[QGIS-Developer] Last call for switching to github issue tracker

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
49 messages Options
123
Reply | Threaded
Open this post in threaded view
|

[QGIS-Developer] Last call for switching to github issue tracker

3nids
Hi all,

I'ld like to raise a last call to switch to github issue tracker before the release.
I'd like to summarize briefly why.

Pros (to me):
- integration with code and mainly with PR
- avoid multiplication of platforms
- so much more powerful (attachment, text formatting, referecing issue/PR, pinging people, fitlering/searching/tagging, possibility to plugin external issue management such as zenhub or others)
- so much more responsive, mobile friendy and easy to use
- lower infrstructure management on QGIS team (would be available for other stuff)
- opportunity to clean the list

A proof that current situation is sub-optimal:
- people are pointing to code issues directly in comments of commits on github rather than creating an issue (high chance of getting lost)
- people send mail to the mailing saying they have created an issue (sic)
- people creates new gitub repo for creating issue lists (sic again): see qgis 3 API or UX changes
- mantra thing is a nightmare for new comers

Why I don't agree on the raised issues to not to move to github:
- migration of existing issues: we have migrated a to a new redmine which should make migration easier [0].
- Github has a close-source: there is an API and existing tools to migrate to other issue tracker [1]. I guess it leads down to pragmatic vs ethical matters.

I would be in favor of using a semi-automatic migration:
- move all open tickets
- propose a tool to migrate others on demand: like a one-click process to move the features.
But this can be decided afterwards.

I know PSC is (was?) against it, but I feel the majority of devs (and new issue posters) are for it.
Sorry to hassle on the topic. Let me know if I shall shut-up forever. Or raise your hand :)

Or can we setup a loomio vote directly???

Best wishes,
Denis



_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Jürgen E. Fischer
Hi Denis,

On Fri, 12. Jan 2018 at 13:52:39 +0000, Denis Rouzaud wrote:
> - mantra thing is a nightmare for new comers

Again: osgeo ids are not a requirement to use our redmine.


Jürgen

--
Jürgen E. Fischer           norBIT GmbH             Tel. +49-4931-918175-31
Dipl.-Inf. (FH)             Rheinstraße 13          Fax. +49-4931-918175-50
Software Engineer           D-26506 Norden             http://www.norbit.de
QGIS release manager (PSC)  Germany                    IRC: jef on FreeNode

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

signature.asc (844 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Borys Jurgiel-4
In reply to this post by 3nids
Well, I have reconciled myself with the fact we're stuck in the darks of
Redmine forever, but if there is still any chance for the migration, I'd not
hesitate for a moment of course.

Borys


On piątek, 12 stycznia 2018 13:52:39 CET Denis Rouzaud wrote:

> Hi all,
>
> I'ld like to raise a last call to switch to github issue tracker before the
> release.
> I'd like to summarize briefly why.
>
> Pros (to me):
> - integration with code and mainly with PR
> - avoid multiplication of platforms
> - so much more powerful (attachment, text formatting, referecing issue/PR,
> pinging people, fitlering/searching/tagging, possibility to plugin external
> issue management such as zenhub or others)
> - so much more responsive, mobile friendy and easy to use
> - lower infrstructure management on QGIS team (would be available for other
> stuff)
> - opportunity to clean the list
>
> A proof that current situation is sub-optimal:
> - people are pointing to code issues directly in comments of commits on
> github rather than creating an issue (high chance of getting lost)
> - people send mail to the mailing saying they have created an issue (sic)
> - people creates new gitub repo for creating issue lists (sic again): see
> qgis 3 API or UX changes
> - mantra thing is a nightmare for new comers
>
> Why I don't agree on the raised issues to not to move to github:
> - migration of existing issues: we have migrated a to a new redmine which
> should make migration easier [0].
> - Github has a close-source: there is an API and existing tools to migrate
> to other issue tracker [1]. I guess it leads down to pragmatic vs ethical
> matters.
>
> I would be in favor of using a semi-automatic migration:
> - move all open tickets
> - propose a tool to migrate others on demand: like a one-click process to
> move the features.
> But this can be decided afterwards.
>
> I know PSC is (was?) against it, but I feel the majority of devs (and new
> issue posters) are for it.
> Sorry to hassle on the topic. Let me know if I shall shut-up forever. Or
> raise your hand :)
>
> Or can we setup a loomio vote directly???
>
> Best wishes,
> Denis
>
> [0] https://github.com/gmontard/issue-sync-redmine-github
> [1] Github importer for Gitlab:
> https://docs.gitlab.com/ee/user/project/import/github.html


