Hi, I'd like to be sure I'm doing the right thing, hence this email. Are we following any rule for deciding whether or not a proposal should be applied or not? For example, I've just sent an email about modifying the XWiki constructor interface. Ludovic has answered but not other committers. Should I go ahead or do I need to wait for others to chime in? Are we following the ASF rules defined here: http://apache.org/foundation/how-it-works.html#management? However this is not enough. Here's an additional rule followed on most Apache projects: " The development process is intentionally lightweight; like other Apache projects, the committers decide which changes may be committed to the repository. Three +1 ('yes' votes) with no -1 ('no' votes or vetoes) are needed to approve a code change. For efficiency, some code changes from some contributors (e.g. feature additions, bug fixes) may be approved in advance, in which case they may be committed first and changed as needed, with conflicts resolved by majority vote of the committers. " Is this too heavyweight? Is it ok? What do you want? Should I wait before committing the xwiki constructor change I have proposed? The reason I'm asking this is because I am a new member of the community and I'm sure other newcomers may also have some doubt when they want to change existing code. Thanks -Vincent _____________________________________________________________________________ D�couvrez le nouveau Yahoo! Mail : 1 Go d'espace de stockage pour vos mails, photos et vid�os ! Cr�ez votre Yahoo! Mail sur http://fr.mail.yahoo.com
Hi Vincent, Most of us are new members, that I think the reason why we all wait for Ludovic advices and follow his decision. If we still want to follow eXtreme programming principles as Ludovic wanted in the beginning of the project, according to collective code ownership principle you should be able to commit your change as soon as they don't break existing tests and that you pair programmed them. As pair programming is not possible, and because we are too few to wait for 3 yes vote, I think that 1 or 2 vote should be enough.
Yes I agree with Alexis.. The way I see it: Rule 1: do not break tests. Only case is if you break it because of external stuff (build process, stuff like that). In this case it's allowed but should be fixed. Rule 2: ask for forgivness better than permission. It's good to inform and ask for other people's opinion because it educates the group about the development process and the code itself. Now it should not slow us down too much. We are not at a size where 3 votes is absolutely necessary. Of course I have the most experience with the code base, so there is a good chance other people won't yet see problems that could occur because of certain changes. But generally speaking I trust the judgement of the fine people in the group. If they would like something change I guess there are good reasons and it is a progress, so unless I see a big problem I'm ok with it. So I would say.. let's vote when we feel it is necessary to vote.. let's not make this mandatory. When you ask for opinions, if nobody screams, let's go ahead. Ludovic Alexis KARTMANN a écrit :
Hi Vincent, Most of us are new members, that I think the reason why we all wait for Ludovic advices and follow his decision. If we still want to follow eXtreme programming principles as Ludovic wanted in the beginning of the project, according to collective code ownership principle you should be able to commit your change as soon as they don't break existing tests and that you pair programmed them. As pair programming is not possible, and because we are too few to wait for 3 yes vote, I think that 1 or 2 vote should be enough.
-- Ludovic Dubost XPertNet: http://www.xpertnet.fr/ Blog: http://www.ludovic.org/blog/ XWiki: http://www.xwiki.com Skype: ldubost AIM: nvludo Yahoo: ludovic
-----Original Message----- From: Ludovic Dubost [mailto:[email protected]] Sent: mercredi 1 juin 2005 22:23 To: [email protected] Subject: Re: [xwiki-dev] [Process] Decision management
Yes I agree with Alexis.. The way I see it:
Rule 1: do not break tests. Only case is if you break it because of external stuff (build process, stuff like that). In this case it's allowed but should be fixed.
Rule 2: ask for forgivness better than permission. It's good to inform and ask for other people's opinion because it educates the group about the development process and the code itself. Now it should not slow us down too much. We are not at a size where 3 votes is absolutely necessary.
Of course I have the most experience with the code base, so there is a good chance other people won't yet see problems that could occur because of certain changes. But generally speaking I trust the judgement of the fine people in the group. If they would like something change I guess there are good reasons and it is a progress, so unless I see a big problem I'm ok with it.
So I would say.. let's vote when we feel it is necessary to vote.. let's not make this mandatory. When you ask for opinions, if nobody screams, let's go ahead.
