[aur-general] Moving noip from [community] to AUR
Ido Rosen
ido at kernel.org
Mon Aug 4 00:12:41 EDT 2014
On Sun, Aug 3, 2014 at 2:48 PM, Alexander Rødseth <rodseth at gmail.com> wrote:
> Hi Ido,
>
>
> 2014-08-02 17:41 GMT+02:00 Ido Rosen <ido at kernel.org>:
>> The noip client is GPL licensed F/OSS software (it's right there in
>> the COPYING file in the tarball). Just because it connects to a
>> commercial service does not mean it deserves to be removed from the
>> official package repository. Take, for example, various GMail
>> clients, Chromium (re:Google Sync and various google services),
>> various AWS/EC2/S3 clients/libraries in [extra]/[community], etc.
>
> There were many reasons for moving noip from [community] to AUR. Not
> being open source was just one of many reasons, while you here present
> it as the sole reason, which is incorrect. Also keep in mind that TUs
> are not bound by rules regarding open source / commercial software, so
> ultimately, this does not matter.
I addressed all of the reasons you listed in the order that you listed
them in the email you're responding to...
>> Additionally, while I don't use noip, some users want choices.
>
> The same users can still install noip from AUR or from the downloads
> provided by upstream.
>
>> Would you remove vim just because emacs exists? You should not remove it on
>> the grounds that there are alternatives. (Let's not make judgments on
>> which is better - you wouldn't want to start a vim vs emacs debate
>> here.)
>
> I think both vim and emacs qualifies, as opposed to noip. They are
> both popular (would result in many votes on AUR if they were not
> official packages), they are open source (which is not a requirement,
> but a plus) and they do not exist only to direct a stream of users and
> money toward a single company (also not a requirement, but a plus).
> They also offer pretty unique approaches to text editing (or did, when
> they were first released), as opposed to noip that does not offer a
> unique approach to dynamic IP adresses. (Having a unique or
> groundbreaking approach is also not a requirement, but a plus in my
> book). In addition to this, just by being a client that needs to
> connect to a commercial service (even though there is a repeatable 30
> day free trial), it is not at all comparable to a standalone
> application like an editor.
The first part of my reply addressed F/OSS packages that benefit a
single company, such as those which interact with gmail, AWS/ec2, etc.
Keeping these in the official repos is good, because it makes Arch
more usable to a wider audience (i.e. users of that service). You're
conflating the two lines of reasoning here.
The paragraph you're responding to addresses your notion that just
because there are better (by your judgment) alternatives to the
package in question, it should be demoted to AUR. If that were
sufficient reasoning to demote a package to AUR, you should demote
emacs to AUR because vim exists, or ruby to AUR because python exists,
or bash to AUR because zsh exists, etc. The point I'm trying to make
is that your taste in what's a better alternative may differ from
others', so if it doesn't cost much to support multiple alternatives
for a certain service, doing so will make Arch more usable to a wider
audience.
> If you are looking for a single official reason for why noip was moved
> from [community] to AUR, the answer is "because an Arch Linux TU/dev
> did it". If you are interested in how I personally judged this
> particular situation, you may ask me why (as has been done in personal
> emails) and I may be kind enough to try to give a lengthy and honest
> answer (as I did), but I am not required to do so.
>
> And don't write "do not do this".
How about "please do not do this"? ;-)
Let's not resort to "because I said so" just yet... If the real
reason you want to move something to AUR is that you just don't want
to maintain it / have better things to do with your time, that's
perfectly okay! FWIW, I don't particularly care about noip, but your
reasoning of "because it benefits a single company" rubbed me the
wrong way, considering it is F/OSS software.
Cheers, -Ido
More information about the aur-general
mailing list