I can no longer be the maintainer of ASDF unless I gain some control over the contents of the repository.
Right now, I am just reacting to changes other people are making. This is ok if I am in the loop, but if other developers are simply wading into the repository and committing changes that I have no idea are coming, then I cannot do my job.
The last straw for me was last night's 3.1.7.23 commits, which completely blind-sided me and also invalidated work-in-progress I had outstanding.
So, we need to either
1. develop some informal means of clearing coming commits through me, perhaps using this mailing list;
2. lock out other developers and change the main repository only through merge requests (this is not my favored alternative, as it requires more busy work on my part); or
3. get a new maintainer.
I decline to continue under the status quo.
On Fri, 2016-09-16 at 12:23 -0500, Robert Goldman wrote:
I can no longer be the maintainer of ASDF unless I gain some control over the contents of the repository.
Right now, I am just reacting to changes other people are making. This is ok if I am in the loop, but if other developers are simply wading into the repository and committing changes that I have no idea are coming, then I cannot do my job.
The last straw for me was last night's 3.1.7.23 commits, which completely blind-sided me and also invalidated work-in-progress I had outstanding.
So, we need to either
- develop some informal means of clearing coming commits through me,
perhaps using this mailing list;
- lock out other developers and change the main repository only
through merge requests (this is not my favored alternative, as it requires more busy work on my part); or
I think this would be the best choice. ASDF is no longer in need of quick changes so delaying commits and pondering longer by way of a code review would be better.
That's all entirely my fault. I should have remembered this work conflicted with a branch you had.
I got carried away. I lazily continued to act the way I always did (i.e. push any supposed improvement to master), even though I'm not maintainer anymore.
I'm backing out, and will not commit anything myself directly to master anymore. I will only put merge requests on gitlab for code review and merge by Robert (or whoever steps forward to replace him).
Robert: I can revert the offending commit, or merge in yours, etc., at your leisure.
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org I have not yet begun to procrastinate
On Fri, Sep 16, 2016 at 6:38 PM, Stelian Ionescu sionescu@cddr.org wrote:
On Fri, 2016-09-16 at 12:23 -0500, Robert Goldman wrote:
I can no longer be the maintainer of ASDF unless I gain some control over the contents of the repository.
Right now, I am just reacting to changes other people are making. This is ok if I am in the loop, but if other developers are simply wading into the repository and committing changes that I have no idea are coming, then I cannot do my job.
The last straw for me was last night's 3.1.7.23 commits, which completely blind-sided me and also invalidated work-in-progress I had outstanding.
So, we need to either
- develop some informal means of clearing coming commits through me,
perhaps using this mailing list;
- lock out other developers and change the main repository only
through merge requests (this is not my favored alternative, as it requires more busy work on my part); or
I think this would be the best choice. ASDF is no longer in need of quick changes so delaying commits and pondering longer by way of a code review would be better.
-- Stelian Ionescu a.k.a. fe[nl]ix Quidquid latine dictum sit, altum videtur.
On 9/16/16 Sep 16 -5:53 PM, Faré wrote:
That's all entirely my fault. I should have remembered this work conflicted with a branch you had.
I got carried away. I lazily continued to act the way I always did (i.e. push any supposed improvement to master), even though I'm not maintainer anymore.
I'm backing out, and will not commit anything myself directly to master anymore. I will only put merge requests on gitlab for code review and merge by Robert (or whoever steps forward to replace him).
Robert: I can revert the offending commit, or merge in yours, etc., at your leisure.
I don't think we need to revert anything right now. But I need to do a review of the latest changes and see how to harmonize them with my UNSUPPORTED-FUNCTIONALITY-ERROR (I am not wedded to my name).
But I would like a brief pause so I can catch up and get results from testing on all three platforms, then a more measured move forward.
Having merge requests for a while will, I hope, be a good change and not too burdensome for anyone.
Good night, r
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org I have not yet begun to procrastinate
On Fri, Sep 16, 2016 at 6:38 PM, Stelian Ionescu sionescu@cddr.org wrote:
On Fri, 2016-09-16 at 12:23 -0500, Robert Goldman wrote:
I can no longer be the maintainer of ASDF unless I gain some control over the contents of the repository.
Right now, I am just reacting to changes other people are making. This is ok if I am in the loop, but if other developers are simply wading into the repository and committing changes that I have no idea are coming, then I cannot do my job.
The last straw for me was last night's 3.1.7.23 commits, which completely blind-sided me and also invalidated work-in-progress I had outstanding.
So, we need to either
- develop some informal means of clearing coming commits through me,
perhaps using this mailing list;
- lock out other developers and change the main repository only
through merge requests (this is not my favored alternative, as it requires more busy work on my part); or
I think this would be the best choice. ASDF is no longer in need of quick changes so delaying commits and pondering longer by way of a code review would be better.
-- Stelian Ionescu a.k.a. fe[nl]ix Quidquid latine dictum sit, altum videtur.
Robert: I can revert the offending commit, or merge in yours, etc., at your leisure.
I don't think we need to revert anything right now. But I need to do a review of the latest changes and see how to harmonize them with my UNSUPPORTED-FUNCTIONALITY-ERROR (I am not wedded to my name).
But I would like a brief pause so I can catch up and get results from testing on all three platforms, then a more measured move forward.
I was just, in my hurry, trying to unblock Elias, trying to throw together a quick functional interface.
Having merge requests for a while will, I hope, be a good change and not too burdensome for anyone.
There are a few merge requests pending already. So far, Elias and I (the only active developers) assume that you are to review and merge them. If you want to delegate more powers to us (e.g. review and merge each other's patches within some limits you define), please tell us your policies.
I believe all current merge requests (at least those not marked WIP) should be completed and merged in before release: !18 !21 !22 !23.
PS: After it appear that Daniel K was misconstruing the intent of many ASDF functions, I decided that I ought to do a pass of documentation of the code in asdf/defsystem. I'll open a merge request for this WIP; I already found a few mistakes to cleanup (some recent, some old), and started !23 for that.
—♯ƒ • François-René ÐVB Rideau •Reflection&Cybernethics• http://fare.tunes.org If freedom makes you think of antisocial behavior, harming other people and not helping anyone — maybe the problem is with you, not freedom.
-
Faré -
Robert Goldman -
Stelian Ionescu