[sakai2-tcc] Proposal: eliminate all indies, re-version trunk to CLE 4.0

[David Adams]

I misunderstood the thing about Github in the original message.
Apparently it was discussed and put off at the conference, but I don't
see how waiting does anything but hurt the project's future. The
collaborative design of git fits much better into the style of
development we need to foster. As for Github, a project in Github is
light-years more friendly to new developers than the current
code-hidden approach with our source spread out in a morass of other
semi-related projects on the current SVN.

Just on a lark, I decided to time cloning Noah's sakai-cle github
project and checking out the sakai full trunk from SVN. Results? From
SVN, 2 minutes, 31 seconds. Cloining from Github? 17 seconds. And the
github clone comes with significant chunks of the revision history.
Clone a few more branches, and it just takes a few more seconds,
whereas in SVN, a second branch would take another 2.5 minutes.
Developers who don't yet know git can learn pretty much all they need
to know to get started in the two minutes they won't be waiting for
the check-out to happen...

David Adams
Director, Systems Integration and Support
Virginia Tech Learning Technologies


On Tue, Jun 18, 2013 at 3:15 PM, Steve Swinsburg
<steve.swinsburg at gmail.com> wrote:
>
>> Dashboard, ElasticSearch, Keitei, LTI 2.0, tincanapi, Samigo improvements, key library upgrades, eliminating indies, simplifying the build, etc.
>
> Yes, this is a great reason to upgrade! I think that with very little effort we can even slot in a few more major and mature improvements like Roster2, Syllabus, Signup and maybe a few more.
>
> Just a subtle note that with a major version switch this is an ideal time to move to github. We could just do it for Sakai 5, though that may be a few more years away, do we want to wait that long to get the efficiency and code quality improvements that we so desperately need?
>
> I don't particularly like Sakai 10 as a version number for the reasons outlined previously, too high a jump and aligning to calendar years locks us in too much.
>
> cheers,
> Steve
>
>
>