12 ms·

> ... the downside is that it's becoming almost impossible to find anyone who is capable of doing this sort of thing without having access to these tools and ma

> ... the downside is that it's becoming almost impossible to find anyone who is capable of doing this sort of thing without having access to these tools and machines. The whole point of the article was nobody can feel that pain even if deliberately, not pointless nostalgia. He's not telling you to start from stone age. He's telling lack of popularity threatens the existence much of the history which was just recently available for the interested parties. Let me put is this way: > On the other hand, some things do tend to niggle at me. One of these things is the sneaking suspicion that we are in danger of losing track of how to do things at the most fundamental levels. If you go back to 1970, for example, there were legions of programmers who could create the most wondrous software out of 8 bit assembler by hand. These days, by comparison, we have incredibly sophisticated compilers, editing environments and runtimes that can do a lot of the "thinking" and the code generation for us -- the downside is that it's becoming almost impossible to find anyone who is capable of doing this sort of thing without having access to these tools and machines. We are poised to lose the experience, history and other trivia that makes us what we are now if not preserved and is easily accessible.

VLM 13y ago
"If you go back to 1970, for example, there were legions of programmers who could create the most wondrous software out of 8 bit assembler by hand." No, first of all he describes what we called writing machine code and calls it "assembler by hand" which is kind of funny. Trust me, on CP/M and TRSDOS and all that on microcomputers we had perfectly good advanced multipass macro assemblers, even (K+R) C compilers, pascal, forth, basic, even some fortran and cobol implementations. In the 60s there were people toggling in the RIM boot loader on their PDP-8 in machine language but not understanding any of it other than an exercise in manual dexterity, to use RIM to load BASIC to actually do stuff in BASIC. For actual machine coders I think you have to go a step or two earlier... IBMs peculiar BCD mainframes in the transistor era and stuff like that. Of course even they had autocoder and RPG for "real" work... Secondly he calls them legions as a turn of phrase but even that is unlikely. Tens of thousands sounds about an order of magnitude too high for the 70s and early 80s. A thousand is probably about right and most of them were debugging assemblers and/or debugging debuggers and symbolic disassemblers and/or cracking copy protection, hand machine coding as an annoyance rather than by choice as the best tool for the job. Finally to this very day you can find people perfectly content to toggle in PDP-8 machine language on a real front panel or 1802 or 8080 code in octal on a H-8 or look at and understand a hex dump about to be slipstreamed into a FPGA's memory for a soft core processor. And guess how many of us there are today... maybe ten thousand... The fact that the percentage has dropped because since then there were added 1M ruby programming CRUD app writing devs in the denominator of the percentage means very little WRT the quantity of the numerator of the percentage or even its growth rate. And we know each other, and we talk with each other, and work on projects together. Not on HN. You want to talk about manual metal lathes and milling machines you go to the HSM BBS online not reddit. You want to look at a repo of FPGA code you go to opencores.org not github. You want to build a machine language front panel for a 1802 in late 2013 you google for lee harts 1802 membership card kit (I have one and it works quite well) not by searching at best buy. The inability of outsiders to successfully very casually google a group does not in any way imply a problem inside a group. "becoming almost impossible to find anyone who is capable" This is an echo of the H1B argument / STEM crisis claiming we are in immanent danger because we can't find any capable coders (willing to work for $7.25/hr or equivalent less on salary). You want someone to write a BIOS or write an ethernet driver, we're here. Its not our problem that you can't find anyone to work for $25K/yr, put up a reasonable offer and we'll come out of the woodwork. Some MBA's inability to provide a reasonable salary hardly proves our non-existence. It would be hilarious to settle religious arguments about the existence or non-existence of ... that way. I find there is an extreme shortage of MBAs willing to work for $7.25/hr in the mail room, so we need to import a couple million to maintain economic dominance, otherwise we'll run out of MBAs and have no one left to screw stuff up or whatever. I would imagine for an english speaker its pretty hard to google for people speaking German. That does not imply no one speaks German anymore or German is dying out and in immediate short term danger of disappearing. Its sort of a dystopian anti-nostalgia which clouds accurate observation.
- artificialidiot 13y ago
  Demoscene is a fine example where people already continue a tradition as the article suggests, even if not gushing but still trickling. I don't claim computing tech faces such a danger yet. I only tried to put a more familiar spin on what the article says for mechanics. You went on a tangent and were overly pedantic for some reason. I won't address every single thing I think you are wrong about but I agree with your depiction of our tech bubble and I disagree with your sentiment that nobody who is not american and willing to work for 25K can write BIOS/driver level software. Your bubble is a threat itself. Pop it or face irrelevance by extinction.
  - VLM 13y ago
    "I disagree with your sentiment that nobody who is not american and willing to work for 25K can write BIOS/driver level software." Ah that is a good point. The problem is businessmen wanting to import H1Bs to maltreat them because they want to pay locals non-local (low) salaries. If I could live somewhere that $25K was a reasonable local wage, I wouldn't mind writing code for $25K. I don't, so I'm not amused at the argument that there are no coders available because there are none willing to live far below the local poverty level.

