Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!cs.utexas.edu!samsung!noose.ecn.purdue.edu!iuvax!copper!rschmidt From: rschmidt@copper.ucs.indiana.edu (roy schmidt) Newsgroups: comp.os.msdos.misc Subject: Re: MS-DOS 4.01 vs. 3.3 Message-ID: <55306@iuvax.cs.indiana.edu> Date: 25 Aug 90 17:13:19 GMT References: <38144@shemp.CS.UCLA.EDU> <2943.26d077ac@cc.helsinki.fi> <1990Aug21.164046.19087@lotus.com> <2209@cernvax.UUCP> <1990Aug22.153913.3961@granite.cr.bull.com> Sender: news@iuvax.cs.indiana.edu Organization: Indiana University, Bloomington IN. Lines: 20 In article <1990Aug22.153913.3961@granite.cr.bull.com> buck@granite.cr.bull.com (Kenneth J. Buck) writes: >In article <2209@cernvax.UUCP> markus@cernvax.UUCP (markus baertschi) writes: >> So, if you don't run *very* old programs you don't need SHARE, but if >> you want to absolutely shure you'll have to pay the price (memory). > >Fine, but what's *real* the penalty for NOT using SHARE? Will the "old" >programs just fail, or is there risk of disk corruption, other_bad_things, etc.? The point is, for "old" (and some "new") programs that use FCBs instead of file handles, you risk scrambling the data files written by the application in question, and possibly overwriting other files. SHARE is there to help the FCB method find the correct area of the extended disk on which to write. The price (15K of RAM) is not particularly high. If you find it a squeeze, take a look at the *low* cost of 1Mbit chips these days.... -------------------------------------------------------------------------- Roy Schmidt | #include Indiana University | /* They are _my_ thoughts, and you can't Graduate School of Business | have them, so there! */