Perhaps I'm just glorifying an age I don't remember, but I'm somewhat jealous of a time when computers weren't black boxes - everybody who owned one had to buy separate components, do bits of physical maintenance, and write code.
Maybe there is room for some kind of 'accelerated history of computing' where you get to replay the last 40 years or so.
Just like in school we start you out with arithmetic and progress from there it might make sense to see the computing world through some kind of compressed time curriculum to increase understanding and possibly avoid the re-learning of some rather painful lessons from the past.
I can relate a bit. Even though assembling components was a thing until early 2K, it's barely not anymore. Only gamer or very demanding people picks the pieces. Any laptop will do now, and even though I love my ThinkPad, I miss the physical interaction with wires and cards. It's not rare to feel that way, we like to sense, so miniaturization is taking something away from us. Also, even in the 90s, assembling a computer wasn't the same as in the 80s, the interfaces were complex and it was transitioning between DIY to plug'n'play so you didn't understand a lot of what was happening. Not the same as really controlling all the pieces with your own code.
I like assembling my own stuff as in micro-controllers and breadboards and things where complexity can be removed by looking at a wiring schematic. I absolutely hate assembling my own stuff as in plugging in an AMD R9 270 and fitting an Intel Core i7 and finding RAM that's compatible with your motherboard and plugging in SATA cables and making sure your case fans are blowing the right way to get proper airflow and figuring out how much of a power supply you need. The only challenge that offers is in selecting compatible parts, and the only reward is having something exactly the same as the person next to you who bought his pre-built.
I'm not surprised to see PC building going away. It's a high risk, low reward enterprise, especially considering that modern games haven't push the minimum spec envelope much further than what you can play on a Macbook Air. Building a PC is like Lego, where the pieces only fit together in one specific way, and each piece costs $250. The good part doesn't start until the building is done and you get into the software.
But there will always be a place for actual hardware tinkerers, the ones who need multimeters and solder and flux. Even if it is a niche.
Hehe, you're already far above the average joe level. As a teen I'd never have been able to really follow schematics, so just Lego playing with PCI cards and such was fun. That said, pre-built configs are never really what you want, I could keep using my old p2 350 because I had opportunity to pick nice enough video, sound cards and drives to improve dataflow in the system. Most on the shelf configs were 'HIGH SPEED CPU + crippled bottleneck coprocessors'.
ps: in the wintel world, I wouldn't call software a good part, unless you enjoy removing windows and drivers crapware, but you probably had clean distributable ghost images of some kind.
I've often thought this. It's something Woz mentions in his book, iWoz. There's something to be said about having a wholistic appreciation of the machine, and there was a time when a somewhat ordinary individual could hold a mental model of all the components -- software and hardware -- of their machine.
It's also a bit reminiscent of Zen and the Art of Motorcycle Maintenance -- the hero rides an older bike, while his buddy rides a newer BMW model that the buddy can't maintain himself. There's less of a connection between the technology and the user; less "feel".
Oh the manuals that came with the machines...they had schematics in them and explained every part of the machine and what it did. It was glorious.
It would also be difficult to do now as the machines are more complex. But if you imagine today that every new PC came with a full MSDN subscription and the Intel system programmer's manuals, that's about how it was.
Of course, it took many minutes to load a program from floppy disk, so there was a lot of not-so-glorious waiting.
Yeah I can understand romanticizing the parts of this guy's experience with computers, but it wasn't all gravy. Did you read the part where he marveled at his new "daisy wheel" printer only having broken down once in a year of use? Or the disk drive that took 10 minutes to write out a few kB?
And how about those prices? Sure, some were $400, but you'd need $3000+ for a machine that has non-volatile storage, can run arbitrary programs, and has an input and an output device. $2000+ dollars for a hard drive storing a few MB? Not to mention the horror of completely undecided OS and instruction set architecture wars.
It's cool to imagine the simplicity of this dude tinkering with his printer made from a converted typewriter and unplugging the entire thing when the thunderheads roll in, but we're definitely better off where we are today.
Indeed! When my first computer, a Nascom I (4MHz Z80A, 16K RAM on an expansion card, and an integer BASIC on eight 2708 EPROMs), died, I was able to have a stab at working out what might have gone wrong by going through the schematics. It proved to be beyond hope - the PSU had fried too much - but it was positively enlightening, going through the video system, and actually understanding, seeing, how the original clock was divided down to give the character generator clock. From there, down to the horizontal sync, and from there, the vertical. With all the logic involved being on just 74xx logic, it was all so readily evident just how this all worked.
