Excerpts from internet.ietf-822: 30-Oct-91 Re: Audio type Marshall
Rose(_at_)dbc(_dot_)mtview (542)
Rather that suffer through a proliferation of
audio/u-law
audio/pcm
...
it's probably better to provide a single-type that can represent them
all and then mandate the basic minimum for interoperability. This is
what the current Audio/Basic definition does.
You know, I'm not an audio expert at all, but I have to disagree with
this. In particular, with the audio/u-law, etc., scheme, at least there
was a way for very simple software (even shell scripts, for example) to
package up an audio message from "standard" components (e.g. raw u-law).
With the audio/basic scheme, you need to slap on a header first, which
complicates matters slightly. Not a big deal, but I don't think it is
as cut & dried as you imply.
I keep thinking that the NETFAX working group provides a better model
for the process of converging on image mail formats than we have for
audio. The IETF had a whole working group that provided us with NETFAX,
and the mail document references that, while providing ways to tag other
image formats. I suspect a similar approach is best for audio -- let
the experts define a format, which is recommended for mail use, but
provide a way to tag other widely desired types, such as sun-format,
next-format, raw u-law, etc. The problem is that there's not currently
a working group for audio formats. This leaves us with two choices:
Write it that way anyway, and hope that an IETF AUDIO group will come
about some day, or simply take Marshall's suggestions in the absence of
any other audio expertise on the mailing list. I believe that if
Marshall weren't so widely and deservedly well-respected, this latter
option wouldn't even be taken seriously. As it is, I'm still very
uncomfortable with it, but I will certainly defer to the will of the
group, whatever that turns out to be. -- Nathaniel
PS -- IN THE ABSENCE OF OTHER INPUT, I will assume that most people are
happy with the "just use Marshall's proposal" approach. Marshall
clearly knows more than I do about audio formats, and I'm not prepared
to try to shoot down his approach -- which, for all I know, is the right
one -- singlehandedly.