Re: nice troll (was: All this resource-fork AKA multiple stream

Theodore Y. Ts'o (tytso@mit.edu)
Thu, 8 Jul 1999 18:58:30 -0400


Date: Tue, 6 Jul 1999 23:42:21 +0200 (CEST)
From: Rik van Riel <riel@nl.linux.org>

[Alan sez: "kernel source tree is compound document"]
> > There isn't a well-defined and obvious way to print the
> > whole thing.
>
> find . -name "(*.[chS]|Makefile|*.dep....)" -exec lpr $* \;

Why the .dep files?
Why not the Documentation/ directory?

I guess these questions just about prove that Albert is right
in that there's no well-defined way to print the whole thing.

Albert's contention that there is no well-defined way to print the whole
thing isn't a good reason to say something isn't a compound document.

There's no well-defined way to print everything in a MS Access
database. Or a Excel spread sheet. Or a Word document. (Remember that
it's more than just the actual text or spreadsheet, there's also the
Visual Basic code, and hidden variables, etc.)

If the sections are fully independant, then almost by definition
it's not a compound document.

Ah, but the sections of the kernel aren't fully independent; each .c or
.h file aren't particularly useful in isolation, since they refer to one
another. If not, what's the difference between that and a TeX document
or a Word document which has embedded GIF images? The embedded GIF
images are independent data streams, but are referenced by the Word
or TeX document.

Why is it OK to demand that the GIF image be bundled with the Word or
TeX document, but say that individual .c and .h files in a kernel source
tree aren't part of a compound document? The point which Alan raises is
a very good one.

- Ted

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/