Re: Seldomly used files [Re: RFC/ITP: Constitution of Finland]
- To: firstname.lastname@example.org
- Subject: Re: Seldomly used files [Re: RFC/ITP: Constitution of Finland]
- From: Kevin Ryde <email@example.com>
- Date: 20 Apr 2000 08:45:34 +1000
- Message-id: <firstname.lastname@example.org>
- References: <20000417171819.A2139@zombie.inka.de> <Pine.LNX.email@example.com> <20000417225243.A18876@melkor.dnp.fmph.uniba.sk> <firstname.lastname@example.org> <20000418005857.B17215@molehole.dhis.org>
Steve Greenland <email@example.com> writes:
> On 17-Apr-00, 17:37 (CDT), Kevin Ryde <firstname.lastname@example.org> wrote:
> > I see the main advantages of a Debian package being that it integrates
> > with the system and can be removed or reinstalled easily. I wonder if
> > a "data-installer" package could do that for all sorts of net grabable
> > documents without them having to be archived on Debian servers.
> It exists already. It's called "wget".
This is what I use, with the urls and local directories in a script.
But it's sometimes tedious checking whether it's one file or several
that need mirroring, whether it has to be untarred and maybe
I'm just thinking it could be nice if random people could contribute
data descriptions in a standard form to smooth out mirroring/installing.
Certainly it'd be better than 200Mb packages with pictures of toenail
clippings in main.