file dependencies and packages and [blocker] bugs

[Andrew Farris]

Nicolas Mailhot wrote:
> Anyway the whole argument stinks. Yes filelists make transactions
> slower. However these particular file deps will only cause the file
> lists to be pulled when one of the aforementionned badly coded marginaly
> used games is installed or updated so
> 1. this won't happen for the vast majority of users or updates

Yes you're right my prior email was a bit broad, but the game packages may not 
be the only ones doing something of this nature with filedeps that could be avoided.

> 2. the costs are paid by the users of the problem packages

No, the 'costs' are paid by mirrors serving bandwidth for users with those 
packages installed.  Yes its only the users with them installed, but the 
filelists are downloaded *many* more times than the offending packages (assuming 
in those cases it is only that game package causing the download and not the 
probably numerous packages with a similar problem).  The filelist can be 
downloaded when the game causing the download is not changed and not being 
downloaded, so I don't see it as a direct tradeoff (correct me if I'm wrong here).

> 3. a single font file at gzip -9 is already more than 300K so the
> savings for a user that installs multiple games are not obvious
> 4. it's a tempest in a teacup
> 
> So I really don't see what's so terrible that requires making Hans
> uglify its specs and reneging on our general modularization and
> de-duplication packaging principles.

I think what Seth Vidal's original email was looking into was a bit more of a 
problem than a surface issue with one or two offending game packages.  But maybe 
it really is not that big an issue.  These just happen to be packages that do 
highlight the issue a bit in a way that can be discussed around, and at least 
for me I have seen its effect.

I think it is a little bit of an issue when my rawhide boxes have pulled the 
filelists at least daily for awhile now... my connection is fast, but is that 
bandwidth really necessary?  Is this idealistic de-duplication principle really 
that valuable in these fringe cases?  If so why?  For the font file case it is 
clearly a choice between install size vs bandwidth, so which 'costs' more?

-- 
Andrew Farris <lordmorgul at gmail.com> www.lordmorgul.net
  gpg 0xC99B1DF3 fingerprint CDEC 6FAD BA27 40DF 707E A2E0 F0F6 E622 C99B 1DF3
No one now has, and no one will ever again get, the big picture. - Daniel Geer
----                                                                       ----