next PSC task

classic Classic list List threaded Threaded
17 messages Options
Reply | Threaded
Open this post in threaded view
|

next PSC task

Helmut Kudrnovsky
Dear PSC,

after the election of the new PSC, before defining the new chair, may I (as a community member) ask you what will be the next task of the PSC?

when will be the new chair announced?

I think an important first/next step to broaden the dev community may be to investigate how new source code management technologies like GIT could be implemented beside/instead of svn.

There are ongoing some OSGeo initiatives to provide GIT infrastructure for OSGeo projects? Will be the GRASS GIS project will use such new technologies?

Python 3 is heavily knocking on the door.

Will there be some lead of the new PSC regarding these challenges?

Thanks for clarifications!
best regards
Helmut
Reply | Threaded
Open this post in threaded view
|

Re: next PSC task

Markus Neteler
On Fri, Sep 16, 2016 at 11:47 PM, Helmut Kudrnovsky <[hidden email]> wrote:
> Dear PSC,
>
> after the election of the new PSC, before defining the new chair, may I (as
> a community member) ask you what will be the next task of the PSC?
>
> when will be the new chair announced?

IMHO we need to have someone besides the current candidate(s) to
organize the election of the chair.
I hoped for some faster pace here :-)

> I think an important first/next step to broaden the dev community may be to
> investigate how new source code management technologies like GIT could be
> implemented beside/instead of svn.

It was on the recent community sprint agenda, with no real outcome.
We do need a git *expert* for this. Anyone?
However, I believe that this topic is more a technical rather than a
political one which belongs to grass-dev.

> There are ongoing some OSGeo initiatives to provide GIT infrastructure for
> OSGeo projects? Will be the GRASS GIS project will use such new
> technologies?

At time there is some experimental setup:

GRASS GIS is obviously already registered there but a git *expert* is
needed. In addition we have to understand how to use git and svn
together (like GDAL appears to do) and when the experimental stage
will be turned into production mode.

> Python 3 is heavily knocking on the door.

IMHO also the Python 3 is a grass-dev topic.

> Will there be some lead of the new PSC regarding these challenges?

For sure the new PSC will do related house keeping but the technical
part should remain in the grass-dev discussion space unless there is a
strong motivation to make it a PSC task.

> Thanks for clarifications!

Urgent is the election of the chair.

Best
Markus
_______________________________________________
grass-psc mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-psc
Reply | Threaded
Open this post in threaded view
|

Re: next PSC task

Helmut Kudrnovsky
>IMHO we need to have someone besides the current candidate(s) to
>organize the election of the chair.
>I hoped for some faster pace here :-)

is there a similar system /infrastructure needed as for the PSC voting?
best regards
Helmut
Reply | Threaded
Open this post in threaded view
|

Re: next PSC task

Martin Landa
In reply to this post by Markus Neteler
Hi,

2016-09-18 18:51 GMT+02:00 Markus Neteler <[hidden email]>:
> It was on the recent community sprint agenda, with no real outcome.
> We do need a git *expert* for this. Anyone?
> However, I believe that this topic is more a technical rather than a
> political one which belongs to grass-dev.

I am willing to help with this issue. Probably others will be
interested too, Vaclav? Ma

--
Martin Landa
http://geo.fsv.cvut.cz/gwiki/Landa
http://gismentors.cz/mentors/landa
_______________________________________________
grass-psc mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-psc
Reply | Threaded
Open this post in threaded view
|

Re: next PSC task

wenzeslaus

On Sun, Sep 18, 2016 at 2:04 PM, Martin Landa <[hidden email]> wrote:
2016-09-18 18:51 GMT+02:00 Markus Neteler <[hidden email]>:
> It was on the recent community sprint agenda, with no real outcome.
> We do need a git *expert* for this. Anyone?
> However, I believe that this topic is more a technical rather than a
> political one which belongs to grass-dev.

I am willing to help with this issue. Probably others will be
interested too, Vaclav? Ma

Yes, to certain extent. However, I think it is not purely technical and it is coupled with OSGeo and other OSGeo projects.

_______________________________________________
grass-psc mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-psc
Reply | Threaded
Open this post in threaded view
|

Re: next PSC task

Markus Neteler
On Sun, Sep 18, 2016 at 8:42 PM, Vaclav Petras <[hidden email]> wrote:

> On Sun, Sep 18, 2016 at 2:04 PM, Martin Landa <[hidden email]>
> wrote:
>>
>> 2016-09-18 18:51 GMT+02:00 Markus Neteler <[hidden email]>:
>> > It was on the recent community sprint agenda, with no real outcome.
>> > We do need a git *expert* for this. Anyone?
>> > However, I believe that this topic is more a technical rather than a
>> > political one which belongs to grass-dev.
>>
>> I am willing to help with this issue. Probably others will be
>> interested too, Vaclav? Ma
>
>
> Yes, to certain extent. However, I think it is not purely technical and it
> is coupled with OSGeo and other OSGeo projects.

