NetNews Usenet Archive 1992 #19/NN_1992_19.iso/spool/comp/sys/hp/9807

Internet Message Format | 1992-08-29 | 3.7 KB

Path: sparky!uunet!ogicse!das-news.harvard.edu!cantaloupe.srv.cs.cmu.edu!crabapple.srv.cs.cmu.edu!andrew.cmu.edu!sean+
From: [e-mail] (Sean McLinden)
Newsgroups: comp.sys.hp
Subject: Re: /usr/lib/X11(R4)/config for 9000/700
Message-ID: <[e-mail]>
Date: 28 Aug 92 04:29:36 GMT
Article-I.D.: andrew.cebVkkG00iUyE1HUsR
References: <[e-mail]>
    <[e-mail]>
Organization: Carnegie Mellon, Pittsburgh, PA
Lines: 58
In-Reply-To: <[e-mail]>


>>Does anyone have a copy of the contents of this directory for X11R4 for the
>>HP. We are running HP-UX 8.02 and this directory seems to be missing from the
>>workstation even though the X developers's kit is installed. It sorta makes
>>imakes a little difficult.
>
>I have the same problem.  I also have all of the developer's kits.
>Still no imake and no config files.  HP is the only vendor that didn't
>supply these minimal tools and the Athena freely distributable
>libraries.  It's pitiful.

Yes. I realize that HP has a concern about Quality Control. But they could
have, at least, included these in an /usr/unsupported directory. And, in
particular, the files in /usr/lib/X11/config since building packages which
utilize X and imake is practically impossible without them.

>Unsupported, but working versions for /config and the Athena
>widgets can be found in hpcvaaz.cv.hp.com via anon. ftp
>in pub/MitX11R4 or pub/MitX11R5.

Not, entirely, correct. The config files for X11R5 are on this system. But
*not* the config files for X11R4 (which is what is currently shipping).

>Yes, it is pitiful (and a major pain to retype this message
>every two days or so :-) )

I believe that most of us appreciate the help given to us by various HP
employees via the netnews and other media. But an ounce of prevention is
worth a pound of cure. Retyping the message would not be necessary if
someone had given some thought to the process of configuring the standard
software distribution. Workstation vendors are starting to take on the
bad habits of the mainframe processor vendors in assuming that users want
shrink wrapped, packaged applications which require little or no thinking
to use and operate. The Unix philosophy is one of TOOLS, not systems. One
can only fully utilize a Unix system when they understand how to use it
as a toolbox in a fundamentally different manner than that with which they
use their PC or even their timeshared mainframe.

The traditional marketing approach made the distinction between end-users
and engineers but such distinctions are becoming more and more blurred as
tools for network information navigation, user interface tools (Tcl/Tk as
an example), and others allow non-programmers to begin to build sophisticated
applications. The speed of desktops systems now makes interpretive versions
of languages practical and this supports rapid prototyping by almost anyone
sophisticated enough to use shell commands and pipes.

Whereas 5 years ago you couldn't find five books on Unix in most bookstores,
there are racks of such documentation, now. Most of these, targeted at end
users, discuss shell programming, tools such as awk, lex, grep... and basically
emphasize the power that the Unix system user has over their system.

This blurring of the distinction between end-users and developers requires
more imaginative marketeering on the part of the vendors. If Sun and HP
think that I don't need lex, yacc, an ANSI-C compatible compiler, and the
files in /usr/lib/X11/config as part of my standard workstation then fine, I
can get them from the GNU effort. What is the point of making people ASK for
the basic tools necessary to make their Unix systems fully functional? Doesn't
that really contradict the philosophy which made Unix what it is, today?