gcc cover wasn't a policy proposal (was Re: PLEASE, ENOUGH)
On Thu, Sep 09, 1999 at 07:18:05AM -0700, Jim Lynch wrote:
> perl invocation per gcc invocation?? You Better Let Users Turn It OFF.
> Do not depend on everyone wanting it, whatever it does (did you notice
> that: I don't KNOW what it does, nor do I CARE.)
>
> You can consider this a second SO LONG AS it can be turned off. If not
> and you force me to use it, then I need to cosider alternatives...
>
> Users have THE FINAL SAY as to what happens and what does not happen
> on their machine. I speak now as a user: YOU BETTER GIVE ME A CHOICE.
Since I suggested the gcc cover:
That was not intended as a policy proposal. That was an example of how
you might speed up autobuilders without a policy change.
[To elaborate slightly: If you were running an autobuilder, you'd set up
a directory with this cover program named gcc in it. If the autobuilder
is building a package whose name doesn't have contain "lib", "-dev",
"-dbg" or any blacklisted names it would arrange $PATH so that the
directory with the cover program is before /usr/bin/. This would yield
an immediate speed improvement for the autobuilder, without waiting
for any packages to be changed, and without a policy change.]
Once again: that was not a policy proposal, so there's no need to
second it.
--
Raul
Reply to: