A product change, like adding the freetext attribute, is subject to lazy consensus. You have the right to make the change, and any other committer has the right to veto it.
http://jakarta.apache.org/site/decisions.html People often announce a short-term plans, not so much because we have to, but because we are fishing for vetos. The committers own the codebase jointly and severally. If any of us do not want to support non-standard attributes, then we cannot support non-standard attributes. This is how Jakarta keeps its product standards high. There are no compromises. We must all agree, or at least not disagree, or it doesn't happen. This way, one person cannot come along, add some difficult features and then walk away, leaving the rest of us holding the bag. We all have the opportunity to veto a change. If we don't exercize our veto, then we are agreeing to support that change in the future. This helps keep the voting lags to a minimum. You don't have to call for a vote, but everything we do is subject to a vote, and the approval of every other committer. Martin's veto includes a number of reasonable justifications; it is valid, and so the changes should be rolled back. At least we have hashed this out, and can refer to this thread in the future, or bring it up in the FAQ (along with why ActionForms are not a *@#$!&% interface ;-) -- Ted Husted, Husted dot Com, Fairport NY USA. -- Custom Software ~ Technical Services. -- Tel +1 716 737-3463 -- http://www.husted.com/struts/ Oleg V Alexeev wrote: > > Hello Ted, > > It is interesting situation now... 8)) I think it was in some letters > before - a number of words about 'custom' or 'literal' attribute. But > only after commit we have a some kind of discussion in list... 8) > Yes, of course, may be my activity is wrong - I think that 'right' > solution with [VOTE] messages don't affect developers so actively. Now > anybody can check it at his own environment and VOTE after this... No > problem if such idea is wrong - I can revert all changes > and make html tags 'clear' again. 8) > > It is eternal problem - to balance between standarts and solution > useful for programmers... > > Monday, December 10, 2001, 9:28:17 PM, you wrote: > > TH> Can we let it ride for a week, to see what else comes up? > > TH> I appreciate input from the developers, but we might want to also see > TH> what the other committers have to say. > > TH> I also just thought of another one: "output" - since that is what we > TH> are really doing, outputting something into the body of tag. > > TH> I do agree with Craig, that compliance has to be the "Prime Directive". > > TH> If that sometimes means making things more difficult for people who > TH> choose to use Struts, because other products choose to be non-compliant, > TH> then so be it. > > TH> I also agree that the framework should encourage proper use, and I do > TH> support a number of other design choices we've made in the place, which > TH> are not always popular with developers. > > TH> Like Craig (I imagine), I would instantly veto something like a wrap tag > TH> for textarea being part of a tag, since that is a vendor-supplied > TH> extension, and not part of the W3C specification. > > TH> This has always seemed like a likely compromise to me, and I've > TH> mentioned doing it several times myself. Though, the argument that there > TH> should be no non-standard attributes, carries some weight with me, since > TH> that is why we have this problem in the first place. People ran around > TH> doing whatever they pleased, standards be dammed. > > TH> The other question I would ask, is how does this fit in with JSPTL? I > TH> don't think any of want to continue supporting tags that overlap with > TH> those. If we allow a literal attribute in all of our tags now, is that > TH> going trip people up later when we migrate. Might the Jakarta Input > TH> taglib's people have any thoughts on this? Eventually, we might want to > TH> hook up, and fill in the JSPTL gaps with a single set of Jakarta > TH> extensions. > > TH> But now that we've let the genie out of the bottle, lets hang back and > TH> see what other people have to say. > > TH> -Ted. > > -- > Best regards, > Oleg 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]>
