On 2011-01-31 15:20, Greg Troxel wrote: > Kevin Reid <[email protected]> writes: >> On Jan 31, 2011, at 2:35, Zooko O'Whielacronx wrote: >> >>> 4. The misunderstanding that Tahoe-LAFS puts secrets into filenames >>> which (if I understand correctly) James Donald was under when he >>> posted http://tahoe-lafs.org/pipermail/tahoe-dev/2011-January/005966.html >> >> This reads to me as non-agreement about "what is a filename" not "what >> does Tahoe-LAFS do with filenames-in-sense-a". > > True, but tahoe breaks the dominant paradigm. Since long ago, > filesystems have had names, and inode numbers. inode numbers have > always been non-interesting; you know they are there and have to be kept > track of, but many users are never aware of them, and even the > uber-nerds look at them very rarely. caps are partly like inodes, but > due to the decentralized nature they are also used as mount points -- > which inodes basically never are. I think it's this use of caps as > mount points (or faux mount points, for WUI/CLI) that confuses people > into thinking they are names.
Caps *are* names, in a more general sense of "name". I think that's the sense James Donald was using. However, the Tahoe CLI almost implements a petname system already, where the aliases are petnames. It does not map the caps back into aliases when printing them, as a full petname system would, but that doesn't matter much because it very rarely prints them. -- David-Sarah Hopwood ⚥ http://davidsarah.livejournal.com
signature.asc
Description: OpenPGP digital signature
_______________________________________________ tahoe-dev mailing list [email protected] http://tahoe-lafs.org/cgi-bin/mailman/listinfo/tahoe-dev
