I truly apologise for any "bad form". I made the call for a few reasons (these should have been in my last mail. Apologise for that too)...
- It's totally separate. If they ended up no good, we could axe them as soon as look at them. No code depends on them at all, no other classes were touched to make them work. Also a part of the reason why I just made its own tld xml file rather than do it with a change to the current build process. - Out of the literally hundreds of people using it ( http://www.keyboardmonkey.com/hitcounter/index.html ), not one has come back to say it doesn't work. Nor have they come back with any suggestion as to how to do it differently to make it better. In regards to "...needed more thought and/or experimentation...". Two full libraries were created to fulfill a business requirement to eventually be superseded by this one. But I've never assumed that I've known everything about making the extension or that it's perfect (if it was perfect, I wouldn't have needed two attempts to get it to this stage), but in this regard I think that it's to be taken that there's a consensus in its favor of it being a stable thing. There has been more than two months to review it, and I've been pushing the thing like crazy. When it was first made and submitted to the group (early november), I thought that it would be jumped at. There was also a release planned, and thought it could be made as it was such a simple extension. Really naive, but anyways. Since then I've truly gone out of my way to make it a mature product, remove all doubt that it was stable. And it has been. The code sat there with no updating for over a month. Last update was a resin fix at the start of the year. As for Craig's contrib request... now I obviously regret it, as there's all that little bit less trust in me now, but all I could see was another release sliding by without it properly in the show. Craig's a busy man, that's really obvious. There were positive comments from Ted and Oleg as to me going ahead (wasn't it was why I was voted in?), especially Ted's "...and entitled to make your own best judgment...". I also see that to be a team, means that possibly a little trust must come from the "group" and of the "group". There was no "in-the-face-glaring-issue" Craig raised as important to get through before committing, seemed he just wanted to get across it. And rightly so, it's a fine lookin' baby he has going here. For a team working towards goals, it should have been the case also, that Craig says... "Ted, you're up on this.... I trust your vote". There were no explicit -1's, and the fact that it can be taken out with less effort than it took for me to get it in, I went for it. And if for some reason Craig came back with a reason why it's invalid or otherwise, get rid of it. Hell, he still can. And IMHO, it hurt nothing by sitting where it is. At any rate.... I *really* don't want to get all this into a controversy of "he said" - "Ted said" :). I do have the projects wishes at heart, as much in the team thing as anything else. I truly believe that this new paradigm for marking-up our objects can really set Struts apart. I value my place in this team above anything else in my CV. For the future.... If people "don't" want something to happen, can I recommend that at least someone speaks points made as clear as if you're the one on other side of the table. For more than two months I've tried to get committing attention for my small contribution, but a comitter has to have a look (And there's nothing wrong with that). Ted did along with heaps of others and I was actually made one of you basically on the strength of what I was trying to contribute. Think about it. I don't think that what I did was totally out of line, but I apologise for any inconvenience or otherwise. ...and this being one mail too late. Arron. Ted Husted wrote: >I agree that it would have been better to have given Craig time to >respond and explain why he "wasn't ready for that". > >But as of yet no one has given an actual reason why we should put it >under acontrib. Ultimately the extension belongs with the other taglibs. >It's been "out there" already, has had it share of downloads, and I'm >unaware of any negative comments. As we move toward 1.1, it's important >that we get things like this into the frontline of the nightly build, so >they can undergo even more use testing. > >We had some things like this in contrib before -- but because we weren't >sure if they should *ever* go into the core. So far, the Validator has >been moved to the Commons and Cedric and I are getting ready to propose >Tiles to Taglibs. Though, Oleg's system manager should be moved over >when he's ready. Scaffolding are things that I feel are too "stylistic" >for the core. Artimus is an example app that isn't ready for prime time >yet. > >If the nesting extension were something that had not already been fully >developed and testing, then, sure, we can use the contrib as a sandbox. >But this taglib has been in circulation for some time, and has had >ardent support on the Dev list. > >So, moving forward, I would like to have discussion of why we would put >something like this in contrib in the first place. > >-Ted. > > >Martin Cooper wrote: > >>I have to say that I'm disappointed to see this committed at this time. >>Craig specifically suggested that this code start out in 'contrib', but that >>request was not honoured. I happened to agree with that suggestion, but did >>not feel that I needed to say so, since I assumed that one dissenting >>comment, especially from Craig, was sufficient to indicate that the proposal >>needed more thought and/or experimentation. >> >>Certainly, any committer can make changes at any time. However, as Ted >>mentioned, "If you believe someone might have a contrary opinon, it's >>helpful to ask first and proactively resolve any >>vetos." Personally, I feel that Craig's comments should have been addressed >>and resolved before any commit was made. I consider ignoring such comments >>to be "bad form". >> >>I'm not going to -1 these changes, because I think the nesting taglibs are a >>useful extension to Struts. However, I'd really like to see us work as a >>team, in the future, rather than as a group of individuals. >> >>-- >>Martin Cooper >> >>----- Original Message ----- >>From: "Arron Bates" <[EMAIL PROTECTED]> >>To: "Struts Developers List" <[EMAIL PROTECTED]> >>Sent: Friday, January 18, 2002 12:28 AM >>Subject: Nested Extension - Committed >> >>>Well... there it is. >>> >>>- Recreated package.html to be more consistent with the rest of them, >>>provide better developer docco etc. >>> >>>- To get things done, simply created the tld struts-nested.xml file of >>>the others, keeping the docco. There's been all this mention of slowing >>>things down in regards to the tags to see what the spec's going to end >>>up with, so there'll be time to refine this if people feel it's >>>warranted. I think something should be done, but this will give time to >>>decide the colour of the bike shed. >>> >>>One idea I had was to put all the tags from the three libraries into one >>>xml file, then run the stylesheet for each taglib, and that way the >>>stylesheet can pick its tags and if a library nees something specific >>>from any of them (or extends them), it can have it's own stylesheet >>>(this could also be handy for other developers to automate the building >>>of their Struts extensions). Naturally this could also be carried into >>>docco. Say a page is needed where we want a list of all tags, we can >>>simply make another stylesheet. I think you get what I'm driving at. >>> >>>Do the taglic.xml files' process do anthything more than just the html >>>pages and the tld's?... if that's it I can spend a little time on making >>>it happen if people think it's a good thing. >>> >>>- It built perfectly. And tested out through all my tests the same, >>>hence the commit. >>> >>>- I haven't updated any of the site's links to include it as I figure >>>that this is only done on a release basis (the api-1.0 docco etc). >>>correct?... >>> >>>- Also cleaned up the main Struts logo (a clean-up was all that was >>>done. I didn't make it rotate or anything :). >>>It's just that 100% black drop shadow was driving me batty :) >>>(the download is actually smaller too!) >>> >>> >>>Arron. >>> >>> >>> >>>-- >>>To unsubscribe, e-mail: >>> >><mailto:[EMAIL PROTECTED]> >> >>>For additional commands, e-mail: >>> >><mailto:[EMAIL PROTECTED]> >> > >-- >To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> >For additional commands, e-mail: <mailto:[EMAIL PROTECTED]> > -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>