[HN Gopher] The untapped potential of human programming (2022)
___________________________________________________________________
The untapped potential of human programming (2022)
Author : jasonnchann
Score : 70 points
Date : 2023-03-19 00:50 UTC (22 hours ago)
(HTM) web link (humanprogramming.substack.com)
(TXT) w3m dump (humanprogramming.substack.com)
| nathan_compton wrote:
| I have to say I'm viscerally turned off by the language here.
| Perhaps even the idea, although there is obviously nothing
| objectionable about helping people reach their own goals, per se.
|
| The problem here is that this whole angle on human life seems to
| have forgotten that efficiency, productivity, etc are all there
| to help us find more time to live in a world where those things
| don't matter. To have leisure. To think unstructured, non-goal
| directed thoughts. You don't need "programming" to be human.
|
| The other thing here is that this stuff is just what humans in
| all societies and organizations have been working on forever. We
| have collective and personal goals and we have all sorts of
| systems to reach them. We do research on how effective they are
| already. We do A/B testing. Not sure what calling this collective
| activity "human programming" accomplishes.
| majormajor wrote:
| The goal of productivity, efficiency, etc, to provide survival
| is a much older one than the goal of it to provide leisure.
|
| What to do when everyone can have survival and their material
| needs met is, I think, actually _not_ something people agree on
| at all.
|
| (And personally I don't think we are at that point yet, in
| terms of health and medical understanding.)
| dmn322 wrote:
| I think that's based purely on your assumptions. Elephants
| chill and play and splash in the water when they reach the
| water on the west coast of Africa after migrating across the
| continent.
|
| They're not the only animals doing this kind of thing.
| Animals not in captivity generally seem to enjoy their lives.
| Deprive an animal of dopamine and it will surely start to
| wither away.
|
| If you're talking about humans, i find it very difficult to
| believe governments predate chilling with your friends with
| some palm wine or something.
|
| Pleasure was an integral part of life until the accidental AI
| that is capitalism started manipulating human behavior for
| its own survival.
| majormajor wrote:
| I would say leisure was "for free" for humans too but was
| often traded for productivity for survival, which was more
| precarious.
|
| I found the book Metropolis to be a good history of that in
| the form of "what did cities look like over the years" -
| cities were dirty, crowded, less healthy, etc, than a lot
| of alternative lifestyles but they also supplied the
| material essentials in predictable and more reliable ways.
|
| If physical survival was _easy_ to do leisurely then there
| wouldn 't have been the motivation for more complex forms
| of society (many of which had very-unfun roles for many
| participants long before capitalism).
| steve1977 wrote:
| > The problem here is that this whole angle on human life seems
| to have forgotten that efficiency, productivity, etc are all
| there to help us find more time to live in a world where those
| things don't matter.
|
| In reality, it's there to help capital owners not care about
| these things anymore.
| rr808 wrote:
| > The problem here is that this whole angle on human life seems
| to have forgotten that efficiency, productivity, etc are all
| there to help us find more time to live in a world where those
| things don't matter. To have leisure.
|
| This is patently untrue. People in rich countries work the same
| or more than people in poor countries. Most people working full
| time could live perfectly well on a tenth their wage. Its a
| human condition that people want more.
| hanoz wrote:
| > Most people working full time could live perfectly well on
| a tenth their wage.
|
| Now that _is_ patently untrue. For most people 10% wouldn 't
| come close to paying the rent alone.
| rr808 wrote:
| I didn't say they would be able to live in the same space,
| they'd have to share with family, and/or cram into one room
| and/or live in a poorly built house, like people in poorer
| countries do, and we did 150 years ago.
| golergka wrote:
| People in rich countries work at much more fulfilling and
| enjoyable jobs. Most of my white collar friends, both in it
| and outside are passionate about what they do.
| amelius wrote:
| Advertising is a type of human programming, and I want to stay
| away from it.
| nickpeterson wrote:
| I've had this argument a lot with people. The goal of
| technology should be to adapt to humans and free up time. I
| feel like instead we keep asking people to further adapt to the
| needs of the system/machine. Keyboards are a great example. We
| should just write as people have done for thousands of years.
| Machines would then read what we write and understand it and
| then use that digitized info.
