Windows 98: The Next Generation

By Stuart J. Johnston

he rumor mill has been grinding loudly in the past couple of weeks to the effect that the U.S. Department of Justice will not challenge Microsoft's right to ship Windows 98. The DOJ has no comment on that.

There are also rumblings that the Department of Justice may instead demand that Microsoft do something similar to what it requested and got -- at least temporarily -- in the current court case. That is, DOJ attorneys may demand that Microsoft ship two versions of Windows 98: one with Internet Explorer 4.0 built into the user interface, and one with the browser hidden from view.

There may be some closure on this whole deal prior to that, of course, since the current case will be initially heard by the federal appeals court on April 21, and Windows 98 is not scheduled for release until late June. If Microsoft wins its appeal of Judge Thomas Penfield Jackson's order, DOJ attorneys may not feel up to going against Microsoft on Windows 98. They may even back off altogether, although that isn't likely.

Even if the Department of Justice does prevail in April, however, I have a hard time figuring out how Microsoft could meet such a demand in regard to Windows 98, considering that the system's help engine runs inside the browser and all of the help files are now written i n HTML.

Besides that, even after Microsoft consented in January to let PC vendors ship versions of Windows 95 with the browser hidden from view, to my knowledge none took them up on the offer. PC vendors are all still shipping the status quo. So it's hard to see to what purpose such a demand by the DOJ might serve in the real world.

Meanwhile, during the U.S. Senate Judiciary Committee's hearing in early March, Netscape Communications CEO Jim Barksdale chose to quote the results of two exclusive InformationWeek surveys to reinforce his point that Microsoft's moves to include the browser as part of the user interface weren't necessarily altruistic. Microsoft chairman Bill Gates has repeatedly stated that features like the browser end up in the operating system because users demand them.

But, as our surveys showed, among IT decision makers, one-quarter who responded actually want the browser merged into the system. A significantly larger percentage -- about 35% -- actively do not want that merger. [Those surveys were discussed in this column on two previous occasions. " Fighting The Enemy Within ," January 12, 1998 and " A Microsoft Decade: Writing And Rewriting History ," February 16, 1998.]

So despite the fact that it is hard to see how Microsoft could comply with a demand to sever the browser from Windows 98, it still doesn't look like Microsoft is going to get off scot-free on that issue.

Finally, after my last column on the differences between history and myth in regard to Microsoft's rise to dominance in the applications arena, I had planned to drop it for a while. But several long-time colleagues sent E-mail regarding that column, all agreeing that the historical version that I presented is the same one they recall.

Additionally, however, one colleague posed the question: "What about Microsoft's aggressive campaigns of bundling features that destroyed the compression market and almost put Quarterdeck (QEM M and DesqView) out of business?"

Good question. And since someone asked, it just so happens that I had already spent some time thinking about that. Quarterdeck's QEMM was a good memory manager, but Microsoft put that function into the operating system with MS-DOS 5.0. The question is should memory management be a "utility" that rides on top of the operating system? Or is it better completely integrated into the operating system as it is in mainframe systems?

Wouldn't you expect your operating system to manage memory for you? Perhaps not in PC operating systems prior to that time, but the historical precedent for it on large-scale systems and on Unix and OS/2 certainly militates for that view.

The same goes for user interfaces. I used Quarterdeck's DesqView, an early, non-graphical application-switching environment, for a year or so in 1988 and 1989. But it was not simple to set up and it was hard to remember all the details required to maintain it. Still it beat the pants off W indows 1.0 through 2.11, which were awful in comparison to DesqView.

At about the same time, I also used GEM, a graphical task switcher from the now-defunct Digital Research Inc. (DRI), the company that originated DR-DOS. But it was also fairly crude and had proprietary programming interfaces which required applications to be written to it -- just as Windows did. However, it never achieved much of a base of either applications or users.

Then along came Windows 3.0 in May 1990. It was relatively elegant in comparison to its competitors and it cut a swath through the market.

Furthermore, in the area of disk compression, were the disk compression utilities companies dragged under by Microsoft adding its DoubleSpace disk compression utility into MS-DOS 6.0? Surely Microsoft's move affected companies such as Stac Electronics.

But what crushed the disk-compression-utilities firms more than anything else was that falling prices for much larger drives obviated the market need for comp ression. Why compress when prices for drives plummeted at the same time that geometrically larger hard drives were coming onto the market?

Today, a new PC with a 166-MHz Pentium comes with 32 Mbytes of RAM and a 3.2 Gbyte hard drive for less than $1000. So, what's responsible for putting the compression utilities companies underwater? I think it's falling hardware prices for drives with more functionality, which is an attribute of this industry--a corollary of Moore's Law.

What I find really ironic, though, is that so many people are decrying the death of the utilities markets at Microsoft's hands. In its most recent quarter, Symantec, the company that acquired Peter Norton Software and some other utilities companies, reported its fifth record quarter in a row. For its most recent quarter, Symantec had sales of $148 million, mostly from sales of utilities software, according to the press releases on their Web site . Profits were at a record high too.

Eight years ago, the entire software utilities market was probably not that large. Companies such as Central Point Software made shareholders happy when they had annual sales of $10 million.

The other major player in the utilities market, Network Associates, which was formed recently by the merger of several companies, including McAfee Software, had revenue of $173 million along with record profits in the last quarter of 1997, according to press releases on its Web site .

These two companies are not by a long shot the only utilities vendors left out there. So I don't think there is really solid proof that subsuming utilities features into the operating system has been the death of that market. Instead, the utilities market appears to be healthier than ever, although it has greatly evolved and shifted in character. One of those changes is that the market has consolidated through buyouts and mergers, which is a function of maturing m arkets -- a trend that is continuing.

AuthorITies Archive

Send Us Your Feedback

Top of the Page

Rich Levin: Run Time Rich fills you in on all of the latest products, issues, and trends in application development.		Stuart J. Johnston: Redmond Watch As our eyes and ears in Redmond, Stuart gives his perspective on the latest events at Microsoft.		Charles Pelton: Eye On IT Charles explores IT management issues and strategies that business and technology managers face.