Considering the number of successful web applications, at first blush it may seem surprising that commonly accepted standards for core elements of a web site’s architecture and design haven’t taken shape to a greater degree. But when you think about it, given that the web is still a form of emerging media and the fact that design tweaks can significantly broaden an application’s appeal–or cause it to become much less attractive–it really isn’t surprising that we’re still in a state of experimentation, even with basic elements of web design.
Take Slashdot. For some time, they have had my favorite commenting system. The site allows readers to choose from a variety of options that determine how comments are displayed: first, you can select the viewing hierarchy (flat, nested, or threaded); second, you can select the ordering scheme (oldest first, newest first, highest scores first, oldest first, or newest first); and third, the system provides a mechanism to restrict what comments are visible based on “ratings thresholds” (so comments that have been marked as spam don’t appear, while comments that people like are visible). Slashdot’s fuctionality is useful given the volume of comments that the site generates. It’s rare to see an article with fewer than a few dozen comments, and popular posts often have 500+. With that much content, it’s good to have tools to help separate signal from noise. For example, if a particular comment is abusive or otherwise uninteresting, readers will likely “demote” it. Future readers won’t see the comment at all due to the threshold system.
Of course, all this functionality comes with a cost: the system is a little complex, and many users probably aren’t aware that these options exist. Some visitors may find the site too difficult to use, and choose to find their news elsewhere. Still, I think that Slashdot’s reputation for generating valuable, thoughtful discussion is due, in large part, to the way they have architected the commenting system. Personally, I spend only a little time scanning Slashdot’s posts, but I frequently spend far more time reading through entire comment threads. This is because Slashdot’s architecture creates incentives for comment posters to generate good content if they want their comments to be seen.
This all came to mind as I was recently on Digg and noticed that they’ve made several significant improvements to their commenting system. In fact, it’s now a little like Slashdot (moderation, hierarchy, banning, etc). I like it. But it isn’t clear how Digg’s user-community will respond. Will Digg start to look more like Slashdot, and generate the same type of discussion? Or, will Digg evolve in a different direction?
The point of all this is that we’re still figuring out the relationships between the technical aspects of web site design and the corresponding utility of a given site as a medium for creating community and for helping people find useful information. That in mind, I think it will take some time before commonly accepted design templates emerge, if they ever actually do.








“Slashdot’s reputation for generating valuable, thoughtful discussion”
… are we talking about the same slashdot?!?
Many of the high-traffic blogs could use a community-mediated comment system to prevent ‘comment overload’. My blog typically gets 1-3 comments (if that) so it would be slightly overkill. I have the feeling that digg, despite sometimes bringing up interesting links, is overrun by screaming monkeys (more so than Slashdot, I mean) and not worth taking the time to comment.
Speaking of comments, if there’s something I would really like is to have some PORTABILITY there. Move to a different blog system > you lose the comments (and sometimes your posts). Move to a different comment system (say haloscan) > lose your comments. That sucks.
Left by Michiel on March 6th, 2006