[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Deleting uncompressed Info/Doc files at upgrades



Hi,
>>"Peter" == Peter S Galbraith <GalbraithP@dfo-mpo.gc.ca> writes:

 Peter> So, at least Craig Sanders and Santiago Vila are so engrained
 Peter> into Debian that they now think the original usable file is
 Peter> the gzip'ed one, not the author's original text.  I disagree.

	Count me in with Criag and Santiago. dpkg should never, ever,
 think it is smarter than the user. It has control over certain files
 on my disk; it should never touch anything else not in its
 jurisdiction.

	There is no reason ever to uncompress a file (lesspipe and
 lessopen make it unnecessary). When I uncomress a file, it is done
 for a reason, and dpkg had bloody well leave it alone.

 Peter>  - A new user won't know about special setups needed for emacs, less and
 Peter>    other program.  

	This should be fixed. The default lessopen script does indeed
 set it up so. 


 Peter>  - A new user may need the docs on a crippled system, or on a
 Peter>    system with only the base system installed.

	The base system has gunzip et al.

 Peter>  - If you are using some docs often on a 486, you end up
 Peter>    uncompressing them because it's too slow otherwise.

	In your particular case. Older machines also tend to be tight
 on disk space, and my 486 is still fast enough to use gunzip on the
 fly. 

 Peter> I'm not arguing that dpkg should handle .aux files files
 Peter> behind after someone has latex'ed docs.  I'm arguing that the
 Peter> `intent' of packaging a compressed file is to have the
 Peter> uncompressed original available on the system.  Debian
 Peter> upgrades should therefore acknowledge the possibility that
 Peter> files have been decompressed.

	I disagree quite strongly. If the intent was to have
 uncompressed originals on the system we would have shipped them as
 such. 

	manoj
-- 
 "Bush has it backwards -- abortion is surgical; bombing is murder."
 sign at anti-war march
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: