[Geomoose-users] incubation meeting in a week

[Jody Garnett]

Great conversation; a few answers inline (however we may wish to make a separate email thread for each discussion to prevent confusion). 

-- 
Jody Garnett

> > - Go through all your source code files (by hand) and check that their headers are correct (this is often why it is a team effort). For GeoTools we had some files without header information which I had to check version control history on to see who created them. One issue is that we are currently in the process of creating GeoMoose v2.6 (90% done?); we'll need to decide when to review the code (at 100% in 6 weeks?).  Or we can start now and do a final check at release time.  Not really a big deal since the code base is small.
That is good news about the code base size.

As for scheduling it is more up to your release schedule (open source is all about making releases so that always has the priority for a development team).

I would recommend doing a quick pass to "try out the review process". And then schedule some time with the team just after your release; so perhaps 2.6.1 could be a patch release with updated headers "released in conjunction with OSGeo graduation" or something. 
> > -- What you are checking for depends on the project? I don't know if you ask developers to assign the code to an organisation? Or is each file considered separately?
> > 
> 
> Interesting question, causing more questions:
>     What should the license text be? Currently it is the same copyright text as Mapserver.
>     Who (or what) should be assigned the copyright?  Currently Dan Little holds the copyright.
That is totally cool; the fact that you have an answer (hi Dan!) puts you ahead of a lot of projects :-) To be clear there is no need to change anything as part of the incubation process; we are only trying to write down what it is you do now.

As long as the project has been upfront about Dan holding copyright; then we should be good when we list the committers to indicate that they knew they were assigning copyright (especially if version history shows them putting the header on the top of the file they worked on).

If you were especially worried, or American, you may ask that each committer (or organisation the commuter works for) sends Dan a letter. I can provide an example if you wish.

For reference here are two other stories:
- For GeoTools copyright had carefully been assigned to the "project management committee". Which unfortunately did not exist; as a result we made arrangements for OSGeo to hold the copyright (and then we had to track down 10 years of committers and get them to write a letter - fun).
- For PostGIS each developer keeps their copyright (and thus the license bit in the header becomes very important as it is the justification on why the files can be distributed together). This is very easy to set up! However the only downside is that the PostGIS project cannot change their license; without contacting all their committers for the last 10 years.
- For uDig (not an OSGeo project) we have the original organisation hold copyright - even though they are not active in the project. We figure they are a legal entity and that we can ask them to change the project license if needed.

So it is a good thing that Dan is a legal entity :-)

Random thoughts - Dan may want to get a letter from his employer or something (as in America an employer can claim work you do on your own time). The letter can just acknowledge that Dan works on GeoMOOSE and they are comfortable with him holding copyright (and I assume their legal council can help them write a pretty letter).

I am not going to bother with INAL - as it is obvious I am not.
> > - Check how you distribute your documentation? Creative Commons?
> > 
> 
> While the Sphinx docs text files are included with the distribution, I'm not sure why.  I'd expect everyone to access the doc on line at geomoose.org (http://geomoose.org).  Currently those doc files are covered with the same copyright  license file included with the source code (we may need to change that).
LoL - that hurts my head :-) So the source code that generates the docs may be able to be treated like code? But the resulting text may be under something like creative commons; or free document license or something. Basically are people allowed to "copy and paste" text? And if so how are they allowed to copy and paste the text...
> > - Check any same data included with your application or documentation. If you don't know where it came from I recommend the natural earth dataset :-)
> > 
> 
> While some of the layers available at Natural Earth may be useful to us, the focus of GeoMoose is with datasets at the county level, especially parcel data.   I expect we'll need to put some effort into making sure we have the rights to distribute the data (especially privacy issues w.r.t. parcel owner's name, etc).  I'm sure this was discussed long before I joined the project, but it will be useful to include a statement somewhere in the project.
Ah good :-)
> Wow, this is a good start to our Review Document...
Yeah it is a good discussion; I am going to CC  Bruce as he has a similar discussion going on with the project he is mentoring :-)

As for the original subject: In the meeting I will let them know GeoMoose has started and is asking great questions about the providence review.

Jody
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/geomoose-users/attachments/20120112/14123a91/attachment.html