I intend to replace Etherpad (opens in new tab)

I've read http://sivers.org/zipit and when I saw this I instantly thought about it. After that, though, I read the blog, tried out the program, and now I'm not so sure. It seems like it might be something Ben Noland is ready to confidently pursue. Also, I would say that it usually doesn't work well, not that it never works well. I'm not big on absolutes when it comes to things like creativity, business, and motivation.

One thing that makes zipit not apply is that it's already partly a reality in two different ways. First, he has a pretty good prototype, with good branding already. Second, his call to action has actually started some serious interest, from what I can tell. There are quite a few encouraging comments and suggestions here and on his blog, for something that was just announced half a day ago.

Ben, I hope you can bring collabedit to fruition!

Related: http://sivers.org/zipit

Yes, but sometimes timing is everything.

In this case I was looking for feedback, ideas and awareness. Given the state of things, the timing has worked out pretty well.

I agree wholeheartedly. The first thing I noticed with Bespin specifically that STILL hasn't been fixed as far as I can tell from using it for the last few minutes was the lack of auto indentation. It is such an important thing for a code editor. It HAS to be there.

EDIT: I see they added the autoindent setting last March. I still can't for the life of me figure out how to actually enable it. It should be on by default, no question about it.

Yeah, TextMate's auto-identing is very smart and usually perfect at reading my mind. I'd like to see something like that in an online collaboration notepad like this because it's a lot of work to do it manually after using TextMate so much ;-)

Does codemirror's indenting work well for you? I actually really like it.

Probably not today, but you try it out: http://collabedit.com/display?id=44959

You can also open multiple browser windows if no one else is there.

Yep I could like to build a Redis-based project doing this stuff.

I don't know if they qualify as startups, but here are two examples:

Mint (http://www.haveamint.com) and FogBugz (http://www.fogcreek.com/FogBugz/)

You can buy and self-host these products, with source code included.

You can (as we see) knock up a bare bones copy reasonably easily. The expertise is already out there. I'd say quite a lot of people are working on realtime collaborative comet stuff, and have been for quite a while now.

The hard bit is getting people to use it.

I wrote a fun multiplayer drawing applet (Java) back in 2000, but no one used it :( sob. Actually a few did, it was fun anyway :) One of my points is, this technology is nothing new.

Fortunately, he's a fan of etherpad, so if the open source product is good (might not be-sometimes open sourcing something doesn't work well), he'll be ready to commercialize it. People will still need support and hosting.