Murphy's Laws of Computing

[Uncle Al]

Enjoy

Murphy's Laws of Computing

1. When computing, whatever happens, behave as though you meant it to happen.

2. When you get to the point where you really understand your computer, it's probably obsolete.

3. The first place to look for information is in the section of the manual where you'd least expect to find it.

4. When the going gets tough, upgrade.

5. For every action, there is an equal and opposite malfunction.

6. To err is human... to blame your computer for your mistakes is even more human, it's downright natural.

7. He who laughs last, probably has a back-up.

8. The number one cause of computer problems is computer solutions.

9. A complex system that doesn't work is invariably found to have evolved from a simpler system
that worked just fine.

10. A computer program will always do what you tell it to do, but rarely what you want it to do.

Uncle Al

[crfriend]

I can vouch for nine and ten above. Wholeheartedly.

I take mild umbrage with two, however, and that's because the simpler machines are wonderful teaching tools -- and those never go "obsolete".

[Dale]

I can vouch for nine and ten above. Wholeheartedly.

I take mild umbrage with two, however, and that's because the simpler machines are wonderful teaching tools -- and those never go "obsolete".

I must agree with the above post 100%. The older machines ARE wonderful teaching machines . Obsolesce is not an issue.

[Brandy]

Murphy, computers and I have a Faustian bargain. I live by computer System Admin/Engineering and enjoy keeping Murphy in his box. Sometimes he cracks his whip in my direction which results in hours or days of pain.

Cut my teeth on Univac 9000's.

-- Brandy

[sapphire]

Come all you young geek-noids
And list' while I sing
For the love of computing
Is a terrible thing
You time slice your soul with a symbolic name
And it makes you a part of the programming game

When I was young
I followed the yen
To delve in the bowels
Of the PDP-10 [what's your favorite model?]
To become a wizard
And bask in that fame
And take my own part
In the programming game

I structured my programs
No gotos in sight
I ordered my life
Down to word, bit and byte
I read all of Knuth
'Till it sort merged my brain
A victim of sorts
In the programming game

And now I am swapped out
My queue is all holes
All that Microsoft code has
Sucked out my soul
But with hacking and cracking
Revenge I will gain
And wipe out their part
In the programming game

[crfriend]

Murphy, computers and I have a Faustian bargain.

Don't we all!

Sometimes he cracks his whip in my direction which results in hours or days of pain.

I understand that completely; I'm still somewhat reeling from a patching session done to the backup infrastructure at work a day or two before Christmas two years ago.

Cut my teeth on Univac 9000's.

Nice. I've seen Univac machines but have never worked on, nor programmed one.

To Dale's comment on "teaching systems", I maintain several fully-operational minicomputers from the 1960s and 1970s for just that purpose. I've found that "the more lights the better" (viz the PDP-12) because it allows those who are learning to get a better grasp of what's happening inside the machine -- and with machines of this class it's perfectly possible for an individual of decent intelligence to completely grasp, at the gate level, what's going on in one at any given instant of time. Lovely instruction sets (the Data General NOVA and Digital Equipment Corporation's DECsystem-10) are also a very good introduction to how computers work, and in simple architectures it's perfectly possible to trace precisely how an instruction is executed.

One of the things that amazes -- and pains -- me is just how many folks really believe that Bill Gates invented computers and, by extension, computing. Of course nothing could be further from either truth or fact, but it persists. I am not amused.

[sapphire]

Hey, I liked the early PDP-11s too!

[Brandy]

Hey, I liked the early PDP-11s too!

Hmm, pdp-11/45, 11/70 and the grown versions the VAX's and VMS. All ancient memories now. These days its Unix of different flavors Solaris 9, 10, 11, Red Hat and CentOS. How many here remember the derivative of Microsoft Windows? Hint nope it is not Apple.

-- Brandy

[crfriend]

Hmm, pdp-11/45, 11/70 and the grown versions the VAX's and VMS. All ancient memories now.

Those are fun machines, but Sapphire was referring to her earliest experiences on an 11/20 -- the very first of the breed. I happen to have one here at home. It's a nice little box, and it was the only -11 that was time-stated; all the others, not to mention the VAXen, were microcode devices -- which are not particularly good for teaching computer-theory with unless you're teaching with the microinstruction set.

These days its Unix of different flavors Solaris 9, 10, 11, Red Hat and CentOS.

It's a shame that the SPARC architecture looks like it may be dying out; once that's gone we'll all be trapped in an Intel-based monoculture that'll be guaranteed to cause proiblems at some future point. Unless somebody does something very interesting with ARM, some derivitive, or brings back the MIPS, POWER, or PA/RISC to general-purpose computing.

How many here remember the derivative of Microsoft Windows? Hint nope it is not Apple.