As for program loading times, that's certainly true as well. One game I wrote, for another system, required two-pass assembly. So I'd edit the source, write it out (a few minutes), rewind, first pass, rewind, second pass, launch! So, about fifteen minutes from editing to seeing the results.
Much as a touch of nostalgia may be fun, I'm happier with my MBP and iPad Air. =:) (Still, the sense of wonder inspired by current devices is probably only enhanced by the sheer contrast in capabilities of ones from that time)
It also meant far fewer people were inclined to own them, and they were far less accessible.
These days we still do the same thing, those of us who are inclined - but we do it with the internet, funky web services, a thousand programming languages, and so on.
And if we want to tinker... we go buy an arduino or something.
I got my first computer in 1983 (a TRS-80), and at that point the old codgers were complaining that computers had gotten too commercialized and that the purity of the old hobbyist days had gone. They were probably right, though it certainly seems hobbyist now.
I wonder if the success of the Raspberry Pi is an indication a lot of people do still like hobbyist computer building, though?
Yeah, they do die. My old IBM PC, put in storage in working condition, came out and when I turned it on there was a snapping noise and the machine died.
This fate has befallen all my machines older than 15 years or so. They won't turn on, or turn on and fail the POST, won't boot, whatever. Sometimes there's that fine smell of smoke being released accompanying the failure.
On the other hand, my Carver amp I bought 30+ years ago still works fine and I run it 16 hours a day, every day. It's the best, by far, piece of electronics I've ever bought when considering bang for the buck. The same goes for the Dahlquist speakers it drives.
Tangentially: anybody have an expectation on longevity of a 2015 machine with decent components vs. one from that era? Are 2015 capacitors still as troublesome? I expect having fewer moving parts (as well as there just being fewer separate (visible) parts in general!) probably will help modern machines last longer.
This raises the philosophical question of where the life of the computer truly lies. Is it within the capacitor? Is it within the network of the individual parts?
Back when I went to university it was normal for my friends to gradually upgrade pieces, leaving after three years with a computer that had none of the the components it had arrived with.
The other common problems in computers as old as that would be leaking NiCad clock batteries (the corrosive gunk coming out of one of those will eat the copper right off the motherboard), tantalum capacitors which like to fail shorted and burn out, and frozen bearings on disk drives.
Modern machines may be more reliable because of the lack of moving parts, but they have other disadvantages in the longevity department. Lead free solder and conductive glue will start to fail after a few thousand thermal cycles. (Conductive glue is especially used to bond flex cables to LCD screens and is not something you could repair yourself.)
The "Capacitor Plague"[1] almost guarantees that machines made 10 to 15 years ago will have shortened lifespans. Would be interesting to peek into the future and see what antique computer collecting is like and if a working 2002 machine would be more expensive because of its rarity.
You're probably right, but replacing them would be many hours of work, and there doesn't seem much point. I was mainly interested in seeing what I'd left on the hard drive :-)
My 486 machine was also dead, but I pulled out the drive and managed to find an old IDE interface on a 10 year old computer that would read it (after trying several), and copied it. (Turns out there was nothing interesting there.) The old disk drive was the only piece of old equipment I could get to work.
I have a collection of old systems, and I've been able to keep my oldest - a 30 year old Oric Atmos - alive quite easily, while teaching myself a bit of electronics to boot.
I'm only very slightly older than this article; my first computer was an XT, so I missed the early hobbyist days of the many competing platforms; I was well into the clone era.
One thing really jumped out at me:
"Of the ones I've seen, a green-tinted monitor by NEC (model JB 1201M) seemed the best bargain, at $210; but patriots should take note that NEC stands for Nippon Electric Company."
We like to marvel at how fast technology moves. But cultural attitudes can change just as fast.
The internationalization of the United States was vastly accelerated by the PC market. It just became too much trouble to be parochial. And in a way, strong IP law was a big cause of the PC market being so international.
Quick, name a car produced by the big three in 1982 that you would actually want to own. "Morning in America" was loaded jingoism. In the eighties and early nineties Walmart filled its stores with a "Buy American" campaign.
That's part one.
The second part is that the audience of the Atlantic 1982 contained a lot of people born before the baby boom who lived through World War II and listened to a lot of anti-Japanese propaganda during their formative years. The casual racism it produced (helped along by the formal racism of their youth) was one reason jingoism was revived as a successful political strategy in the 1980, 82, and 84 election cycles.
Jingoism is still successful, but it had been before the early '80's, the 1960's and 70's were the exception, not the norm. What nobody would have bet on in 1982 is a non-white president or same-sex marriage. Old attitudes toward Japan die off with the people who hold them.
