Information on computers

I'd like to commend The Jakarta Post for publishing a very interesting and rather advanced computer article, Fast lane computing with new CPUs and OSes, by James O. Scharf (May 23, 1994). However, there are various errors within the article I'd like to point out.

My first concern is Mr. Scharf's claim that Microsoft's "Chicago" operating system is the same thing as Windows 4.0, dubbed "Cairo." Chicago and Cairo are not the same; they are two different variants of the Microsoft NT operating system. One is meant to be a "pure" 32-bit Window-based environment (unlike the complicated NT, with various kludgy 16-bitisms to support older applications), while the other is meant to be the "next logical step" for current Windows 3.1 users, complete with a separate DOS system.

On another note, perhaps many computerists would agree with Mr. Scharf's claim that Microsoft "...could have brought NT to the market long before it did..." but refrained because they did not want to"...down play Windows 3.0." Personally, I do not see any facts that would justify such a claim. Last year, after struggling continuously with the pre-release NT March-Beta prototype operating system, there were so many bugs and design misses I was surprised that Microsoft released NT as early as it did.

Furthermore, there was little doubt that the NT market is significantly different from the DOS based Windows 3.x market, that any concerns that NT would down play Windows 3.0 could not have been justified. Even today, with general availability of the NT release, NT has had only marginal impact on the Windows 3.x market.

Mr. Scharf also remarks that the IBM "Workplace OS...is really OS/2 version 3.0 (Although IBM isn't calling it that yet)." I suspect Mr. scharf misses the significance of why IBM is not calling it OS/2 version 3.0. The IBM OS/2 version 2.x operating systems contain Windows code copyrighted by Microsoft. As such, for every copy of OS/2 sold, IBM is required to pay heavy licensing fees to Microsoft, enough that IBM has problems making OS/2 competitive. To avoid paying massive royalties to Microsoft, IBM chose to pursue two avenues: the first is to ship OS/2 without Windows binaries, in the form of a product called IBM OS/2 for Windows. The second avenue of pursuit is to rewrite the entire OS/2 operating system, taking advantage of newer technologies, and without Microsoft code. The rewrite, an ambitious object-oriented operating system with a micro-kernel variant, is called Workplace OS. Workplace OS is not OS/2 version 3; it exists because IBM wants to get away from OS/2 and Microsoft royalties.

Later in the article, Mr. Scharf describes Win32 as "...a software layer that lets you run applications using the Win32S subset API on Windows 3.1." Win32 is a full API for the NT operating system, and has little to do with Windows 3.1. Win32 does not let applications run on Windows 3.1. However, if NT application programmers restrict themselves to a limited subset of the Win32 interface, called Win32s, then the resulting binary is runable under extended versions of Windows 3.1. The extension is done by installing Win32s drivers within Windows 3.1. I suspect confusion between Win32 and Win32s.

I also find the comment, "...System 7 is Apple's only PC operating system..." rather misleading. Many Macintosh users, especially those familiar with the Internet, use a variant of the UNIX operating system (A/UX) instead of System 7.

In fact I'm surprised that there is only a passing mention of UNIX. Clearly, an article discussing high technology operating systems should have mentioned UNIX as an important player in the field. Regardless, please do not be discouraged from publishing other computer technology papers in the future. As a whole, I felt the article was generally good and entertaining. I hope to see other in-depth articles in The Jakarta Post soon.

ADE BARKAH

Jakarta