And the claim is that introducing "file system internal to the file"
is somehow going to make that faster.
And the claim is that using the existing file support is *bad* because
it will confuse the newbie?
For this model to even vaugely make sense, you're talking about internal
hard links from one document component element *into* another document.
And writing an application smart enough to know how to generate proper
backups vs. proper updated documents taking advantage of this complex
file concept.
[But what we're really talking about is making MS-Word run faster,
and probably about making Samba run faster, and for that we should be
focussing on WINE and Samba and maybe on kernel tweaks based on careful
analysis of WINE/Samba performance tests.]
If you're not talking about specifically enhancing MS-Word, and all
you're concerned about is write latency, note that it's relatively
trivial to fork and save in the background. It might be necessary
to enhance the kernel to reserve a certain amount of disk space (or
fail) right away, but that sounds a lot simpler than moving MS-Word
semantics into the kernel.
But maybe I'm just stupid.
-- Raul- 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/