Re: An alternate view of PEAR QA

From: Date: Mon, 19 Jan 2004 20:14:54 +0000
Subject: Re: An alternate view of PEAR QA
References: 1 2  Groups: php.pear.qa 
Request: Send a blank email to pear-qa+get-379@lists.php.net to get a copy of this message
On Mon, Jan 19, 2004 at 10:03:38AM +0100, Lukas Smith wrote: > The reason why I proposed a different system was not because your > proposal is not what we ideally want, but that its simply unrealistic > with the means we have today. We dont have enough people that will > dedicate their time to PEAR QA. Furthermore packages dont have test > suites (and we havent made that a requirement either). Packages dont > even have to reside in any CVS atm. Thats why I dont see that we can > really monitor the changes, test the changes and provide feedback within > a reasonable timeframe. My proposed system intentionally has no connection to CVS. That's why I emphasized the fact that the QA process is release-based. From my original email: In order to achieve the best results, quality assurance should operate on a per-release. QA members should not be concerned with development-specific systems such as CVS. Instead, their view into the package world should focus on pearweb's interface. (http://news.php.net/article.php?group=php.pear.qa&article=368) > The system I proposed is more geared to be realistic solution and not so > much the ideal. However we shouldnt give up that ideal and maybe one by > one we will be able to increase our standards to also require test > suites, everything in (atleast synced into) PEAR CVS and tons of willing > and able people that will make sure that this QA process doesnt turn > into a hassle but more into a great source of people making sure that > your package works as intended. I'm not trying to be unrealistic, but I don't think there's another way to perform proper quality assurance. For reference, here are the goals (http://qa.php.net/projects.php) of the PHP QA team: - Release Candidate Build - Automated Smoke Tests - Automated QA Checklists - Automated Regression Tests If we're really just talking about forming a group whose job is to be on alert for potential problems that occur in released packages, it's no longer a quality assurance group; instead, it's a live maintainance group. -- Jon Parise (jon@php.net) :: The PHP Project (http://www.php.net/)

« previous php.pear.qa (#379) next »