PHP PEAR Questions and Answers - Commit to repository without authorization.

Author

Commit to repository without authorization.

David Coallier

2007-01-04, 7:02 pm

http://news.php.net/php.pear.cvs/43987

Refering to this commit, Bertrand added himself with explicitely no
authorization to commit to this package even though he has full karma.

He has asked BOTH Amir and I and we asked him a good reason to be lead
again (even though he did quit pear) and we said unless you have a
good patch or something good then we are not adding you back.

Even though we told him no, he used his full karma and commited
without permission to CVS.

Although I know this is simply adding himself back as an inactive
lead, this is, imho, ridiculous if everyone starts coming back from
their "retirement" and adding wherever they want.

They Bertrand stated that I have done the same before:
http://cvs.php.net/viewvc.cgi/pear/...r1=1.76&r2=1.77

The only difference is that I had permission from the package lead.

No matter this is not speaking of me, the issue here is that rules are
being respected, please use correct authority to resolve this
situation please.

Thanks for taking care of this issue.

--
David Coallier,
Founder & Software Architect,
Agora Production (http://agoraproduction.com)
51.42.06.70.18

Amir Mohammad Saied

2007-01-04, 7:02 pm

Hi,

On Thu, 4 Jan 2007 17:52:48 -0500
davidc@php.net ("David Coallier") wrote:

> http://news.php.net/php.pear.cvs/43987
>
> Refering to this commit, Bertrand added himself with explicitely no
> authorization to commit to this package even though he has full karma.
>
> He has asked BOTH Amir and I and we asked him a good reason to be lead
> again (even though he did quit pear) and we said unless you have a
> good patch or something good then we are not adding you back.
>
> Even though we told him no, he used his full karma and commited
> without permission to CVS.
>
> Although I know this is simply adding himself back as an inactive
> lead, this is, imho, ridiculous if everyone starts coming back from
> their "retirement" and adding wherever they want.

>
> They Bertrand stated that I have done the same before:
> http://cvs.php.net/viewvc.cgi/pear/...r1=1.76&r2=1.77
>
> The only difference is that I had permission from the package lead.

I Confirm it.

>
> No matter this is not speaking of me, the issue here is that rules are
> being respected, please use correct authority to resolve this
> situation please.
>
> Thanks for taking care of this issue.
>
> --
> David Coallier,
> Founder & Software Architect,
> Agora Production (http://agoraproduction.com)
> 51.42.06.70.18

Gregory Beaver

2007-01-04, 7:02 pm

David Coallier wrote:
> http://news.php.net/php.pear.cvs/43987
>
> Refering to this commit, Bertrand added himself with explicitely no
> authorization to commit to this package even though he has full karma.
>
> He has asked BOTH Amir and I and we asked him a good reason to be lead
> again (even though he did quit pear) and we said unless you have a
> good patch or something good then we are not adding you back.
>
> Even though we told him no, he used his full karma and commited
> without permission to CVS.
>
> Although I know this is simply adding himself back as an inactive
> lead, this is, imho, ridiculous if everyone starts coming back from
> their "retirement" and adding wherever they want.
>
> They Bertrand stated that I have done the same before:
> http://cvs.php.net/viewvc.cgi/pear/...r1=1.76&r2=1.77
>
> The only difference is that I had permission from the package lead.
>
> No matter this is not speaking of me, the issue here is that rules are
> being respected, please use correct authority to resolve this
> situation please.
>
> Thanks for taking care of this issue.

Hi Amir and David,

I am taking pear-dev off of this, as it is strictly a QA concern.
Without having seen the interaction between you two and Bertrand Gugger
(toggg), it is hard to know what the conversation was like.

Although defying the request not to commit is always a bad thing, in
this case, the commit itself is harmless and accurate. Bertrand Gugger
was indeed one of the leads [1], and is indeed inactive. He also added
a note that as Amir does not want him as a lead, he won't be active,
assuming my "translation" is accurate.

I would recommend that in this case, because of the mitigating factors
above, an official warning should be given that Bertrand's cvs karma
privileges will be revoked if committing again without permission. Is
this solution acceptable to both of you?

Greg

[1] http://cvs.php.net/viewvc.cgi/pear/...r1=1.63&r2=1.64

David Coallier

2007-01-04, 10:02 pm

On 1/4/07, Gregory Beaver <greg@chiaraquartet.net> wrote:
> David Coallier wrote:
>
> Hi Amir and David,
>
> I am taking pear-dev off of this, as it is strictly a QA concern.
> Without having seen the interaction between you two and Bertrand Gugger
> (toggg), it is hard to know what the conversation was like.
>
> Although defying the request not to commit is always a bad thing, in
> this case, the commit itself is harmless and accurate. Bertrand Gugger
> was indeed one of the leads [1], and is indeed inactive. He also added
> a note that as Amir does not want him as a lead, he won't be active,
> assuming my "translation" is accurate.
>
> I would recommend that in this case, because of the mitigating factors
> above, an official warning should be given that Bertrand's cvs karma
> privileges will be revoked if committing again without permission. Is
> this solution acceptable to both of you?
>
> Greg
>
> [1] http://cvs.php.net/viewvc.cgi/pear/...r1=1.63&r2=1.64
>

Yes that sounds ok.

The thing is he quit, so active or not he left the package, then he
said he needed to be added back for a job he wanted to have.

We proposed him to add him back even on the main page if he does a
public apology to the mailing list but he of course refused..

Although, this solution sounds ok to me. But we have to make sure that
if this happens again, he's revoked..

--
David Coallier,
Founder & Software Architect,
Agora Production (http://agoraproduction.com)
51.42.06.70.18

Amir Mohammad Saied

2007-01-05, 4:03 am

On Thu, 04 Jan 2007 18:06:38 -0600
Gregory Beaver <greg@chiaraquartet.net> wrote:

> David Coallier wrote:
>
> Hi Amir and David,
>
> I am taking pear-dev off of this, as it is strictly a QA concern.
> Without having seen the interaction between you two and Bertrand Gugger
> (toggg), it is hard to know what the conversation was like.
>
> Although defying the request not to commit is always a bad thing, in
> this case, the commit itself is harmless and accurate. Bertrand Gugger
> was indeed one of the leads [1], and is indeed inactive. He also added
> a note that as Amir does not want him as a lead, he won't be active,
> assuming my "translation" is accurate.
>
> I would recommend that in this case, because of the mitigating factors
> above, an official warning should be given that Bertrand's cvs karma
> privileges will be revoked if committing again without permission. Is
> this solution acceptable to both of you?
>
> Greg
>
> [1] http://cvs.php.net/viewvc.cgi/pear/...r1=1.63&r2=1.64

Hi,

Well, there's no doubt that "Bertrand Gugger" has did a lot for this
package and we all appreciate it, and I'm not against his back as
lead, but I just told him it has bureaucracy. He quit once, and I just
told him, if you want to return show us some codes or even ideas
(opening a bug/feature-request) and you'll be the lead then.

But he did none of them and just went his way and added himself as a
lead in a non pleasant way (although the comment that I'm against his
back as lead couldn't be true), he could easily open a bug and put what
he has wrote as commit log (... I get suspicious about pcre ...) into
it, then there was no reason for furthermore wars and annoyances.

The solution sounds OK to me, not it toggg's case, but for every one
with privileges who touch other developers files w/o permission or
strong reasons.

-- Amir Mohammad

Sponsored Links