Financial identity is *dangerous*? (was re: Fake companies, real money)
Roy M. Silvernail
roy at rant-central.com
Fri Oct 29 11:45:41 PDT 2004
Dave Howe wrote:
> Roy M. Silvernail wrote:
>
>> I'd thought it was so Microsoft could offer an emulation-based migration
>> path to all the apps that would be broken by Longhorn. MS has since
>> backed off on the new filesystem proposal that would have been the
>> biggest source of breakage (if rumors of a single-rooted, more *nix-like
>> filesystem turned out to be true).
>
> To be fair to MS, that is already here - you can "mount" NFS volumes
> as subfolders in 2K and above, just like unix. however, MS don't
> really seem to want to crow about that - just in case someone points
> out unix did this literally decades ago....
I was thinking more of the rumor that Longhorn's filesystem would start
at '/', removing the 'X:' and the concept of separate drives (like unix
has done for decades :) ). When I first saw this discussed, the
consensus was that it would break any application that expected to use
'X:\PATH'-style filenames or chdrive() (or whatever that lib call to
change the default drive is). Someone suggested that MS might ship an
emulator to handle translation (at some non-trivial cost in performance,
else no one would have an incentive to refactor) until the vendors could
rewrite their apps to use the new native filesystem.
--
Roy M. Silvernail is roy at rant-central.com, and you're not
"It's just this little chromium switch, here." - TFS
SpamAssassin->procmail->/dev/null->bliss
http://www.rant-central.com
More information about the Testlist
mailing list