[sakai2-tcc] Planning for Sakai 2.10 and beyond

[Matthew Jones]

I believe that most of the items that Mark suggests could be possible. IF
we had 1-2 developers dedicated to working on issues that were of highest
importance to all members of the CLE community, similar to what we existed
on OAE. Right now, all we have are developers interested in local efforts,
nearly all of which amount to either very incremental improvements or
wholesale replacements prospects that 4 times out of 5 fizzle out before
completion. Profile2 being the best exception of one that was completed.

Three core "LMS" functions of the CLE that were missing from the OAE:
Grading, Assessments and Assignments have essentially been unchanged since
the beginning of the project. That's not to say attempts were made to
improve them in the CLE.

- Assignment 2 (A2) was started by IU years ago across multiple developers
and has some features that Assignment doesn't. (Revision history, improved
UI, better backend) However over the years, the gap has gotten wider. (as
it had between CLE/OAE) Many people (specifically outside vendors) have
believed A2 to be the eventual replacement and have contributed effort
toward that. At the conference we had a session where we talked about
needing to make one of them the *only* Assignments and seemed like A2 was
the consensus. However, new gaps including group assignments appeared and
nobody put up the specific resources (people/money) to close the gap. So
we're left where we are with that.

- Everyone always talks about Samigo, however it always *seemed* like Mneme
was the idea if the simpler more updated Samigo. The problem with this is
Etudes was not interested in getting it into core, and it was difficult to
get any features or changes into their code. (Similar to other open source
LMSes) So for us, it would have required maintenance a fork if we
considered Mneme to be a replacement assessment, which it was never debated
(and debatable), and there is/was still an active team on Samigo. I only
knew Mneme from starting on Sakai at Michigan (they didn't use Samigo) and
did prefer it overall, though heard Ambrosia (it's custom framework) did
have a learning curve.

- Gradebook 2 felt like it was solving the problem people wanted solving.
It was giving instructors a way to do grading like they've been grading in
the past both digitally in Excel and on paper; spreadsheet style in columns
and rows. However the usage of a GPL library prevented that from ever
realistically making it into core. And Canvas speed grader made me wonder
if this was really the the problem people were having. (Difficulty, slow
digital grading) Perhaps GB2 slightly solves the problem better of an
instructor manually entering grades for a students work submitted
physically but doesn't really change the workflow for electronic
submissions.

What would have really helped grading more is there was a way for external
tools to create more complex outcomes (LTI: Whole groups of grades for
separate items dedicated to that external service than just the one
pre-created item) And external api's to the Gradebook in general to allow
for custom UI's

Anyway, I'm not sure if in 2008/2009 when all of these replacement efforts
started a TCC/PC or anything could have better influenced the direction of
these. We'll certainly have to figure out what to do with large effort,
duplicate functionality projects like this so the same mistakes aren't made.

On Mon, Sep 24, 2012 at 10:47 AM, Charles Severance <csev at umich.edu> wrote:

> This list form Mark is a perfect example of what I am talking about :)
>
> At our current resource levels, it is 7-8 years of work - and this is the
> list of ideas from one stakeholder who is not even a teacher or student
> using Sakai on a daily basis.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://collab.sakaiproject.org/pipermail/sakai2-tcc/attachments/20120924/d98c8182/attachment.html