Listen up, Microsoft et al--quit giving us upgrades that aren't

A lot of people still use Microsoft Windows 3.x. I do, given a choice. But I'd really
rather go back to DesqView from Quarterdeck Corp. of Marina del Rey, Calif., running under
MS-DOS 6.20.

DesqView and MS-DOS are both simple and versatile, and by now I know every bug in both.
There are good reasons why DesqView is still selling a decade after its introduction.

Given a choice, I'd turn to Windows for Workgroups 3.11 only when I need an Internet
browser or a PostScript printing job. Plus, of course, when I review Windows software.

But I dislike the general protection faults (GPF) in Windows, which often force a
complete cold reboot to clear everything out.

GPFs bother me a dozen or more times a week. I can't remember the last time MS-DOS
needed a reboot, and when it did, I generally could find the text I was working on in the
form of a lost file segment.

What really bothers me is that I have no idea why Windows 3.11 keeps crashing. The same
programs do the same tasks just fine for hours, then all of a sudden start crashing. When
I reboot, they go back to work with absolutely no changes.

Windows 95 is even buggier. New and exciting bugs are still being discovered by
users--how come Microsoft never seems to find them first?

So you can imagine how happy I felt when Compaq Computer Corp. announced in July that
it had stopped shipping Windows 3.11 preloaded on its PCs, and IBM Corp. announced it
would stop doing so by fall. Can other major vendors be far behind?

Yep, I was happy to hear that, because I make a living out of hardware and software
that don't work right. Real-world users may feel differently about the announcements.

The handwriting is on the wall: You will use Win95 or Windows NT, like them or not.

Microsoft will continue to support Windows 3.x, but that's small comfort to quantity
buyers who must choose from whichever OSes the hardware vendors give them.

Now that Windows 98 is on the horizon, I expect it to be worse.

Marketing departments push the bottom line so hard that software goes to market long
before it's ready.

Although you may disagree, this is a true statement: Virtually no U.S.-produced
software works well out of the box.

The situation is so bad that only a fool, a masochist, an optimist or a software tester
like me would buy software in x.0 versions. It always amazes me how many people fit those
categories. I bet most of them are optimists.

The software situation today parallels the American automotive industry back in the
1960s. Cars were too big and haphazardly built, and consumers deserted them for better
foreign models.

Now software buyers are complaining about poor quality control, and the companies keep
expecting them to make repairs after the products are delivered.

As with the auto industry, could foreign suppliers with low-cost labor be eyeing a ripe

Publishers make a big deal about how many beta testers they recruit, but you notice
they don't say they fix all the reported bugs, only that they test for them.

I recently downloaded the beta version of Netscape Communications Corp.'s latest
Communicator browser. After waiting 24 minutes for the file to download and expand, I
watched the installation run smoothly for about five minutes until, at the end, there was
a GPF.

I tried again, then deleted the file--about 45 minutes totally wasted.

I can't really complain because, after all, it was a beta version, but I think we
should at least get to install software before we begin logging bugs.

Will the next decade see Silicon Valley go the way of Detroit in the 1970s? Of course
not, says the conventional wisdom. Microsoft is far too dominant to falter.

Can you say hubris? Just ask General Motors' management team about its perspective.

Win98, due in mid-1998, just possibly could be the first Microsoft OS to fail in the
marketplace. Win95 has had more than two years to displace Windows 3.x, but most analysts
say the two OSes hold about equal shares of the installed base.

Will all those systems administrators who have enjoyed the exciting bugs in Win95 and
endured its upgrades really want to support a third major desktop OS, in addition to
whatever percentage of users clamor for NT?

I have my doubts. As 1999 dawns, I think we will still see a lot of Windows 3.x users,
even more Win95 users, a rapidly growing interest in NT and little sign of Win98.

I have just one question: If Microsoft doesn't merge NT and Windows soon, would it
really name an operating system Win00?

On the hardware front, Compaq, IBM and Hewlett-Packard Co. are all feeling the pressure
from Gateway 2000 Inc. and Dell Computer Corp., so PC prices are falling faster than the
autumn leaves will in a few weeks.

Pinnacle Micro Inc. of Irvine, Calif., also is feeling pressure from other
removable-media drive makers, because it recently cut prices of its Apex 4.6G
magneto-optical drive by 37 percent and disk prices from $170 to $100.

Of course, you can buy a 4.0G IDE hard drive from Western Digital Corp., also of
Irvine, for only about $340 or a hard-drive swap kit for $90, so some people still
hesitate to pay $1,200 for a 4.6G MO drive.

Let's see, 16G of storage with one MO drive and some disks works out to about $1,500.
Four separate hard drives that add up to 16G would cost $1,360. Which would you rather

John McCormick, a free-lance writer and computer consultant, has been working with
computers since the early 1960s.

Stay Connected

Sign up for our newsletter.

I agree to this site's Privacy Policy.