Re[2]: image3D

Alan Peltzman ([email protected])
Thu, 23 Mar 95 13:16:21 EST


Keith,

What would you recommend as the
proper MIME type for 3D geometry?
Do you think we should have a new
MIME type?

I would very much appreciate
a cogent recommendation.

- Alan Peltzman

______________________________ Reply Separator _________________________________
Subject: Re: image3D
Author: Keith Moore <[email protected]> at smtp
Date: 03/23/95 02:21 AM

> >Now what about the existing mime types
> >
> > text: your basic terminal
> >
> > audio: an audio output device
> >
> > image: a display device (monocular)
> >
> > video: moving image capability (monocular)
>
> Philosophically speaking, things have already broken down.
> If you have image, you've got video capability.

Actually, no. Fax machines have image capability but not video
capability.

> I am aware of patented & proprietary techniques that
> enable binoccular vision on ordinary displays (i.e.
> give you illusion of stereo viewing & depth, but require
> no specialized hardware to acheive the effect).

not even 3-d glasses? :)

Actually, I'd think that 3d models require considerably more hardware
support than (say) stereoscopic images. For instance, you would want
to be able to change the observer's position relative to the model,
which you can't really do for a stereoscopic image.

> > video3D: moving 3D image capability
>
> And, since moving images are "just" a sequence of stills,
> this is redundant as well.
>
> So the real question is, why did the IETF think that video is,
> in some "fundamental" way, different than image?

a) you can't usefully display video on a fax machine or a printer.
b) video typically requires the ability to update at the rate of at least
a few frames a second, image doesn't require that kind of frame rate.
c) if you have the option, you probably want to use different protocols
to transfer video (for real-time display, anyway) than you do to
transfer still images (i'm not talking about the organization of
the bits here -- I'm talking about how the transport protocol handles
packet delay and loss)

All of these are important differences. You could easily have the
hardware capability to accept images, but not video, via email or www.

If you're sitting on the nether end of a lossy 14.4kbps SLIP link, you
might well want to tell your upstream mail gateway to discard any body
part that has content-type video/* (or tell your web browser to not
accept them).

Keith Moore