Mentor availability (was: GSoC 2016)

[Urs Liska]

Hi all,

as mentioned I have done a little "survey", asking all those currently
listed as (potential) mentors for GSoC projects. I asked them if they
are still available and if they consider "their" projects still current
and important. The results are not really surprising but still not
really encouraging.

Starting with the good news: Werner Lemberg is still ready and willing,
and he considers the project of adding glyph variants still valid.

The other side of the spectrum: Reinhold Kainhofer and Mike Solomon are
*not* available, at least for this summer.
Joe Neeman would - if at all - serve as a backup mentor if necessary.
Carl Sorensen seems ready to do something but doesn't feel being the
right person to act as primary mentor for the project he's listed for.
Instead he suggested a new project where he could be the mentor (copied
at the end of this message).

So what do we have?

  * There are six project ideas listed on the website.
  * All of them are still considered valid, but only one of them (glyph
    variants) is equipped with a mentor.
  * Three projects (grace notes, beams, build process) have someone who
    might consider being a secondary, backup mentor, but no main mentor
  * Two projects (slurs and ties, MusicXML) don't have any support at all

In addition we have one project (ScholarLY) with a mentor (me) and a
website patch and one additional suggestion (chord structure) with a
mentor (Carl).
This makes three "living" projects and five "zombies", which is far from
being satisfactory.

I think the fact that the situation has been so unclear is also a reason
why our GSoC ideas have so little impact - it simply doesn't feel right
when skimming through that page.

I suggest we substantially review the page, giving "actual" projects
some more room and condense the other valid ideas more to a sort of list.

Any opinions?

Best
Urs


PS:
Carls suggestion:

"
I would like to propose a project on adding chord structures to
LilyPond.  I envision a scheme definition for a chord, that would
capture the structural elements used in the composer's intent.

We've discussed this a bit on lilypond-user:

https://lists.gnu.org/archive/html/lilypond-user/2015-01/msg00065.html

I envision that the first thing to do is to define the data structure
(like the fret-diagram-verbose data structure).

The second thing to do is to develop a scheme procedure fill the
structure (kind of like fret-diagram-verbose).

The third thing to do is to define a routine that can convert from the
data structure to a set of pitches (this allows us to use the data
structure to generate notes, e.g. for tabulature or regular staff notation).

At this point, we would actually be able to use the chord structure in
music (although it wouldn't yet be convenient). 

Probably the fourth step is write a formatter (perhaps for
Brandt-Roemer, or maybe for Fake Book).

The fifth step would be to define an input format that could be
implemented in the parser as part of chord mode, along with the parsing
necessary to create the structures easily as part of chord mode.

I'd expect a GSOC student to do steps 1-3 for sure, and at least get a
beginning on step 4.  I would hope a student could get started on step
5, but I wouldn't expect it from a novice.  Of course, during the
bonding period I might change my expectations.
"




Am 21.01.2016 um 17:02 schrieb Urs Liska:
> Starting the next round somewhat earlier ...
>
> Surprisingly there also will be a Google Summer of Code 2016, and I
> think we should be better prepared this time and not miss a slot due to
> lack of students.
>
> Mentoring organizations apply between 8-19 February, students 14-25 March.
> IIRC in our case the "mentoring organization" wouldn't be LilyPond but GNU.
>
> But I think *now* is the time to start thinking about possible projects
> for this instead of waiting for some students to show up out of the blue.
>
> What would be a suitable approach?
>
>   * Thinking about a number of tasks that
>     - would be good for us to promote and
>     - feasible as a GSoC project and
>     - would have a suitable mentor available
>   * Advertising these ideas (along with encouraging own ideas) on our
>     website and mailing list, other locations for developers, personal
>     relations ...
>
> Any opinions?
>
> Urs
>
> _______________________________________________
> lilypond-devel mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/lilypond-devel