Right- But the PSC is only involved once the technical aspects are clear.

E.g., we (at grass-dev) may check first how it works for GDAL.

Markus
_______________________________________________
grass-psc mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-psc
Reply | Threaded
Open this post in threaded view
|

Re: next PSC task

Moritz Lennert
On 18/09/16 20:50, Markus Neteler wrote:

> On Sun, Sep 18, 2016 at 8:42 PM, Vaclav Petras <[hidden email]> wrote:
>> On Sun, Sep 18, 2016 at 2:04 PM, Martin Landa <[hidden email]>
>> wrote:
>>>
>>> 2016-09-18 18:51 GMT+02:00 Markus Neteler <[hidden email]>:
>>>> It was on the recent community sprint agenda, with no real outcome.
>>>> We do need a git *expert* for this. Anyone?
>>>> However, I believe that this topic is more a technical rather than a
>>>> political one which belongs to grass-dev.
>>>
>>> I am willing to help with this issue. Probably others will be
>>> interested too, Vaclav? Ma
>>
>>
>> Yes, to certain extent. However, I think it is not purely technical and it
>> is coupled with OSGeo and other OSGeo projects.
>
> Right- But the PSC is only involved once the technical aspects are clear.

One of the questions I would think we should think about from the
beginning is whether we really want to run with a double system (svn and
git) or whether it might be easier in the long run, to completely switch
over to git...

Moritz
_______________________________________________
grass-psc mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-psc
Reply | Threaded
Open this post in threaded view
|

Migration to git/gog

"Peter Löwe"
In reply to this post by Markus Neteler
Dear PSC,

I would like to propose two topics for discussion by the board:

a) Option to migrate to a git/gog-based repository. This proposal is being forwarded from Helli (cc), based on the current successful migration by the Postgis project ( https://lists.osgeo.org/pipermail/postgis-devel/2016-October/025981.html), to ensure that the developers can use a state of the art repository.

b) Option to use persistent identifiers (Digital Object Identifiers (DOI)) for software, documentation and data in the GRASS project. DOI are currently mostly known to reference and cite scientific articles ("like a ISBN number for books"). DOI can also be assigned to scientific data, including software. The benefit in doing this is providing scientific citability, which implies scientific credit and recognition for the authors of the software or data. DOI for software/data are provided/minted by the DataCite non profit organisation (https://www.datacite.org/ https://en.wikipedia.org/wiki/DataCite). OSGeo, being a non profit organisation itself with significant IT infrastructure, could apply to DataCite to receive DOI (has already been positively checked by the DataCite folks). However, this requires a need within the OSGeo projects to use DOI. I believe this could be a great driver to foster the authoring and publication of add-on modules for GRASS, for which the authors could be credited via a DOI upon "publication" within the GRASS community. FWIW, DOI are already assigned to all OSGeo (and GRASS!)-related FOSS4G conference videos in the TIB AV-Portal. Example: http://dx.doi.org/10.5446/20425 (DOI also serve as "unbreakable" weblinks.)

Best,
peter
_______________________________________________
grass-psc mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-psc
Reply | Threaded
Open this post in threaded view
|

Re: Migration to git/gog

margherita
Dear Peter, Helli, All,

On Tue, Nov 1, 2016 at 7:41 PM, "Peter Löwe" <[hidden email]> wrote:
Dear PSC,

I would like to propose two topics for discussion by the board:

a) Option to migrate to a git/gog-based repository. This proposal is being forwarded from Helli (cc), based on the current successful migration by the Postgis project ( https://lists.osgeo.org/pipermail/postgis-devel/2016-October/025981.html), to ensure that the developers can use a state of the art repository.

I'm in favor, provided that a volunteer is found for migrating the history and repository. In the case of Postgis, as far as I understand, strk took care of that. 

b) Option to use persistent identifiers (Digital Object Identifiers (DOI)) for software, documentation and data in the GRASS project. DOI are currently mostly known to reference and cite scientific articles ("like a ISBN number for books"). DOI can also be assigned to scientific data, including software. The benefit in doing this is providing scientific citability, which implies scientific credit and recognition for the authors of the software or data. DOI for software/data are provided/minted by the DataCite non profit organisation (https://www.datacite.org/ https://en.wikipedia.org/wiki/DataCite). OSGeo, being a non profit organisation itself with significant IT infrastructure, could apply to DataCite to receive DOI (has already been positively checked by the DataCite folks). However, this requires a need within the OSGeo projects to use DOI. I believe this could be a great driver to foster the authoring and publication of add-on modules for GRASS, for which the authors could be credited via a DOI upon "publication" within the GRASS community. FWIW, DOI are already assigned to all OSGeo (and GRASS!)-related FOSS4G conference videos in the TIB AV-Portal. Example: http://dx.doi.org/10.5446/20425 (DOI also serve as "unbreakable" weblinks.)

