7 ms·

I think the difference is that the entire software/hardware stack is a world created entirely by humans, untouched by "reality" for the purposes of all these an

I think the difference is that the entire software/hardware stack is a world created entirely by humans, untouched by "reality" for the purposes of all these annoyances, so it feels like we should be able to wrangle it better after so many decades. It's entirely our own creation, and we decide every iota of it, and yet it bites us (justifiably or not - turns out thousands of people each creating different layers of a gigantic state machine is hard to perfectly coordinate, but we may have been able to do better by now if we had been more thoughtful and patient throughout).

ihadfun 6y ago
We probably could make gas stink less just like we could fix that bug. The ROI just isn’t there.
- TeMPOraL 6y ago
  Gas stinking is a feature, not a bug. It's a safety measure. Particularly when we're talking gas (and not gasoline), which is naturally odorless and made to stink on purpose.
Out_of_Characte 6y ago
Reality creeps in from us creating tech-utopia trough "legacy" systems where the internet, cryptography, multi-core architecture and full program isolation didn't exist yet. Software has a nasty habit of iterating on yesterday's ideas instead of rewriting for tomorrow. Not that there's anything wrong with that, it seems to be the path of least resistance thusfar.
- dublin 6y ago
  I disagree - by and large, we don't need to "rewrite for tomorrow". Almost every significant problem in Computer Science and Software Engineering was solved (well) decades ago, often in the 1960s. Security is a bigger issue now, but it was largely solved years ago. The problem is that we do engage in so much "rewriting", instead of leveraging known good, quality code (or at least fully-fleshed out algos, etc.) in our "new, modern, elegant, and trendy" software edifices of crap. To me, this may be the one really good thing to come of the cloud (as opposed to the re-mainframe-ication of IT): the "OS" of the 21st century, allowing plumbing together proven scalable and reliable cloud/network services to build better software. (Again, not a new idea, this was behind the "Unix Philosophy" of pipes, filters, and making each program do one thing well. Eventually, it will be in fashion to realize this really is a better way...) We need smaller, better software, not the latest trendy languages, insanely complex platforms that no one understands, and toolchains of staggering complexity that produce crap code so bloated that it requires computers thousands of times faster than the Crays of the 1990s just to do ordinary desktop stuff. (Seriously, folks, the Raspberry Pi 4 on the next table is a rough match for the Cray we had in Houston when I worked for Chevron in the early 90s! Think about that, and then think about how little data you really need to deal with to do the job, vs what you're actually shuffling around.)
hrktb 6y ago
I hear you, but feel like we are biased by what we accepted as normal in our formative years, and that filter doesn’t apply on what we are discovering now that we’re grown up professionals. For instance books have been with us for centuries, and honestly most of them suck. Paper pages are thin and sometimes cut your finger (how many times did you get cut by an ebook ?), most are weak to liquids yet our world is filled with liquids everywhere, sometimes coming down from the sky. Updates are painful and costly and non scalable. Font sizes can’t be changed, you have to use an external device to deal with it. Not saying there are perfect alternatives or that the tradeoffs don’t make sense. Just that we learned very early that books have these limitations and we’ll need to live with them to be a member of society. And we can agree all of these aspects could be and sometimes are fixed, but most people are just ok with books ‘sucking’ in those ways.
- adamsea 6y ago
  We’ve also had centuries to improve the technology of books and I think that makes a difference. Although the weaknesses you cite seem like problems in search of a solution. No one ever expected books to have variable font size ... why would they? Lastly let’s recall the book five hundred years ago is dramatically different from the book of today. For example your point about liquids is now in many ways resolved by the cheapness and ubiquity of books. 500 years ago, not so much.
  - TwelveNights 6y ago
    While nobody expected books to have variable font sizes, the fact that ebooks do allow it to be adjusted means that people with deteriorating vision may still read them.
    - jschwartzi 6y ago
      A magnifying glass was the original solution to this problem. They never run out of battery.
      - 6y ago
acituan 6y ago
> I think the difference is that the entire software/hardware stack is a world created entirely by humans, untouched by "reality" for the purposes of all these annoyances, so it feels like we should be able to wrangle it better after so many decades. Humans, as the makers of these systems, are part of that reality, which was not created by us. The reality is that we are great apes writing precise machine instructions with our general intelligence that was not purpose built for being that precise but selected for survival. Our cognitive machinery cannot exhaustively predict all the possibilities of failure of what we write, if we are working in teams, we have transfer most of our technical ideas still through natural language, in a combinatorially increasing manner as the team size increases etc. None of this is user hostile, it is just human fallibilities and limitations in play. And since we can't alter our cognitive capacity drastically, we can only make more machines against these (e.g. unittests) with their own limitations. I think the scale of what we have been achieving despite these limitations are just fantastic. If anything users are becoming too egocentric, expecting the world to conform to their comfort, with a dash of construal level fallacy, underestimating from a mile away how easy it would be to write bug free programs with perfect designs in a real world, by real people, with real budgets etc.
- titzer 6y ago
  > If anything users are becoming too egocentric, expecting the world to conform to their comfort, with a dash of construal level fallacy, underestimating from a mile away how easy it would be to write bug free programs with perfect designs in a real world, by real people, with real budgets etc. Selection bias. You only hear from users who want new features. You rarely hear from users who don't want new features and just want software to stop being buggy and acting like a goddamn haunted house.
  - acituan 6y ago
    I was talking about the “users who don't want new features and just want software to stop being buggy and acting like a goddamn haunted house.” so the selection bias is yours. Most bugs are just annoying, consequences are not catastrophic if your favorite tuner app forgets your settings, your word processor messes up the formatting, your pdf reader crashes. You can recover with some frustration and wasted time. The perception of these being catastrophic failures shows the sense of entitlement users have because they are used to a certain smoothness in their experience and expect everything to go their way. This doesn’t match the underlying realities of the task; it is very easy to construe a sense of a working program in one’s mind but it is exponentially difficult to make the implementation actually bug free, usable and functional the way user wants.
Negitivefrags 6y ago
You reminded me of this quote. “Einstein repeatedly argued that there must be simplified explanations of nature, because God is not capricious or arbitrary. No such faith comforts the software engineer.”

