🚀 go-pugleaf

ag129@cus.cam.ac.uk (Alasdair Grant) writes:
> My problem is
...
>   (b) that I seem to be unique in thinking that MIME could be a general
>       solution for all kinds of file structuring applications, such as
>       those currently handled by ODA, SGML and various ad hoc encodings,
>       or by context, or by making guesses based on file content or file
>       name, in applications where the concept of "headers" is not present.
>       For example I see no reason why program source code should not use
>       the MIME tagging mechanism to indicate which language it is in, and
>       cause the appropriate compiler to be invoked on its content.

Well.. if you'd mentioned this initially it might've saved some problem.. ;-)

It's been pointed out that the gophers and waisers are using MIME in
non-mail applications.

MIME could be a file structuring thing as you're suggesting.  It's was
designed with mail systems in mind but I can't think of many things which
are unique to them.  The most obvious is encoding anything not 7-bit clean.

But why is MIME *uniquely* suited to this purpose?  The MacIntosh
environment, for example, has its own way of tagging which compiler
to use.  I personally would like to use SGML to edit my source
files, have an SGML->particular-language translator, etc.  It would
do Knuth's ``literate programming'' one better by using a widely accepted
standard markup language at least.  But this is orthogonal to the issue
of whether there should be a header in the file marking what kind of
file it is.


<- David Herron <david@twg.com> (work) <david@davids.mmdf.com> (home)
<-
<- "That's our advantage at Microsoft; we set the standards and we can change them."
<- Karen Hargrove of Microsoft quoted in the Feb 1993 Unix Review editorial.