Good idea. 
Any opinion from the others?
Thanks for the proposals

Regards,

--
Margherita Di Leo

_______________________________________________
grass-psc mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-psc
Reply | Threaded
Open this post in threaded view
|

Re: next PSC task

NikosAlexandris
In reply to this post by Moritz Lennert
* Moritz Lennert <[hidden email]> [2016-09-19 11:12:09 +0200]:

>On 18/09/16 20:50, Markus Neteler wrote:
>> On Sun, Sep 18, 2016 at 8:42 PM, Vaclav Petras <[hidden email]> wrote:
>>> On Sun, Sep 18, 2016 at 2:04 PM, Martin Landa <[hidden email]>
>>> wrote:
>>>>
>>>> 2016-09-18 18:51 GMT+02:00 Markus Neteler <[hidden email]>:
>>>>> It was on the recent community sprint agenda, with no real outcome.
>>>>> We do need a git *expert* for this. Anyone?
>>>>> However, I believe that this topic is more a technical rather than a
>>>>> political one which belongs to grass-dev.
>>>>
>>>> I am willing to help with this issue. Probably others will be
>>>> interested too, Vaclav? Ma
>>>
>>>
>>> Yes, to certain extent. However, I think it is not purely technical and it
>>> is coupled with OSGeo and other OSGeo projects.
>>
>> Right- But the PSC is only involved once the technical aspects are clear.
>
>One of the questions I would think we should think about from the
>beginning is whether we really want to run with a double system (svn and
>git) or whether it might be easier in the long run, to completely switch
>over to git...

From my point of view, git is easier.  I have experiences now, like many
in the GRASS community, controlling almost everything ascii using git
(not only code).  My experiences in resolving conflicts, though, when it
comes to merge branches, are, indeed, limited.  I did a few stuff, but
not enough.

But we should just go for it, in the beginning, at least, experimental
and playing around in a sandbox-like repository.

Cheers, Nikos
_______________________________________________
grass-psc mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-psc
Reply | Threaded
Open this post in threaded view
|

Re: Migration to git/gog

Moritz Lennert
In reply to this post by margherita
On 02/11/16 11:42, Margherita Di Leo wrote:

> Dear Peter, Helli, All,
>
> On Tue, Nov 1, 2016 at 7:41 PM, "Peter Löwe" <[hidden email]
> <mailto:[hidden email]>> wrote:
>
>     Dear PSC,
>
>     I would like to propose two topics for discussion by the board:
>
>     a) Option to migrate to a git/gog-based repository. This proposal is
>     being forwarded from Helli (cc), based on the current successful
>     migration by the Postgis project (
>     https://lists.osgeo.org/pipermail/postgis-devel/2016-October/025981.html
>     <https://lists.osgeo.org/pipermail/postgis-devel/2016-October/025981.html>),
>     to ensure that the developers can use a state of the art repository.
>
>
> I'm in favor, provided that a volunteer is found for migrating the
> history and repository. In the case of Postgis, as far as I understand,
> strk took care of that.

Contrary to others here I'm complete ignorant of git, and so have no
experience whatsoever which would allow me to judge the advantage of
moving. But I'm willing to read "git for dummies" if everyone else
thinks it is better... ;-)

>
>
>     b) Option to use persistent identifiers (Digital Object Identifiers
>     (DOI)) for software, documentation and data in the GRASS project.
>     DOI are currently mostly known to reference and cite scientific
>     articles ("like a ISBN number for books"). DOI can also be assigned
>     to scientific data, including software. The benefit in doing this is
>     providing scientific citability, which implies scientific credit and
>     recognition for the authors of the software or data. DOI for
>     software/data are provided/minted by the DataCite non profit
>     organisation (https://www.datacite.org/
>     https://en.wikipedia.org/wiki/DataCite
>     <https://en.wikipedia.org/wiki/DataCite>). OSGeo, being a non profit
>     organisation itself with significant IT infrastructure, could apply
>     to DataCite to receive DOI (has already been positively checked by
>     the DataCite folks). However, this requires a need within the OSGeo
>     projects to use DOI. I believe this could be a great driver to
>     foster the authoring and publication of add-on modules for GRASS,
>     for which the authors could be credited via a DOI upon "publication"
>     within the GRASS community. FWIW, DOI are already assigned to all
>     OSGeo (and GRASS!)-related FOSS4G conference videos in the TIB
>     AV-Portal. Example: http://dx.doi.org/10.5446/20425
>     <http://dx.doi.org/10.5446/20425> (DOI also serve as "unbreakable"
>     weblinks.)
>
>
> Good idea.
> Any opinion from the others?


