Path: utzoo!attcan!uunet!lll-winken!ncis.llnl.gov!ncis!helios.ee.lbl.gov!pasteur!ucbvax!decwrl!labrea!rutgers!tut.cis.ohio-state.edu!osu-cis!killer!elg From: elg@killer.DALLAS.TX.US (Eric Green) Newsgroups: comp.edu Subject: Re: Student Preparedness Message-ID: <6838@killer.DALLAS.TX.US> Date: 19 Jan 89 04:05:16 GMT References: <5382@pdn.UUCP> Organization: The Unix(R) Connection, Dallas, Texas Lines: 70 in article <5382@pdn.UUCP>, reggie@pdn.UUCP (George W. Leach) says: > In article <13432@bellcore.bellcore.com> duncan@ctt.bellcore.com (Scott Duncan) writes: >>Should very many be expected to commit to such an expense -- they have many >>courses and buying "every" book on the subject is a tremendous commitment! I >>would expect people at the very least in their Senior year to have at least >>looked at or read "every" book in a specific area of interest. > Lots of folks in graduate school don't go this far! I think one That's why they invented libraries. If you intend to specialize in, e.g., compilers, you ought to have read the "classics" of the field, and bought what books you found to be up-to-date and good reference works. Or, if you intend to specialize in, e.g., low-level kernel work, you should have at least read (but preferably bought) the Bach book and the Minix book, as well as having read more general type references and looked at all the low-level code you could beg, borrow, or pilfer. (as an aside, has the book about the 4.3 kernel been released yet?). If you're going to be a "Computer Scientist", you simply have to have the reference materials to back it up. > the student. Undergradutes are presented with a great variety of material, > but at an introductory level. In graduate school the student should concentrate > on a few areas, but in greater depth. Even undergrads can have areas that they "like" more than others. At junior-senior levels, they ought to look into those areas in more depth than topics they don't particularly care for. But rarely will you find someone who does. > [Commenting on Eric Green's suggestion to check out the literature on a topic] >>I think that's a good idea, only many introductory books (first year) don't >>reference many "papers" that students might easily find available. I think an Note that I was not referring to introductory-level courses. I was specifically referring to upper-level courses, where the student has some skill in programming and some knowledge about computer systems. For example, in a computer architecture course, you might tell your students to go check out SigARCH, IEEE Computer, and IEEE Micro... in particular, if your text is more than 2-3 years old, you might want them to check out e.g. Patterson's summary of RISC in the '85 CACM. One of the most important courses I ever took was Technical Writing. Not for the writing part itself (which certainly is an important thing to know), but, rather, because a good TW course will introduce an undergrad to the indexes and journals available at the college library. In other words, the "HOW TO" of researching. > The introductory level student will not be as willing to read a lengthy paper > as a junior or senior. Also, many papers in the literature require some basic > knowledge in the subject matter that these students just don't have at this > point! Many tutorial or survey papers, while written for the novice in a > particular subject area, require some level of basic general CS knowledge. I'll agree about the "basic knowledge" part. When I was a freshman, the typical impenetrable prose and lofty vocabulary of your average paper befuddled me to no end... when I got the "Boyer-Moore" paper for my grad student (utilized me for free research :-), I had a heckuva time making heads or tails out of it (even after it was translated into English :-). But as for the "willing to read a lengthy paper": In my view, the people serious about CS will be quite willing to do it. The rest will change major to Business Administration as soon as they find out that CS not "easy", so why pander to them? -- Eric Lee Green ..!{ames,decwrl,mit-eddie,osu-cis}!killer!elg Snail Mail P.O. Box 92191 Lafayette, LA 70509 Netter A: In Hell they run VMS. Netter B: No. In Hell, they run MS-DOS. And you only get 256k.