3 ms·
Something to remember with that book is that it's a highly selective perspective. Gerstner did do a lot to turn IBM around, possibly the best thing he did was t
by epc 9y ago
Something to remember with that book is that it's a highly selective perspective. Gerstner did do a lot to turn IBM around, possibly the best thing he did was to cut through tradition and corporate poltics with a chainsaw to make decisions. IBM in the early 90s was catastrophically stuck in between knowing the world was shifting and an inability to address the changing world.
Decisions you don't hear much about are things like selling off the IBM Global Network at the height of the dot com era, probbaly short term brilliance, but the sale included much of IBM's commercial web hosting, which left IBM blind to the shift to cloud a decade ago.
Or the decision to effectively axe OS/2 almost immediately after the Warp release, not so much because Windows 95 was better, but because the continued existence of OS/2 was inconvenient to the PC division.
Not saying it's a bad book, but like much history it tells a story from the perspective of the victor.
- vram22 9y agoI get your points. Incidentally, I had read good reviews of OS/2 Warp in a computer magazine, PC Quest. I think they included a trial or full copy of Warp with that issue on a CD. (They used to do that now and then, and I got to try out QNX for a while that way - very fast and light OS.) It was a while back, but I remember a few things, like they said it (Warp) was very stable, maybe fast, and probably a few more good points. I remember terms like System Object Model and Workplace Shell, although I probably did not understand all of those concepts fully - a bit too junior at the time, maybe. https://en.wikipedia.org/wiki/IBM_System_Object_Model https://en.wikipedia.org/wiki/IBM_System_Object_Model https://en.wikipedia.org/wiki/Workplace_Shell https://en.wikipedia.org/wiki/Workplace_Shell https://en.wikipedia.org/wiki/OS/2 https://en.wikipedia.org/wiki/OS/2
- epc 9y agoI’m obviously biased but I thought it was a decent O/S. App support was minimal but with the WinOS2 support you could run most Windows code. I thought the TCP/IP stack was years ahead of Windows, not quite comparable to what you'd find on SunOS/Solaris or whatever DEC ran on the microvax workstations. IBM, being IBM, charged an arm and both legs for the development kit, even to internal employees. I managed to use it until June 1998 when I found it was getting to be too much of a pain to dual boot to Windows to open Netscape (there was a Netscape browser for OS/2 but it was web years behind what they were distributing on Windows and *nix).
- vram22 9y agoInteresting. >I thought the TCP/IP stack was years ahead of Windows, not quite comparable to what you'd find on SunOS/Solaris or whatever DEC ran on the microvax workstations. I remember back in the day, people saying that the TCP/IP and Internet layers on Windows were not good. There was this software called Trumpet Winsock, which may have tried to make things better. https://en.wikipedia.org/wiki/Winsock#Other_implementations https://en.wikipedia.org/wiki/Winsock#Other_implementations You may be interested in "Mike's PC Jr. page" which is linked to in this blog post: Lissajous hippo, retrocomputing and the IBM PC Jr.: https://jugad2.blogspot.in/2012/09/lissajous-hippo.html https://jugad2.blogspot.in/2012/09/lissajous-hippo.html He wrote a TCP/IP stack for his IBM PC Jr. and is using it :)