On Tue, 5 Feb 2002 20:04, Stefan Bodewig wrote: > On Tue, 5 Feb 2002, Peter Donald <[EMAIL PROTECTED]> wrote: > > On Tue, 5 Feb 2002 19:12, Stefan Bodewig wrote: > >> I don't really understand this, sorry. > > > > What you dont understand why people or using it or why we should > > should support these people? > > Why people would want to use one nightly build and adapt to it but not > want to adapt to yet another nightly build.
the majority of people don't know they are using a nightly ? > > It is trivial to update the projects at Apache and some of the ones > > at sourceforge. however it is not so easy to update others that are > > not opensource or that we can't contact. > > Is it unreasonable to expect that people who use nightly builds of Ant > instead of released versions follow ant-dev? I would say yep ;) > > what exactly is the problem with these 3 lines? > > That we are binding our own hands if we extend backwards compatibility > to the level of nightly builds. <apply> would still be <transform> as > well and <antcall> would be <calltarget>. <zipfileset> has been named > <fileset> and <prefixedfileset> and ... > > It is not the problem of three lines, but a problem of how we want to > develop Ant. If every change I make to Ant's code that adds an > attribute or nested element or task means I cannot change the name or > meaning of this thing while we are in alpha cycles, things will slow > down considerably. The problem is the length of time that it has been like that. The change was made prior to 1.4 being released. If it had been a lesser time and fewer people had already adapted to the change then I would have no problem with it. However the change has been too long in codebase to completely remove. -- Cheers, Pete "The ability to quote is a serviceable substitute for wit." -- Maugham -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
