Ports with GUI configs
chuckr at chuckr.org
Thu Nov 15 19:00:06 PST 2007
> On Thu, 15 Nov 2007 16:00:55 -0500
> Chuck Robey <chuckr at chuckr.org> wrote:
>> I've already deleted the message that kicked me off, but it looked to
>> me that you were talking about the 10,000 ports I was talking about,
>> and that meant you were referring to new installs, not upgrades.
> Why would anyone want to configure ports they don't want to install?
>> BUT if you were talking only about upgrades, then I kinda think,
>> personally, that you probably should instantiated a new thread, not
>> used this one. Hmm?
> Is that supposed to irony, because before you hijacked this thread it
> was about preventing options screens being brought up at build-time,
> and pausing the build.
> Your ideas do absolutely nothing to address this issue because, they
> would only reduce the number of options, not eliminate them - unless
> you are intending to radically dumb-down the system. As a case in point
> take a look at the options for www/squid, I don't believe for a moment
> that your scheme could handle more than a small fraction of them.
> If people want an easier desktop system, they already have the pc-bsd
> and DesktopBSD versions of FreeBSD.
I need to take exception to that. My claim (and I have the messages in
which I made it) is that the setting of options needed these changes:
(1) To move the time that they need to be set, from ports compile time
to system install time, and
(2) To always phrase the questions in a form that non-technical users
can field, without extensive research that they are not equipped for, and
(3) To find a way to urge both ports-writers and ports users to share
the same notion of what the options refer to.
I think (I may be wrong, correct me if I am) that you were taking
exception, above, to my first point, right? You may correct me on that,
but on whether or not it will actually succeed in this is what all this
discussion is about. I did not bring this up without bringing the idea
past local friends, and defending it there, so I think I can do that.
Do i need to requote all of my arguments about that here (and really, by
now, boring folks to sleep) or can you look up the older posts? If
you've lost them, I've always had problems myself getting really recent
posts out of the archives, so write me privately, I will be glad to send
them to you. I do believe that it will perfectly accomplish exactly
what you claim "do absolutely nothing to address this issue", they will
100% move the work from ports build-time to system install-time. This
is pretty simple to prove, so I can't follow your assertion, and one of
us must have a disconnect here.
My points #2 and #3 are more arguable; I believe in them, but I guess
an argument could be made against this. There was a second part of my
argument, also (a list of regular-exceptions that ports, if they match,
would be rejected from), but that part would have somewhat less effect
(some, but less obvious). Both parts of my suggestion are things I'm
pushing, but at this point, I'm only at the point that I am completely
convinced that enough folks do agree with it to justify my writing the
Makefile mods. This isn't to say that the work I do will be approved
and brought in, you should know the FreeBSD development pattern well
enough to realize this, so lets not lose our cool.
Save that for the next time it's brought up, ON THE PORTS LIST, not
More information about the freebsd-questions