ihadfun 6y ago

We probably could make gas stink less just like we could fix that bug. The ROI just isn’t there.

TeMPOraL 6y ago
Gas stinking is a feature, not a bug. It's a safety measure. Particularly when we're talking gas (and not gasoline), which is naturally odorless and made to stink on purpose.

Out_of_Characte 6y ago

Reality creeps in from us creating tech-utopia trough "legacy" systems where the internet, cryptography, multi-core architecture and full program isolation didn't exist yet. Software has a nasty habit of iterating on yesterday's ideas instead of rewriting for tomorrow. Not that there's anything wrong with that, it seems to be the path of least resistance thusfar.

dublin 6y ago
I disagree - by and large, we don't need to "rewrite for tomorrow". Almost every significant problem in Computer Science and Software Engineering was solved (well) decades ago, often in the 1960s. Security is a bigger issue now, but it was largely solved years ago. The problem is that we do engage in so much "rewriting", instead of leveraging known good, quality code (or at least fully-fleshed out algos, etc.) in our "new, modern, elegant, and trendy" software edifices of crap. To me, this may be the one really good thing to come of the cloud (as opposed to the re-mainframe-ication of IT): the "OS" of the 21st century, allowing plumbing together proven scalable and reliable cloud/network services to build better software. (Again, not a new idea, this was behind the "Unix Philosophy" of pipes, filters, and making each program do one thing well. Eventually, it will be in fashion to realize this really is a better way...) We need smaller, better software, not the latest trendy languages, insanely complex platforms that no one understands, and toolchains of staggering complexity that produce crap code so bloated that it requires computers thousands of times faster than the Crays of the 1990s just to do ordinary desktop stuff. (Seriously, folks, the Raspberry Pi 4 on the next table is a rough match for the Cray we had in Houston when I worked for Chevron in the early 90s! Think about that, and then think about how little data you really need to deal with to do the job, vs what you're actually shuffling around.)

