CVSUP and 5.2.1 RELEASE

[Poul-Henning Kamp]

In message <p06002065bd1a0a3116a6@[10.0.1.3]>, Brad Knowles writes:
>At 11:49 PM +0200 2004-07-13, Poul-Henning Kamp wrote:

>	The way to fix this problem is to fix it.

Correct.  Removing the "world" target as the original poster
proposed is not a fix.

>	All that stuff is wrong anyway.  There's no reason we should 
>continue to support that.

Hey, pc98, ia64 and IPv6 are "just wrong anyway" too :-)

There are countless people out there with working scripts which use
make world and there is no reason why we should rip the carpet out
from under them either.

>	If you've got one, I'd love to hear it.  So far, I haven't heard 
>of anything better.

I havn't heard any workable solutions so far.

The fact that "make world" is used as the canonical example of the
benefit of FreeBSD in countless piecs of advocacy means that the
only way this can be fixed is by making make world fail when it
should.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.