[was originally "RE: standalone patch - patched" but Byron's post prompted me into a long-winded discussion I'd been sitting on for awhile]
(The short answer to Byron's post is that the patch works for me, some other guy had troubles with it. I'd really like to take the time to improve it, but I haven't for reasons elaborated below). The patch does work, but not as elegantly as it should be. I was going to sit down a few days ago and try to make it as elegant as possible, but got absolutely no feedback from this list on the issue (specifically whether my choice of renaming properties to make controlling where Torque reads/writes stuff much more customizable, while still retaining backwards compatibility). Which is a bit of a sore point, for personal reasons, of course, but I'm certainly not the only one who receives an eerie silence whenever anything except a bug fix is posted. Case in point is the "Status of Criteria patch" that was posted just before this. I'm just getting into Torque and don't know much about the technical viability of the patch. It seems good to me, but I can't really say whether it should be committed or not (nor is it my place to say). But the people who do have a say in whether it's getting or not (e.g. Jon, John, Jason, Martin, etc.) haven't said a word about it. Another case is a few Turbine-3 patches I've posted to turbine-dev. Most are basic NullPointerException-type fixes that just get ignored. Now, I know everyone is busy. Jon, John, Jason, etc., work on a zillion different projects. Scarab, Turbine-3, Turbine-2 maintenance, Betwixt, Avalonizing Fulcrum, Maven, JCS, etc., etc., are all projects you guys are actively involved. Plus I'd imagine you also have to do real work on top of all of that. It just seems like Torque is getting left out a bit as you're spreading yourselves too thin. Granted, it's fairly stable and mature, and ...John...or Jason, I forget who, committed some great new data source code awhile ago. Which is awesome, I really appreciate your work, but it'd also be nice if you could open up a little bit to people trying to help out. At least take a few minutes to respond and tell us why our patches suck or aren't spiffy enough to make it into CVS. I can take criticism...tell me what's wrong in anything I submit and I'll try and fix it. Or just tell me to go to hell and stop wasting my time and yours. Any type of response is better than none. Perhaps this is the result of Turbine's sub-sub-project abuse mentioned on the avalon-dev list the other day. I wasn't around before the decoupling, but I imagine it was a fairly strong community. But now with the decoupling, plus other project you guys are interested in, having you guys spread out across all of the sub-sub-projects is weakening the community. I don't know the exact requirements for Jakarta projects...2-3 active developers per sub-project? But I feel that if some of the current Turbine sub-sub-projects where to apply for sub-project status, they'd be denied. I don't mean to lay all the blame on you guys and say you're horrible project owners; you're not. And you could make a very valid argument that the user community is at fault for not having other people step up and contribute more when you guys get busy. And that makes sense, but my tongue-in-cheek comment is that it's hard for people in the user community to step up when the doors aren't exactly being held open. You guys do awesome work; I can't stress that enough. And we probably do need to have a few of you guys spread across several projects so that brain-sharing occurs and they all work really well together (e.g. Turbine-3 becoming avalonized and using Excalibur/Fulcrum components). I dunno, just venting a bit. Please, let me know what's going through your guys' head. If I'm full of crap, tell me. If I have a few good points, a few bad points, tell me. This is also my first real foray into an open source community...is this just normal and I haven't adjusted yet? I'll also agree to the argument that not every patch is a good patch, even if it works for whatever problem the author was having, that doesn't necessarily mean it belongs in the repository. You guys do a great job of screening such patches, though I think it's more a passive process then active, but the thing that pisses me off is that the screening method consists of just ignoring it. I don't have hard statistics about how many patches are/aren't ignored, so maybe I'm over-exaggerating the problem because I'm emotionally involved. Because there is still support, you committers haven't left us completely high and dry. Though I don't think it's the case, maybe I'm looking for that personal response to every minute post that I chastised that guy over in scarab-dev for expecting...wouldn't that be ironic. Dunno, I don't mean to be too harsh and I almost decided not to post this. But what the hell, the worst you guys can do is not respond. :-| Thanks, Stephen Ps. Please, do really respond. I'm young (it probably shows) and impressionable and will learn by whatever response is given, good or bad. > -----Original Message----- > From: Byron Foster [mailto:[EMAIL PROTECTED]] > Sent: Friday, June 21, 2002 6:43 PM > To: Turbine Torque Developers List > Subject: Re: standalone patch - patched > > > Hello Stephen, I wanted to try out this patch but I have a few > questions about it. > > We've found it combersome to try and use Torque in our projects from a > configuration standpoint. We've had to make application specific > versions of build-torque.xml file to use it in our apps. Ideally We > would only work with a build.properties file and within that file we > would specify the necessary directories such as the torque home > directory, ouput directories for sql and OM java files, etc... I know > that some of these properties can already be specified, but not to the > point that they are independent of torque-build.xml. The application > would then only contain the torque-build.properties file. > > Does this patch address these issues? > > Thanks, > Byron > > Stephen Haberman wrote: > > Okay, for some stupid reason I assumed that the last standalone patch I > > posted would work because it worked for my maven project. I ran into > > some misc. problems today and in the process realized the patch probably > > never worked for either non-classpath or non-maven project. > > > > I've now tested it with/without maven and with/without the -classpath > > targets. I'm humbly resubmitting it to the list if anyone like to try it > > again (or for the first time). > > > > If I'm not approaching it the right way, am missing something, or should > > just stop trying altogether, let me know. > > > > (Btw, rather than having targets for both classpath and non-classpath > > tasks, wouldn't it be a heck of a lot easier to just have it be a > > property in the build.properties and get rid of all the duplicate > > targets?) > > > -- > To unsubscribe, e-mail: <mailto:turbine-torque-dev- > [EMAIL PROTECTED]> > For additional commands, e-mail: <mailto:turbine-torque-dev- > [EMAIL PROTECTED]> -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
