commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Gilles <gil...@harfang.homelinux.org>
Subject Re: [ALL] The Commons Math issue (Was: [DRAFT] Board report)
Date Tue, 11 Apr 2017 22:33:20 GMT
Hi Gary.

On Tue, 11 Apr 2017 12:56:09 -0700, Gary Gregory wrote:
> On Tue, Apr 11, 2017 at 3:54 AM, Gilles 
> <gilles@harfang.homelinux.org>
> wrote:
>
>> Hello.
>>
>> Gary,
>>
>> Thank you for mentioning the CM issue.
>> But...
>>
>> [...]
>>>
>>> ## Health report:
>>>  [...]
>>>  - We are still experiencing some growing pains toward Commons Math 
>>> 4.
>>> There
>>>    is a backlog of issues in JIRA (not unlike other components) but 
>>> no
>>> clear
>>>    concensus in the community. Do-ocracy is likely to prevail.
>>>
>>
>> ... This being an Apache community, is everyone at ease with using
>> "likely" in the above sentence?
>>
>>    There is a
>>>    proposal on how to move the code base forward and having Math 
>>> depend on
>>>    Commons Numbers and Commons RNG, and then dropping the 
>>> corresonding
>>>    Math code. All of which can be seen on the developer's mailing 
>>> list.
>>>
>>
>> Several board reports have mentioned the "growing pains" and lack of
>> of consensus.
>> At what point are they going to reflect that the problem must be 
>> dealt
>> with?
>>
>
> If you are looking to involve the board, then you are asking for a
> sledgehammer. I do not see any rules or guidelines being broken, or 
> the
> Apache Way being blocked. I do not see egregious behavior or 
> disrespectful
> communication. I see busy people doing what they want when the want 
> in all
> of Commons, respectfully and diligently.

I do not make any generalization, but I asked a simple question.
Unless I'm mistaken, no PMC member answered.

>>
>> PMC members do not have time to contribute to all components, but is
>> it fine that this obvious fact prevents further development of some
>> components (including _parts_ of current CM) that could be 
>> maintained
>> with much less manpower?  [IMHO, the development of "Commons RNG"
>> somewhat proved that last point.]
>>
>
> For me, the PMC's main job is to vote on releases. PMC members do not 
> have
> to code or participate in the bits and bytes of all components. Sure, 
> it
> helps to know that some PMC members agree with one's direction, which
> should facilitate pushing releases through. We all know the voting 
> rules by
> this point, the barrier to release is not very high IMO.

The barrier might not be high but it is sufficient to prevent the
viability of the CM codebase.
I know that my opinion is dismissed by some PMC members, but the
issue is that the "do-ocracy" is not applied.

>
> Commons is more open than any other Apache Project, anyone with 
> commit
> rights can commit to Commons. That's pretty open. Our job is to 
> attract
> like minded folks for the components we need help.

I've noted that ignoring bug reporters, not keeping codes up-to-date,
and not making releases (even though work has been done) is damaging.
No acknowledgment, no alternative proposed.

>>
>> Must do-ocracy prevail or not?  And, if not, why?  [Let the board
>> answer if the PMC members won't.]
>>
>
> Yes doing the work gets you most of way there. Proposing a path 
> forward and
> engaging fellow developers gets you most of the way there. Hopefully 
> PMC
> members can offer some wisdom along the way and point out what they 
> think
> is best. No one is malicious here. But, we all have POVs of course 
> ;-)

Opinions opposing mine have been given.
But no action taken.

> Discussions, building consensus, over time, is what I feel will get 
> Math 4
> out of the mud.

Slogging through mud is an apt image!
For at least 14 months.
Again, development is possible, but the PMC does not support it,
which should be its role IMO.

>> My proposal concerning the future of the CM codebase has been on the
>> table for almost a year, without anyone else working on the 
>> alternative
>> that had been discussed (that is, keep working on CM "as-is"), as 
>> can
>> be viewed from the lack of communication with users (cf. JIRA 
>> issues).
>>
>> Readers of the ML could infer that I tried to come closer to a
>> "compromise", by suggesting a release of CM 4.0 (thus mitigating my
>> disappointment that a large part of that code would still have no
>> support from any of the "Commons" active developers).
>> However, the compromise only makes sense if those on the other side
>> of argument will participate in its implementation.
>>
>
> I appreciate you efforts over time. Let's keep things moving. 
> Hopefully
> more folks will become engaged.

Wishful thinking.
And then?

How much more "efforts" must I put in order to deserve the
application of "do-ocracy" (in the sense that the PMC members
_support_ work being done, even if the route taken is not
their preferred choice)?

Everyone is entitled to his opinions.
What I ask is whether, in this particular situation, the PMC
is doing all it could/should.
I know that I cannot request time or interest from anyone; but
isn't the role of a PMC to solve a _project management_ issue?

I wish the report to suggest that advice from outside "Commons"
might be useful to get "things moving".


Thanks,
Gilles

>
> Gary
>
>
>>
>> Thanks for your attention,
>> Gilles
>>
>>
>>> [...]
>>>
>>


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org


Mime
View raw message