On Thursday 23 December 2004 10:43 pm, you wrote: > On Thu, 23 Dec 2004, Michael C. Shultz wrote: > > Way back when, I used to get the INDEX when running cvsup unless > > I put it in the refuse file. Of coarse it was old and useless so > > one was better off building their own INDEX. So why cant we just > > receive the INDEX like we used to, but unlike before, l have it > > updated as often as is being done now? > > Because the way we used to do it required tremendous churn in the > CVS repository. And it was _still_ out of date.
We in user land don't see that end of things :) Why does INDEX have to go into the CVS repository anyways? Guess that is why you all changed things around? > > The fact that you can't refuse it is a bug, and has been reported > to jdp. In the meantime, a cron script that does something like a > fetch and then a copy is your best bet. It's really not a bother, and with DSL it takes like 30 seconds to download it, but as I'm sure you can see many new users are getting tripped up by the change because now there is this extra make fetchindex step. I really think making the INDEX files into a port would be the best solution because every one understands ports and it could easily be done without creating churn on the CVS repository, just don't use a distinfo file and then you could update the source tarball at will without bothering anybody. my 2 cents worth -Mike > > There's really quite a bit of email in the mailing list archives > about why this was done. The short summary is that in the short > run there is some pain but in the long run this is a win for the > project, and the users. > > mcl _______________________________________________ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "[EMAIL PROTECTED]"
