Not "totally" infeasible, but just incredibly biased towards a
Unix approach and the universal email problem.
Why is it biased towards UNIX? Couldn't we do the same on the Macs?
You could make a program for the Mac that allows a user to look at the
ASCII header of a file. Whether that header is derived from the Mac
file's data part or resource part seems to me to make little
You could build an NFS system that allows a UNIX system to mount a
Mac's files, in such a way that, from the UNIX system's viewpoint, the
file looks like a stream of bytes with an ASCII header. Whether the
Mac has this information in one block or in two (the data and the
resources), should not matter.
Also, to come back to the problem of migration again, if we update an
application and then prepend an ASCII header to its files, we will not
be able to use different, unupgraded programs on these files. So, for
UNIX, we should probably add a new system call (new_open, or
something), which reveals the whole file (header included) to new user
programs, while the old open() system call in such a new kernel only
reveals the "body" (no header) to old user programs. We would probably
need some sort of unique header (a long, funny string) that indicates
that the header is there, so that the kernel can tell whether or not
any particular file has one of these fancy new headers.
I'm afraid the
net result of that would be to position Unix even farther out
in hackers' left field, rather than helping to make it disappear
under a decent GUI.
I think that these new headers would allow you to build far more
powerful GUIs on UNIX than was previously possible. After a while you
wouldn't even recognize it as a UNIX system, since all the
applications will be doing fancy things with user interfaces.
...the new extended email
format is intended to allow a sender to transmit to a receiver,
everything the receiver ever wanted to know about the data.
Hmm. I thought that was what the DATA was for...
Exactly! The data describes itself.
This idea is *so* simple, that I can't possibly imagine that noone
else has thought about this... (I would like to know why those people
gave up. Does anybody know?)