|
| The problem is that it's quite difficult, and machines decades
| ago simply couldn't do that, so now we all type on keyboards
| all day (digital or physical). It's reached the point where my
| children no longer have to learn cursive handwriting in school,
| because what is the point since everything goes into the
| computer via a keyboard? This strikes me as incredibly short
| sighted and backwards. I think it's hurting us long term.
| nine_k wrote:
| > _Keyboards are a great example. We should just write as
| people have done for thousands of years_
|
| Nope, nope, nope. Keyboards are a great thing that save a lot
| of effort and muscle strain. I did my fair share of
| handwriting, and I experienced the strain and tiredness
| firsthand (ha). This is to say nothing of the speed.
|
| Ergonomic keyboards make things even easier.
|
| So no, we don't always want faster horses, and the heritage
| of thousands of years often happens to be a yoke.
| nickpeterson wrote:
| I don't know, I'm pretty dubious that the problems of the
| future are grossly aided by faster input. I feel like I
| think better when I write than when I type, I find that
| _really_ hard to ignore.
| majormajor wrote:
| I think and write much better on a modern computer than
| when writing by hand. And much better than I would on a
| typewriter.
|
| Don't underestimate the ability to revise multiple times.
|
| At the same time, I design better with a pencil than with
| Sketchup or LucidCharts or something, for similar "draft
| then modify" reasons. But that pencil could be "real" on
| a paper or it could be on a tablet.
| nine_k wrote:
| Writing by hand is just slower, so it gives more time to
| think.
|
| Jotting and doodling, on the other hand, _is_ important
| to improve thinking, from my experience. It 's just
| mostly not about text.
| ruined wrote:
| you're shortsighted--handwriting also is a technological
| compromise.
|
| natural language interaction is perhaps the main focus of
| modern machine learning research.
| layer8 wrote:
| Physical keyboards are at a near optimum for symbolic input,
| for the same reason that many musical instruments are
| operated by keys, and the standard interface for entering
| notes is a (musical) keyboard. Due to how our fingers work
| and having ten of them, this is the most efficient and
| precise way for humans to enter symbolic information.
| h_r wrote:
| I have mixed feelings about the various arguments I see
| raised in the comments. It seems crazy to me to insist that
| efficiency and productivity gains via technology have, as
| their proper goal, a world in which none of that matters.
| Freeing up time in this way benefits us as individuals as
| well. Just staying alive and entertaining ourselves requires
| goal-directed behavior. I like not having to spend hours
| cooking food daily on an open fire.
|
| Your argument about keyboards struck me in just this way -
| it's a mistake to assume that we should stick with the status
| quo and have machines adapt to us. After all, writing on
| paper (or with a digital stylus) is just another iteration of
| improving the technology. Nobody wants to pound symbols into
| stone with a chisel, for example.
|
| I can type much faster than I can write cursively and it
| would be incredibly painful to revert to such writing.
| Natural language speech input can improve a lot of things vs
| typing, but I think writing code - for as long as it lasts -
| would be tricky to implement well using our voices.
| 1MachineElf wrote:
| You may find Neal Stephenson's _Snow Crash_ cyberpunk novel an
| interesting read. It could be considered a cautionary tale
| exploring the concept of human programming.
| orgelve wrote:
| [dead]
| Keeo wrote:
| Every time I am cooking sometig from the recipe I first need to
| translate it from the long text to a chart where one axis is time
| and second is what to do.
| susrev wrote:
| Regarding the introduction to this essay..
|
| Isn't some of the magic of being an agent in this world taken
| away when you are following a set of instructions to a T?
|
| Unable to make decisions for yourself without consulting your
| "virtual assistant" about something as trivial as if you can use
| y milk in place of x milk seems like a sad reality to me
| copymoro wrote:
| untapped? ahahahhaah
|
| it's not untapped. it's just slightly more effective if I don't
| tell you I'm trying to program you.
|
| in any case, in terms of "ultimate principles" all meaningful
| information is in the end an expression of some form of control;
|
| more precisely, we can only observe (as a 3rd "objective" party)
| the results of meaningful exchanges between interacting entities
| by noticing (measuring) changing behaviors whenever "meaning"
| gets transferred from one to another entity; i.e. I'm pointing
| out that (similarly to electricity/magnetism) we can never
| observe "meaning" directly, but can measure its effects on beings
| interacting with the "fields of meaning" ahahhaha.
