commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Gilles <gil...@harfang.homelinux.org>
Subject Re: [VOTE] Move Commons Math to TLP (again)...
Date Mon, 13 Jun 2016 21:11:57 GMT
Hi.

On Mon, 13 Jun 2016 01:30:20 +0100, Niall Pemberton wrote:
> On Sun, Jun 12, 2016 at 11:08 PM, Gilles 
> <gilles@harfang.homelinux.org>
> wrote:
>
>> On Sun, 12 Jun 2016 14:33:58 -0700, Dennis E. Hamilton wrote:
>>
>>> -1 (non-binding)
>>>
>>> Reason for objection:
>>>
>>>  I think the framing of this vote is confusing.
>>>
>>>  1. There appears to be less ability to go to TLP than there was at
>>> the time the previous motion passed.
>>>
>>>  2. The discussion (but not the [VOTE]) speaks of going to TLP via
>>> the incubator.  It has to be one or the other.  Propose a podling 
>>> to
>>> Incubator or propose a TLP to the Board.  There is no assurance 
>>> that a
>>> podling will graduate and it doesn't fit to make that a condition.
>>> One could raise the special circumstances at general-incubator, but 
>>> I
>>> think that works best with something specific (but malleable) in 
>>> hand.
>>>
>>>  3. The Incubator is reluctant to start podlings from scratch, as
>>> Niall observes.
>>>
>>
>> Could you please expand on how 3 Commons PMC members and 3 would-be
>> contributors are assimilated to "scratch"?
>>
>
> It would be good if all those wanting to be part of a Math TLP could
> indicate that here and cast a vote for a Math TLP. Including yourself
> Gilles, since so far I don't remember seeing whether you that you 
> were in
> favour of this.

In another thread, I indicated what was my preferred way to get
back on track after the fork announcement.

I know that I repeat myself but I am *totally* convinced that some
part of what was CM can become one or a few bona fide components.

Getting after this "low-hanging fruit" (meaning: release *early*)
can have a positive effect.
At the opposite, letting code that could be released rot until we
can fix all the issues in CM, is depressing.

And this is mostly independent of a larger-scale effort aimed at
creating a TLP.

It's a pity that the only way to make any decision about the CM
codebase, that does not amount to not deciding anything, seems
to depend on taking this code elsewhere.

Whether here at Commons, in a new TLP, or in the incubator, my
proposal would be the same: rebuild a project around the competences
available *now*.
Personally, I'll not participate in a project based on the messianic
belief that subject matter experts (a.k.a. "Mathematicians") will
suddenly figure out that they want to contribute to Commons Math.
[For better or worse, CM never worked that way.]

The question is: Why do some non-contributors try to force
contributors to continue releasing all the abandoned codebase?

I understand that some people in the Commons PMC might want to
evaluate the usefulness of each new component.
But why not discuss each one proposed, specifically?


Regards,
Gilles

> Niall
>
>
>
>>
>> Thanks,
>> Gilles
>>
>>
>>  4. It seems to me that the best first-step on whether incubation is
>>> feasible is to do the work to create an incubation proposal.  This
>>> will require certain key factors to be addressed.  Not the least is
>>> how the code base will be imported and, because it is from an 
>>> Apache
>>> Project, how it will be left behind too.  That definition can start
>>> here and then be refined on the general-incubator list where one 
>>> will
>>> need to find a champion (perhaps), mentors, and a sufficient body 
>>> of
>>> initial committers.  It is important for those who would form the
>>> initial core for a podling to learn enough about how incubation 
>>> works.
>>>
>>>  - Dennis
>>>
>>> Disclosure:
>>>
>>>  I have no idea how this might go.  I am not a Commons Math
>>> subject-matter expert, even though computational mathematics has 
>>> some
>>> appeal for me.  I still have my bound "Collect Algorithms from ACM,
>>> Volume 1: Algorithms 1-220."  I did not hold onto the microfiche of
>>> later algorithms that were published in conjunction with the ACM
>>> Transactions on Mathematical Software (TOMS). The latest (Algorithm
>>> 959) is interesting although I have no idea where to find the code 
>>> and
>>> am dismayed that it is a library under the GPL.
>>>
>>> -----Original Message-----
>>>> From: Niall Pemberton [mailto:niall.pemberton@gmail.com]
>>>> Sent: Sunday, June 12, 2016 11:56
>>>> To: Commons Developers List <dev@commons.apache.org>
>>>> Subject: Re: [VOTE] Move Commons Math to TLP (again)...
>>>>
>>>> On Sat, Jun 11, 2016 at 10:39 AM, James Carman
>>>> <james@carmanconsulting.com>
>>>> wrote:
>>>>
>>>> > We would take math through the incubator in order to build 
>>>> community
>>>> around
>>>> > it first. If we fail to do so, then we can decide its fate at 
>>>> that
>>>> time. We
>>>> > haven't done a good job attracting new people to math here at 
>>>> all. It
>>>> has
>>>> > always been maintained primarily by a select few.
>>>> >
>>>>
>>>> It made sense to me when there were 6 committers working on Math, 
>>>> but I
>>>> think given the exodus of most of those people to hipparchus then 
>>>> it
>>>> would
>>>> be better to wait a while for the dust to settle to see what 
>>>> happens
>>>> with
>>>> Math.
>>>>
>>>> I also don't think the incubator is a good place for starting a
>>>> community
>>>> from scratch (i.e. one or two man projects) - if you have a 
>>>> nucleus of
>>>> at
>>>> least a few people, then it has much better chance of success.
>>>>
>>>> So for me, I'm -1 unless there are enough Mathematicians who want 
>>>> to
>>>> work
>>>> on the code to give it a chance as an incubator project.
>>>>
>>>> Niall
>>>>
>>>>
>>>> >
>>>> > On Sat, Jun 11, 2016 at 1:36 AM Ralph Goers
>>>> <ralph.goers@dslextreme.com>
>>>> > wrote:
>>>> >
>>>> > > -1 (binding)
>>>> > >
>>>> > > At least until there are enough people to have a viable PMC.
>>>> > >
>>>> > > Ralph
>>>> > >
>>>> > > > On Jun 10, 2016, at 8:47 PM, James Carman
>>>> <james@carmanconsulting.com>
>>>> > > wrote:
>>>> > > >
>>>> > > > Since it has been suggested that the previously passing vote
>>>> should be
>>>> > > > voided, I propose we vote again to move Commons Math to a

>>>> TLP:
>>>> > > >
>>>> > > > +1 - Yes, move Commons Math to a TLP
>>>> > > > -1 - No, do not move Commons Math to a TLP
>>>> > > >
>>>> > > > The vote will remain open for 72 hours.
>>>> > > >
>>>> > > > Thank you,
>>>> > > >
>>>> > > > James Carman
>>>> > >
>>>> > >
>>>> > >


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


Mime
View raw message