[Gt-eos] gtk source repo location

[Oliver Keeble]

Hi,

On 13/10/17 07:26, Frank Scheiner wrote:
> Hi Brian, all,
> 
> On 10/12/2017 11:13 PM, Brian Lin wrote:
>> Hi all,
>>
>> Apologies if this has been decided on already but has there been a 
>> decision made on the location of the post-EOL gtk source? I saw some 
>> slides from a GDB meeting but it wasn't clear if it resulted in a 
>> final choice.
> 
> Can you post these slides on the list or post a link to them if they're 
> public?

These are the slides, I hope they are publicly accessible;

https://indico.cern.ch/event/578990/#2-globus-toolkit-status

This presentation was simply a report on the phoneconf organised by EGI 
and an assurance that a coordinated response to the situation was 
underway. Nothing further has been decided since then, to my knowledge.

> 
>   If this hasn't been decided, I suggest creating an open-gtk
>> organization for which we can add EGI, OSG, WLCG, etc. teams.

I think we need a unique reference repository with the usual tooling 
(issue trackers etc), communication channels (such as this list), and 
hopefully some names against key responsibilities (source curation, 
packaging). Does that constitute an "organisation"?

> The issue with "gtk" is that there's already a well-known toolkit ([1]) 
> available that uses that name and adding "open" to it might even more 
> confuse people which aren't familiar with the Globus Toolkit.
> 
> [1]: https://www.gtk.org/

Indeed, I don't think we can use the gtk acronym.

> And actually the Globus Toolkit is already open in the sense of open 
> source code, but you're right that this will change with what Globus 
> plans for the future with their "Globus Community License", if they 
> still want to call their code base Globus Toolkit.
> 
> I'd like to propose another name:
> 
> Some time ago I created a GitLab group named "gnext" for this ([2]).
> 
> [2]: https://gitlab.com/gnext/
> 
> Let me explain the ideas behind the name:
> 
> * It's short, it's easy to pronounce - at least for central Europeans 
> like me - and it has some meaning in it.
> 
> * It starts with a "g" which relates to "Globus (Toolkit)" but doesn't 
> use the "Globus" trademark (btw "Globus Toolkit" and "Globus alliance" 
> are also trademarks, see e.g. [3] at the bottom). I have some preference 
> to use the letter "g" for my Globus related tools (e.g. gtransfer, 
> gsatellite). So I'm a little biased here, of course.
> 
> [3]: http://toolkit.globus.org/toolkit/
> 
> * The "next" should make clear that it's the next thing after the Globus 
> Toolkit or simply the future of the Globus Toolkit (in its current 
> form). People should instantly recognize it as the future source of the 
> Globus Toolkit after Globus abandoned it.
> 
> Not sure what you and others think of the name. Don't hesitate to 
> discuss this. :-D
> 
> Sadly the name is already taken by an assumed empty GitHub account, but 
> we could still ask GitHub to "free up" that account for use by us.

I'm fine with the name, although I can think of at least 3 ways to 
pronounce it! However I think we need a pretty good reason not to use 
GitHub, so this could be an issue.

> GitLab and/or GitHub also offer wikis and issue tracking, the latter 
> could also be used to support users, in addition to a mailing list like 
> the current gt-user list ([4]), which will most likely vanish in the 
> future.
> 
> [4]: https://lists.globus.org/mailman/listinfo/gt-user
> 
>   There also
>> seemed to be a question about licenses -- from the OSG point of view, 
>> I believe the Apache 2.0 license should suffice as that's what we use 
>> for our own software.
> 
> Yes and I also think a license change for future additions would only 
> complicate things, although I personally prefer GPL licenses.

I agree that a new license for future additions should be avoided if at 
all possible (CERN is fine with Apache2).

I would say it's time to organise a follow-up phoneconf.

Cheers,

Oliver.

> Cheers,
> Frank
> 
> 
> 
> _______________________________________________
> Gt-eos mailing list
> Gt-eos at mailman.egi.eu
> http://mailman.egi.eu/mailman/listinfo/gt-eos
> 

-- 
Oliver Keeble                      Information Technology Department
oliver.keeble at cern.ch                                           CERN
+41 75 411 5965                                    CH-1211 Geneva 23