On Friday 20 June 2008 01:04:03 pm Noah Kantrowitz wrote: [snip] > installing AccountManager). I would propose an alternate solution, the > creation of a "new" project that is officially recognized by t.e.o in some > way as being the turnkey solution so many people want. Perhaps something > like bitnami+oforge+some common options and config settings. "Trac" remains > the minimalist system we all know and love, and anyone installing it will > continue to get the same experience. People wanting the turnkey version can > get $NEWPROJECT, and not care about what particular components it pulls in.
I don't think we should start $NEWPROJECT. Trac has a lot of name recognition... and we need to keep that. Just for the record, the plugins I use on most every Trac I setup: AccountManager and Graphviz. The latter I don't see as an item for 'core', but graphviz is an awesome tool. :) Other comments regarding this thread; primary audience is the other devs: I think that AccountManager needs to migrate to Trac core just as WebAdmin did. And I think the form-based login needs to be the default. Since 0.12 is "translation & internationalization", I don't think we should include it in our next release... but I think we need to get 0.12 out ASAP. Then add AccountManager to core, and call it 0.13, and again, get that release out ASAP. We need to improve our "momentum" (for lack of a better term). I think we (the developers) need to take this discussion as something of a 'wake-up call'. I don't think we should go into a panic, but I think we need to go back and challenge some of the things we "know" about our users' needs. Our user base has been growing... but I think it has been growing in different proportions to what it started as, so our user base as a whole is probably very different than it was. But if I tried to specify how the user base has changed, or what the most pressing needs are, I would just be guessing. We need real data here. And I think we have some to look at -- trac-hacks and t.e.o. For trac-hacks, we need to figure out what the most downloaded plugins are... that will give us insight into our users' needs. Then, we need to start at the top, and for each plugin answer these questions (and do it publicly, on a t.e.o wiki page, dev-editable only): * How well does the plugin live up to its goal? If it doesn't do what it claims, or does it poorly, this is a portion of our userbase that is either unhappy, or went to something else since it wasn't good enough. (I'm guessing that one of these will be the Gantt plugin.) * What ugliness in the plugin could be removed by adding an extension point (or whatever) to Trac? If we have a popular plugin that has to use ugly tricks to do what it needs, we need to fix Trac to support it well... and prioritize that fix based on the plugin's popularity. * What is the quality of the plugin's code? If the code is horrible, but it is popular, we need to get the code improved. We'll probably want to try: track down the maintainer, see if others can help out, try to energize the community for that plugin so it improves. For t.e.o, we need to find the most "popular" tickets. What are people clamoring for in our buglist? Can we find someone to champion those? Other things we need to look at are: * For each extension point in Trac, is there a plugin that uses it? If not, can we articulate the reason for that extension point, or should we maybe remove it? (Should be openly trumpetted on the mailing lists if we think the latter.) * What are the most common search terms on t.e.o? Are people finding what they need, or are they just getting frustrated? * Same as the above, but for t-h. * It may also be enlightening to find out what the most common Google searches that include "trac" are, if there is a way to find that out. It may also tell us something surprising. * Trac code quality. While increasing the testing line coverage, I found a number of bugs, and in general, found my opinion of our codebase decline. (And let me state for the record: much of it was looking at code I had a hand in.) Wow. Essay. And yikes, there's a lot of work to be done here. Hm. I have an idea, and I'm going to try to lead by example on this.... separate email to follow. Thoughts? Eli ------------------. "If it ain't broke now, Eli Carter \ it will be soon." -- crypto-gram [EMAIL PROTECTED] `------------------------------------------------- --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Trac Development" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/trac-dev?hl=en -~----------~----~----~----~------~----~------~--~---