_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Matthias Kuhn 🌍
On 01/12/2018 03:19 PM, Borys Jurgiel wrote:
> Well, I have reconciled myself with the fact we're stuck in the darks of
> Redmine forever, but if there is still any chance for the migration, I'd not
> hesitate for a moment of course.

I couldn't have said it better.
+1

Matthias
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Andreas Neumann-4
In reply to this post by 3nids

Hi,

I think the biggest question: who does all the work to migrate from Redmine to Github while not loosing all our history?

The migration attempts so far haven't been very convincing.

And I don't think that Redmine is so bad usability wise. I agree, the UI is not as sexy as Github, but it does its job well. It is much more powerful for finding and filtering. The issues we had with the slow performance was fixed when moving to a new Machine at Hetzner. The main argument for github is in my opinion not ease of use - but the integration with the code base - that is definitely a strength of github issue reporting.

To undermine that github is not easy to use for non-dev users: last Wednesday at the Swiss QGIS user meeting I was in the meeting for QWC2. Users had no idea at all how Github works and could be used for issue reporting. One had to explain it to them and teach them how to use it. To assume that using github is self-explanatory for non-tech users is a wrong assumption.

Andreas

On 2018-01-12 14:52, Denis Rouzaud wrote:

Hi all,
 
I'ld like to raise a last call to switch to github issue tracker before the release.
I'd like to summarize briefly why.
 
Pros (to me):
- integration with code and mainly with PR
- avoid multiplication of platforms
- so much more powerful (attachment, text formatting, referecing issue/PR, pinging people, fitlering/searching/tagging, possibility to plugin external issue management such as zenhub or others)
- so much more responsive, mobile friendy and easy to use
- lower infrstructure management on QGIS team (would be available for other stuff)
- opportunity to clean the list
 
A proof that current situation is sub-optimal:
- people are pointing to code issues directly in comments of commits on github rather than creating an issue (high chance of getting lost)
- people send mail to the mailing saying they have created an issue (sic)
- people creates new gitub repo for creating issue lists (sic again): see qgis 3 API or UX changes
- mantra thing is a nightmare for new comers
 
Why I don't agree on the raised issues to not to move to github:
- migration of existing issues: we have migrated a to a new redmine which should make migration easier [0].
- Github has a close-source: there is an API and existing tools to migrate to other issue tracker [1]. I guess it leads down to pragmatic vs ethical matters.
 
I would be in favor of using a semi-automatic migration:
- move all open tickets
- propose a tool to migrate others on demand: like a one-click process to move the features.
But this can be decided afterwards.
 
I know PSC is (was?) against it, but I feel the majority of devs (and new issue posters) are for it.
Sorry to hassle on the topic. Let me know if I shall shut-up forever. Or raise your hand :)
 
Or can we setup a loomio vote directly???
 
Best wishes,
Denis
 
 

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer



_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Tom Chadwin
In reply to this post by Matthias Kuhn 🌍
A huge +1 from me. I was too intimidated to add issues to Redmine for some
time through not understanding what some fields meant.

Tom



-----
Buy Pie Spy: Adventures in British pastry 2010-11 on Amazon
--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-Developer-f4099106.html
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Matthias Kuhn 🌍
In reply to this post by Andreas Neumann-4
On 01/12/2018 03:31 PM, Andreas Neumann wrote:
>
> Hi,
>
> I think the biggest question: who does all the work to migrate from
> Redmine to Github while not loosing all our history?
>

I started working on it some time ago and thought it looked quite
promising. I stopped the work because of lack of conviction that there's
a perspective for this.

