Re: Removing packages without DESTDIR support

[Aleksej Saushev <asau%inbox.ru@localhost>]

Jonathan Perkin <jperkin%joyent.com@localhost> writes:

> * On 2013-10-09 at 17:08 BST, Aleksej Saushev wrote:
>
>> Joerg Sonnenberger <joerg%britannica.bec.de@localhost> writes:
>> 
>> > On Wed, Oct 09, 2013 at 12:14:03AM +0400, Aleksej Saushev wrote:
>> >> I strongly object to removal of packages on arbitrary grounds.
>> >
>> > It has been discussed often enough. It is far from arbitrary.
>> 
>> Yes, it has been discussed previously, and you have given the same
>> arbitrary reasons.
>> 
>> > You have had enough time to fix packages you are interested in.
>> 
>> You have had enough time to "fix" them as well.
>
> In a voluntocracy it is up to those who care about these packages to
> fix them so that they work correctly, otherwise they should be moved
> to wip.
>
> The current situation is untenable, and very confusing to users - it
> took me long enough and a number of blind alleys before figuring out
> that PKG_DESTDIR_SUPPORT=none was the cause of the problems I was
> seeing.

Sorry? I have just typed "make install" in lang/ghc, and it built.
I have typed "make package", and it has built binary package.
Thus these packages (at least GHC) work correctly.

In general, I find such reasoning (<<you had time to "fix" what is not
broken>>) fundamentally wrong. It makes you appear as if you consider pkgsrc
your personal playground. There're some packages you care about, and
everything else is just a ballast to make some sort of advertising
by appeal to number of packages. Except most users don't care of number
of packages, they care of whether important ones are present or not.
At least GHC and R are very important in modern world, if you throw
them away, you discourage others from using pkgsrc.

In addition, do you think that it is polite to introduce changes that
don't bring any real value but break code that works for all others?


-- 
BCE HA MOPE!