Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 22 Jan 1997 11:53:55 -0800
From:      "Jordan K. Hubbard" <jkh@time.cdrom.com>
To:        John Fieber <jfieber@indiana.edu>
Cc:        Larry Lee <lclee@primenet.com>, chat@freefall.freebsd.org, config@freebsd.org
Subject:   Re: Commerical applications 
Message-ID:  <25086.853962835@time.cdrom.com>
In-Reply-To: Your message of "Wed, 22 Jan 1997 14:16:41 EST." <Pine.BSI.3.95.970122141140.1465S-100000@fallout.campusview.indiana.edu> 

next in thread | previous in thread | raw e-mail | index | archive | help
> The best interface designers are often lousy programmers, if
> programmers at all, thus a designer/programmer team is required. 
> The FreeBSD development model does not *appear* to support such
> teams very well; everybody wants to "own" their project and the
> designer/programmer arrangement can so easily turn into a
> master/slave arrangement.... 

Well, I like to think of myself as a fairly good interface designer
(and please don't take sysinstall as an indication of said skills
since that was a quick hack using UI tools equivalent to stone knives
and bearskins :-) and a reasonable programmer, but I don't mind being
involved in either role.  I simply gravitate towards design because
I've seen and done more than my share of both good and bad interfaces,
and I think I have a reasonable idea of how to do it right now.

However, we still suffer from a bad case of "tools gap", where there
is really no reasonable framework for allowing the UI people to easily
practice their craft.  Since we can't count on an X server being
present (either due to configuration hassles, non-support for a given
card or other hardware issues), we need to have support for some kind
of CUI paradigm which also doesn't lock us into CUI forever and make
us look crude to those people who do have access to X displays and
tools.  We've talked plenty about various sorts of universal forms
translators or HTML based approaches, but the fact remains that there
are simply too few bodies willing or able to provide these tools.

I also don't think that design is really the issue here since I'm
pretty sure that we all basically know what we need and Windows, for
better or for worse, has already shown us much of the way.  We need a
hardware configuration tool for configuring, creating and installing
kernels.  We need a user configuration tool for adding, deleting and
otherwise managing user accounts and groups.  We need a login
configuration tool for setting desktop preferences, login defaults,
quotas, etc.  We need a network configuration tool for managing
interfaces, DNS, routing, IP aliases, etc.  We need a network services
tool for turning services like telnet, rlogin, ssh, etc on and off and
configuring their individual settings.  We need a printer setup and
queue management tool.  I could go on, and these are just the work of
about 2 minutes thought - if I really put my mind to it, I could
probably list another 3 or 4 major categories in which we have no
configuration coverage at all. :-)

And tying all of this together, we need a hypertext help system (Hi
John! :) which enables the user to actually find out how to drive all
this crap.  That's kind of why I like the idea of an HTML based tools,
even with all the limitations of HTML - it lets you integrate the
whole concept together at a level previously unheard of, basically.

Also, every time this subject comes up, about 2 or 3 people pipe up
unnecessarily saying "Argh!!  Please don't remove the ascii
configuration files, please please!"  Relax, folks.  Never in my
wildest dreams have I ever conceived of adding such ease-of-use
services on top at the expense of the mechanisms below, and I don't
think that anyone else here has either.  One of the big strengths of
UNIX is that you can really get down into its guts and tweak it when
you have to, rather than simply being left high and dry as you are
with Windows ("No button for that, sorry!").  Nobody's talking about
giving that up, simply automating the configuration process for those
who wish it and putting a pretty face on the whole thing.

However, there's just a lot of work there, plain and simple.  We don't
even have the tools for writing all those nice looking configuration
screens yet, and we need to get that situation taken care of before
any progress will ever be made in this area.

					Jordan



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?25086.853962835>