> And I don't think that Redmine is so bad usability wise. I agree, the
> UI is not as sexy as Github, but it does its job well. It is much more
> powerful for finding and filtering. The issues we had with the slow
> performance was fixed when moving to a new Machine at Hetzner. The
> main argument for github is in my opinion not ease of use - but the
> integration with the code base - that is definitely a strength of
> github issue reporting.

Agreed, it's better than before.
Because no more mantra, usable performance.
On the other hand I don't prefer it's search and filtering options.

> To undermine that github is not easy to use for non-dev users: last
> Wednesday at the Swiss QGIS user meeting I was in the meeting for
> QWC2. Users had no idea at all how Github works and could be used for
> issue reporting. One had to explain it to them and teach them how to
> use it. To assume that using github is self-explanatory for non-tech
> users is a wrong assumption.
>

I wouldn't claim that for either of the two platforms. There's a
learning curve for both. Myself, I happen to meanwhile embrace markdown
but struggle with the redmine syntax. Not because it's worse, just
because I am too lazy to learn another one.

Matthias

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Jorge Gustavo Rocha
Hi,

Good point, Denis!

Mathias already have done a lot to support the migration.
I volunteer to help on this task.

Regards,

Jorge Gustavo

On 12-01-2018 14:49, Matthias Kuhn wrote:

> On 01/12/2018 03:31 PM, Andreas Neumann wrote:
>>
>> Hi,
>>
>> I think the biggest question: who does all the work to migrate from
>> Redmine to Github while not loosing all our history?
>>
>
> I started working on it some time ago and thought it looked quite
> promising. I stopped the work because of lack of conviction that there's
> a perspective for this.
>
>> And I don't think that Redmine is so bad usability wise. I agree, the
>> UI is not as sexy as Github, but it does its job well. It is much more
>> powerful for finding and filtering. The issues we had with the slow
>> performance was fixed when moving to a new Machine at Hetzner. The
>> main argument for github is in my opinion not ease of use - but the
>> integration with the code base - that is definitely a strength of
>> github issue reporting.
>
> Agreed, it's better than before.
> Because no more mantra, usable performance.
> On the other hand I don't prefer it's search and filtering options.
>
>> To undermine that github is not easy to use for non-dev users: last
>> Wednesday at the Swiss QGIS user meeting I was in the meeting for
>> QWC2. Users had no idea at all how Github works and could be used for
>> issue reporting. One had to explain it to them and teach them how to
>> use it. To assume that using github is self-explanatory for non-tech
>> users is a wrong assumption.
>>
>
> I wouldn't claim that for either of the two platforms. There's a
> learning curve for both. Myself, I happen to meanwhile embrace markdown
> but struggle with the redmine syntax. Not because it's worse, just
> because I am too lazy to learn another one.
>
> Matthias
>
> _______________________________________________
> QGIS-Developer mailing list
> [hidden email]
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>

J. Gustavo
--
Jorge Gustavo Rocha
Departamento de Informática
Universidade do Minho
4710-057 Braga
Tel: +351 253604480
Fax: +351 253604471
Móvel: +351 910333888
skype: nabocudnosor
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Régis Haubourg
Hi all,
+1 too.
My main concern is that I now fail to search efficiently for redmine issues, because there are so many now. And this leads to duplicates, or even worst, no declaring the issue because the process is too long.
Github issues are very well indexed by google, which doesn't seem to be the case for redmine. (but this probably can be improved - not sure how)

Régis

2018-01-12 16:06 GMT+01:00 Jorge Gustavo Rocha <[hidden email]>:
Hi,

Good point, Denis!

Mathias already have done a lot to support the migration.
I volunteer to help on this task.

