lilypond-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Google Summer of Code 2015


From: David Kastrup
Subject: Re: Google Summer of Code 2015
Date: Thu, 19 Mar 2015 09:10:16 +0100
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/25.0.50 (gnu/linux)

David Garfinkle <address@hidden> writes:

> Okay, thanks David for the quick reply. I have put my application in a google
> doc here: 
> <https://docs.google.com/document/d/1adMWBHlaKKFjznk-2AWly2YzPcjmTsHL-_d9DUxIYCw/edit?usp=sharing>
>
> If you could briefly look it over, that would be great. I will send it in to
> the google-melange website on 
> Friday.

Well, the bottom first: you'll don't need "fluency in MusicXML".  Then
you write

    The skills that I need to learn include understanding of the
    differences between ML and Scheme, obtaining a good understanding of
    Guile’s SXML parser, and becoming fluent in the MusicXML format.

I don't really know what you mean by "ML".  Guile does not really have
an "SXML parser" as far as I can tell: the point of Scheme is that it
has an easily readable/writable syntax for lists and various data types,
and both programs as well as data are represented in it.  SXML is just a
Scheme-compatible rendition of the data stored in XML.  You can write
and read it with the normal Scheme input/output in which case it will
look like "SXML": GUILE has no particular parser for that.  It's just
the result of calling the normal Scheme input/output on the SXML data
structures.  Or you write and read it with the functions provided by
GUILE's XML libraries in which case it will be proper XML.

> At this point I think I need some clarification in terms of the overall plan
> of action. I have some questions for you:
> First, what are your milestones for me over the course of the summer; can
> you give me a general timeframe for these goals?

Not really since I have no idea about your previous contact with Scheme,
LilyPond, and your learning style.

> Second, when we reach the midterm, in what way will you assess and
> determine a pass?

We are talking about tutoring here, not a university course.  You sound
like you expect a predefined schedule, exams, course material, tests.

This is not how it works: the main work and responsibility lies with the
student.  The tutor's task is to make the student work efficiently by
having the student avoid to get stuck unnecessarily for lack of possibly
readily available or at least known information.  It is to provide help
crossing hurdles, not dragging the student across the track.

By midterm I'd expect the capability of exporting an XML rendition of
LilyPond music input and a tenable strategy for how to export actual
MusicXML, likely by having a way to make SXML conversions.

> Third, if I am unable to complete the project, is my work recyclable
> (and if so, what parts) by the Lilypond community?

That depends on its quality, design documentation, and how far you
managed to get.

> Fourth (in response to Paul) I hadn't realized that this project had
> been in the works since 2008.

It hasn't.  Just because somebody registers some functionality as
desired in the issue tracker does not mean that any work will
automatically commence.  We don't have a surplus of developers with no
projects of their own on our hand.

> Out of curiosity, what has been the major challenge for developers
> over the years?

Starting.

> Would I be starting anew or recycling previous attempts?

You'd be starting from scratch.  As far as I know, there are no previous
attempts.  Three years ago, there was a LilyPond developer meeting at my
house and some developers might have experimented with accessing
LilyPond internals.  I don't really remember whether this was with Nils
Gey (Laborejo), Patrick Schmidt (Philomelos) or someone else.  I don't
know whether this resulted in any code currently in use and it would not
likely be worth it to hunt after it (it would probably a matter of
catching up with a day of mostly but not entirely related work).

> Finally, over the summer I think I would benefit from a few voice
> chats over Ekiga. Can I suggest a schedule consisting of weekly IRC
> chats? This is also something they want to see in the
> application.

I don't see a problem with weekly IRC chats but I'd usually expect most
communication to happen by Email right when required (I tend to check
every few hours at least).

> Also, should I store my code on github for communication?

I'd prefer a branch in the LilyPond repository (that way nothing will
get lost and experimenting with it becomes trivial for LilyPond
developers), but you can obviously also put stuff on GitHub if you have
other people interested in that.

-- 
David Kastrup



reply via email to

[Prev in Thread] Current Thread [Next in Thread]