VLM 13y ago

"If you go back to 1970, for example, there were legions of programmers who could create the most wondrous software out of 8 bit assembler by hand." No, first of all he describes what we called writing machine code and calls it "assembler by hand" which is kind of funny. Trust me, on CP/M and TRSDOS and all that on microcomputers we had perfectly good advanced multipass macro assemblers, even (K+R) C compilers, pascal, forth, basic, even some fortran and cobol implementations. In the 60s there were people toggling in the RIM boot loader on their PDP-8 in machine language but not understanding any of it other than an exercise in manual dexterity, to use RIM to load BASIC to actually do stuff in BASIC. For actual machine coders I think you have to go a step or two earlier... IBMs peculiar BCD mainframes in the transistor era and stuff like that. Of course even they had autocoder and RPG for "real" work... Secondly he calls them legions as a turn of phrase but even that is unlikely. Tens of thousands sounds about an order of magnitude too high for the 70s and early 80s. A thousand is probably about right and most of them were debugging assemblers and/or debugging debuggers and symbolic disassemblers and/or cracking copy protection, hand machine coding as an annoyance rather than by choice as the best tool for the job. Finally to this very day you can find people perfectly content to toggle in PDP-8 machine language on a real front panel or 1802 or 8080 code in octal on a H-8 or look at and understand a hex dump about to be slipstreamed into a FPGA's memory for a soft core processor. And guess how many of us there are today... maybe ten thousand... The fact that the percentage has dropped because since then there were added 1M ruby programming CRUD app writing devs in the denominator of the percentage means very little WRT the quantity of the numerator of the percentage or even its growth rate. And we know each other, and we talk with each other, and work on projects together. Not on HN. You want to talk about manual metal lathes and milling machines you go to the HSM BBS online not reddit. You want to look at a repo of FPGA code you go to opencores.org not github. You want to build a machine language front panel for a 1802 in late 2013 you google for lee harts 1802 membership card kit (I have one and it works quite well) not by searching at best buy. The inability of outsiders to successfully very casually google a group does not in any way imply a problem inside a group. "becoming almost impossible to find anyone who is capable" This is an echo of the H1B argument / STEM crisis claiming we are in immanent danger because we can't find any capable coders (willing to work for $7.25/hr or equivalent less on salary). You want someone to write a BIOS or write an ethernet driver, we're here. Its not our problem that you can't find anyone to work for $25K/yr, put up a reasonable offer and we'll come out of the woodwork. Some MBA's inability to provide a reasonable salary hardly proves our non-existence. It would be hilarious to settle religious arguments about the existence or non-existence of ... that way. I find there is an extreme shortage of MBAs willing to work for $7.25/hr in the mail room, so we need to import a couple million to maintain economic dominance, otherwise we'll run out of MBAs and have no one left to screw stuff up or whatever. I would imagine for an english speaker its pretty hard to google for people speaking German. That does not imply no one speaks German anymore or German is dying out and in immediate short term danger of disappearing. Its sort of a dystopian anti-nostalgia which clouds accurate observation.

artificialidiot 13y ago
Demoscene is a fine example where people already continue a tradition as the article suggests, even if not gushing but still trickling. I don't claim computing tech faces such a danger yet. I only tried to put a more familiar spin on what the article says for mechanics. You went on a tangent and were overly pedantic for some reason. I won't address every single thing I think you are wrong about but I agree with your depiction of our tech bubble and I disagree with your sentiment that nobody who is not american and willing to work for 25K can write BIOS/driver level software. Your bubble is a threat itself. Pop it or face irrelevance by extinction.
- VLM 13y ago
  "I disagree with your sentiment that nobody who is not american and willing to work for 25K can write BIOS/driver level software." Ah that is a good point. The problem is businessmen wanting to import H1Bs to maltreat them because they want to pay locals non-local (low) salaries. If I could live somewhere that $25K was a reasonable local wage, I wouldn't mind writing code for $25K. I don't, so I'm not amused at the argument that there are no coders available because there are none willing to live far below the local poverty level.