Re: [OctDev] moving Octave Forge mailing list to core's mailman server

[Juan Pablo Carbajal]

On Mon, Nov 26, 2012 at 1:45 AM, Carnë Draug <address@hidden> wrote:
> On 26 November 2012 01:01, Daniel J Sebald <address@hidden> wrote:
>> On 11/25/2012 04:10 PM, Carnė Draug wrote:
>>>
>>> On 25 November 2012 21:44, Daniel J Sebald<address@hidden>  wrote:
>>> At the moment, the decision whether a thread belongs to the help or
>>> octave-dev mailing list is whether the reply is "use package X from
>>> octave forge". I'll argue that most Octave users already use at least
>>> one of the Octave Forge packages. And I'll also argue that no one in
>>> Octave Forge uses all the Octave Forge packages. So if the question is
>>> how to use a function from an Octave Forge package, users on the help
>>> mailing list already are the right people to answer it. Keeping them
>>> separated makes no sense anymore.
>>
>> So there will be changes to the Octave webpage descriptions that
>> consequently (or at least intend to) direct the bulk of OctDev to the
>> "address@hidden" mailing list?
>
> Yes. That's why this is being discussed in the maintainers mailing list.
>
>>>>> There's plenty of applications and packages for Octave that are not
>>>>> part of Forge.
>>>>
>>>>
>>>> That doesn't mean Octave Forge isn't primarily about packages and
>>>> applications.
>>>
>>>
>>> What is this applications you keep talking about? There's only packages.
>>
>> You are thinking of applications as in hunk of software, I suspect.  I'm
>> speaking in terms of applied science, e.g., signal processing, civil
>> engineering, image processing, statistics.
>
> Damn you homophones. Causing trouble since monkeys learned to talk.
>
>>>> Yes and no.  I often see discussions of bugs.  Some bugs are
>>>> straightforward
>>>> and remain on the tracker.  Some are either vague and difficult to solve
>>>> and
>>>> warrant help from others, hence discussion list.  Some bugs expose an
>>>> underlying weakness in design and warrant discussion about design
>>>> modifications.
>>>
>>>
>>> That may be true in core. I do not remember that ever happening in
>>> forge. Considering the way development is done in Forge, I wouldn't
>>> consider this to ever be a problem.
>>
>>
>> "install package" would be the conceptual development there--now stable.
>
> "install package" would already belong to the maintainers mailing list
> since it's handled by pkg, itself part of core. It is, however, a very
> good example of a maintainers discussion that developers of forge
> should be involved.
>
>>> Yes it is. Not one big change though, but slowly slowly seems to be
>>> the direction it's taking. It doesn't make sense to make that question
>>> yet, maybe it never will. But in the mean time, when things start to
>>> overlap, such as in the case of the mailing lists, it makes sense to
>>> merge them. We are not discussing more than just that, mailing lists.
>>
>>
>> Getting rid of an active mailing list is more than a name change.  That
>> traffic has to go somewhere.  I doubt the package concept is going away.
>
> We are merging 3 mailing lists, whose subjects have been overlapping
> too much and too often, into 2.

I do agree with Carnë idea. In particular with the refinement proposed
by jwe were everything gets merged to the current mailing lists.

I do not really understand, the complication observed or proposed by
Daniel (no ofense!). I think the issue is quite simple, so a simple
solution should be enough.

Cheers