Kelly's picks for 1997 and 1998: What's hot and what's not IMHO

Hindsight is always 20/20. It's easy to look back at 1996 and say which new products
and technologies shouldn't even have been driven out of the garage.

 As for 1997 and 1998, I predict new technologies will find more potholes and land

 This hasn't stopped me before and won't now. If you disagree with my predictions or
have ideas I haven't thought about, feel free to e-mail me at the address at the end of
the column.

 The first of my hot picks is Microsoft Windows NT, which underlies several other hot
picks for the near future. 

Windows NT Workstation 5.0, now under development for the desktop, will make smaller
resource demands on memory and processing power when it arrives in 1998.

 That positions NT 5.0 to take over the PC desktop, replacing Windows 95 and Windows
98. As a one-size-fits-all desktop operating system, I expect it to meet the needs of
everybody from budget crunchers to engineers. 

Meanwhile, Intel Corp.'s microprocessors will face increased competition from Advanced
Micro Devices Inc. and Cyrix Corp. chips. I've heard reports that the Cyrix chipsets are
real screamers, and I recently tested a 200-MHz MMX multimedia chip from AMD. I usually
work on a 200-MHz Pentium Pro, and let me tell you, the AMD was really fast. For more
information, see and  

I predict your agency will be well served by these clone chips at prices $200 to $350
lower per machine than the Intel variety. If you can't find such machines on General
Services Administration schedule, whip out your IMPAC credit card.

 My next hot pick is DVD, or digital video disk. It will start a multimedia
revolution. You can access massive amounts-up to 17 gigabytes-of media-rich content on a
single CD-ROM-sized disk. In a speech I gave in Tokyo last December, I predicted DVD will
take us way beyond CD-ROM and give better reliability, too. A PC-DVD drive can even play
your existing CD-ROMS.

 Creative Labs Inc. of Milpitas, Calif., is selling its PC-DVD drive for $499 retail,
and you can expect that price to fall. For more information, see

Unlike other columnists, I don't see the Internet collapsing. The Internet is tomorrow's
network. Today's Internet hardware is co-ruled by Unix and Windows NT Server. According to
surveys I've seen, NT Server seems to be capturing virtually all the new Internet
business. I expect this trend to continue and in fact accelerate.

 I'm not debating the two operating systems' technical merits but merely forecasting
where the market will go in 1997 and 1998. Exactly what other software will be running on
Internet servers is a bigger question.

 Netscape Communications Corp.'s SuiteSpot server package shows a lot of promise.
I've been looking at Netscape's Calendar Server 1.0, only a first release but, in my
humble opinion, the best Internet calendaring and scheduling software around.

 If you haven't seen it yet, download a free evaluation copy from It could be the answer to
your agency's calendaring needs.

 The next version of Microsoft's Internet Information Server (IIS), which is included
with Windows NT Server, will dramatically change the way we view the work done by Internet

IIS will incorporate the Microsoft Transaction Server and give Internet servers the
underlying muscle for large-scale transactions. This scalability has broad business
implications for agencies.

 The other order of business on the Internet will be development tools for making
Internet applications more compelling. Microsoft has fired its shot with the recent
introduction of Visual Studio and Visual InterDev.

 Visual Studio is a compilation of Microsoft development tools into an integrated
environment. You can mix and match components developed in Visual C++, Visual J++ or
Visual Basic into a smashing World Wide Web application. ActiveX and ActiveServer pages
tie all this together.

 Visual InterDev makes integration of Active Server Pages and databases on the Web
much less daunting. For information, see

Java is hot-for everyone except me. I don't see how taking a subset of the C
language-effectively gutting it of powerful stuff like the Microsoft Foundation Classes
library-and repackaging it makes Java the messiah of development. Maybe I'm just jaded.

 Hotter even than Java in 1997 and 1998 will be its integration into other software
through JavaBeans and ActiveX. This sounds like nirvana for Web developers: No more
writing to multiple platforms, just writing and letting the integration technology sort it
all out.

 So there you have my decidedly biased view of what's hot for the next two years.
Send me your thoughts and predictions, and I'll roll them into a later column. 

Charles S. Kelly is a computer systems analyst at the National Science Foundation.
You can e-mail him on the Internet at [email protected].
  This column expresses his personal views, not the official views of NSF.

Stay Connected

Sign up for our newsletter.

I agree to this site's Privacy Policy.