Re: `du' control files
Hi,
>>"Hamish" == Hamish Moffatt <hamish@debian.org> writes:
Hamish> On Sun, Feb 15, 1998 at 08:13:29PM -0500, Adam P. Harris
Hamish> wrote:
>> I have to heartily agree with Manoj. It's a fundamental tenet in
>> data quality analysis that data *use* should drive data collection,
>> not vice versa. So until we really find how we need this file, how
>> we use it, we shouldn't bother collecting the data.
Hamish> Surely your argument applies equally to the md5sums file. Can
Hamish> you please post arguing for these to be removed also?
I would tend to agree. Maybe this shall prompt Kai to come up
with a solution for that too ;-)
Hamish> There is more effort in removing the files and adding them
Hamish> again later to every package than there is in just adding them
Hamish> to the packages that don't have them.
Yes, but the emphasis is on utility and correctnes. Even in the
simple case of checking file sizes, it was discovered that the format
was not the one wnated -- despite all care having been taken to
visualize what a non-existant tool might require.
Yes, doing things correctly often takes more effort (and, for
Debian, talk). In the log run, it takes more time (now we have to
wait till the older format of the du file is flushed out of packages,
ar [better] decree a new standard file name; in any case, providing
du files already saved no time at all).
manoj
who really should put on flame retardant under clothing
--
"It's a dog-eat-dog world out there, and I'm wearing Milkbone
underwear." Norm, from _Cheers_
Manoj Srivastava <srivasta@acm.org> <http://www.datasync.com/%7Esrivasta/>
Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05 CC 2D 27 12 1D F5 E8 6E
Reply to: