Well, it was just a suggestion, so granted there are some minuses, but...
it could - *if* we want a sloppy and inefficient implementation. ACLs have arbitrary length and are in the vast majority of cases inherited from the parent directory: NTFS, in fact, collapses identical ACLs. And
And the collapsed binary version couldn't be stored in a binary attribute? Attributes of parent nodes couldn't be checked for inheritance?
that's just an example. Just speaking feature-wise, NTFS has DOS attributes and DOS short filenames, OS/2 extended attributes, sparse files, compression,
various filenames and OS/2 attributes could be stored in extended attributes (easily), and even indexed (BFS) for efficient searching capabilities. As for compression, you can still compress the data and use an attribute to store the compression status; same with encryption. Named streams could either be dropped, or supported by storing in multiple files. object IDs could be stored in attributes.
encryption, named streams, per-file object ids, reparse points and god knows how many minor ones I've left out. But that's just overlooking the by far biggest issue: what about user ids? Windows NT has globally unique user ids of arbitrary length, vs the fixed-length, locally unique user ids
SIDs could be stored in attributes, and the attributes could be indexed for efficiency -- this is all supported quite nicely by BFS.
normally found elsewhere. Before you ask: yes, files need to be indexed by SID. Quota management requires it. Good luck implementing all of that with extended attributes
Compression and encryption are separate issues. BTW, the suggestion was to support a filesystem OTHER THAN existing ones, so expect different advantages AND disadvantages...
All of those features of NTFS imply added storage overhead, too, which makes it impractical for use on smaller volumes (floppy/zip disks, for example). Of course, this is true of many modern filesystems, which is probably the REAL reason that FAT is still so commonly used on those devices.
===== ======= Frank D. Engel, Jr.
Modify the equilibrium of the vertically-oriented particle decelerator to result in the reestablishment of its resistance to counterproductive atmospheric penetration.
__________________________________ Do you Yahoo!? Protect your identity with Yahoo! Mail AddressGuard http://antispam.yahoo.com/whatsnewfree