Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.1 (Fortune) 6/7/84; site dmsd.UUCP Path: utzoo!watmath!clyde!bonnie!akgua!sdcsvax!dcdwest!ittvax!decvax!decwrl!amd!fortune!hpda!dmsd!bass From: bass@dmsd.UUCP (John Bass) Newsgroups: net.cse Subject: Re: learning unix or not learning unix Message-ID: <152@dmsd.UUCP> Date: Fri, 26-Oct-84 02:58:13 EDT Article-I.D.: dmsd.152 Posted: Fri Oct 26 02:58:13 1984 Date-Received: Mon, 22-Oct-84 06:48:52 EDT References: <392@sdcsvax.UUCP> Lines: 45 John Van Zandt raises some good points ... teaching a how to use UNIX class is clearly a waste of student/teacher time. However, I believe he totally missed the point of discussion. At hand is the issue of many university CS/ENGR departments still only providing 1960's computational resources which ARE NOT good examples of 1980's how to do things right (or the principles behind good system design). For a CS Department to teach principles without up to date facilities is equally a waste of student/teacher time. It is easy for staff in the UC system to accept UNIX as an every day tool that everybody just USES. UNIX access is provided as an important resource to most CS students in the UC system. However in many other campuses (like the Cal State system) UNIX machines are non-existant or locked away for Prof's and Grad students -- maybe a single 11/4? or 11/750 on the whole campus. In most of these schools CDC NOS or IBM 370 OS's are used and taught. This was fine in 1970, but it is like teaching VLSI design in the class room and vacumun tube technology in the lab. The resulting VLSI design major has missed 2 generations or more of technology and is a poor risk to industry without a LOT of additional training. The text book learning of VLSI was not backed up with practical experience. Only a program with CAD, 3 or 4 inch foundry access and real student chip projects allows the principles of VLSI design to be learned. Likewise a 1980's CS student who only has access to CDC NOS resources will miss all the fine points of two generations of Computer Science progress. Their personal tool kit and experience base is so limited that a LOT of additional training (or personal learning) is required to adapt to design issues of micro-computers, CAD/CAM workstations, personal workstations, networks, and distributed systems designs. Unfortunately, most of the micro computer systems built with 1980's hardware still use 1960's software technology ... at least providing UNIX on campuses gives the students access to late 1970's technology. Hopefully network and workstation technology will follow soon. UNIX is not the end product ... but is a minimum tool set to address the principles and problems associated with system design in the 1980's and beyond. John Bass