Tags:
create new tag
, view all tags
Are more people authoring twiki sites then are interesting in developing (t)wikis?

-- DavidLeBlanc - 24 Aug 2001

Probably, and isn't that to be expected and probably the desired situation? I'm not sure what you're getting at -- do you want to run a survey or are you disappointed with quantity of developers, or ??

-- RandyKramer - 25 Aug 2001

I agree, I think is natural because people are pursuing their self-interest, trying to solve a need. The point at which they become interested in developing or, like me, interested in the development of TWiki is at that point that they want to influence the direction of the project, again for self-interest purposes or that they have a modification that they want to make available to others not least because they donīt want to reapply that modification for every new release of TWiki.

For some time Iīve been thinking that we should advertise of SourceForge for more developers. There are a lot of people out there looking for something cool to be working on. The wider the net the more diverse the skillset, this in turn makes TWiki good at solving broader problems.

Also, I am still of the opinion that using more CPAN modules would make the project progress faster. I say this primarily to the extent that it allows the CoreTeam to continue to enrich the core architecture whilst pulling in quality components and hopefully their authors.

(re-written v1.5) -- MartinCleaver - 26 Aug 2001

I am the administrator of a couple of twiki sites. As a user, what I'd like to see is a stable release of the current code base. My perception is that there are too many TWiki developers moving in too many different directions, which is preventing that sort of stability from coming about. Bringing in more developers wouldn't improve the situation.

I honestly do appreciate the work you all do, and I really like TWiki. Please don't look on this as a complaint. But I wanted to share my opinion as someone "on the outside."

-- KristopherJohnson - 27 Aug 2001

Actually there are only four people with commit rights to the cvs at the moment, which is hardly too many. Given that I've been swamped with other work there is really only three people work on the HEAD branch. (Although it might be said that both Andrea and John probably multiplex a bit. wink Just because there is a lot of activity in Codev, doesn't mean that discussion results in code that gets into the cvs. Peter is pretty good at keeping to the TWikiMission statement.

Its true that TWiki being an OSS probably suffers a little (recently) from the one more feature before release creep. Still a large amount of new stuff has been put in the code since the last sta ble release. Its bound to take a little while longer for this to be good production quality code.

-- NicholasLee - 27 Aug 2001

From where I sit, TWiki is like an operating system; I use it as a substrate on which to develop my (corporation-specific) applications. I share these enhancements, but I don't force anyone else to adopt them.

I believe we should keep the TWiki core very lightweight while encouraging growth of a sharing economy among developers around the outside of the core. A specific example; we have our own section numbering and index generation scheme for book authoring; but not many people want or need this - it's a minority requirement. I don't want to see section numbering/indexing system in the core, but I do want to publish what I've done in this area. I'd also like to be able to pick and choose among extensions developed by other people.

The core functionality is rich enough that I'd like to see fewer new features, and a bit more invested in a fully unit-tested API to the core. We might even look at moving functionality out of the core and into plug-ins. (This would also make other stuff simpler, like compiling wiki into Apache.)

- CrawfordCurrie - 28 Aug 2001

One reason we'd like section numbering in TWiki is so it can used for TWiki documentation. I think it would be a good idea to widen the Plugin hooks so some of what is now core can move out. This is some unit testing available in the tools area on CVS - I've yet to receive any feedback on it though.

I'm puzzled by your references to compiling wiki into Apache. You can run TWiki under modPerl, but I don't see how you can compile into Apache.

-- JohnTalintyre - 28 Aug 2001

mod_perl basically runs a perl interpreter embedded into apache. However it would be possible to generate a new apache module (e.g. mod_wiki) which would be compiled native into apache (i.e. not running via the interpreter). This came up here when our wiki server started to really suffer performance problems. We haven't attempted to do it, because we thought the performance advantage would be slight.

I was very pleased to see your tests, and it was partly that that triggered my remark. It's a step in the right direction, but we need much more - both unit tests and client-side system tests. Aside from anything else, well written tests help you refine and define the specification much more clearly.

-- CrawfordCurrie - 29 Aug 2001

mod_perl gives a very big speed up as Perl code is only compiled once, rather than for every script execution. I've found this gives a big performance boost. I guess it would be possible to compile all the Perl to a native form and build that into an Apache module, but I doubt you'd get much speed advantage over mod_perl for the substantial effort you'd have to put in. Give mod_perl a try, it only took me a few minutes to configure.

I agree we need more tests, but it would also be good to see feedback from people running the ones already supplied.

-- JohnTalintyre - 29 Aug 2001

Many authors and a few developers is a good thing IMO. Authors bring lots of good ideas and help in defining new directions for TWiki. A small and focused dev team can produce better quality code then lots of developers. (Well this summer we had some stability issues in Alpha and Beta releases, more at DevOpinionPoll). The Plugin API will help to broaden the developer base, independently of the core code and CoreTeam.

And yes, KISS and a stable code base are high on my list of priorities for the core code.

-- PeterThoeny - 31 Aug 2001

Edit | Attach | Watch | Print version | History: r13 < r12 < r11 < r10 < r9 | Backlinks | Raw View | Raw edit | More topic actions
Topic revision: r13 - 2001-12-29 - TWikiBot
 
  • Learn about TWiki  
  • Download TWiki
This site is powered by the TWiki collaboration platform Powered by Perl Hosted by OICcam.com Ideas, requests, problems regarding TWiki? Send feedback. Ask community in the support forum.
Copyright © 1999-2017 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.