Over at my O’Reilly blog: A few months ago, one of the instructors I work with asked me to attach a wiki to his class’ website. I’d been meaning to start testing various wiki systems for a while, and this was the perfect opportunity to dig in.
Short story: The wiki world desperately needs a product with the kind of vision, direction, and momentum of Movable Type. There are mezzo-mezzo wiki packages out there, but nothing I’ve seen yet that really nails the category the way MT does for blogs.
Music: David Thomas & the Pedestrians :: Confuse Did
Agreed – the wiki world has a pretty high entry curve right now. Instiki – http://www.instiki.org – has been getting some attention lately, thanks in part to a mention by Merlin over at 43Folders.
But damn, I’d love an MT or WordPress for the Wiki World.
There are some Wiki developers who have recognized the target as important – namely, at least one: myself. But I’m not very far along. Everyone out there seems to respond to the not-quite-rightness of existing Wikis by coding their own, leading to too many codebases that aren’t easy enough to coax out of their own idiosyncracies. I’m trying not to contribute to that problem… but then again, I am writing a new Wiki. Maybe I should contribute to MediaWiki instead.
I frankly haven’t played with this much, and it’s a hosted application, but it’s new and getting a fair bit of a buzz.
http://www.jot.com/
Doh! Shoulda read the O’Reilly piece fully, esp this:
“Yes, there are wiki services out there we could take advantage of, such as SocialText and SeedWiki, but I’m committed to keeping all school content hosted on school servers – I’ve been burned going down that road before when services went belly up.”
Nevermind.
Check out my reply (of sorts) on blog.taospace.com
I would have posted this to the ORA site – but some sort of redirect loop (I think relating to logging in to my ‘account’) prevents me.
I’d had a conversation some weeks back with a colleague about needing a ‘collaboration space’ on our internal network at the office. So this weekend I installed TWiki (http://www.twiki.org), installation was pretty painless on a Mac OS X 10.3 server (aside from dealing with Apple’s funky apache config files).
It works well so far for the moderate amount of testing I’ve done, and the plugin mechanism seems pretty useful. The twiki.org website had one of the most ‘polished’ feels and some good user testimonials which gave me a better warm and fuzzy feeling than other wiki projects.
One thing that I do wonder about wiki’s generally (and this may or may not be an issue with twiki) is how they hold up and scale with large amounts of content and a lot of users. Wikipedia seems to prove it can be done but Twiki does at times seem a little sluggish – even on a fast G4 XServe with no real users.
Tao:
If XWiki is less polished, it is mainly due to time constraints. The whole thing is currently being built by one guy. Contributors are very welcome indeed.
That’s an unfortunately common refrain in the open source world, and is pretty much exactly what leads to us having dozens of small projects without any real backing, and always in danger of dying from neglect, no matter how good they are at core.
Thanks for the review Tao – you definitely covered some territory that I didn’t. Would be great if all this discussion started a meme that resulted in Six Apart or someone picking up the ball and filling this hole in wiki world.
Andrew – you’re not the first person I’ve heard complain that they couldn’t log in to the ORA site – seems to go into a loop with some browsers.
Scalability is definitely an issue. MediaWiki seems perfectly snappy, but I would still bet that WikiPedia does a ton of load balancing etc.
Thanks for your thoughts on Twiki – will have to give that one a closer look.
Joi ito (http://joi.ito.com/) one of the investors in Six-Apart has been working with Wiki’s personally for a while. He just recently moved away from using moin moin to use SocialText – but I can’t read anything in to that other than an ability to afford SocialText (it strikes me as expensive) and his feeling that it’s better than moin moin. He might have blogged about why he switched – I don’t recall.
If anyone is likely to drive some of the convergence he might !
It likely won’t be long before somebody builds a Wiki on top of the plug-in architecture in Movable Type 3… is that what we’re waiting for?
Andrew, Misuba –
Six Apart has a great opportunity waiting for them here. Although at the same time, I’m reluctant to see them diversify if they can’t handle the growth elegantly, or if it would mean bloating MT.
Scott, if you have 10 minutes you might want to check out Efurt Wiki, http://erfurtwiki.sourceforge.net/.
I’ve deployed it several times with success. It has a nice plugin architecture, plays very nicely with WordPress (it is even available as a WordPress plugin), and it simple to install and setup. The developers are very responsive to comments.
I don’t know if it would meet your Chinese and PHP5 requirements — I haven’t had to deal with those issues.
Tim
P.S. Given what I have seen, 6A’s is having enough difficulties with what is already on their plate. It also isn’t clear to me if there is a business model for a standalone wiki package, unless it is as sophisticated as something like Jot.
That was Erfurt Wiki – not the most euphious name in English. Nice licence too — do whatever you want with it.
Tim, thanks for the tip on Erfurt – I’ll give it a whack. From the 3,000 beta testers Jot has gotten, mostly from business clients, I think it’s clear that there is a business case for a good wiki. And I would say that a combination blog/wiki package is probably the best route to a good biz case, rather than standalone. SixApart’s momentum in the space could translate pretty well.