Dos / windows 3.1 how did programs install?

[Bottom_Feeder]

the turds of windows 2000 and windows ME need no mention.

Windows 2000 was far from a 'turd'.

 

[bunnspecial]

Windows 2000 was far from a 'turd'.

Agreed, Windows 2000 was a superb OS. It was basically NT 4.0 with a lot of modern additions like proper full USB support(including mass storage devices) and just generally rock solid. Unlike 98 or ME, it also had proper multi-user support. I still have a computer in regular use in the lab running Windows 2000-it's the newest OS that will reliably run a certain piece of software that needs to access an ISA interface card.

Windows ME definitely is a different story. It was basically Microsoft's attempt to put together one last version of DOS with a nice GUI over it, but they stretched the platform too far and it showed. It's analagous to Mac OS 9, Apple's last iteration of the "classic" OS before transitioning completely to BSD/Unix based OS X. Unlike Windows ME, though, OS 9 tends to be fondly remembered, although it certainly has its quirks. They did manage to pull off some nice touches, though, like full proper USB support. It's relatively stable and overall just a good OS. Unfortunately was limited by a lot of legacy code that I've been told no one at Apple fully understood by the time OS 9 came around, and 1990s Apple had gotten sloppy and left some key parts of the OS running on the Motorola 68K emulator that was baked deep into the OS, despite OS 9 being the second major release to not support 68K system...

 

[OVERKILL]

I remember back in middle school kind of coming into my own, so to speak, and learning the ins-and-outs of working on my own computers.

My dad had always been a computer person-he did a few stints in IT in his professional career, and I never remember us not having multiple computers around the house.

In any case, my dad was an MS "Action Pack" subscriber(not sure if that's still a thing) and they sent him a beta version of Windows 2000(which I believe you can find called NT 5.0 in some references). He had zero interest in it, so with his okay I installed the beta on my computer and then the release version when they sent it. I LOVED it and thought it was amazing. I couldn't believe that my computer almost never crashed-it was such a different experience than Windows 98.

Meanwhile, he went all in on Windows ME, and had no end of trouble with it. That has to be the single worst version of Windows ever created. Vista was decent after a few service packs and even though I used Windows 8 very little, I found it was fundamentally a sound OS once I installed Classic Shell to get rid of MS's stupid UI "revolution." Windows ME, though, I just remember being a mess.

To the OP's question about transplanting Windows software, though, I've painstakingly done it by copying the program folder, launching the program, noting the DLL or any other files it asked for, going and finding them, copying and placing them in the same place on the destination system as I found them on the source system, and repeating until the program finally worked. I wasn't always sucessful, and it took a lot of time, but I have done it. It really makes me appreciate that MacOS typically has everything needed for a program in a single package that is easily moved around, and installation of a lot of programs often really is just copying the package to where you want it....

This mirrors my experience. I was in first year Uni when NT5 came out (and later renamed to 2000) and I remember ME coming out not long after with about the only good thing about it being the 2K GUI.

 

[OVERKILL]

Agreed, Windows 2000 was a superb OS. It was basically NT 4.0 with a lot of modern additions like proper full USB support(including mass storage devices) and just generally rock solid. Unlike 98 or ME, it also had proper multi-user support. I still have a computer in regular use in the lab running Windows 2000-it's the newest OS that will reliably run a certain piece of software that needs to access an ISA interface card.

Windows ME definitely is a different story. It was basically Microsoft's attempt to put together one last version of DOS with a nice GUI over it, but they stretched the platform too far and it showed. It's analagous to Mac OS 9, Apple's last iteration of the "classic" OS before transitioning completely to BSD/Unix based OS X. Unlike Windows ME, though, OS 9 tends to be fondly remembered, although it certainly has its quirks. They did manage to pull off some nice touches, though, like full proper USB support. It's relatively stable and overall just a good OS. Unfortunately was limited by a lot of legacy code that I've been told no one at Apple fully understood by the time OS 9 came around, and 1990s Apple had gotten sloppy and left some key parts of the OS running on the Motorola 68K emulator that was baked deep into the OS, despite OS 9 being the second major release to not support 68K system...