This one I do not know -- and I'm curious. I know that NT (modern Windows) is very VMS-ish under the covers; which is not surprising when one recalls that it was done by Dave Cutler who also did -- wait for it -- RSX-11 and VMS!

[Brandy]

Hmm, pdp-11/45, 11/70 and the grown versions the VAX's and VMS. All ancient memories now.

Those are fun machines, but Sapphire was referring to her earliest experiences on an 11/20 -- the very first of the breed. I happen to have one here at home. It's a nice little box, and it was the only -11 that was time-stated; all the others, not to mention the VAXen, were microcode devices -- which are not particularly good for teaching computer-theory with unless you're teaching with the microinstruction set.

These days its Unix of different flavors Solaris 9, 10, 11, Red Hat and CentOS.

It's a shame that the SPARC architecture looks like it may be dying out; once that's gone we'll all be trapped in an Intel-based monoculture that'll be guaranteed to cause proiblems at some future point. Unless somebody does something very interesting with ARM, some derivitive, or brings back the MIPS, POWER, or PA/RISC to general-purpose computing.

How many here remember the derivative of Microsoft Windows? Hint nope it is not Apple.

This one I do not know -- and I'm curious. I know that NT (modern Windows) is very VMS-ish under the covers; which is not surprising when one recalls that it was done by Dave Cutler who also did -- wait for it -- RSX-11 and VMS!

Crfriend,

The PDP models listed are just the old one's I worked on. The earliest PDP I worked on was a PDP-8, I never had a chance to work on other PDP preceding the 11/45.

Some good news for you then Sparc's are not dead yet. Oracle has dumped the Ultra xx desktop line, which has left allot of people trying to recover including me. Oracle/Sun have moved into the server room. A little Sparc(ky) reading; http://www.oracle.com/us/corporate/press/1386731. We need desktops for our applications and "Solaris 11 x86 will run on any Intel x86 system" well not true, leave off any hardware more than one year old. Any way just before Xmas break we got Solaris 11 running our ide and application suite on a HP 8100 SFF PC. Good news the replacement hardware is 1/2 the cost of the Sun Ultra xx box.

To answer the quiz both Apple OS X, which is based on Steve Jobs "NeXT" AND Windows NT are based on the Carnegie Mellon Mach kernel. http://en.wikipedia.org/wiki/Mach_%28kernel%29 AND http://en.wikipedia.org/wiki/Windows_NT scroll down to the "Development" section.

--Brandy

[crfriend]

The PDP models listed are just the old one's I worked on. The earliest PDP I worked on was a PDP-8, I never had a chance to work on other PDP preceding the 11/45.

In retrospect, I consider myself exceptionally fortunate to have taken the path I did through my career. The first machines I did anything meaningful on were the Data General NOVA 840 (on which I learned several programming languages and made my first forays into really understanding hardware in a practical sense) and the Interdata Model 3 (which was donated to the school I was at in an inoperative condition, and I was given the task of resurrecting it). From there I passed through Control Data Institute and got my first professional gig working on PDP-10s (all ilks ever introduced including the 1966-vintage asynchronous-logic KA-10, the 1978-vintage KS, 1973-vintage KI, and 1975-vintage KL, in increasing order of performance) and PDP-11/40s. (This after having been turned away by DEC as, quote, "not being qualified to work on pdp11s". Their loss.) This set in motion a sometimes grudging respect for history that later on blossomed into something truly magical.

Over the years, I've worked professionally on all sorts of kit ranging from discrete-transistor logic to modern (and boring) microprocessor systems; in my personal life I've worked on an amazing array of technologies ranging from bit-serial delay-line-memory computers (the PB-250) through 12-bit parallel systems (which early on I looked down my nose at, until I actually encountered just how much fun they can be to play with), through the 36-bit beasts, and on into a bewilderingly large number of workstation types. It is for those reasons, and more, that I mourn the coming monoculture.

Some good news for you then Sparc's are not dead yet. Oracle has dumped the Ultra xx desktop line, which has left allot of people trying to recover including me. Oracle/Sun have moved into the server room.

Unfortunately, the transition caused much angst at many shops, including the one I currently toil in, and that's led to a wholesale shift away from Solaris atop SPARC and to Linux atop Intel. It's getting to the point that the only non-Intel-based kit I routinely see is in collections here and there.

To answer the quiz both Apple OS X, which is based on Steve Jobs "NeXT" AND Windows NT are based on the Carnegie Mellon Mach kernel.

There's a little bit of Mach in NT (not surprising given Gordon Bell), but the overwhelming bulk of it is Mica which was DEC's next-generation VMS; to this day I find it absolutely astounding and sometimes incomprehensible just how badly Microsoft set the world of computing back. Virtually everything they're hooting about now was being routinely done in the 1970s; so I think it is not inappropriate to say that the company set the computing world back by 30 years. Or more.