[gmpi] Re: Reqs 5, 6, 11 for debate

  • From: Paul Davis <paul@xxxxxxxxxxxxxxxxxxxxx>
  • To: gmpi@xxxxxxxxxxxxx
  • Date: Thu, 20 Nov 2003 13:13:55 -0500

>Here's what I've become convinced of, at this point.
>
>1) There are two layers of API being discussed.
>   a) A host-provided API for the host to trap some common actions
>   (hereafter called the host services API).
>   b) A generic portability layer which may or may not be host-centric but
>   is surely platform specific (hereafter called the portability layer).
>
>2) The host services API is needed, but needs to be minimized and justified.
>
>3) The portability layer sounds like a good idea at first.  Upon further
>examination, it is NOT the job of GMPI to force you to write portable code.
>Plugin coders who want to be portable will write portable code.  Those who
>don't won't.  Trying to mandate it is a waste of time.
>
>4) While we can not mandate a portability layer, we can endorse one of
>the existing portability layers, or we can fork a sub-group to write one.
>Our endorsement is merely a reccomendation.
>
>Do these sound right?

sounds right to me. 

however:

i would say though that our "recommendation" should be very strongly
put. existing proprietary plugin and host authors might be primarily
interested in GMPI as a way of reducing the number of plugin APIs they
support (and perhaps the control issues too). realistically, this is a
long way off. VST/DX/TDM/MAS etc. will not just evaporate once a
working GMPI appears.

otoh, a working GMPI means that in theory it should be possible to get
a working (if not entirely optimized) plugin for any platform that
supports GMPI merely by recompiling. from my perspective, this is the
biggest draw of GMPI. the other plugin APIs pay lip service to the
idea of not directly accessing system services, but they do nothing to
enforce it. with GMPI, we have a *chance* to create plugins that steve
could write on a linux system, yet could be easily recompiled for some
other platform.

i know that from my perspective as a dwindlingly-philanthropic Libre
software author, *this* is the biggest draw of GMPI to me. leaving the
issue of accessing system services on the sidelines may eventually
help reduce the number of plugin platforms, and will certainly provide
one controlled by an independent body, but it won't do anything to
encourage platform neutrality in code that *can* and IMHO *should* be
platform neutral.

--p

ps. someone reported ardour compiled and running on mac os x a couple
    of days ago. cross-platform issues are on my mind :)

----------------------------------------------------------------------
Generalized Music Plugin Interface (GMPI) public discussion list
Participation in this list is contingent upon your abiding by the
following rules:  Please stay on topic.  You are responsible for your own
words.  Please respect your fellow subscribers.  Please do not
redistribute anyone else's words without their permission.

Archive: //www.freelists.org/archives/gmpi
Email gmpi-request@xxxxxxxxxxxxx w/ subject "unsubscribe" to unsubscribe

Other related posts: