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][Math] New component: "Commons Geometry"?
Date Fri, 01 Dec 2017 13:26:54 GMT
Hello Amey.

On Thu, 30 Nov 2017 23:45:45 +0530, Amey Jadiye wrote:
> Pardon me for pulling this thread up again, I havent read anything 
> about
> "Commons Geometry" since long

Thanks for your renewed interest.

> (or may be I missed any other disscussion? ).

Probably not.

> is someone working on this ?

It would be a surprise.

> what is the final decision ?

There hasn't been any progress towards a decision.
There isn't even a consensus on one of the central tenets of
Apache ("Those who do the work..."): how sad/strange (?).

> I'm having good
> amount of time to spend on this now, appreciate If someone direct me 
> to
> correct disscussion thread

IIRC, the one below is where we left off...

> I think I can help here.

Thanks for the offer!

> It took me half hour to read all old mails but dont see final 
> verdict,
> though I was in favour with Maven modules but after reading all again 
> I
> think Gilles approch is more practicle here and If no one is working 
> I can
> submit something to review.

IMHO, the priority would be to review the status of "Numbers"
(i.e. what is preventing a first release?).

Best regards,
Gilles


> Regards,
> Amey
>
> On Wed, Sep 13, 2017 at 4:44 AM, Gilles 
> <gilles@harfang.homelinux.org>
> wrote:
>
>> On Tue, 12 Sep 2017 13:07:24 +0200, Jochen Wiedmann wrote:
>>
>>> On Sat, Sep 2, 2017 at 12:50 AM, Gilles 
>>> <gilles@harfang.homelinux.org>
>>> wrote:
>>>
>>> Because of "Commons" rules, it is not "equivalent": There was
>>>> a long thread concluding that all modules must be released
>>>> _together_, and with the same top-level package name and version
>>>> number.
>>>> It is very "maintainer(s)-unfriendly" because of the quite
>>>> different subject matters that coexist in CM.
>>>>
>>>
>>> I wouldn't count that rule "*all* modules must be released" as a 
>>> mantra:
>>>
>>
>> I found the idea attractive, but Stian (link to older discussion
>> in a previous post) advised that maven would not easily "support"
>> it.
>>
>> Has that changed since the discussion took place (10 months ago)?
>>
>> a) In case of an emergency release (fixing a CVE, for example), I'd
>>> clearly consider pushing out the module as more important than 
>>> waiting
>>> for a full release. (Of course, one must be careful to maintain
>>> compatibility when pushing out just a module, but that goes without
>>> saying.)
>>> b) I'd like to hear others experiences on that topic (maybe VFS).
>>> Anyways, my personal experiences with Rat are clear: Releasing 
>>> *all*
>>> together is causing nothing but pain, and tends to defer releases
>>> indefinitely. OTOH, releasing a submodule can be done at all times,
>>> and without overly much preparation.
>>>
>>> In conclusion, I'd definitely support the release of a single
>>> submodule, if the need would arise.
>>>
>>
>> How can one reconcile what you say here with what was said in
>> that old thread?
>>
>> Would the PMC accept that a component contains independent modules
>> (where "independent" means that each module can have its own version
>> number, irrespective of the component's version)?
>>
>> Arguably (cf. thread referred to above), a "Commons" component
>> should be simple enough that multiple versions are not necessary.
>> [Chorus:] This is not the case with "Commons Math", hence separate
>> components for independent contents (such as "Geometry", "RNG",
>> "Numbers" and "SigProc") is the simplest solution.
>>
>> Gilles
>>
>> Jochen
>>>


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


Mime
View raw message