Regards,

Jorge Gustavo

On 12-01-2018 14:49, Matthias Kuhn wrote:
> On 01/12/2018 03:31 PM, Andreas Neumann wrote:
>>
>> Hi,
>>
>> I think the biggest question: who does all the work to migrate from
>> Redmine to Github while not loosing all our history?
>>
>
> I started working on it some time ago and thought it looked quite
> promising. I stopped the work because of lack of conviction that there's
> a perspective for this.
>
>> And I don't think that Redmine is so bad usability wise. I agree, the
>> UI is not as sexy as Github, but it does its job well. It is much more
>> powerful for finding and filtering. The issues we had with the slow
>> performance was fixed when moving to a new Machine at Hetzner. The
>> main argument for github is in my opinion not ease of use - but the
>> integration with the code base - that is definitely a strength of
>> github issue reporting.
>
> Agreed, it's better than before.
> Because no more mantra, usable performance.
> On the other hand I don't prefer it's search and filtering options.
>
>> To undermine that github is not easy to use for non-dev users: last
>> Wednesday at the Swiss QGIS user meeting I was in the meeting for
>> QWC2. Users had no idea at all how Github works and could be used for
>> issue reporting. One had to explain it to them and teach them how to
>> use it. To assume that using github is self-explanatory for non-tech
>> users is a wrong assumption.
>>
>
> I wouldn't claim that for either of the two platforms. There's a
> learning curve for both. Myself, I happen to meanwhile embrace markdown
> but struggle with the redmine syntax. Not because it's worse, just
> because I am too lazy to learn another one.
>
> Matthias
>
> _______________________________________________
> QGIS-Developer mailing list
> [hidden email]
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>

J. Gustavo
--
Jorge Gustavo Rocha
Departamento de Informática
Universidade do Minho
4710-057 Braga
Tel: <a href="tel:%2B351%20253604480" value="+351253604480">+351 253604480
Fax: <a href="tel:%2B351%20253604471" value="+351253604471">+351 253604471
Móvel: <a href="tel:%2B351%20910333888" value="+351910333888">+351 910333888
skype: nabocudnosor
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

lbartoletti
In reply to this post by 3nids
On 12.01.2018 14:52, Denis Rouzaud wrote:
> Hi all,
Hi,

> I'ld like to raise a last call to switch to github issue tracker
> before the release.
> I'd like to summarize briefly why.
>
> Pros (to me):
> - integration with code and mainly with PR
> - avoid multiplication of platforms
> - so much more powerful (attachment, text formatting, referecing
> issue/PR, pinging people, fitlering/searching/tagging, possibility to
> plugin external issue management such as zenhub or others)
> - so much more responsive, mobile friendy and easy to use
Totally +1
When I show in formations where you have to / can make reports, ask
questions, etc., the users - who are not all power users, far from it...
- prefer user-friendly interfaces and a single source easily googleable.
Apart from the different forums or stackexchange, GitHub is the best
place to do this.

> - Github has a close-source [...] I guess it leads down to pragmatic
> vs ethical matters.
Open question: How do you feel, (Have you ever had any thoughts), about
migrate to gitlab (and using github as a mirror)?

Regards.

Loïc
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Régis Haubourg
Hi Loïc,
using both gitlab and github, I'd definitly would be in favor of gitlab, which would allow us to host our own servers. That would imply a big task switching travis to gitlab-ci.
Cheers
Régis


2018-01-12 18:21 GMT+01:00 L.Bartoletti <[hidden email]>:
On 12.01.2018 14:52, Denis Rouzaud wrote:
Hi all,
Hi,
I'ld like to raise a last call to switch to github issue tracker before the release.
I'd like to summarize briefly why.

Pros (to me):
- integration with code and mainly with PR
- avoid multiplication of platforms
- so much more powerful (attachment, text formatting, referecing issue/PR, pinging people, fitlering/searching/tagging, possibility to plugin external issue management such as zenhub or others)
- so much more responsive, mobile friendy and easy to use
Totally +1
When I show in formations where you have to / can make reports, ask questions, etc., the users - who are not all power users, far from it... - prefer user-friendly interfaces and a single source easily googleable.
Apart from the different forums or stackexchange, GitHub is the best place to do this.

