"Do we really need a site for each individual content management system?"
It's interesting that CMS specific sites were even proposed. In my experience (mostly with the Drupal community) CMS advocates tend to be pretty tribal and desire a narrow focus. A cross-CMS Stack Exchange site would probably seem like it was full of noise to everyone involved. What would deemed to be signal VS noise would depend entirely on one's CMS of choice.
My first reaction as well. If that's true though, how do we explain that Stack Overflow itself has done so well? People are at least as tribal about languages.
I program professionally in Ruby, Javascript, Erlang, and some Java. Add in the pieces of each, HAML, JQuery, GWT, and the languages I like reading about, such as Haskell and Lua, and thats a good % of SO. It would be much more annoying if it was separate sites.
Also, lots of info about most languages is useful, even to other languages. I imagine its much less so for CMSs.
I agree with "Ugh." There are way too many stack exchange sites. And moreover, the pedantic "on-topic/off-topic" debates work well with the programmer culture, but it gets tiresome really fast when you want to just ask a question about how to do something simple on a basic web app.
Hopefully the core team will make strides to eliminate the duplication and over-segmentation of all these Q&A sites.
Well, a part of it (the comp.* hierarchy), perhaps. In fact, this seems to be a pattern—Facebook is really just a new soc.*, etc.
I wonder if there's a good pointer to potentially-profitable market gaps here: pick some part of Usenet, and turn it into a website that tries to centralize all discussion on that topic to itself.
Yes, much better model imho - put it all under one roof, and make it easy to find and expose only topics you're interested in. Similar to Reddit's subreddits in that respect, both work well.
Interesting that he mentions "we made one big site for all programmers and told them to use tags and, lo and behold, it worked." One of the most frustrating things about using stackoverflow.com is finding a good question and having it be marked off topic or moved to another site to die.
I was recently trying to compare ereaders (with the iPad in mind) for technical PDF's. I thought stack overflow might have other programmers asking the same question, so I searched and found the page/question I was looking for. However, it was moved to superuser.com. Then, it was moved to some gadgets stack exchange. By the time it was on the gadget site, nobody was adding to it. Now, the gadget site is closed due to inactivity and I can't find the question's active URL (if there even is one anymore).
For reference, I think this was the original question on stack overflow:
I had high hopes for the SE platform back when you could pay to host your own. I've lost fascination with it now. StackOverflow is great but everything else ... Meh.
Browsed the cooking one and even submitted a question. Most of the best best answers are simply links out to other popular food sites/blogs. If that's going to be the case, it's doomed IMO. At least with SO you get solid answes with code in the answer.
While scanning this article I was struck by an uncanny likeness of Spolsky's article to very old Philip Greenspun posts about photo.net and the ArsDigita system, complete with nice random photos of wolf-like dogs. In my awareness, and to my interest, frameworks to meet psycho-social challenges in software development are still so lacking. I'm glad that Spolsky and Atwood are working on this.
I'm really confused: after all that insightful discussion of social site size, why is there suddenly a "programmers" site, in addition to "stackoverflow"?
It's interesting that CMS specific sites were even proposed. In my experience (mostly with the Drupal community) CMS advocates tend to be pretty tribal and desire a narrow focus. A cross-CMS Stack Exchange site would probably seem like it was full of noise to everyone involved. What would deemed to be signal VS noise would depend entirely on one's CMS of choice.