| treetoppin wrote:
| While I appreciate the subject matter, I think ignoring the prior
| art of checklist design and human factors engineering is a big
| oversight. I don't know enough about either field to know what
| tools they are using in the process of designing instruction
| manuals, time critical check lists, or diagnostic and
| troubleshooting guides, but I imagine that would be a very good
| place to start. I will say though that bringing analog checklists
| into a multimedia world is an intriguing direction, since being
| able to access an expanded checklist that enables you to see
| details, infographics, etc when you want to dive into a specific
| step while also allowing you to track your current place in a
| process would be pretty cool.
| eternityforest wrote:
| The problem is essential complexity. Or at least that would be if
| we were computers. "The thing you want to do is hard" might be a
| better description.
|
| When was the last time an automated flowchart script help line
| actually helped you?
|
| If something is hard it's probably because one or more steps are
| hard by themselves, or because actually doing it requires
| multiple simultaneous actions in real time without gaps to look
| things up, or because there's an insane number of steps. People
| have learned from books for centuries, if you can't learn it from
| existing media,it's probably just a really hard task.
|
| Like, guitar is hard because it's all about repeatable physical
| motions, you don't have time to carefully inspect your fingers to
| see whether it will sound acceptable when you strum.
|
| Drawing is hard because it seems to involve a mental image that
| is so clear and stable one can use it as a reference, plus the
| ability to translate points in (Real or imagined 3D) space to
| points on a page.
|
| Troubleshooting tech is hard because of the number of things to
| go wrong, almost always not covered in the manual, because if the
| designers knew they couhave prevented it, leading to Google being
| the best tool, and full teardown, deep understanding, and reverse
| engineering often being needed if that fails.
| DeathArrow wrote:
| Human programming is easy until you reach the async, concurrent
| and parallel parts. Then you have to catch lots of exceptions and
| log the call stack. :)
| dr_dshiv wrote:
| And the copy-pasting of errors into chatGPT can result in
| carpal-tunnel syndrome really quickly!
| navane wrote:
| How about design the coffee maker so I don't need a program to
| use it?
|
| We have a water dispenser at work which can dispense regular,
| chilled, carbonated and cooking water. It has a printed out
| laminated instruction sheet next to it, because it all works with
| a combination of twisting a ring left or right, pushing it up or
| down, and I believe there is a handle too. That's bad design.
|
| This is the reverse of "automate anything that can be automated",
| this is bringing the humans back in a step by step process.
| mikrl wrote:
| A coworking space I visit has a little (digital) foot pedal for
| their water fountain
| esperent wrote:
| > How about design the coffee maker so I don't need a program
| to use it?
|
| The coffee machine is just an example, don't get hung up on it.
| If you think all possible coffee machines should be made so
| simple that they don't need instructions (1), then mentally
| replace it with a machine that you can accept will need
| instructions, like a 3D printer or a CNC lathe or whatever.
|
| (1) a fallacy in itself because even if you are a genius who
| finds every well designed machine obvious to use, there are
| plenty of us who are not geniuses and require instructions even
| for the simplest of mechanisms.
| textread wrote:
| Educators, generals, dieticians, psychologists, and
| parents program. Armies, students, and some societies are
| programmed.
|
| by Alan J Perlis, the first Turing Award recipient.
| zshrdlu wrote:
| Nice :) Do you know any books or anything that address
| "programmability", as distinct from "computability"?
|
| I've been mulling on it for a while now.
| dmn322 wrote:
| http://spacechimplife.com/societies-use-machine-language/
| r-bar wrote:
| I think human programming is not untapped at all. This is
| describing every line of business application in existence. A
| great example of both the power and limitations of this is phone
| trees for customer support.
|
| A product team and dev team encode business knowledge and flows
| into code and leverage a human to make judgement calls when
| necessary. The outcome is a program that can either be used by
| skilled workers to multiply their output or allow unskilled
| workers to perform tasks that would have formerly required a
| skilled worker to accomplish.
|
| There are already (arguably) optimized flows and design patterns
| for application UX. Companies have already spent years trying to
| build and optimize this "human programming". Dev teams have
| developed many DSLs to make it easier to encode business logic
| into their applications more quickly.
|
| I am not saying line of business applications are good or near
| some optimal final form, but to call "human programming" untapped
| is taking a very narrow view of the definition.
___________________________________________________________________
(page generated 2023-03-19 23:03 UTC)