Re: Ripping out GLib bindings, depending on GLib

[Rob Browning]

Andreas Rottmann <address@hidden> writes:

> Well some argue you save a bit of disk space without all those
> .arch-id/*.id files, but that's not too convincing.

Ahh, you mean b/c of all the extra inodes.  I hadn't thought of that,
although I had the impression that arch's current implementation was
fairly unconcerned about disk usage anyway.

FWIW looks like we do use about 138k for the .id files (acc to du, on
my system with 2k blocks).  That's excluding {arch}/++pristine-trees
which requires 790k (including its .id files).

> Also, it shouldn't be a problem to switch to tagline from explicit
> tagging, if we ever happen to change our minds.

I didn't think you could do that right now and preserve the file
history, but I seem to recall a discussion about implementing
something to make that easy in the future.

In any case, I'm still leaning toward explicit tags unless anyone's
opposed, but if so, please feel free to argue the point.  I have a
re-done tree that's about ready to go (including all the
.arch-inventory files to cover the autotools and build detritus), but
it'd be easy to re-do it with arch-tags:s right now.

-- 
Rob Browning
rlb @defaultvalue.org and @debian.org; previously @cs.utexas.edu
GPG starting 2002-11-03 = 14DD 432F AE39 534D B592  F9A0 25C8 D377 8C7E 73A4