an appropriate aggregator site out there tending to it (and including commentary and discussion). For example, I go to Daily Python-URL for Python links, Slashdot for technology, and browse a few newsgroups and other sites less frequently for more specific resources. Yet everything I find on del.icio.us I've either already read elsewhere or find extremely uninteresting.
I'll probably edit this post later with more thoughts, I'm just writing off the top of my head for now.
Comments
I like the tags. I subscribe to tags like del.icio.us/tag/miata, del.icio.us/tag/shibboleth, and del.icio.us/tag/internet2.
Plus, it is also nice as a way to collect my "to read" links and be able to access those links from my work desktop, home desktop, and laptop.
Personally I find it much more useful and interesting to look at my inbox--consisting only of selected peoples' recently added links--than the del.icio.us homepage. I agree that as a search service it's no competitor to the existing ones, but it's nice to have another "shared attention" space on the web, and once I filter it by "people whose links I tend to find interesting", it does turn up enough good links for me to find it worthwhile.
I figured it used a collaborative filter mechanism to determine what links you would like based on the links that others have. Like the way Amazon has that not-so-useful-to-me "other products to buy" feature. Just making a big list isn't so great, but if you have enough data, you can exploit similarities to recommend bookmarks to people. So if you and say, someone named Omega have very similar bookmarks, then the system can recommend that you view the Omega's bookmarks that you don't have, and vice versa.