- Github has a close-source [...] I guess it leads down to pragmatic vs ethical matters.
Open question: How do you feel, (Have you ever had any thoughts), about migrate to gitlab (and using github as a mirror)?

Regards.

Loïc

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer


_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

berndv.
In reply to this post by Régis Haubourg
For me as a simple user, the main obstacle of posting issues in redmine is, that I fear to produce a duplicate, cause I didn't even learn over the years how to efficiently track an issue down if it already exists.
No idea if github or any other solution is better, but redmine just sucks.

My 2 cents,
Bernd

Am 12.01.2018, 16:30 Uhr, schrieb Régis Haubourg <[hidden email]>:

Hi all,
+1 too.
My main concern is that I now fail to search efficiently for redmine issues, because there are so many now. And this leads to duplicates, or even worst, no declaring the issue because the process is too long.
Github issues are very well indexed by google, which doesn't seem to be the case for redmine. (but this probably can be improved - not sure how)

Régis

2018-01-12 16:06 GMT+01:00 Jorge Gustavo Rocha <[hidden email]>:
Hi,

Good point, Denis!

Mathias already have done a lot to support the migration.
I volunteer to help on this task.

Regards,

Jorge Gustavo

On 12-01-2018 14:49, Matthias Kuhn wrote:
> On 01/12/2018 03:31 PM, Andreas Neumann wrote:
>>
>> Hi,
>>
>> I think the biggest question: who does all the work to migrate from
>> Redmine to Github while not loosing all our history?
>>
>
> I started working on it some time ago and thought it looked quite
> promising. I stopped the work because of lack of conviction that there's
> a perspective for this.
>
>> And I don't think that Redmine is so bad usability wise. I agree, the
>> UI is not as sexy as Github, but it does its job well. It is much more
>> powerful for finding and filtering. The issues we had with the slow
>> performance was fixed when moving to a new Machine at Hetzner. The
>> main argument for github is in my opinion not ease of use - but the
>> integration with the code base - that is definitely a strength of
>> github issue reporting.
>
> Agreed, it's better than before.
> Because no more mantra, usable performance.
> On the other hand I don't prefer it's search and filtering options.
>
>> To undermine that github is not easy to use for non-dev users: last
>> Wednesday at the Swiss QGIS user meeting I was in the meeting for
>> QWC2. Users had no idea at all how Github works and could be used for
>> issue reporting. One had to explain it to them and teach them how to
>> use it. To assume that using github is self-explanatory for non-tech
>> users is a wrong assumption.
>>
>
> I wouldn't claim that for either of the two platforms. There's a
> learning curve for both. Myself, I happen to meanwhile embrace markdown
> but struggle with the redmine syntax. Not because it's worse, just
> because I am too lazy to learn another one.
>
> Matthias
>
> _______________________________________________
> QGIS-Developer mailing list
> [hidden email]
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>

J. Gustavo
--
Jorge Gustavo Rocha
Departamento de Informática
Universidade do Minho
4710-057 Braga
Tel: <a href="tel:%2B351%20253604480" value="+351253604480">+351 253604480
Fax: <a href="tel:%2B351%20253604471" value="+351253604471">+351 253604471
Móvel: <a href="tel:%2B351%20910333888" value="+351910333888">+351 910333888
skype: nabocudnosor
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer




--
Bernd Vogelgesang
Siedlerstraße 2
91083 Baiersdorf/Igelsdorf
Tel: 09133-825374

_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Sandro Santilli-4
In reply to this post by Matthias Kuhn 🌍
On Fri, Jan 12, 2018 at 03:49:54PM +0100, Matthias Kuhn wrote:

> I wouldn't claim that for either of the two platforms. There's a
> learning curve for both. Myself, I happen to meanwhile embrace markdown
> but struggle with the redmine syntax. Not because it's worse, just
> because I am too lazy to learn another one.

