Re: ctime vs. Create Time

["Joseph Galbraith" <galb-list%vandyke.com@localhost>]

> unix doesn't have a concept of "creation time".  within the posix
> "stat" API, "ctime" is the "last metadata change" time whereas "mtime"
> is "last contents change"; "atime" is "last access time".
> 
> Note that update of atimes is commonly disabled for performance
> reasons.
> 
> If you are aiming for "least common denominator", the best you can do
> is a single timestamp for the mtime value.
> 
> If you are aiming for "model every known filesystem feature relating
> to dates", then we'll be here for a while...

What I'm really aiming for is "model the most commonly available
and used filesystem features relating to dates."

Both Windows and VMS store creation dates.  Windows and unix
optionally store acces time, but it is commonly disabled for
performance.  (I think VMS also can record access time,
but usually does not.)

So I would argue that Creation time is part of the
"medium common denominator"

What I wasn't sure of is if "metadata change" as
seperate from "content change" is interesting.

Windows does not track that data. I don't know if VMS
does.  I don't know if people use that data when operating
on files.

- Joseph