4 ms·

An administrator is updating drivers? So maybe not a developer job at all, updating means installing not coding? And the Windows 3.11 on the train has not dece

An administrator is updating drivers? So maybe not a developer job at all, updating means installing not coding? And the Windows 3.11 on the train has not decent networking or remote maintenance system it needs to be done manually somehow? But then it's a remote position. Doesn't seem to make sense to me. But running something as unreliable as Windows 3.11 in a production environment does not seem to make sense anyway. Some say recertification of a new system would be too expensive. But if this system is certified, I'd say it's a certification theater not worth the paper it is printed on.

tauchunfall 3y ago
>But running something as unreliable as Windows 3.11 in a production environment does not seem to make sense anyway. They used and still use MS-DOS, Windows 3.11, Windows 95, and Windows NT for different systems over the years in the older trains starting from 1989 to estimated 2030.
fsiefken 3y ago
I can't remember Windows 3.11 crashing so much, dos 6.22 neither. Windows 9x and ME did crash more. Anyone know what happened to the Win98 presentation guy and where he works now? https://www.youtube.com/watch?v=yeUyxjLhAxU https://www.youtube.com/watch?v=yeUyxjLhAxU
- usr1106 3y ago
  I remember it hanging regularly. Maybe not every day, but several times a month I would guess. Of course it depends on what software you are running. I remember running mostly Humingbird eXceed X-Windows server to access our Unix workstations. I don't remember any BSOD on 3.11, was that a new NT feature? Maybe also "backported" to W95 and/or W98. Well, it's been a while and I don't miss them...
  - lmm 3y ago
    Maybe you had bad hardware, or bad drivers. Windows driver certification didn't even exist in those days. But 3.11 by itself was rock solid, I don't remember ever crashing it from windows proper.
    - usr1106 3y ago
      IIRC 3.11 had no process isolation and the IP stack was an afterthought. So every application bug could crash or hang the whole system. And networking problems could freeze everything. It's rather unlikely that something complicated like an X-server was bug free. So the system you describe as rock solid, can only be rock solid as long as no real world application is running on it. IIRC NT 3.51 was the real first kernel where a buggy application could not affect the whole machine. But at least in early days BSOD was pretty common, maybe depending a bit on drivers you needed. (Did not do any archeology now, just from increasingly vague memories.)

tauchunfall 3y ago

>But running something as unreliable as Windows 3.11 in a production environment does not seem to make sense anyway. They used and still use MS-DOS, Windows 3.11, Windows 95, and Windows NT for different systems over the years in the older trains starting from 1989 to estimated 2030.

fsiefken 3y ago

I can't remember Windows 3.11 crashing so much, dos 6.22 neither. Windows 9x and ME did crash more. Anyone know what happened to the Win98 presentation guy and where he works now? https://www.youtube.com/watch?v=yeUyxjLhAxU https://www.youtube.com/watch?v=yeUyxjLhAxU

usr1106 3y ago
I remember it hanging regularly. Maybe not every day, but several times a month I would guess. Of course it depends on what software you are running. I remember running mostly Humingbird eXceed X-Windows server to access our Unix workstations. I don't remember any BSOD on 3.11, was that a new NT feature? Maybe also "backported" to W95 and/or W98. Well, it's been a while and I don't miss them...
- lmm 3y ago
  Maybe you had bad hardware, or bad drivers. Windows driver certification didn't even exist in those days. But 3.11 by itself was rock solid, I don't remember ever crashing it from windows proper.
  - usr1106 3y ago
    IIRC 3.11 had no process isolation and the IP stack was an afterthought. So every application bug could crash or hang the whole system. And networking problems could freeze everything. It's rather unlikely that something complicated like an X-server was bug free. So the system you describe as rock solid, can only be rock solid as long as no real world application is running on it. IIRC NT 3.51 was the real first kernel where a buggy application could not affect the whole machine. But at least in early days BSOD was pretty common, maybe depending a bit on drivers you needed. (Did not do any archeology now, just from increasingly vague memories.)