RE: AW: [PEAR-DEV] BC breakage results

> From: Jan Schneider [mailto:jan@horde.org] > Sent: Friday, September 19, 2003 4:18 PM > > Zitat von Alexey Borzov <borz_off@cs.msu.su>: > > > Hi! > > > > Daniel Khan wrote: > > >>Now, we have a dilemma: either remove them in 3.2 (which was > > >>intended initially), make the version 4.0 instead of 3.2 > > >>(which will be misleading for users after 3.0) or create an > > >>abomination of QuickForm4. > > >>Removing them will shave 10kb from the main include file, > > >>which is Good. > > > > > > > > > So wouldn't here a slot based solution a party solution? > > > So you at least don't have to release a 4.0 but a 3.2..? > > > > I think that slot-based solution is better than mandatory installation > > of different versions into different directories. Thus we let users > > decide what to do with the packages. > > One can install the package into slot 1, test it, then upgrade the > > package in the main slot. Or one can install older version of the > > package to the slot 1 and install the package that depends on it. > > More power to the users. > > Yeah, cool, especially for those users (and there are lots of them) who > don't even understand the meaning of include_path. > Remember: while developers are *using* PEAR packages, end users have to > *install* them. And you can't compare these users with geeks running > Gentoo. Also this will not allow you to run two versions in one script. I have a framework with tons of modules which can call each other. If I want to move to a version that breaks BC I can only do that if I port all my modules at once. This might not be possible due to time constraints often. So this means I even have to code new modules with the old (possibly buggy, possibly slow, possibly limited) version until I can port all modules at once. Therefore I am for: 1) making major versions bumps with every release (the current mess can never serve as a reason not to move to that) 2) make each major version a new package so that different major version can run in one script <rant> Not talking to anyone particular here. It is been a while since I have read anything new in this thread. Some developer seem to care about users, some developers seem to have unlimited time, etc. Repeating that over and over will just prevent us from going anywhere. </rant> regards, Lukas