Re: Major problem with 6.0.2.1
- To: mathgroup at smc.vnet.net
- Subject: [mg89401] Re: Major problem with 6.0.2.1
- From: David Bailey <dave at Remove_Thisdbailey.co.uk>
- Date: Sun, 8 Jun 2008 02:33:37 -0400 (EDT)
- References: <g2dbg6$9i4$1@smc.vnet.net>
Michael Morrison wrote: > Very recent I upgraded from 6.0.1.0 to 6.0.2.1. I have found this latest > version of Mathematica almost totally dysfunctional. Now when I load > notebooks that previously worked fine, almost invariably I get one or more > messages warning of an error as a result of an internal self-check. (I've > never seen such messages with earlier versions.) The messages send me to a > Wolfram site that solicits details about my problem, which I provide, but > which doesn't explain the message, the problem or how to deal with it. The > site says that the problem may not affect the execution of my notebook or > results. But something is affecting execution of a lot of my notebooks! > Since installing 6.0.2.1, notebooks that used to run fine now cause > Mathematica to crash (!)---a problem I have not had since about version 4. > Even when that doesn't happen, trying to run a notebook typically generates > still more of these irritating self-check errors. I have tried everything I > can think of to ensure that the problem is not with my PC (cleaned the > registry, various internal maintenance checks, even defragmented by hard > disc, etc.) No improvement or change. Moreover, the problems appear on a > second PC I use where I loaded 6.0.2.1 a couple of hours ago. > > I've sent urgent pleas for help to the appropriate Wolfram site but so far > have received no reply. (I'm facing an urgent deadline in about 8 hours and > am suddenly unable to do anything until this problem gets fixed, since most > of my notebooks won't execute any more or will do so only amidst a blizzard > of errors.) If anyone can offer constructive advice I would be most > grateful. I'm running on a conventional Windows XP machine. > > I feel compelled to add that this is only the latest (and by far most > severe) problem I've had since the release of Version 6. I've used > Mathematica since about version 2 (very primitive non-GUI interface, very > limited help info, couldn't evaluate every integral known to man > correctly---but it did what you told it to almost always correctly, and it > rarely crashed). Every subsequent version through 5 improved the product, > eliminated errors, enhanced usability, and maintained the high standards of > reliability and freedom from bugs that has made Mathematica the choice of a > lot of scientists and teachers, including those at my institution. > > Then came 6. Bugs all over the place. Commands that don't work even in the > help browser (try TableAlignments in TableForm). A new help browser in > which a search now provides reams of links somewhere in which you might find > buried the information you want. Unless the information you need isn't > there: So far as I can figure out, it's still not possible, now months after > the release of 6.0 and however may subsequent releases, to find a list of > what is in the standard packages; neither can I find a single place to see > all the backward-consistency problems that plague notebooks written > previously. And so forth. Important commands that used to work fine now > sometimes work and sometimes don't (PlotLegend, which I use throughout > myriad notebooks, now doesn't work at all with many types of plot commands > where it used to work fine, in versions before 6.). For a while I reported > these to Mathematica, and always got very constructive replies. However > usually those replies indicated that the problem is a known bug that would > be fixed soon. That's fine, but it's now months after the release of 6 and > many of these problems are still problems (unless they were fixed in > 6.0.2.1, which I can barely get to work at all). Moreover, some bugs should > require user reports (e.g., things that don't work even in the Help > Browser). > > Several teachers I know are so frustrated by 6.0 that they're trying to > persuade their institutions to switch from Mathematica to a competitor > So far I've argued strongly against that change here. But this latest > debacle with 6.0.2.1 is the straw that broke my camel's back. I don't > know what's happened at Wolfram, but whatever has happened is a real > shame. Mathematica is fast degenerating from a vital research tool > into software with problems I expect only from a Microsoft product. > > Frustrated, furious, and fed up. > > Michael A. Morrison > David Ross Boyd Professor of Physics & General Education > Dept. Physics & Astronomy > University of Oklahoma > Norman, OK 73019 > Phone: 405 325-3961 > FAX: 405 325-7557 > EMAIL: morrison at mail.nhn.ou.edu > WEB PAGE: http://www.nhn.ou.edu/~morrison > > Have you tried moving Mathematica to another machine? I can't help wondering if your machine has developed a fault. Another thing to try would be to deinstall Mathematica and reinstall it from scratch. If you developed your notebooks under a beta version of Mathematica, of course, all bets are off! If you post a link to a notebook that worked at 6.0.01, I am sure someone will check it out. The windows and Linux 6.0.02 software works very smoothly for me. David Bailey http://www.dbaileyconsultancy.co.uk