On Thu, 11 Apr 2013, Frank Shearar wrote:

> Well, to be fair this is the right time to push longer term
> work-in-progress stuff in trunk. Frustrated is my usual state :)

Code which has known issues should go into the Inbox. The Trunk is for 
code which is ready for consumption. It doesn't mean that work-in-progress 
stuff shouldn't be in the Trunk, but it should be done in a way that 
doesn't break stuff intentionally, unless it's unavoidable.

http://squeakboard.wordpress.com/2009/07/02/a-new-community-development-model/


Levente

>
> frank
>
> On 11 April 2013 21:43, Nicolas Cellier
> <nicolas.cellier.aka.nice at gmail.com> wrote:
>> I understand how this can be frustrating for Frank :(
>> His efforts for having the CI server working could supposedly be rewarded by
>> correction of errors and regressions soon after their introduction...
>> Unfortunately, it's precisely the time chosen by Bert and Colin for
>> introducing longer term work in progress in trunk.
>> We are less patient than we used to be :)
>>
>>
>> 2013/4/10 Bert Freudenberg <bert at freudenbergs.de>
>>>
>>> On 10.04.2013, at 05:45, Frank Shearar <frank.shearar at gmail.com> wrote:
>>>
>>>> On 10 April 2013 13:42, Frank Shearar <frank.shearar at gmail.com> wrote:
>>>>> On 11 March 2013 00:11,  <commits at source.squeak.org> wrote:
>>>>>> Bert Freudenberg uploaded a new version of EToys to project The Trunk:
>>>>>> http://source.squeak.org/trunk/EToys-bf.94.mcz
>>>>>>
>>>>>> ==================== Summary ====================
>>>>>>
>>>>>> Name: EToys-bf.94
>>>>>> Author: bf
>>>>>> Time: 11 March 2013, 1:11:38.945 am
>>>>>> UUID: 4abe21cf-5215-40c9-8371-07b6a9e09b81
>>>>>> Ancestors: EToys-fbs.93
>>>>>>
>>>>>> Bring in some classes and methods from the Etoys image so that project
>>>>>> loading gives a meaningful error, not just 'unknown class'. See
>>>>>> ReleaseBuilderSqueakland>>buildInitialScreen
>>>>>>
>>>>>> =============== Diff against EToys-fbs.93 ===============
>>>>>
>>>>> This has several missing classes, I just discovered: SExpElement,
>>>>> EToyExpressionTransformer2 and SugarPropertiesNotification.
>>>>>
>>>>> I only just noticed now because
>>>>> Tests.Release.ReleaseTest.testUndeclared often fails for spurious
>>>>> reasons (discussed at length elsewhere).
>>>>
>>>> I hit "send" too soon. It bothers me that Bert didn't get told that he
>>>> was about to add missing class definitions. I understand that they
>>>> weren't missing in his image, but it'd be really nice if something
>>>> somewhere could have told Bert "you've forgotten some classes". The
>>>> current "commit and see what CI says" mechanism is suboptimal.
>>>
>>>
>>> They're not in my image either. The mechanism to inform me is the
>>> undeclared warning that gets logged to the transcript. And I ignored it
>>> because this is work in progress.
>>>
>>> - Bert -
>>>
>>>
>>>
>>
>>
>>
>>
>
>