On Fri, 2002-03-08 at 11:20, Mitch Pirtle wrote:
On Fri, 2002-03-08 at 11:51, seb bacon wrote: More importantly, forget about the commercial competitors and look at an excellent OSS rolemodel: PHP. Pick on their implementation, hairstyles or whatever, but their documentation just plain rocks the house. I could be a complete moron (unfurtunately a common debate) but with PHP I will most definitely find the answer to all my questions. Even if I cannot figure it out, there is example code that I can cut-n-paste to muck about with.
With Zope, folks, we just aren't there yet. The ZDP (zdp.zope.org) is trying to get a handle on things, but it seems more of a past-tense thing that just petered out. E.g., last document change was 02/07/2001. And the mailing list archive for February was 50% spam (ok, 2 out of 4 messages, that is).
Have to disagree about the spam. The mailing list is actually pretty fine, IMO. All the best Zope wisdom is dispensed there. If only we could harness that...
What is it about Zope that makes it so difficult to document?
I don't know. I suspect it's something to do with the nature of the community, rather than the software. I think it's because there is too few people able to participate in any meaningful way. I think the ZDP people did a great job but were ultimately frustrated by the fact that they were pretty much doing it on their own, without much support from the rest of the community. Another example is ZopeLabs. It's a great idea, and a useful resource, but Alan is constantly frustrated by the lack of input from the rest of the community. Examples from my personal experience: Jon Edwards and I started doing a CMF news. We found other commitments were making it hard for us to do it. I asked for help about 3 times and never got a response. Finally, a few months ago there was lively debate about how rubbish the zope.org site is, and everyone had an opinion about how it should be improved. I've done a survey and Paul Everitt's done some mockups, but the debate generated since then has been tiny. I've got a list of names of people who have said they are willing to commit fixed hours to the project; I'm going to try to hold them to it soon :-) I'm not complaining, or pointing the finger. Everyone has their own reasons to do what they do, and their own priorities. I'm not suggesting it's the duty of every person on the mailing list to be intimately involved in Zope. Also, there is a small number of people who put a *lot* of energy into various things. If there were perhaps 200 people putting regular effort into the platform, things would be very different. My suspicion is that Zope has not reached the critical mass required. For 200 people to be fairly deeply involved, we probably need another 10000 people using Zope regularly. I suspect only 5% of the people reading this who would like to help actually have the time or resources to do so. We need to make that 5% be a proportion of a much larger pool of people. There is also a bootstrapping problem here, because newcomers will be put off Zope by its lack of developer polish. Another difference between Zope and PHP, that may explain the documentation thing, is that the commercial entity behind PHP makes money out of it by selling it as a product. Zope Corp makes money by selling consulting based on their product. Those are my guesses, anyway. Here are my questions: 1) Why do so many community efforts struggle to get momentum? 2) What is the secret to PHP's excellent documentation? seb