minor slim-down in LFS-3.2-RC1 (e2fsprogs)
William L. Maltby
billm at wlmlx0.wlmcs.com
Wed Feb 13 13:53:46 PST 2002
Back in _THE_ _DAY_ (PWB V6/7, System III/IV) there were _no_ _symlinks. There-
fore, there was no need to distinguish hard/soft links. Although every directory
entry containing a valid inode reference was a "link", a single reference was not
referred to as a link. Only the result of an ln command was called a link. It
was _not_ hard or soft - just a "link" or "linked".
When symlinks became available, a distinction was required. In _standard_ jargon
the original file is not referenced in terms of "link" even though the Section 2/3
calls add/remove files with "link" activities. _Only_ the results of an ln command
warrant that distinction.
In pursuit of effective communication, we need common semantics. Although technically
correct, refering to a single _normal_ file as being linked promotes confusion.
BTW, the use count referenced proves to be very handy. The system knows not to
reuse blocks associated with a file until the use count achieves 0.
billm at wlmcs.com
In article <1013626299.1162.98.camel at gimli>,
marc at koelkast.net (Marc Heerdink) writes:
|> Op wo 13-02-2002, om 19:39 schreef Gerard Beekmans:
|> > On Wed, Feb 13, 2002 at 07:33:03PM +0100, Marc Heerdink wrote:
|> > > Op wo 13-02-2002, om 19:06 schreef Don Smith:
|> > hm perhaps you'd want to change that to 'write' instead of 'schreef' ;)
|> I'd love to, I've voted for that bug in Evolutions bugzilla like 20
|> times or so, but nobody's listening :(
|> > > Aren't all files just hardlinks? After all, all "files" just refer to a
|> > Yes in essence all files are links. We generally only talk about hard links
|> > though when two or more files are involved that point to the same physical
|> > data blocks on the harddisk.
|> "Generally" isn't always accurate :) so I wanted to point that out for
|> those who want an exact description.
|> - A Cow.
|> Marc Heerdink <marc at koelkast.net>
|> Unsubscribe: send email to listar at linuxfromscratch.org
|> and put 'unsubscribe lfs-dev' in the subject header of the message
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-dev' in the subject header of the message
More information about the lfs-dev