Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 16 Jun 1998 00:42:34 -0700
From:      "Jordan K. Hubbard" <jkh@time.cdrom.com>
To:        Jay Nelson <jdn@acp.qiv.com>
Cc:        Sue Blake <sue@welearn.com.au>, ben@rosengart.com, Darren Reed <avalon@coombs.anu.edu.au>, hackers@FreeBSD.ORG
Subject:   Re: 2.2.6 CD-ROM : Package dependencies up the creek ? 
Message-ID:  <4550.897982954@time.cdrom.com>
In-Reply-To: Your message of "Mon, 15 Jun 1998 21:43:43 CDT." <Pine.BSF.3.96.980615194034.1224A-100000@acp.qiv.com> 

next in thread | previous in thread | raw e-mail | index | archive | help
> I'm more concerned about the clamor to make FreeBSD newbie friendly. I
> don't think Unix, of any flavor, is appropriate for a newbie who
> hasn't paid their dues and earned a basic understanding of the dirty

Erm, now I'm going to have to jump in on the other side of the
argument. :-)

Arguing that Unix shouldn't be user-friendly or that an expert-only
FreeBSD could be good or desirable is an argument you're only doomed
to lose since it's ultimately the users who will determine the size
and shape of this OS, and it's the natural tendency of every user to
want to make repetetive or complex tasks easier.  Saying that it's
somehow "wrong" to evolve the system along these lines is like saying
that we should all still be relying on horses for transportation
because automobiles pollute the air.  Don't confuse flaws in the
implementation for flaws in the ideal. :-)

FreeBSD *will* evolve, and user-friendliness is just one of the many
directions in which we'll see changes made.  If those changes are well
thought-out and do not _supplant_ existing features of the system,
then we will have the best of all possible worlds, something which
Win95 can never have since there's nothing worth-while under the hood,
so to speak.  If such evolution comes without careful thought and
design then of course we'll be a lot worse off, but are changes to the
VM system or device drivers any different in that respect?  A well
done UI "front end farm" which leverages off of existing tools and
provides easy configuration for a wide variety of system services
would be more than possible without seeking to replace "the old ways"
in any way (visualize multiple roads to Rome rather than one road
replacing another :).

If most attempts to provide decent front-ends to all these various
tasks in Unix have been rather unsuccessful, I think it's largely
because the designers didn't approach the problem with the same zest
that they may have approached, say, the task of writing the ultimate
ethernet card driver or a 3D game.  It's just not really all that
"cool" to write user front-ends and, to really do it right, you have
to be seriously into the idea of designing a comprehensive and highly
extensible framework from the ground up.  You also need to be very
much in-touch with what the users are really asking for vs what you
think it might be fun to design (though, hopefully, the two ideas are
not entirely mutually exclusive :) and you need to have a fairly good
grasp of human factors and good UI design, knowing just which parts of
Unix make good front-end targets and which ones simply _don't_.

You also say you don't like /usr/local/etc/rc.d, for example, but
don't indicate a better way that add-on packages for FreeBSD (and
there are now well over 1,000 of them) can add themselves to the
startup sequence, if necessary.  The need is there and it's not even
hard to imagine a point in time where even the most Exalted Guru had a
very hard time keeping his /etc/rc.local file up to date if it's your
suggestion that everything simply be tossed in there, "the old way."

Evolution will have its way, and the only item open to question is
whether its progress is deliberate or simply left to chance. :-)

- Jordan

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4550.897982954>