Re: Introducing mnxconverter (MusicXML-to-MNX converter)

Hi Adrian, Michael, Daniel, all,

My microphone wasn't working yesterday, so I was unable to say anything 
and came away a bit frustrated. Chatting via keyboard is no substitute 
for real conversation.

I'd like to iron out a little wrinkle (please correct me if I get any of 
this wrong):
As I understand it, we currently have the following chain of dependencies:
1. Michael is working on MusicXML 3.1 (or 4.0). This is fully backward 
compatible with previous versions of MusicXML, and is intended to be the 
last backward-compatible version of MusicXML ever.
2. Features that don't fit into (1) are being postponed to a new, 
breaking (i.e.non-backward-compatible) version of MusicXML. I'm not sure 
if this is intended to be a real (maybe unofficial) specification, or 
just a list of features that Michael thinks need fixing.
3. MNX-Common, as described in the "MNX-Common By Example" pages. Those 
examples are "cast in stone", but The Draft Spec contains proposals that 
still have to be agreed and finalized. /Some/ but not all of those 
proposals are /highly/ controversial, and need to be discussed.

Adrian is, I think, currently working to convert (1) into MNX-Common. 
That's fine, as long as we are just doing simple examples, but there 
will come a time when the aspects of (1) that need reviewing will need 
to be taken into account. (One of those is, judging by the Draft Spec,  
the absence of Digital Time)

As I said in a previous post, I'm currently working on an application 
that converts MNX-Common (from the "MNX-Common By Example" pages) into 
SVG. This project is going well, and I'll soon be making the code public.
So Adrian is currently the only programmer /writing/ MNX-Common files, 
and I'm the only one /reading/ them. I think this CG needs feedback not 
only from both these directions, but also from those having experience 
with other formats (e.g. MEI)
BTW 1: when I go public, can/should my repository be hosted by he W3C?
BTW 2: once my SVG-Writer is working, I should be able to provide some 
hand-written MNX-Common code demonstrating non-controversial proposals 
from the Draft Spec (e.g. the way multi-staff Parts work).

all the best,
James

https://james-ingram-act-two.de
https://github.com/notator


Am 30.04.2020 um 20:22 schrieb Adrian Holovaty:
> Hi all,
>
> Thanks to everybody who joined our community group meeting today! 
> We'll post the meeting minutes and video ASAP.
>
> In the meantime: One of the things I discussed was our new project 
> mnxconverter, a free and open-source Python library for converting 
> MusicXML to MNX-Common.
>
> Unfortunately the GitHub repository wasn't yet publicly available 
> during the meeting — but immediately afterward we got word from the 
> W3C that they were able to add the appropriate permissions for us. So 
> the library is now public!
>
> https://github.com/w3c/mnxconverter
>
> Yes, the MNX-Common format is still being designed, so this is a bit 
> of a chicken-and-egg thing. But my hope here is that having working 
> code will (1) help us conceptualize MNX-Common more easily, (2) 
> recruit project involvement from developers who are more comfortable 
> with reading/writing code than reading/writing specifications, and (3) 
> eventually serve as a production-ready tool to help notation 
> applications import and export MNX.
>
> There's much more context in the readme, which is visible directly at 
> that link.
>
> One more thing: During today's meeting, I also shared "MNX-Common By 
> Example," which is a website we've had for a while, but several 
> attendees weren't aware of it. This is a collection of simple notation 
> examples encoded in both MusicXML and MNX-Common, to help you get to 
> know MNX-Common. Here's the link:
>
> https://w3c.github.io/mnx/by-example/
>
> Adrian
>
> --
> Adrian Holovaty
> https://www.soundslice.com/
> http://www.holovaty.com/