You could have "something like" StackOverflow in a day the way a 10-speed bike is "something like" a Honda, in the sense that they've both got wheels and they get you around.
Are we taking bets here? let's talk shop; get a check for $2k ready, list everything from SO you would like to see copied and have a stop-watch ready.
Sorry, didn't mean to put you in the spot like that, but yeah, sometimes it's just that easy, specially when you have a check list and well specified requirements.
Sometimes it takes longer negotiating the deal than writing the actual software.
Maybe you should go back and re-read the article. The whole point of the article is that programmers see a few database queries and the major, obvious functional points and say "I could do that in ..." Whereas, to me, saying "something like" means recreating a comparable user experience.
I think any decent web programmer could hack together a bare bones, functional Q&A site in a few hours. I have no reason to doubt you couldn't implement a lot more features in 24 hours. But to claim you could provide a comparable user experience, I'm just not buying it, even if you have a lot of the generic building blocks already written.
No, you are exaggerating. And the things you WOULDN'T have done are precisely the things that make the app stand out from the crowd and become a success, which is the thesis of the post.
Stop talking, start writing! Seriously, if you can do it in two or three days, what are you waiting for?
These guys are experienced businessmen and have analysed and guessed that they can charge $$$ (per month!) for StackExchange and that it will be a success.
If you can put out a clone and charge $$ for it (one time), and allow people to host on their own servers (which StackExchange disallows AFAIK), I'm sure that people will be interested.
Seriously though guys, I don't think it's that difficult to do in a week. Simply because; you've already got a perfect specification and model in StackOverflow. You've just got to copy it and every single feature until it works exactly the same. You even have an open source (CNProg) copy available for code reference if needed.
Do it, I'm waiting for a thread on HN from you and from ktharavaad on monday.
Pretty much all of the functionality (minus the stuff directly related to the questions) alreayd exists for my framework of choice.
The only "custom" from scratch bit I can really see is a GUI (arguably a lot of that might pre-exist in ones personal toolkit) the Q/A segments and the search.
Not that Im agreeing it can be done in a weekend: but I definitely think he is assuming a lot would be made from scratch, when the beauty of open source is you dont have to do that! :)
