Sender:	crough45 AT amc DOT de
Message-Id:	<97Apr1.113615gmt+0100.16656@internet01.amc.de>
Date:	Tue, 1 Apr 1997 11:35:08 +0100
From:	Chris Croughton <crough45 AT amc DOT de>
Mime-Version:	1.0
To:	eliz AT is DOT elta DOT co DOT il
Cc:	djgpp AT delorie DOT com
Subject:	Re: ANNOUNCE: GNU Hello 1.3 uploaded to SimTel.NET

Eli Zaretskii wrote:

> It's just a non-DJGPP question, that's all.  The GNU standards are 
> described in the document called `standards.info'; you can get from 
> one of the GNU ftp sites, in the standards/ directory.

Ah, I didn't realise it was a GNU standard.  The GNU programs I've
downloaded and built on Unix don't seem to have much consistency
between them - not all of them even have a 'configure' script to 
select which compiler to use, and the directory structure seems
to be fairly random.  The DJGPP ones, on the other hand, seem to
be much more consistent.

> I don't see any problems with that.  In fact, one of the machines where I
> work on these ports has its gnu/ subtree on drive d:, whereas DJGPP is
> installed on c:.  What you need is just unzip a package that you need to
> build on the other drive, but there should be nothing in the source
> distribution that requires, say, gnu/hello-1.3 be a subdirectory of
> %DJDIR%.  If you see any problems with that, please describe them.  I
> usually make a point of specifically testing that the package builds even
> in another directory. 

The only one I remember offhand was libc, that explicitly uses include
files from ../../include and puts its libraries in ../../lib.  I'll
have a try with some of the others when I get home this evening.

> I just ftp to a US mirror early in the morning, when North America is 
> asleep.  The link is very fast then.

Where's that from?  My link from work (we refer to it as a "piece of wet 
string"!) objects to a lot of sites, some even the German ones.  Trier
is
good, ftp.coast.net (the original Simtel stuff) is reasonable, 
wcarchive.cdrom.com is reasonable; ftp.simtel.net and most other NA and
UK sites are bad.  Sometimes it refuses to admit that a whole top-level
domain even exists - it did that a few weeks ago with all the .uk
sites...

(Oh, to me 'reasonable' means over 1K per second - from Trier I can 
sometimes get 6K per second.  I realise that by the standards of most
university links this is a crawl...)

But I suspect that, as with the dancing bear, the surprise is "not how
well it dances but that it dances at all"...

Chris

- Raw text -