My first "computer" was an IBM 407 Calculating Punch, programmed by placing jumper wires on a board about twice the size of today's mobos. By the time I went to college, I found they had a Clary DE-60, also programmed with jumper wires and a General Precision LGP-30. Back in '65, this computer had 64K bytes - of rotating drum memory, no RAM, not even "core" memory. I/O was only through the Freiden Flexowriter, a huge typewriter (85 lbs.) with a paper tape punch. And it was programmed in hex machine language. I'll never forget debugging hex where the codes were 0-9,F,G,J,K,Q and W. (They represented the codes from the Flexowriter for 10-15.) Finally, the school got an advanced computer, a CDC-8090 with 4K 12 bit words of core memory. But it came with tape drives, a punched card reader and a FORTRAN compiler! So we really coded up a storm!!
My first operating system was tape based because there wasn't any room left in memory after that compiler loaded its run time libraries for the execution. (It took two of us three weeks to write.) And it was reloaded after every job! The OS was really just a job control system but it was designed to interpret only the first two characters of each word on the job control card so we had contest to come up with the weirdest sentences that would still specify the right job parameters.
After graduation, I got to work on an IBM 1401 and one of the custom machines (AN/FSQ-31) IBM built for the military just before they designed the 360. By the time I bought my own home computer (a Commodore 64), I'd already been programming for over 15 years. But that C-64 was capable of graphics and sprites that the mainframe at work couldn't touch. When the C-128 came out and offered CPM, I though it was a wonderful alternative to OS/360.
My favorite magazine was Dr. Dobb's Journal but that was back when it was titled "Dr. Dobb's Journal of Computer Calisthenics and Orthodontia; Running Light Without Overbyte". Things have changed a bit since. I'm playing with Beowolf clusters at home and designing distributed comm networks for world wide deployment now but thanks for the opportunity to stroll down Memory Lane! ---- VP Unmarketing, Product Confusion and Linux Distributions
If your doctor were going to operate on you, would you prefer him to use a screwdriver or a scalpel? Most of us would prefer him to use the appropriate (depends on whether you're organic or cybernetic) tool. The tools have evolved for the application and don't cross over easily to new applications. Neither tool would work well in adjusting the response time of a web server.
So what application are we discussing here? Reading? The "words on paper" tool has evolved over the last few centuries to support the "reading" application - which includes several unstated but required features such as "in a comfortable chair," "propped up on my lap," and "in a warm incandescent light." Clearly, few electronic books meet those user requirements. Clearly, we are talking about using a "book" for some application other than what it was designed for. If we want "words on phosphor" to replace "words on paper" then we will also have to change some of the other requirements of "reading."
The application "enjoying a good story" has different requirements from "reading" and both differ from "locating parameter info for API." "Reading" is assumed when we're discussing the use of "words on paper" but electronic media including "words on phosphor," "books on audio tape," "DVD movies" and other multimedia presentations are distinctly different tools and should be used with applications other than "reading." Which has been a very round-a-bout way of saying we need to have a common frame of reference when we compare paper and electronic distribution of a collection of words - what's the intended audience and what's the intended application? These are vastly different and have little overlap. ---- VP Unmarketing, Product Confusion and Linux Distributions
I'm not surprised to see (again!) that "old" programmers are discriminated against. Same thing is true for "old" engineers. In some places! But who do you go to design a system that will take 200 to 1000 man years to finish? Or to find weak points in someone else's design for a billion dollar satellite system? Or to plan the integration and test for a nation-wide communications network? Or to fix a project that's way behind schedule and over budget and has more bugs than a long int can count? Do you go to the programmer with 2 years experience or the one with 20? The smart folks go for the experience, even with smaller projects. The ones who won't pay for experience eventually find themselves up the creek without a paddle, then without a canoe, then without even water in the creek. And the project (or even the entire business) gets canceled.
I've been a programmer since 1964. (Hmmm, that may be even more than 20 years . ..) Also a designer, analyst, project lead, program manager, systems integrator, instructor, consultant and trouble shooter. I've had the opportunity to work on accounting systems, operating systems, database systems, satellites, communications, device drivers, web sites, compilers, etc. I've seen projects staffed with kids (because they're inexpensive) and old fossils (because they're here) and seen just about every software design mistake possible - several times. And I know how to avoid them (mostly!)
These days I'm a programmer in my spare time (trying to play with Beowolf clusters on my home network) because my experience is too valuable to the company for bigger problems than coding a new application. I've successfully transmuted the joy of building a new program and seeing it work right into the joy of building full systems and seeing them work well.