When I think I'm finished with an article,
I set the print speed to Slow. This runs the
printer at about 100 words per minute, or roughly
the pace of reading aloud. I stuff my ears with
earplugs and then lean over the platen as the
printing begins. Watching the article printed
at this speed is like hearing it read; infelicities
are more difficult to ignore than when you are
scooting your eye over words on a page.
The author forces himself to read slowly to make his mind reevaluate his words. Does any one else do this or perhaps pass their compositions through a text 2 speech software? It seems like it could potentially be a useful writing aid.
I loved my old Diablo 630 daisywheel printer. The sound it made was great - it sounded like a machine gun when hammering out your document. Watching it furiously type was quite a sight. It was bulletproof, too. I couldn't find a computer museum to take it. So sad.
My mother used to be a primary school teacher; she has the habit when reading carefully (correcting homework) to read while dragging a ruler over the paper in order to occlude the next line.
After many years of reviewing and debugging code, I can't read books comfortably fast anymore, I'm too focused on catching every single word and punctuation. I'll often go back half a sentence if I thought I might have seen some kind of error. Can't help it...
Yes and no. The biggest writing task I do is film scripts (which run about 20-25,000 words). I'm used to editing as I go and I have the whole story plus a lot of the language in my head while I'm writing (or more often, typing - the composition part usually takes place away from the keyboard).
But once I'm done with a draft, the next day I go get it printed and then I go through it with a pencil, crossing things out and scribbling notes on it. Not being able to edit the paper version makes some errors suddenly visible (like misplaced subject errors that a spelling/grammar checker can't easily catch) and more importantly it gives me a much better sense of the narrative pacing and suchlike.
I do very much appreciate the capabilities of a good word processor. The idea of writing in linear fashion using a typewriter isn't appealing at all. Incidentally I'd like to mention I use Fade In, which is a one-man project by Kent Tessman who answers all his own support queries. It's well worth the $50 he asks and vastly more pleasant to use than Final Draft.
Yes, absolutely. This is standard practice among my peers in creative writing; the last step of editing is to read your piece aloud (to the cat, or the dog, or to the mirror, or a human victim, or just to yourself), slowly and carefully.
My friend who is finishing up his novel is currently in the reading-aloud stage. He's had various friends in and out of his house to listen to chapters as he does it, too.
As for myself, I've been caught by my friends standing in back stairwells reading papers aloud thirty minutes before class more times than I care to admit.
Keep doing it, you're likely an auditory thinker and reading aloud activates different structures from silent reading. Personally I've been known to think in actual text, which is somewhat unusual, but if I'm ever going to speak words I'm writing, I say them out loud beforehand, usually several times.
...For any of these systems (not including the Osborne), I'd spend no more than
$6,000, or half as much as for the Displaywriter.
Godspeed as you follow this advice; meanwhile, I'll be spending nothing, sticking
with SOL and The Electric Pencil, and hoping for a world in which my sons can
grow up to have a better computer than their father had.
I remember my dad writing a book in the 70's. My mom would type up each new draft. Over and over, she'd type the whole book. Getting a word processor is like getting a sewing machine in the 1800s.
I enjoyed this line: "The new machines will require different disk-operating systems, and may therefore inspire another DOS war. [...] many people suspect that IBM will wage a counteroffensive with a DOS of its own."
Close, but not quite! As it turned out, choosing to not exactly make a DOS of its own had rather large consequences for the computing landscape...
48 comments
[ 6.0 ms ] story [ 116 ms ] threadJust like in school we start you out with arithmetic and progress from there it might make sense to see the computing world through some kind of compressed time curriculum to increase understanding and possibly avoid the re-learning of some rather painful lessons from the past.
I'm not surprised to see PC building going away. It's a high risk, low reward enterprise, especially considering that modern games haven't push the minimum spec envelope much further than what you can play on a Macbook Air. Building a PC is like Lego, where the pieces only fit together in one specific way, and each piece costs $250. The good part doesn't start until the building is done and you get into the software.
But there will always be a place for actual hardware tinkerers, the ones who need multimeters and solder and flux. Even if it is a niche.
ps: in the wintel world, I wouldn't call software a good part, unless you enjoy removing windows and drivers crapware, but you probably had clean distributable ghost images of some kind.
It's also a bit reminiscent of Zen and the Art of Motorcycle Maintenance -- the hero rides an older bike, while his buddy rides a newer BMW model that the buddy can't maintain himself. There's less of a connection between the technology and the user; less "feel".
