It's been a number of years (15 or more) since I used VMS, but as the OP indicated, it was a nearly bulletproof operating system. The on-line help was extensive and all the commands reflected the fact that this operating system was designed vs. evolved. All the commands had the same syntax and were intuitive. One also had significant control with the command line instructions like Unix and unlike Microsoft OS's, but the commands also made sense and were consistent, unlike Unix.
That's about all I remember anymore; I just remember being very frustrated when I had to move to DOS and Unix OS's after working with VMS. Now, I'm comfortable with Unix, it was just hard-won experience.
"Nearly crash-proof" as a start... :) (In fact, I don't believe the the VAX when I was using it ever had a fault/failure that was OS related in the eight years I was there...one hardware failure that I remember was the only non-scheduled downtime I remember ever.)
Re-read my post. It follows the paragraph discussing the "snipped and trimmed for content and length" broken logic.
When you change the subject line, you have *got* to make sure that you include everything point you can from the replied-to post, otherwise you've potentially started a new thread with a statement from someone else devoid of the context of that statement.
I isolated it below, but of course, this doesn't flow properly because of your top post.
Most of the Windows crashes I've experienced since 2K shipped have been hardware, usually RAM. System crashes still occur but they're rare and when they do occur they're often driver crashes. One area of confusion is that many "crashes" aren't really--the keyboard or display driver is hung while the kernel is still up and running. On a single-user machine with only the one keyboard and display it is difficult to clear these other than by reboot (although briefly pressing the power button may lead to an orderly shutdown--by briefly I mean tap it, don't hold it to force a hardware shutdown), but the same sort of problem on a server can often be corrected by logging in from another terminal and killing the offending session.
Windows has been multiuser since 2K Server shipped. There's a Unix client called "rdesktop". Prior to 2K Server Microsoft had a separate product called "Terminal Services" that added the capability to NT. The multiuser support is disabled on the single-user versions though so you won't see it unless you have access to one of the server products and the administrator has enabled the capability--by default it's turned off.
Err, I resemble that statement. Our department has a two-course series:
ECEN 5837 - Mixed-Signal IC Design
3 credit hours Catalog Description: (1) Design of core analog circuits in mixed analog and digital systems, including data converters and sampled-data circuitry, and (2) system level IC design methodologies and CAD based circuit design and layout techniques in mixed analog and digital IC's. Prerequisite: ECEN 5827, Analog IC Design Textbook: ? Allen and Holberg, CMOS Analog Circuit Design, Oxford, 2002. Course objectives: This course is the second in a two-course series (the first course is ECEN 5827) on integrated circuit design, which together provide a complete set of fundamental concepts and skills for the growing number of students who wish to pursue a career in the semiconductory industry. Graduate students who wish to specialize in research projects related to IC design and power electronics will be required to take both courses in the sequence. Topics: 1. Fully-differential op-amsp, simulation and layout 2. Comparators 3. Switched capacitor circuits 4. Nyquist rate DAC 5. Nyquist rate ADC 6. Over-sampling converters
To be sure, they're graduate-level courses, primarily because most data acquisition systems are either designed into custom ICs, as the courses above, or, just as common, engineers these days use "black box" solutions -- slap together a commercial sample-and-hold, a commercial ADC and output 'em into a circular buffer.
These days at the undergraduate level our kids get courses that teach them how to design systems that *use* the black boxes: Circuits I, II, and III, "Computers as Components," Embedded Systems, and, the topic most often ignored to the peril of the EE, Digital Signal Processing: what to do with all that digital data streaming off at MHz or even, these days, GHz rates.
In the capstone lab they even learn how to solder and wire-wrap. But, as I tell them, even after they graduate they still won't be able to fix their dad's stereo amp. :-)
Oh, right, this is the wreck. I better say 73 and get back to the workshop where I'm fighting with a piece of zebrawood.
VMS also had a great command line language. (Called "DCL", IIRC). You could automate many tasks. Even create simple apps with a little more work. Also, saying that most crashes are due to , is missing the point. The point is that VMS did not crash. Windows is where the money is, and VMS is extinct, so that's where I spend my time out of the woodshop. Mac, with it's unix underpinning might be interesting to try...
Give that man a silver dollar. I was installing that big honker LV
10-1/2" steel bench vise on my new bench with hard maple blocks to offset the vise from the bottom of the bench. I used 3/8" lag screws, and followed the advice in one of those tables drilled 9/32" pilot holes. the first screw sheared off when it was 1/2 way in.
Windows has several different scripting languages available. Don't assume that the DOS shell is the only shell available.
On bad hardware it did. Most Windows 2K/XP/Vista crashes are hardware.
What you're saying, whether you realize it or not, is that a VAX or Alpha with VMS didn't crash. You're ignoring the reliability of the hardware. Put OpenVMS on a non-ECC Itanic with crap RAM and work it hard and you're likely to get a surprise.
VMS was capable of taking hardware subsystems off-line (while keeping the system up and running) when soft error rates exceeded a certain threashold. Heuristic diagnostics, although only part of the diagnostic suite) turned out to be very effective.
So you don't consider the display driver in a single-user computer to be part of the system? Even if you can't fault MS for the dispaly driver code, it is the O/S architecture that enables that scenario.
On stability:
Secure memory management has been an afterthough in the Windows world. The legacy of relying on applications to be good citizens is what system craches and viruses possible.
By contrast, the VAX memory management architecture (the hardware architecture) was specifically developed to enable secure (that is, a one process can't access another process' memory data,code,stack) memory access and paging.
I rather like that apple (Jobs) decided to give up their broken NIH syndrome and go with unix.
Shows levelheadedness, with a touch of chutzpah...
FTR, I started out on DEC machines, and loved VMS until I ran into Unix, which was a horrific culture shock at the time, but I grew to love it over VMS.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.