This is the main reason why I'd fighting for people to have an
alternative to GitHub. Learning is an hard process. Finding yourself
with knowledge about a closed software at the end of that process is
a huge disservice to yourself (IMHO). You'll be stuck with that
dependency.

How about learning something you can keep using, sharing and
adapt to your needs ?

--strk;
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Sandro Santilli-4
In reply to this post by Régis Haubourg
On Fri, Jan 12, 2018 at 07:44:57PM +0100, Régis Haubourg wrote:

> That would imply a big task switching travis to gitlab-ci.

How about *adding* gitlab-ci, rather than *switching* ?
With PostGIS and GEOS we have both Travis adn Gitlab-CI
and Drone and Jenkins watching after the code...

--strk;
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Tim Sutton-6
Hi All

[QGIS-Developer] Last call for switching to github issue tracker

Given our history on the topic, I doubt very much this is true :-P

I know PSC is (was?) against it, but I feel the majority of devs (and new issue posters) are for it.

Actually this is not true (correct me if I am wrong PSC): the PSC was against two things:

1) losing all history in issue manager. I think this is reasonable since there is a lot of very useful history in the issue tracker and it was felt that a 'clean slate start ‘ would be a disservice to all those who have raised issues in the past.
2) calling for a migration without a fully tested, fully fleshed out plan to manage the migration from start to finish. This means someone taking responsibility for the *whole* process which so far hasn’t been forthcoming.

In Essen we voted against migrating to GitHub issues (even though many of use were in favour of to idea) because of these reasons. If items 1 and 2 above can be resolved, I don’t think you will find too much resistance from the PSC (though I will let other PSC members speak for themselves if this is the case or not).


@sandro: the wiki markup of GitHub is Markdown (an open format) which is widely used in many places so it is definitely a re-useable skill beyond GitHub. The converse is true for Redmine’s markup language which seems pretty arcane after using Markdown.

Regards

Tim



On 14 Jan 2018, at 00:33, Sandro Santilli <[hidden email]> wrote:

On Fri, Jan 12, 2018 at 07:44:57PM +0100, Régis Haubourg wrote:

That would imply a big task switching travis to gitlab-ci.

How about *adding* gitlab-ci, rather than *switching* ?
With PostGIS and GEOS we have both Travis adn Gitlab-CI
and Drone and Jenkins watching after the code...

--strk;
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

 



---

Tim Sutton
QGIS Project Steering Committee Chair





_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer

smime.p7s (2K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Richard Duivenvoorde
On 14-01-18 00:17, Tim Sutton wrote:

> Hi All
>
>> [QGIS-Developer] Last call for switching to github issue tracker
>
> Given our history on the topic, I doubt very much this is true :-P
>
>> I know PSC is (was?) against it, but I feel the majority of devs (and
> new issue posters) are for it.
>
> Actually this is not true (correct me if I am wrong PSC): the PSC was
> against two things:
>
> 1) losing all history in issue manager. I think this is reasonable since
> there is a lot of very useful history in the issue tracker and it was
> felt that a 'clean slate start ‘ would be a disservice to all those who
> have raised issues in the past.
> 2) calling for a migration without a fully tested, fully fleshed out
> plan to manage the migration from start to finish. This means someone
> taking responsibility for the *whole* process which so far hasn’t been
> forthcoming.

That is exactly what it was in my memory too.

Mathias and Juergen both tried some of those migrating scripts OR
created something else. Hitting for example Github-thresholds for max nr
of requests etc, loosing comments and inner linking, had problems with
user mapping etc
Myself I also tried something, and besides this tried to invent some
tag-system to be a alternative for the 'categories' we have now
(https://github.com/rduivenvoorde/temp/issues/18).

Some conclusions (at that time):
- our Redmine version was (at that time) too old to be used for newer
migrate scripts
- Github was not very friendly for migrating our huge issue list
- Some dev's wanted to hold full history, while others were ok with
starting from 3/scratch
- Nobody took the time to further investigate, and people told us that
speed was the main issue on the old Redmine, so we upgraded Redmine AND
put it on a dedicated fast server now.