Yeah, ME was effectively 98 SE with the 2K GUI and more effort placed into trying to hide the MS-DOS underpinnings while adding a whole pile of half-baked and poorly tested features that never worked right. Its development timeline was far, FAR shorter than 2000 (NT5) and it showed.

 

[Brons2]

Agreed, Windows 2000 was a superb OS. It was basically NT 4.0 with a lot of modern additions like proper full USB support(including mass storage devices) and just generally rock solid. Unlike 98 or ME, it also had proper multi-user support. I still have a computer in regular use in the lab running Windows 2000-it's the newest OS that will reliably run a certain piece of software that needs to access an ISA interface card.

Windows ME definitely is a different story. It was basically Microsoft's attempt to put together one last version of DOS with a nice GUI over it, but they stretched the platform too far and it showed. It's analagous to Mac OS 9, Apple's last iteration of the "classic" OS before transitioning completely to BSD/Unix based OS X. Unlike Windows ME, though, OS 9 tends to be fondly remembered, although it certainly has its quirks. They did manage to pull off some nice touches, though, like full proper USB support. It's relatively stable and overall just a good OS. Unfortunately was limited by a lot of legacy code that I've been told no one at Apple fully understood by the time OS 9 came around, and 1990s Apple had gotten sloppy and left some key parts of the OS running on the Motorola 68K emulator that was baked deep into the OS, despite OS 9 being the second major release to not support 68K system...

Wait you have a production application running off of an ISA card on Win2K? I don't think I've had anything I needed that ran off of an ISA card in 25 years.

 

[Rand]

2000 was a turd for me because I got a free copy at an Intel conference and had high expectations
I forget what it was exactly but I had major issues running programs I needed.
windows XP came along about a year later.....was much much better and painless.

Now if you were already using NT for NT things and moved to 2000 I'd expect it was ok.
TO me: Windows ME was taking windows 98SE and giving it issues.. with no positives.

I really wanted windows 2000 to work because I couldnt boot 98SE with over 512MB of ram. 768MB installed and you would get out of memory errors

 

[bunnspecial]

Wait you have a production application running off of an ISA card on Win2K? I don't think I've had anything I needed that ran off of an ISA card in 25 years.

The software is HP/Agilent MSD Chemstation G1701BA B.02.03 and the computer has an HP 82341 IS HPIB card. It controls an HP 5890 gas chromatography.

When I first set this system up, it was primarily to run an HP 6890/5973 gas chromatograph/mass spectrometer. I was able to get the money to upgrade that one to LAN connectivity, which allows me run a much newer version of the software, G1701EA E.02.01, which can run on Windows 10. That lately has been running on average about 60-80 hours a week(I often load up 20-30 samples at a time, which run in triplicate and sometimes a 4th additional run to get a bit more data, and take a half hour or so each to run, so that can turn into a lot of instrument continuous time). Thank goodness I have the newer and faster computer.

BTW, there's a PCI card, HP PN 82350, or alternatively Agilent PN 82350b(interestingly, for the real geeks out there, the 82350 is labeled as an "HPIB" card, where the Agilent branded 82350b is labeled as a "GPIB" card since, well, Agilent no longer had the rights to brand anything HP after they were spun off). AFAIK, it's an 82341 with a PCI-ISA bridge, and in fact when you set it up you tell the software it's an 82341c card. A few years ago those were $200-300 cards, so I stuck with the older ISA 82341 and made it work. Not too long ago, I was offered a half dozen of them along with a huge pile of other parts for $300...

If you think that's old, on the other side of the lab I have an HP 8452 Diode array UV-VIS spectrophotomer that's run in Windows 95 using an ISA HP 82335 HPIB card. The computer it's on is a DEC branded Pentium Pro, quite a good computer for what it is. I bought this computer already set up with a bunch of HP Chemstation licenses installed(and I have the HDD imaged a half dozen times over because some of those licenses are nearly irreplaceable) and it's just sitting there doing its thing. I actually would like to upgrade it to Windows 98 so that, if nothing else, I can connect a USB printer and/or get functional LAN for a LAN printer(also in the lab) but I'm afraid to mess with it too much.

 

[davemunson]

In the day I got really good at batch file programming. Worked at network pc; a multimillion dollar computer mail order place.