Am I a zillionaire yet? Nah. Am I deeply enthralled with my current assignment? Not really. Would I change any of the choices I've made over the past three decades? Never!
So the question is: how do you go from being a newly graduated programmer eagerly sought after to an older but wiser fart who is still eagerly sought after? And the answer is in four parts:
Never stop learning!
Always be intensely curious about new ideas and techniques!
Never be afraid to make a mistake!
Always be willing to admit that it was your mistake - and learn from it!
---- VP Unmarketing, Product Confusion and Linux Distributions
Slashdot Mirror
My first "computer" was an IBM 407 Calculating Punch, programmed by placing jumper wires on a board about twice the size of today's mobos. By the time I went to college, I found they had a Clary DE-60, also programmed with jumper wires and a General Precision LGP-30. Back in '65, this computer had 64K bytes - of rotating drum memory, no RAM, not even "core" memory. I/O was only through the Freiden Flexowriter, a huge typewriter (85 lbs.) with a paper tape punch. And it was programmed in hex machine language. I'll never forget debugging hex where the codes were 0-9,F,G,J,K,Q and W. (They represented the codes from the Flexowriter for 10-15.) Finally, the school got an advanced computer, a CDC-8090 with 4K 12 bit words of core memory. But it came with tape drives, a punched card reader and a FORTRAN compiler! So we really coded up a storm!!
My first operating system was tape based because there wasn't any room left in memory after that compiler loaded its run time libraries for the execution. (It took two of us three weeks to write.) And it was reloaded after every job! The OS was really just a job control system but it was designed to interpret only the first two characters of each word on the job control card so we had contest to come up with the weirdest sentences that would still specify the right job parameters.
After graduation, I got to work on an IBM 1401 and one of the custom machines (AN/FSQ-31) IBM built for the military just before they designed the 360. By the time I bought my own home computer (a Commodore 64), I'd already been programming for over 15 years. But that C-64 was capable of graphics and sprites that the mainframe at work couldn't touch. When the C-128 came out and offered CPM, I though it was a wonderful alternative to OS/360.
My favorite magazine was Dr. Dobb's Journal but that was back when it was titled "Dr. Dobb's Journal of Computer Calisthenics and Orthodontia; Running Light Without Overbyte". Things have changed a bit since. I'm playing with Beowolf clusters at home and designing distributed comm networks for world wide deployment now but thanks for the opportunity to stroll down Memory Lane!
----
VP Unmarketing, Product Confusion and Linux Distributions
If your doctor were going to operate on you, would you prefer him to use a screwdriver or a scalpel? Most of us would prefer him to use the appropriate (depends on whether you're organic or cybernetic) tool. The tools have evolved for the application and don't cross over easily to new applications. Neither tool would work well in adjusting the response time of a web server.
So what application are we discussing here? Reading? The "words on paper" tool has evolved over the last few centuries to support the "reading" application - which includes several unstated but required features such as "in a comfortable chair," "propped up on my lap," and "in a warm incandescent light." Clearly, few electronic books meet those user requirements. Clearly, we are talking about using a "book" for some application other than what it was designed for. If we want "words on phosphor" to replace "words on paper" then we will also have to change some of the other requirements of "reading."
The application "enjoying a good story" has different requirements from "reading" and both differ from "locating parameter info for API." "Reading" is assumed when we're discussing the use of "words on paper" but electronic media including "words on phosphor," "books on audio tape," "DVD movies" and other multimedia presentations are distinctly different tools and should be used with applications other than "reading." Which has been a very round-a-bout way of saying we need to have a common frame of reference when we compare paper and electronic distribution of a collection of words - what's the intended audience and what's the intended application? These are vastly different and have little overlap.
----
VP Unmarketing, Product Confusion and Linux Distributions
I've been a programmer since 1964. (Hmmm, that may be even more than 20 years . .
These days I'm a programmer in my spare time (trying to play with Beowolf clusters on my home network) because my experience is too valuable to the company for bigger problems than coding a new application. I've successfully transmuted the joy of building a new program and seeing it work right into the joy of building full systems and seeing them work well.
Am I a zillionaire yet? Nah. Am I deeply enthralled with my current assignment? Not really. Would I change any of the choices I've made over the past three decades? Never!
So the question is: how do you go from being a newly graduated programmer eagerly sought after to an older but wiser fart who is still eagerly sought after? And the answer is in four parts:
Never stop learning!
Always be intensely curious about new ideas and techniques!
Never be afraid to make a mistake!
Always be willing to admit that it was your mistake - and learn from it!
----
VP Unmarketing, Product Confusion and Linux Distributions