RE: [PEAR-DEV] Duplicated/redundant package for PEAR::Message -> PEAR::Crypt_HMAC

> From: Derick Rethans [mailto:derick@php.net] > Sent: Friday, April 25, 2003 11:21 AM > > On Fri, 25 Apr 2003, Lukas Smith wrote: > > > > From: Michael Bretterklieber [mailto:mbretter@jawa.at] > > > Sent: Thursday, April 24, 2003 8:13 PM > > > The problem/question is another one: > > > Does the voting system apply for all php/pear-developers, or do have > > > some guys the privilege to add silently packages?! > > > > It applies to all developers. > > I am not sure if it is in the manual but Stig said it quite clearly in > > his recent article in the php mag: if you don't go through the voting > > system the package might be removed. > > That was unknown to me at the time it was added. I do remember asking > about what the voting was for, and at that time it was explained that it > was because of the NAME, not the approval of the package. (Yes, I know > there was no vote for the name either, I'm not sure why not, have to dig > into my IRC logs); but if you can come up with a better name, please > tell me. > > > Anyways if indeed the package is even redundant I am for removing it. > > However lets hear what Derick has to say about this and maybe we can > > find a solution to works for all parties involved. > > Removing is not an option, as > 1) it was downloaded by 267 people who might now depend on it [1] This cannot be a reason. Otherwise we would be setting a horrible precedent. > 2) I don't think it is redundant, as the code is very light, and about > 250% faster. (See attached test script), output: > > [derick@kossu mhac]$ php crypt_hmac.php > 11:16:07 > 11:16:29 > 11:17:25 > > The HMAC solution takes 22 seconds, the Message one 56 seconds. This starts to sound like a valid reason. > 3) the licence is not compatible with PEAR::Message (well, it's not the > same... ) > > [1] > http://pear.php.net/package-stats.php?pid=177&rid=&cid=6 I think this could be worked out seeing that each package only has one developer. Again 1) does is not a valid reason. However it does make it important to act quickly on this. What I would like to see is maybe see that the package is disabled for now. Users that already rely on this package would then need to mail pear-general if they need the source, but new users would not start to rely on the package if it may be removed. Regards, Lukas