+1
Moritz
_______________________________________________
grass-psc mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-psc
Reply | Threaded
Open this post in threaded view
|

Re: Migration to git/gog

Luca Delucchi
On 2 November 2016 at 13:48, Moritz Lennert
<[hidden email]> wrote:
>
> Contrary to others here I'm complete ignorant of git, and so have no
> experience whatsoever which would allow me to judge the advantage of moving.
> But I'm willing to read "git for dummies" if everyone else thinks it is
> better... ;-)
>

I Think is better (now you have to read "git for dummies" :-P )

The main problem is to have/find a dedicated person knows really well GIT

>
> +1
> Moritz
>


--
ciao
Luca

www.lucadelu.org
_______________________________________________
grass-psc mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-psc
Reply | Threaded
Open this post in threaded view
|

Re: Migration to git/gog

Markus Neteler
On Wed, Nov 2, 2016 at 4:11 PM, Luca Delucchi <[hidden email]> wrote:
...
> The main problem is to have/find a dedicated person knows really well GIT

... and who is willing to dedicate time for a while to keep everything
up and running.
This was one of the obstacles so far. Anyone?

Markus
_______________________________________________
grass-psc mailing list
[hidden email]
http://lists.osgeo.org/mailman/listinfo/grass-psc
Reply | Threaded
Open this post in threaded view
|

Re: Migration to git/gog

Helmut Kudrnovsky

>... and who is willing to dedicate time for a while to keep everything
>up and running.
>This was one of the obstacles so far. Anyone?

maybe asking in user ML if there may be some interest and support?
best regards
Helmut
Reply | Threaded
Open this post in threaded view
|

Re: Migration to git/gog

Markus Neteler
On Wed, Nov 2, 2016 at 4:55 PM, Helmut Kudrnovsky <[hidden email]> wrote:
>>... and who is willing to dedicate time for a while to keep everything
>>up and running.
>>This was one of the obstacles so far. Anyone?
>
> maybe asking in user ML if there may be some interest and support?

Apparently it didn't happen yet :-)

FYI:
In any case, I have contacted Even from GDAL, they wrote some
decumented (slightly outdated he said):

https://trac.osgeo.org/gdal/wiki/UsingGitToMaintainGDALWorkflow

At time they manually merge pull requests back to the master SVN.
Doing that (semi-)automatically requires git expertise.

Markus
_______________________________________________
grass-psc mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/grass-psc
Reply | Threaded
Open this post in threaded view
|

Re: Migration to git/gog

Markus Neteler
In reply to this post by "Peter Löwe"
n Tue, Nov 1, 2016 at 7:41 PM, "Peter Löwe" <[hidden email]> wrote:
> Dear PSC,
>
> I would like to propose two topics for discussion by the board:
>
> a) Option to migrate to a git/gog-based repository. This proposal is being forwarded from Helli (cc), based on the current successful migration by the Postgis project ( https://lists.osgeo.org/pipermail/postgis-devel/2016-October/025981.html), to ensure that the developers can use a state of the art repository.

.... since this is a recurring question: I'd suggest that we make some
progress here.
Maybe set up a task force collecting some knowledgeable git
enthusiasts supporting us here?

thanks,
Markus
_______________________________________________
grass-psc mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/grass-psc
Reply | Threaded
Open this post in threaded view
|

Re: Migration to git/gog

NikosAlexandris
Dear all,

I would like that we set-up a "test" repo where users like me (I guess
there are others like) who like to mess up working things, can play
around and improve skills in complex git merging scenarios.

Thank you, Nikos


* Markus Neteler <[hidden email]> [2017-06-15 11:12:20 +0200]:

>n Tue, Nov 1, 2016 at 7:41 PM, "Peter Löwe" <[hidden email]> wrote:
>> Dear PSC,
>>
>> I would like to propose two topics for discussion by the board:
>>
>> a) Option to migrate to a git/gog-based repository. This proposal is being forwarded from Helli (cc), based on the current successful migration by the Postgis project ( https://lists.osgeo.org/pipermail/postgis-devel/2016-October/025981.html), to ensure that the developers can use a state of the art repository.
>
>.... since this is a recurring question: I'd suggest that we make some
>progress here.
>Maybe set up a task force collecting some knowledgeable git
>enthusiasts supporting us here?
>
>thanks,
>Markus
>_______________________________________________
>grass-psc mailing list
>[hidden email]
>https://lists.osgeo.org/mailman/listinfo/grass-psc

--
Nikos Alexandris | Remote Sensing & Geomatics
GPG Key Fingerprint 6F9D4506F3CA28380974D31A9053534B693C4FB3
_______________________________________________
grass-psc mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/grass-psc