Re: Comments on MIME/SGML

[Steve Dorner <[e-mail]>]

At  9:10 AM 3/9/94 -0500, Jim Conklin wrote:
>Representing (I believe) lots of relatively unsophisticated users, I'd like
>to push for keeping text reserved for only the very simple things, please.

We have a type for very simple things; text/plain. I think "very simple
things" is not the right test for the entire text type.  Any non-trivially
formatted text/enriched body part will fail that test.  (Indeed, because of
the FIFO requirement on directives, it's likely to get unpleasant very
fast.)

I think the proper tests for text should be:

1. Can the typical user get reasonable use out of reading a typical
instance of the part in its raw form?

2. Can the typical text display system handle a typical instance of the part?

These tests are of their nature fuzzy and subjective, I agree; but so is
our concept of "text".

Particular users or systems with more severe restrictions can always key
off the subtype, if they don't want to display somewhat more difficult text
types.

I think we're seeing too much readable stuff being dumped into application.

--
Steve Dorner, Qualcomm Incorporated
 "Their speech is a combination of three technical terms I don't
  understand, and of one or two breathtakingly banal ideas." - Milan Kundera