[GRASS-PSC] Re: [Incubator] Motion: GRASS to Graduate Incubation

[Markus Neteler]

On Jan 24, 2008 6:39 PM, Frank Warmerdam <warmerdam at pobox.com> wrote:
> Arnulf Christl (OSGeo) wrote:
> > Hello IncCom,
> > after spending a lot other people's time and energy - especially Markus' -
> > I cannot find any more reasons to further postpone the graduation of
> > GRASS. Find all the details of the work in the Wiki and the logs to the
> > code repositories linked from there. The developer community is
> > outstanding in every respect (also in procrastinating tedious incubation
> > issues...) and I am happy to finally be able to recommend GRASS for
> > graduation.
> >
> > I motion that the Incubation Committee recommends GRASS to the OSGeo Board
> > of Directors for graduation from incubation with Markus Neteler as project
> > liaison officer.
> >
> > I expect the next board meeting to take place February 1st or 8th (it is
> > not yet scheduled) and suggest that comments should come within one week
> > so that we can potentially proceed with graduation festivities on February
> > first.
> >
> > Please feel free to comment by email voting -1, -0, 0, +0, or +1.
>
> Arnulf / Markus,
>
> I've review various documents and two things come to mind as modest
> issues.  The primary docs are:
>
> http://wiki.osgeo.org/index.php/GRASS_Incubation_Progress
> http://wiki.osgeo.org/index.php/GRASS_Provenance_Review
>
> The issues were:
>
> 1) I don't see any mention of OSGeo on the main GRASS page or the "About
> GRASS" page.  I'd like some mention in one or both places of the projects
> connection to OSGeo.

It was already mentioned in the various release announcements
but not yet in the indicated pages. Now there is, too.

> 2) I looked at the code provenance review.  It seems clear that an
> automated review of copyright headers took place but it isn't clear
> to me how much manual review was done.

It was indeed a manual, interactive procedure. The automated part
included:
- search of main.c files without copyright statement
  (those with GPL statement have been vetted already 1-2 times earlier,
   see grass-dev archive)
- extraction of ChangeLog for the actual main.c file from CVS
- extraction of authors from Changelog file and draft insertion into
actual main.c
- open of text editor with following files:
  - main.c in process
  - related description.html (which contains authors at page bottom; in order
    to check for unmentioned authors from the pre-CVS times)
  - local Changelog file

Then the reviewer browsed main.c for offensive code (essentially incompliant
copyright notices and such), added missing authors to the copyright header from
description.html and polished the layout of the copyright header. Then
submission
to CVS (now OSGeo SVN).

Kudos to Schuyler Erle for writing the PERL script which quite helped
(tools/copywrite.pl). The expansion of CVS names was done by reading
in contributors.csv. I feel that this is a valid level of automatisation.

Again: we had already done rigorous code vetting when GRASS was
made GPL'ed software (1999).

Note:
All contributors with SVN access have agreed to RFC2:
http://download.osgeo.org/grass/grass6_progman/rfc/rfc2_psc.html
RFC 2: Legal aspects of code contributions

> The provenance review has a long
> list of directories which contain source but no indication if anything was
> found, or if they were even reviewed.

Right. I/we didn't do so since all is documented in ChangeLog (now trac).
http://grass.osgeo.org/grass63/source/snapshot/ChangeLog.gz

grep -i copyright ChangeLog
        * : copyright header added
        * : copyright header added
        * : copyright header added
        * : copyright header added
        * : copyright header fixed
...

grep -i copyright ChangeLog  | wc -l
86
grep -i provenance ChangeLog  | wc -l
27
grep -i review ChangeLog  | wc -l
31

Certainly above contains some false positives but you can see that
code has been revisited.

If you insist, I'll populate the Wiki page with notes.

> That said, I did a spot check of quite a few files and I see no problems.
> Headers seem fairly well maintained and I see no signs of outside code
> incorporated that would need deeper review.

Excellent.

> I vote +1 in favor of the motion for GRASS graduation.
>
> Hopefully some followup on the two points could be given but my vote
> is not conditional on that.
>
> Best regards,
> --
> ---------------------------------------+--------------------------------------
> I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
> light and sound - activate the windows | http://pobox.com/~warmerdam
> and watch the world go round - Rush    | President OSGeo, http://osgeo.org


Best regards,
Markus