Gates back on stand in Utah in $1B antitrust trial
google.com
google.com
When Win95 came on the scene, we were hesitant at first -- let others figure out the kinks -- but we were rolling out new desktops within 6 months. We had also been waiting for a Windows version of WordPerfect. The DOS version was ok, but it was awkward running it in Windows.
So we waited. And waited. And waited some more. Finally, after I don't know how long, we got our hands on WordPerfect 6.0 for Windows. It royally sucked. The WP folks had spent approximately zero time imagining their product in a Windows environment -- GUI, mouse click, that kind of thing -- and instead just pushed their F-commands into a menu system. It didn't flow at all, and our staff complained mightily.
So, we took a look at MS Office (I think it was version 4.x or something around then.) I didn't like MS Office, and knew we would have re-training issues (especially going from Quattro Pro to Excel). Nonetheless, we decided to give it a try.
Staff loved it. Not jumping-up-and-down loved it, just that it let them do their job quite easily. The analysts in our department raved over Excel, however. And personally, I could rid myself of Paradox, which seemed to dislike me more than others.
Summary: it's a small sample size, but from where I stood WordPerfect basically blew it. I can't speak to the MS/WP relationship, but I can without a doubt say that WP didn't put any thought into their product. Why they released their Windows version, I'll never know.
Was MS a tough competitor? Of course. Did they make it difficult on WP to provide their version on Windows? Maybe. But being completely unimaginative about the application and workflow? That's nobody but WordPerfect's fault.
Seems forever ago.
That statement is only even remotely true if one interprets it to mean that there was no alternative to Windows at the time that ran on the same hardware. Macintosh computers were alive and well, no?
"And you'll see listed here are a number of boot images. Each one of these boot images contains the configuration of a kernel that will probably match one of your systems. Let's go ahead and use the DOS program 'edit' to view the images.idx file now."
"Here it shows in the left hand column the last four numbers of the boot images of the files we just saw earlier. On the right side is a description of the type of system we need to install it onto. For example if you look on the first line it'll show that it has IDE/ATAPI or SCSI. In other words if I take the boot image 0000 then I will be able to utilize the boot image on a system that has an IDE CD-ROM drive."
Heh. It sounds like installing Linux was as easy as 0001, 0002, 0003...
Recompiling the kernel manually was something you just did; it was expected. Downloading a bunch of source code and compiling the dependencies and managing everything yourself; also completely ordinary.
The best thing was going between Pillbox, Yggdrasil, Mandrake, Slackware(!) ... those were the days!
There were two floppy disks that you would boot from, the `Boot` and the `Root` if I remember right. If you had a system up and running you could use dd, otherwise there was a DOS tool that would do as much. Interestingly, this was when the first thing you'd see at boot was
LILO
LILILILILILILILI
or
LIL
CDs originally weren't bootable and lots of BIOSES didn't know how to deal with them. There was a standard that went out called the "El Torito" or something that basically talked about how CDROMS would be bootable. I think Windows 98 had this; in fact, I think it came out of Redmond (but I haven't done my homework in the Goog)
So in 199x, even if you had a fancy bootable CD, you were still making floppy images, then inserting the CD when instructed to. In fact, I still have a box of floppies just in case I need to deal with hardware pre-1999 ... which is becoming less and less relevant (I think I used one last in 2006). Then again, I used a 5 1/4 in 2003 ... so I guess it's worth keeping around another 4-6 years.
The worst memory I have is when I found out just how bad swap could be. I had to compile the NetBSD kernel on a laptop I had (IBM 360C, active color 640x480 display, NiCd battery, 2.88MB floppy, SCSI disk, huge docking station with speakers ... this is about 1999 (more of 1999 below)) and it had all of 8MB of RAM. I had to recompile the kernel to make use of an early wireless (which felt like pure magic) PCCARD I had (mind you, 16 bit, not 32-bit CardBus) and the kernel took 4.5 days(!) to compile. In fact, I remember starting it Christmas eve before eating dinner and it went on through Christmas day and then a few days after that.
I had mastered screen by this time so I was able to see that the machine was really chugging away. The reason was that the compiler used up all of my RAM and was hitting swap. After that I upgraded the machine and the new compilation took 40 minutes.
40 minutes then, seemed like getting DSL for the first time and then clicking on that 10MB file and thinking "What is going on here? A whole megabyte downloaded right in front of my eyes? In mere seconds?! I am never going outside again".
I long for these exciting days. The mobile world has it, that's about it.
More about 1999;
Speaking of 1999, two other big things, the BeOS 5 Demo, booted from windows and was awesome, then there was the "Bootable Floppy Challenge" which was done by QNX ... it was quite impressive in the day.
Lord help you if you tried to use a web browser while running another application.
http://www.operating-system.org/betriebssystem/_english/bs-m...
Edit: Apple also had an official licensing program for Mac system software between 1995 and 1998. I'm not sure if it was before or after the lawsuit, though.
http://en.wikipedia.org/wiki/Macintosh_clone#Official_Macint...