Re: an HTML_Template_IT[X] fork is available

From: Date: Wed, 12 Feb 2003 09:19:38 +0000
Subject: Re: an HTML_Template_IT[X] fork is available
References: 1 2 3  Groups: php.pear.dev 
Request: Send a blank email to pear-dev+get-13161@lists.php.net to get a copy of this message
12 Февраль 2003 01:28, Markus Wolff написал: > > Christian Dickmann wrote: > > > We don't have to prove, that your new package breaks BC, > > > you have to prove, that it does not! > > > > This is a *perfect* example of faulty logic. > > > > Consider: I don't have to prove that the world was created just > > yesterday by The Great Almighty Foobar. You have to prove that it was > > not! > > The earth has existed for longer than a day, otherwise I wouldn´t be > turning 30 by the end of this year (sigh). The majority of people living > on earth are under the impression that they are older than one day. > > As you represent the minority of one-day-believers and deliver a theory > that´s brand-new, it´s your turn to prove all the people who are older > than one day wrong. That's simple: the Great Almigthy Foobar is almighty and created you with all the memories in your head. ;) Now back on topic. I am a mathematician, and learned quite well that you just don't prove negative statements. I can't prove that my package does not break BC, for I have to iterate over *all* possible templates to do this. If we have either specification or regression tests, I will be able to prove a *positive* statement: a) My package conforms to specification --- or --- b) My package passes regression tests. As we currently don't have them, the only remaining option is to actually test my package with *some* of the existing templates, and I urge you to do so. I did post the links to classes already. > On the other hand, proof is not really the point here - any application > developer is free to keep using the old version. And if that works for > him, why shouldn´t he? I tend to bundle a specific version of PEAR > classes with my apps anyway to prevent some system administrator to > break my app by upgrading to a newer PEAR version that will break BC - > and I would advise any other developer to do the same. There were lots of BC-breaking changes in lots of PEAR packages, some of which were widely used. That's what major version changes are for...

« previous php.pear.dev (#13161) next »