[Qgis-developer] Dubt on plugins without repo and bugtrack

Paolo Cavallini cavallini at faunalia.it
Wed Feb 18 03:13:40 PST 2015


Il 18/02/2015 10:53, Matthias Kuhn ha scritto:

> While a bugtracker and a code repository are definitely a good thing to
> have I do not think that it should be mandatory from the beginning.
> 
> It should be required that a way to contact the developer (mail), the
> source code and license are there.
> 
> At the beginning (experimental) state of a plugin the dev may be worried
> about other things than the bugtracker and telling him by mail about
> possible defects is fine (it's him who has to deal with this).
> 
> Plugins need to be open source, but the way they handle project
> management and community should be up to the plugin author. So the best
> thing we can do is send him a notice with "what has been tested by and
> proven helpful for others in the past".
> 
> Putting too many requirements in place may scare people off.
> 
> I think being minimal with regard to requirements is important for
> experimental plugins and IMHO I would treat it the same way for stable
> plugins but there is a bit more space to discuss requirements I think.
> 
> How do others feel?

Let me recapitulate my experience, after several months of managing the
publication process:

* most developers add repo and bugtracker anyway
* most use github for the code, so thie is really a non-issue for them
* when gently asked if they can fill the metadata, almost all adhere,
and usually show gratitude for the attention
* if I remember correctly, nobody ever found adding a repo and a
bugtracker a stumbling block (occasionally they need help for this, but
I can do that), nor scaring
* having the fields as optional in the plugin builder does not encourage
them to fill the fields
* before publishing a plugin I check it; not having a bugtracker makes
things more difficult and far less reliable to keep track of the process
(I send a mail, the I have to remember, check the thread etc.)

So in short I think making them mandatory will make the process more
reliable, with a minimal overhead for authors; alternatively, making
them recommended would be a step in the good direction.

Of course I will not start a flame about this, I can keep on managing
like this.

All the best.

-- 
Paolo Cavallini - www.faunalia.eu
QGIS & PostGIS courses: http://www.faunalia.eu/training.html
*New course* "QGIS for naturalists":
http://www.faunalia.eu/en/nat_course.html

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 181 bytes
Desc: OpenPGP digital signature
URL: <http://lists.osgeo.org/pipermail/qgis-developer/attachments/20150218/8e405b18/attachment.pgp>


More information about the Qgis-developer mailing list