It would also be difficult to do now as the machines are more complex. But if you imagine today that every new PC came with a full MSDN subscription and the Intel system programmer's manuals, that's about how it was.
Of course, it took many minutes to load a program from floppy disk, so there was a lot of not-so-glorious waiting.
And how about those prices? Sure, some were $400, but you'd need $3000+ for a machine that has non-volatile storage, can run arbitrary programs, and has an input and an output device. $2000+ dollars for a hard drive storing a few MB? Not to mention the horror of completely undecided OS and instruction set architecture wars.
It's cool to imagine the simplicity of this dude tinkering with his printer made from a converted typewriter and unplugging the entire thing when the thunderheads roll in, but we're definitely better off where we are today.
As for program loading times, that's certainly true as well. One game I wrote, for another system, required two-pass assembly. So I'd edit the source, write it out (a few minutes), rewind, first pass, rewind, second pass, launch! So, about fifteen minutes from editing to seeing the results.
Much as a touch of nostalgia may be fun, I'm happier with my MBP and iPad Air. =:) (Still, the sense of wonder inspired by current devices is probably only enhanced by the sheer contrast in capabilities of ones from that time)
These days we still do the same thing, those of us who are inclined - but we do it with the internet, funky web services, a thousand programming languages, and so on.
And if we want to tinker... we go buy an arduino or something.
I wonder if the success of the Raspberry Pi is an indication a lot of people do still like hobbyist computer building, though?
This fate has befallen all my machines older than 15 years or so. They won't turn on, or turn on and fail the POST, won't boot, whatever. Sometimes there's that fine smell of smoke being released accompanying the failure.
On the other hand, my Carver amp I bought 30+ years ago still works fine and I run it 16 hours a day, every day. It's the best, by far, piece of electronics I've ever bought when considering bang for the buck. The same goes for the Dahlquist speakers it drives.
If you replace the capacitor did you resurrect the dead or just fix a broken part? https://en.wikipedia.org/wiki/Ship_of_Theseus
Modern machines may be more reliable because of the lack of moving parts, but they have other disadvantages in the longevity department. Lead free solder and conductive glue will start to fail after a few thousand thermal cycles. (Conductive glue is especially used to bond flex cables to LCD screens and is not something you could repair yourself.)
[1] https://en.wikipedia.org/wiki/Capacitor_plague
My 486 machine was also dead, but I pulled out the drive and managed to find an old IDE interface on a 10 year old computer that would read it (after trying several), and copied it. (Turns out there was nothing interesting there.) The old disk drive was the only piece of old equipment I could get to work.
You can't do that with an iPhone!
One thing really jumped out at me:
"Of the ones I've seen, a green-tinted monitor by NEC (model JB 1201M) seemed the best bargain, at $210; but patriots should take note that NEC stands for Nippon Electric Company."
We like to marvel at how fast technology moves. But cultural attitudes can change just as fast.
That's part one.
The second part is that the audience of the Atlantic 1982 contained a lot of people born before the baby boom who lived through World War II and listened to a lot of anti-Japanese propaganda during their formative years. The casual racism it produced (helped along by the formal racism of their youth) was one reason jingoism was revived as a successful political strategy in the 1980, 82, and 84 election cycles.
Jingoism is still successful, but it had been before the early '80's, the 1960's and 70's were the exception, not the norm. What nobody would have bet on in 1982 is a non-white president or same-sex marriage. Old attitudes toward Japan die off with the people who hold them.
My laserjet is so much better, but still.
But once I'm done with a draft, the next day I go get it printed and then I go through it with a pencil, crossing things out and scribbling notes on it. Not being able to edit the paper version makes some errors suddenly visible (like misplaced subject errors that a spelling/grammar checker can't easily catch) and more importantly it gives me a much better sense of the narrative pacing and suchlike.
I do very much appreciate the capabilities of a good word processor. The idea of writing in linear fashion using a typewriter isn't appealing at all. Incidentally I'd like to mention I use Fade In, which is a one-man project by Kent Tessman who answers all his own support queries. It's well worth the $50 he asks and vastly more pleasant to use than Final Draft.
My friend who is finishing up his novel is currently in the reading-aloud stage. He's had various friends in and out of his house to listen to chapters as he does it, too.
As for myself, I've been caught by my friends standing in back stairwells reading papers aloud thirty minutes before class more times than I care to admit.
Close, but not quite! As it turned out, choosing to not exactly make a DOS of its own had rather large consequences for the computing landscape...