Ok, my email wasn't clear. What you have said is exactly the way it happens at Apache for example. My question was to clarify the "if nobody screams". For example, I can send an email, wait 1 minute and then commit. Nobody has screamed. Then someone says, "hey hold on, I don't agree with this". Taking it back can take time, etc. But ok, I hear you. At this point you don't care too much. I'll continue doing what I've done so far by publishing my intent, wait a few hours and do it. Thanks -Vincent _____________________________________________________________________________ D�couvrez le nouveau Yahoo! Mail : 1 Go d'espace de stockage pour vos mails, photos et vid�os ! Cr�ez votre Yahoo! Mail sur http://fr.mail.yahoo.com
Hi Alexis, Ludovic is the creator sure but he's only one person and more and more people are contributing. I'm sure Ludovic wants a rule where committers are equal and that decisions be taken collectively. We just need to make the rules clear (hence my email). XP simply recommend collective code ownership. It does not mention anything about the process for deciding on code refactoring/proposals. When the change is simple and you master the topic you just go ahead. The question arise when you change an important public API (as is my case) or when you don't know enough about the code you're modifying. Thanks -Vincent _____ From: Alexis KARTMANN [mailto:[email protected]] Sent: mercredi 1 juin 2005 22:07 To: [email protected] Subject: Re: [xwiki-dev] [Process] Decision management Hi Vincent, Most of us are new members, that I think the reason why we all wait for Ludovic advices and follow his decision. If we still want to follow eXtreme programming principles as Ludovic wanted in the beginning of the project, according to collective code ownership principle you should be able to commit your change as soon as they don't break existing tests and that you pair programmed them. As pair programming is not possible, and because we are too few to wait for 3 yes vote, I think that 1 or 2 vote should be enough.
Yes indeed.. the public APIs are the most sensitive parts.. Basically anything in com.xpn.xwiki.api should be supplemented but not changed.. We have little control about which the code that uses these APIs so we would break existing applications.. We can of course depreciate APIs. The more collective our decisions are the better it is. Now it should not slow us down.. Trust is very important.. Basically it's about each person's judgement to feel when it's needed to ask. Ludovic Vincent Massol a écrit :
Hi Alexis,
Ludovic is the creator sure but he’s only one person and more and more people are contributing. I’m sure Ludovic wants a rule where committers are equal and that decisions be taken collectively. We just need to make the rules clear (hence my email).
XP simply recommend collective code ownership. It does not mention anything about the process for deciding on code refactoring/proposals. When the change is simple and you master the topic you just go ahead. The question arise when you change an important public API (as is my case) or when you don’t know enough about the code you’re modifying.
Thanks
-Vincent
------------------------------------------------------------------------
*From:* Alexis KARTMANN [mailto:[email protected]] *Sent:* mercredi 1 juin 2005 22:07 *To:* [email protected] *Subject:* Re: [xwiki-dev] [Process] Decision management
Hi Vincent,
Most of us are new members, that I think the reason why we all wait for Ludovic advices and follow his decision.
If we still want to follow eXtreme programming principles as Ludovic wanted in the beginning of the project, according to collective code ownership principle you should be able to commit your change as soon as they don't break existing tests and that you pair programmed them.
As pair programming is not possible, and because we are too few to wait for 3 yes vote, I think that 1 or 2 vote should be enough.
-- Ludovic Dubost XPertNet: http://www.xpertnet.fr/ Blog: http://www.ludovic.org/blog/ XWiki: http://www.xwiki.com Skype: ldubost AIM: nvludo Yahoo: ludovic
On 6/1/05, Ludovic Dubost <[email protected]> wrote:
Yes indeed.. the public APIs are the most sensitive parts.. Basically anything in com.xpn.xwiki.api should be supplemented but not changed.. We have little control about which the code that uses these APIs so we would break existing applications.. We can of course depreciate APIs.
The more collective our decisions are the better it is. Now it should not slow us down.. Trust is very important.. Basically it's about each person's judgement to feel when it's needed to ask.
For this kind of change I think the better choice is 3 +1 containing at least one from the contributors and one from the XpertNet staff. what do you think about this? -- http://www.sport42.net - http://fr.runnerslifestyle.com Tel : 06.84.50.33.49 skype: jeremi23
participants (4)
-
Alexis KARTMANN -
jeremi joslin -
Ludovic Dubost -
Vincent Massol