Interesting idea. Keep going with your prototyping.
But it's addressed at quite a different problem than "desktop" user interface. A mass-adoption user interface has to be a solution to how a lot of people can understand what things they can do and how to find those things and how to get them to work. You can offer a lot of functions, but it has to be possible for people to have a mental model of what's possible and how to do it. For that, often uniformity and simplicity (and lack of choice) are the keys to success. Or choice as an option (hidden away somewhere for the curious to discover), but not required to for the average user to succeed.
Things like - common language, common visual language, common process-design language, and other UX principles are determinants of success.
You can offer a lot of things, but offer them through a narrow, one-size-fits-all pipe. And make the things just work, with very little specialized knowledge needed.
If you have lots of microservices, what is the common ontology of process concepts and entities that allow their APIs to "just work" together, to form a coherent whole? You can't rely on there being smart programmers in the middle to glue A to B to Z.
If you're offering complicated stuff, then in today's world, you probably need a very good AI to "explain it like I'm 5" to every new user, and to remove unnecessary choices (that each require knowledge) from the end users.
End users want something they already understand, and they want it to just work.
If wide adoption is what you want, you need a SINGLE, UNIFORM thing with a strong "design opinion", user experience simplicity, aimed at mass market user, yet evolving, but with principled change management, as lessons are learned etc. etc.
Mass-market user interfaces are about mass-market users. Their purpose is not to serve the tinkering needs of geeks. The geeks can keep playing to their heart's content on alt-UIs, but there needs to be a benevelent-dictator-controlled single-themed mass-market UI, if mass-market desktop UI is one of the goals.
If you have a completely open-entry apps market, it will turn into a cesspool of malware, misleading fraudware. i.e. like the total set of websites on WWW.
If user choice of what app does every key function is not only allowed but required by regulators, the complexity of the user experience will rise beyond many peoples' comprehension. The utility they get out of the ecosystem will decline.
Quality-control (and avoidance of technical debt on a platform) is a legitimate GOOD that should be seriously weighed by market regulators.
User-experience complexity-creep prevention (and prevention of increase of choice-knowledge requirements of users) is a legitimate GOOD that should be seriously weighed by market regulators.
If I had any faith that market regulators could be that enlightened when balancing their decisions, I would say, sure, regulate to ensure competition in that market. Right now, I have no faith that anyone other than software architects would know what the hell I'm talking about.
When iPhone first came out in 2007, it was a completely closed computing eco-system with nothing but a handful of Apple apps.
It wasn't til more than 3 years later that Apple announced availability of 3rd party developer apps, tightly quality controlled by Apple.
Now Apple is going to be regulated in this market that they are entirely responsible for creating in the first place.
Not sure what I think of this. Is it regulatory overreach?
In theory, there's nothing except network effect stopping new company X from creating new mobile device line and OS X1 with completely new market for apps Y with arbitrarily different rules for participation in that market. Oh wait, that already happened. X1 = Android. X1b = Windows Phone (oh never mind),....
So how is this abuse of a monopoly. When the first real AR platform comes out, it might eventually overtake screen-phones. And it will have its own controlled app market. Isn't that a new competitor in the wider definition of the ubiquitous mobile apps market? Innovative disruption defeats potential monopoly, as far as I know. Isn't it fine to leave these markets to do what they will?
Ok so you too know how to critique. Really hard to tell where you are coming from politically, other than you have something against socialism (and capitalism? but it's hard to tell)
Perhaps you would be so kind as to tell what kind of political solutions you think we do need? Give me a target:-)
I did say that those who publish illegal material (that is, the poster onto the public network) should be held accountable. Although it is currently unclear as to in which country's jurisdiction and standards - is it determined by: a) where you post from (which could be random), b) where the poster is a citizen of, c) where the filmed act took place, or d) where it was copied to by web browsers i.e. everywhere in the world?)
If you're going to hold middleware responsible, which of the above a) b) c) or d) should determine whether it should be filtered and who should pay what penalty, to which jurisdictional authority?
Is the principle, for example, that a citizen of any country, and publishing from anywhere (else), should be charged with an offensive/illegal material publication offense separately by each jurisdiction in which at least one user of the Internet / middleware was able to see the content? So charged in some countries and not others, according to different standards and laws? If we do it that way, the Internet itself would become a de-facto illegal concept, and would have to be divided into siloed national intranets instead.