But I'm pretty sure that IF somebody takes lead and:
- migrates all our current issues with history and comments and users
- convinces all/most devs (it is not a PSC descision in my view)

That we can move to Github or Gitlab or Gitea

First step would be to try out one of those scripts, see if you get
those to work...
OR
convince people that we should start with an empty slate for 3.0...

Regards,

Richard Duivenvoorde
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Jorge Gustavo Rocha
Hi all,

This move from redmine to gitlab (and from transifex to weblate) is not
a technological decision. It's a move towards a more open source
project, open to more people to participate and contribute.

If the argument against this move is "we don't want to lose anything
from the previous workflow" we will never move on. This is an easy
argument to keep the things as they are. It is an argument just based on
tools and technology. We should be more concerned about the community.
QGIS is the community. A dynamic and inclusive worldwide QGIS community
will keep an awesome QGIS application.

We need to decide if we prefer to loose something in favour of a much
more open source project. We all know to be "open source" is more than
release the sources. If it is difficult for users to report bugs, the
most basic thing in a open source project, the project is not so open.
Being an open source also means be dependent on other open source
projects. If there are things we can contribute to gitlab, to weblate,
etc, we will do it, as we already do with FOSS4G related projects.
Others are also dependent on QGIS and trust QGIS community to keep the
project up and running.

This is mostly a PSC decision. To be less or more open is a major
decision, related with principles. That is why PSC members don't have to
be developers, because this is not a technological decision.

If we want to move, we will move and there will be no technological
constraints we can't solve.

Regards,

Jorge Gustavo

On 14-01-2018 11:11, Richard Duivenvoorde wrote:

> On 14-01-18 00:17, Tim Sutton wrote:
>> Hi All
>>
>>>  [QGIS-Developer] Last call for switching to github issue tracker
>>
>> Given our history on the topic, I doubt very much this is true :-P
>>
>>>  I know PSC is (was?) against it, but I feel the majority of devs (and
>> new issue posters) are for it.
>>
>> Actually this is not true (correct me if I am wrong PSC): the PSC was
>> against two things:
>>
>> 1) losing all history in issue manager. I think this is reasonable since
>> there is a lot of very useful history in the issue tracker and it was
>> felt that a 'clean slate start ‘ would be a disservice to all those who
>> have raised issues in the past.
>> 2) calling for a migration without a fully tested, fully fleshed out
>> plan to manage the migration from start to finish. This means someone
>> taking responsibility for the *whole* process which so far hasn’t been
>> forthcoming.
>
> That is exactly what it was in my memory too.
>
> Mathias and Juergen both tried some of those migrating scripts OR
> created something else. Hitting for example Github-thresholds for max nr
> of requests etc, loosing comments and inner linking, had problems with
> user mapping etc
> Myself I also tried something, and besides this tried to invent some
> tag-system to be a alternative for the 'categories' we have now
> (https://github.com/rduivenvoorde/temp/issues/18).
>
> Some conclusions (at that time):
> - our Redmine version was (at that time) too old to be used for newer
> migrate scripts
> - Github was not very friendly for migrating our huge issue list
> - Some dev's wanted to hold full history, while others were ok with
> starting from 3/scratch
> - Nobody took the time to further investigate, and people told us that
> speed was the main issue on the old Redmine, so we upgraded Redmine AND
> put it on a dedicated fast server now.
>
> But I'm pretty sure that IF somebody takes lead and:
> - migrates all our current issues with history and comments and users
> - convinces all/most devs (it is not a PSC descision in my view)
>
> That we can move to Github or Gitlab or Gitea
>
> First step would be to try out one of those scripts, see if you get
> those to work...
> OR
> convince people that we should start with an empty slate for 3.0...
>
> Regards,
>
> Richard Duivenvoorde
> _______________________________________________
> QGIS-Developer mailing list
> [hidden email]
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>