hrktb 6y ago

I hear you, but feel like we are biased by what we accepted as normal in our formative years, and that filter doesn’t apply on what we are discovering now that we’re grown up professionals. For instance books have been with us for centuries, and honestly most of them suck. Paper pages are thin and sometimes cut your finger (how many times did you get cut by an ebook ?), most are weak to liquids yet our world is filled with liquids everywhere, sometimes coming down from the sky. Updates are painful and costly and non scalable. Font sizes can’t be changed, you have to use an external device to deal with it. Not saying there are perfect alternatives or that the tradeoffs don’t make sense. Just that we learned very early that books have these limitations and we’ll need to live with them to be a member of society. And we can agree all of these aspects could be and sometimes are fixed, but most people are just ok with books ‘sucking’ in those ways.

adamsea 6y ago
We’ve also had centuries to improve the technology of books and I think that makes a difference. Although the weaknesses you cite seem like problems in search of a solution. No one ever expected books to have variable font size ... why would they? Lastly let’s recall the book five hundred years ago is dramatically different from the book of today. For example your point about liquids is now in many ways resolved by the cheapness and ubiquity of books. 500 years ago, not so much.
- TwelveNights 6y ago
  While nobody expected books to have variable font sizes, the fact that ebooks do allow it to be adjusted means that people with deteriorating vision may still read them.
  - jschwartzi 6y ago
    A magnifying glass was the original solution to this problem. They never run out of battery.
    - 6y ago

acituan 6y ago

> I think the difference is that the entire software/hardware stack is a world created entirely by humans, untouched by "reality" for the purposes of all these annoyances, so it feels like we should be able to wrangle it better after so many decades. Humans, as the makers of these systems, are part of that reality, which was not created by us. The reality is that we are great apes writing precise machine instructions with our general intelligence that was not purpose built for being that precise but selected for survival. Our cognitive machinery cannot exhaustively predict all the possibilities of failure of what we write, if we are working in teams, we have transfer most of our technical ideas still through natural language, in a combinatorially increasing manner as the team size increases etc. None of this is user hostile, it is just human fallibilities and limitations in play. And since we can't alter our cognitive capacity drastically, we can only make more machines against these (e.g. unittests) with their own limitations. I think the scale of what we have been achieving despite these limitations are just fantastic. If anything users are becoming too egocentric, expecting the world to conform to their comfort, with a dash of construal level fallacy, underestimating from a mile away how easy it would be to write bug free programs with perfect designs in a real world, by real people, with real budgets etc.

titzer 6y ago
> If anything users are becoming too egocentric, expecting the world to conform to their comfort, with a dash of construal level fallacy, underestimating from a mile away how easy it would be to write bug free programs with perfect designs in a real world, by real people, with real budgets etc. Selection bias. You only hear from users who want new features. You rarely hear from users who don't want new features and just want software to stop being buggy and acting like a goddamn haunted house.
- acituan 6y ago
  I was talking about the “users who don't want new features and just want software to stop being buggy and acting like a goddamn haunted house.” so the selection bias is yours. Most bugs are just annoying, consequences are not catastrophic if your favorite tuner app forgets your settings, your word processor messes up the formatting, your pdf reader crashes. You can recover with some frustration and wasted time. The perception of these being catastrophic failures shows the sense of entitlement users have because they are used to a certain smoothness in their experience and expect everything to go their way. This doesn’t match the underlying realities of the task; it is very easy to construe a sense of a working program in one’s mind but it is exponentially difficult to make the implementation actually bug free, usable and functional the way user wants.

Negitivefrags 6y ago

You reminded me of this quote. “Einstein repeatedly argued that there must be simplified explanations of nature, because God is not capricious or arbitrary. No such faith comforts the software engineer.”