Re: Request to add the SQL Server PHP extensions to PECL

On Fri, 11 Feb 2011 11:31:18 -0000, Pierre Joye <pierre.php@gmail.com> wrote:

On Thu, Feb 10, 2011 at 11:31 PM, Ashay Chaudhary <ashayc@microsoft.com> wrote:

We have received numerous requests from the PHP community to put this extension into PECL and this request is a direct response to that feedback. In addition to the extension source code, we are ready to include tests and documentation to PECL and actively maintain the project.

Right, but "put an extension in PECL" usually comes with a couple of things, they are not requirements but they are key part of a successful contribution. One of them is an open repository. I'm sad that you did not get the OK to do it.

[...]

It is not a problem to keep the code repository where it is now (codeplex) and do the releases through pecl. But we won't create read only code repositories under svn.php.net. [...] I however have to say that it is somehow bad to do this proposal under this circumstance, it is against the spirit of pecl and the way we should work.

It appears I'm the only dissenting voice here. What I read you and the others are saying is: 1) It is OK to keep the code repository in codeplex (and I assume also in github, like mongo). 2) Is is inadmissible to keep a read-only dump-on-release repository on PHP's SVN repository. 3) It is inadmissible to keep a read-only dump-on-release repository on codeplex or wherever. 4) It is irrelevant who can commit to the repository. In sum, where the code is is irrelevant as long as the repository where development is made is public, as that is not "truly open". Basically, this is a rejection of the Cathedral model. The Cathedral model may be outdated but saying it isn't "open" or is against free software principles is a non-sense. I'd like to reiterate having (even read-only) projects in PHP's repository is beneficial because it allows us to keep track which APIs are used and how they're used. -- Gustavo Lopes