J. Gustavo
--
Jorge Gustavo Rocha
Departamento de Informática
Universidade do Minho
4710-057 Braga
Tel: +351 253604480
Fax: +351 253604471
Móvel: +351 910333888
skype: nabocudnosor
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

pcav
In reply to this post by Richard Duivenvoorde
Hi all,

Il 14/01/2018 12:11, Richard Duivenvoorde ha scritto:
> On 14-01-18 00:17, Tim Sutton wrote:

>> Actually this is not true (correct me if I am wrong PSC): the PSC was
>> against two things:
>>
>> 1) losing all history in issue manager. I think this is reasonable since
>> there is a lot of very useful history in the issue tracker and it was
>> felt that a 'clean slate start ‘ would be a disservice to all those who
>> have raised issues in the past.
>> 2) calling for a migration without a fully tested, fully fleshed out
>> plan to manage the migration from start to finish. This means someone
>> taking responsibility for the *whole* process which so far hasn’t been
>> forthcoming.
>
> That is exactly what it was in my memory too.

I can confirm. I Still believe loosing all our history of open and
closed issues would be a serious damage to the project.

> Some conclusions (at that time):
> - our Redmine version was (at that time) too old to be used for newer
> migrate scripts
> - Github was not very friendly for migrating our huge issue list
> - Some dev's wanted to hold full history, while others were ok with
> starting from 3/scratch
> - Nobody took the time to further investigate, and people told us that
> speed was the main issue on the old Redmine, so we upgraded Redmine AND
> put it on a dedicated fast server now.
>
> But I'm pretty sure that IF somebody takes lead and:
> - migrates all our current issues with history and comments and users
> - convinces all/most devs (it is not a PSC descision in my view)
>
> That we can move to Github or Gitlab or Gitea

I see the first a a serious problem, the second will be far easier or
even non existent IMHO.

All the best.
--
Paolo Cavallini - www.faunalia.eu
QGIS & PostGIS courses: http://www.faunalia.eu/training.html
https://www.google.com/trends/explore?date=all&geo=IT&q=qgis,arcgis
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Nyall Dawson
On 14 January 2018 at 23:19, Paolo Cavallini <[hidden email]> wrote:

> Hi all,
>
> Il 14/01/2018 12:11, Richard Duivenvoorde ha scritto:
>> On 14-01-18 00:17, Tim Sutton wrote:
>
>>> Actually this is not true (correct me if I am wrong PSC): the PSC was
>>> against two things:
>>>
>>> 1) losing all history in issue manager. I think this is reasonable since
>>> there is a lot of very useful history in the issue tracker and it was
>>> felt that a 'clean slate start ‘ would be a disservice to all those who
>>> have raised issues in the past.
>>> 2) calling for a migration without a fully tested, fully fleshed out
>>> plan to manage the migration from start to finish. This means someone
>>> taking responsibility for the *whole* process which so far hasn’t been
>>> forthcoming.
>>
>> That is exactly what it was in my memory too.
>
> I can confirm. I Still believe loosing all our history of open and
> closed issues would be a serious damage to the project.

I agree. I'd find this a very very sore loss. Personally, I'm
referring to historic issues almost on a daily basis, both for
checking what closed issues actually were and also checking for open
issues before working on an area of code.

Nyall
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Reply | Threaded
Open this post in threaded view
|

Re: Last call for switching to github issue tracker

Nyall Dawson
In reply to this post by Sandro Santilli-4
On 14 January 2018 at 08:33, Sandro Santilli <[hidden email]> wrote:
> On Fri, Jan 12, 2018 at 07:44:57PM +0100, Régis Haubourg wrote:
>
>> That would imply a big task switching travis to gitlab-ci.
>
> How about *adding* gitlab-ci, rather than *switching* ?

I'm sure if someone volunteers to set this up and maintain it (****AS
A MIRROR TO GITHUB****), including reviewing all PRs and such which
comes via the mirror, this move would be welcome and you'll find no
objections.

It's all about managing available manpower.

Nyall
_______________________________________________
QGIS-Developer mailing list
[hidden email]
List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
123