Re: MIDI spec questions

Hmm.  I'm just going off my Macs, which all stopped having a GM synth
pre-installed.  Could be a config problem.



On Mon, Mar 9, 2015 at 10:12 PM, Tom White <twhite@midi.org> wrote:

>  Chris,
>
>  Windows (desktop) still has a GM synth. And though MS has not announced
> all of the features of upcoming Windows versions, I don't believe there is
> any reason to assume they won't also have a GM synth.
>
>  OSX did lose the QuickTime GM synth around 10.6, but IIUC there's still
> one in the OS (CoreMIDI DLS Synth) which I assume could be accessible to
> the browser if someone wanted...
>
>  Tom White
> MMA
>
>
>  Sent from my Sprint phone.
>
> ----- Reply message -----
> From: "Chris Wilson" <cwilso@google.com>
> To: "public-audio@w3.org" <public-audio@w3.org>
> Subject: MIDI spec questions
> Date: Mon, Mar 9, 2015 1:54 PM
>
>  I did some updates to the spec this morning.  We still have six open
> issues:
>
>    1. Cross-platform persistent identifiers:
>    https://github.com/WebAudio/web-midi-api/issues/124 - I think this
>    should be solved along with VSTis and virtual MIDI ports.  I think this can
>    be closed as essentially a duplicate of #45 below.
>    2. MIDI input/output disconnection:
>    https://github.com/WebAudio/web-midi-api/issues/79 - This is pending
>    some final wrangling mostly between myself and Takashi; should be resolved
>    in the next day or two.
>    3. General MIDI support:
>    https://github.com/WebAudio/web-midi-api/issues/110 - I would like to
>    won't-fix and close this.  The only thing I could see us doing here is some
>    way of the user specifying a "default General MIDI device".  As these are
>    usually no longer present (Windows and OSX don't ship a default device
>    anymore, IIUC), and features that require user configuration tend to almost
>    never work (and the user could config it in the web app anyway), I would
>    strongly suggest we won't-fix this.
>    4. Web MIDI should be available from Workers:
>    https://github.com/WebAudio/web-midi-api/issues/99 - I think we need
>    to decide if this is critical for v1 or should simply be a future
>    enhancement.
>    5. Does MIDIPort.id need to be origin-scoped:
>    https://github.com/WebAudio/web-midi-api/issues/79.  I think the
>    answer to this is "no", as per outlined in issue comments.
>    6. Virtual MIDI ports:
>    https://github.com/WebAudio/web-midi-api/issues/45 - I think we agreed
>    at the FTF this should be v2.
>
> As I said, I think point 2 will be worked out in the next couple of days;
> the rest of these I am actively proposing resolutions for (1 resolved as a
> dupe of 6, 3 and 5 won't-fixed, and 4 moved to v2.)  I would welcome input
> on these issues.
>
>  -Chris
>
>