They've done a fine job of keeping the site responsive, but that's only a small part of what we normally demand from web apps.
I don't fault them for spam. It's a very, very difficult problem, and since I don't have a magical solution to propose, I'm not going to complain.
But there are a few areas where CL could have made major improvements at a reasonable cost, especially considering their revenue. They could have at least experimented with some of these ideas, and tossed them out later if they turned out to be bad. Here are some of the main opportunities I think CL has missed:
-Richer data, tailored to each category. CL started down this road, e.g. by adding the rent field. But there's so much more they could do in every category.
-Better searching. With the huge volume of posts to sort through--and the large number of duplicate posts--the basic search features currently implemented fall flat. Improvements in this area would probably have to tie in with the richer data set mentioned above.
-Abandon the serial model, where the database is just a chronological list of posts. This model creates an arms race to be the person who posts the most frequently without getting flagged. In other words, it incentivizes the fine art of sneaking what amounts to spam just below the radar.
The core problem, as I see it, is that CL still views their service as an extension of its original incarnation: a mailing list. In the CL model, a listing is just arbitrary text and maybe some metadata if you're lucky, and the listings simply appear in reverse chronological order. Sounds like a mailing list archive to me.
Sure, that's how CL got started, but why not grow into something better? There's clearly enough usage and revenue to justify a realignment. Rather than "sticking with what works," I think this is a case of inertia.
There also seems to clearly be enough usage and revenue, and profit, for the owners not to want to take it any further and be happy with what they have rather than risk change.
They are under no obligation to do anything with the site other than what they want to. They don't have to monetize it, or go public, or improve anything. If they're happy with what they've got, it's theirs to be happy with.
The only thing that would drive change would be an exodus of people - not just people complaining about how things could be better, but an actual clear, obvious drop in usage with a clear cause and a need for redesign.
(Look how well Digg's redesign worked out for them... it all but killed the site overnight.)
All good points, and I agree with them in the abstract. However, in this specific case, I don't entirely agree. The post was about how various other sites were taking aim at individual Craigslist categories, and this might be a good reason for CL to change in the long run.
"The only thing that would drive change would be an exodus of people - not just people complaining about how things could be better, but an actual clear, obvious drop in usage with a clear cause and a need for redesign." Maybe, or maybe the perception that an exodus is impending would be enough. In any case, that exodus may very well happen. Which, I think, is what the post is suggesting. I for one have stopped using CL for almost everything, due to the very reasons under discussion.
I don't fault them for spam. It's a very, very difficult problem, and since I don't have a magical solution to propose, I'm not going to complain.
But there are a few areas where CL could have made major improvements at a reasonable cost, especially considering their revenue. They could have at least experimented with some of these ideas, and tossed them out later if they turned out to be bad. Here are some of the main opportunities I think CL has missed:
-Richer data, tailored to each category. CL started down this road, e.g. by adding the rent field. But there's so much more they could do in every category.
-Better searching. With the huge volume of posts to sort through--and the large number of duplicate posts--the basic search features currently implemented fall flat. Improvements in this area would probably have to tie in with the richer data set mentioned above.
-Abandon the serial model, where the database is just a chronological list of posts. This model creates an arms race to be the person who posts the most frequently without getting flagged. In other words, it incentivizes the fine art of sneaking what amounts to spam just below the radar.
The core problem, as I see it, is that CL still views their service as an extension of its original incarnation: a mailing list. In the CL model, a listing is just arbitrary text and maybe some metadata if you're lucky, and the listings simply appear in reverse chronological order. Sounds like a mailing list archive to me.
Sure, that's how CL got started, but why not grow into something better? There's clearly enough usage and revenue to justify a realignment. Rather than "sticking with what works," I think this is a case of inertia.