Slashdot Mirror
Bill Gates: Windows 95 Was 'A High Point'
BobJacobsen writes "CBSnews.com has an article about Bill Gates and Steve Balmer answering questions at the 'All Things Digital' conference. When asked about 'high points' in his time at Microsoft, Gates replied 'Windows 95 was a nice milestone.' The article continues 'He also spoke highly of Microsoft SharePoint Server software, but didn't mention Vista.' Was there really nothing else that Gates considered a high point?"
Come on! When Win95 came out, with preemptive multitasking, Macs were still using "cooperative" multi-tasking, which is really just a toy by comparison. In many ways Win95 was quite an advance as a true preemptive multi-tasking OS that ran on off-the-shelf hardware. And it also maintained very good compatibility with the old DOS and 16-bit Windows applications (games) at the same time. Quite an achievement actually.
The more you regulate a company, the worse its products become.
>"it was a decently advanced OS for the time."
Only by Mocrosoft standards.
At the time 95 was launched, SGI was putting 64-bit IRIX machines on people's desktops.
OS/2 3.0 ("Warp") released in 1994 was better then Win95.
Then there was NeXTSTEP, Apple Mac, etc. - all better then Microsoft.
Microsoft "won" because they ran on cheaper hardware. In no way was their software superior.
No sig today...
The advantages (pentium support, better 32 bit support) were outweighed by its stability problems.
Are you insane? Windows 95 may have crashed every week or so on average, and it certainly crashed every 49.7 days if you were ever lucky enough to make it that far, but we're comparing it to Windows 3.1 here! Even if you disregard the bugs in Windows 3.1 code itself, the thing used cooperative multitasking and unprotected memory, so your computer crashed every time the buggiest program you ran had a particularly bad flaw. It would freeze up multiple times a day, under any kind of heavy use.
I think it's clear that if your criterion is "improvement over best previously available version", Windows 95 really was the high point of Microsoft development. Stability doesn't outweigh that conclusion, stability is one of the reasons for it.
Actually, the original quote is accurate. Apple considered licensing the NT kernel to run under their own interface.
http://www.x86.org/ddj/aug98/aug98.htm
With the Pentium, Intel introduced a 94-entry, two-way set associative cache of segment-descriptor cache entries. Therefore, the phrase "segment-descriptor cache" is now ambiguous, with two possible meanings. Making matters worse, the new segment-descriptor cache was removed from the Pentium Pro design, but reintroduced in the Pentium II. (The lack of the new segment-descriptor cache in the Pentium Pro largely accounted for its poor 16-bit performance.)
When designing the PPro Intel thought that Windows NT would take over from 16 bit Windows. Windows NT doesn't do many segment loads. Threads use FS for thread local data so that is presumably loaded every time the scheduler switch threads, every 10 to 100ms. But that is a very small percentage of instructions. All code and data use the same values for CS and DS - base address 0 and limit 4GB. So Intel removed the segment descriptor cache. But since 16 bit OSs were still popular and those OSs load the CS and DS segment registers much more frequently. In fact they have to, since they were designed to work on the 286 back when 64K was the maximum possible limit. Since datasets and code sizes were way bigger than 64K, the segment registers are loaded very frequently. So in the Pentium 2 Intel reintroduced the cache. It's not a hack, just bad crystal ball gazing.
Actually most of Intel's mistakes are like that. They predict the future badly because of a strange mix of wishful thinking, a desire to get rid of legacy stuff and outright hubris.
echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
Well, let's be fair: Windows 95 was supposed to be able to scale down to 386 CPUs, which were capable of 32-bit code but thrived on 16-bit code. How well it did this is a matter of some debate, and generally you didn't want to do anything "serious" with the OS on less than a 486, but at the time there were a lot more potential customers using a 386 than there were using 686 CPUs, and the codebase indicates as much. :)
At the time, Intel decided to market the Pentium Pro as a server chip, so it was not meant to run Windows '95. It was meant for NT and OS/2 exclusively. The Pentium Pro was supposed to compete with the big iron servers running Unix, and Intel gambled that 32-bit software would replace 16-bit software in time. They were right: But they were ahead of their time. The market was not ready to get rid of the cheap desktop OSs and the vast quantities of 16-bit software.
So Windows '95 was indeed a high point for Microsoft. They were the first to deliver a stable 32-bit-ish graphical OS to Intel PCs. And it was the first OS to integrate well enough with DOS to replace it. Windows 3.1 was more of a graphical shell than an operating system. Windows '95 is why we use the term "wintel" and it is why IBM and OS/2 did not win the operating system wars. Back to the thread; So there was so much 16 bit code in the "new" 32bit Windows 95 that a new CPU optimized for 32bit code ran the software way slower than the old 16bit optimized Pentium CPU. Exactly what you'd expect from a company where marketing is job #1. IMO. Microsoft optimized Windows '95 to run on the CPUs available at the time, not the Pentium Pro which wasn't even released yet. If you wanted true a protected-mode 32-bit OS, Windows NT was the target. And it ran well on a Pentium Pro. Perhaps, had Microsoft done what you are suggesting, then OS/2 might be dominating the desktop today.
I don't know why you named those four people; at least three of those four have been or are currently being compensated for their most famous "free" projects.
A common thread among those people is that they all started their major projects during college or grad school and found financial backing as they were leaving academia. Or in Larry Wall's case, he had a day job at JPL while working on Perl. I think you'll admit that college/grad student life can't realistically go on forever. Eventually your parents will stop giving you money and/or the university will stop paying your room and board, and you'll have to find a "real job" to support yourself and your family. I think lots of people in the open-source community are employed by the likes of IBM, Red Hat, Oracle, OSDL, etc. for their work. No, I don't feel like finding more references.
The message might be that we need to fund more people in grad school to work on pet projects, or that Microsoft needs to fund them, but in general I agree with Mr. Gates - development on large-scale projects can't continue indefinitely without some sort of compensation.
He enrolled in Helsinki in 1988, announced Linux in 1991, got the BSc in 1995, and the MSc in 1997 (having worked odd jobs at the University) and only then moved to the "real money job". I guess we missed the part where he moved back from silicon valley to finland "much later" to study for a few more years?