If middleware is to be held accountable for user-posted content, what are the standards of reasonableness of response-time? Given that there is as yet no AI that can reliably diagnose the really bad stuff, and that a human complaint-then-review process takes time and imperfect and somewhat arbitrary, and jurisdiction-relative decision-making, what is reasonable there. Given that content is edited / morphed specifically to avoid detection while being replicated around the web or re-posted on the middleware platforms, to what degree should the middleware owner be held accountable for reliably keeping the essence of the content off their platform at all times?
Things like capitalism, hierarchical governance, taxes etc just natural emerge, as they are evolutionarily stable, and not co-incidentally thermodynamically efficient, solutions in social game theory.
As soon as one person says: I have an idea, why don't you let me organize some collective activity so we can do it better with less wasted effort, and here, chip in some of your collected resources / effort-payment tokens to help with it, and get some benefit out of it, they've basically re-invented capitalism.
As soon as some good organizer / deal maker / resource aggregator says to the next rung of not quite as excellent organizers/deal makers / aggregators; look, it's more stable and economically efficient if I'm in charge without quarrel for a while, they've basically re-invented hierarchical governance by "elite-at-organizing" organizers.
And as soon as the central organizer(s) tip of the pyramid realizes it can only do its organized-initiatives thing by collecting a percentage of the resources / effort that the rest of the population is gathering / putting out, then they've basically re-invented taxation.
All of these things are inevitable in human society, or, indeed, probably, in any society of resource-constrained, environment-threatened semi-autonomous intelligent agents trying to optimize their energy-expended-per-survival-probability-increment, and able to communicate with each other and influence each others' behaviour.
Re: populism - One problem with it is that that populist leaders tend to exploit their superior grasp of psychological manipulation to amp up the anger by whatever rhetorical tricks will do the trick. This often means cynically riling people up about the wrong issues (spurious issues, or entirely mischaracterized or fictional issues) with (then, obviously) wrong diagnoses for a fix. Rationality and good information seem to be in extremely short supply in populist movements and leadership, as does general good will toward all of humanity, never mind the ecosystems / biosphere.
So yes, I'm opposed to that kind of irrational, mean-spirited, xenophobic populism that is all the rage these days.
Populism today has a clear connotation of manipulation and amplification of peoples' worst instincts (misdirected fear, anger, willful ignorance, denialism, inappropriate blaming, emphasis of insider/outsider division based on some arbitrary criterion, fanning flames of conflict etc etc).
--------------- We don't need: When the going gets tough, the bullies get going. That's chimpanzee-pack level. We don't need MMA/WWE-style politics where the loudest roared bullshit wins support. That's not going to help anything. That's just a symptom of the shitstorm, not a cure.
When the going gets tough, we need knowledgeable and compassionate leadership to implement rational, fair changes and solutions that have a chance of effectively improving the situation. We need eloquence and charisma in service of informed, decent, intelligent ideas that could actually work in our complex world.
1) Well there are many countries in the world where the top 10% or so live a "western standards" wealthy or upper middle class lifestyle and 90% essentially rot in a subsistence black/grey market economy. That pattern is already well established. And the top 10% there do fine, (as long as they can afford adequate security measures to fend off the moderately embittered opportunistic underclass.)
I'm not saying that's good. I'm just saying it's a historically well established pattern that still persists in various places. Not implausible at all.
4) The voting power of the disenfranchised will eventually trump (no pun intended heh heh) that of the shrinking number of beneficiaries of the automation. The results of this are unpredictable but could polarize to: a) Uber-Trump Moron Populist Fascism lays waste to pretty much everything, or b) Sensible leader puts in good universal basic income program (and then is probably assassinated, unfortunately - see a)
5) Or of course, there WAS the French Revolution kind of scenario. Of course, after having decapitated the problem, you have to figure out what to do next.
The web as a whole is event larger, and with only a tweak to search tech, an unmoderated "virtual, distributed" facebook-like browsing pattern could be supported.
So how are you going to police the whole web? And why would you even try?
I really think people need to get thicker skins, and accept that the full, ugly range of human behaviour exists. People who don't want to be able to see it all should start becoming patrons of some new filtered-search service that works FOR THEM, according to their own specified standards.
The onus should be on the viewer not to look.
And of course, posters of illegal material should also be identified and prosecuted.
Middleware is just middleware, a platform, and new ones can spring up like hydras heads in the Internet infrastructure as a whole. Middleware is not the problem. It's a mistake to focus on that. That leads to totalitarian censorship.
Yes. AI is being used to try to police/filter it quickly these days (because with a billion possible contributors, how else could you do it fast?) but the challenge for that AI to recognize really bad content (the only kind that should be filtered, right?) is very complex. Human moderators often can't even do it reliably.
So realistically, it has to be a system that relies partly on human users to flag bad content for immediate review. Then a large team of moderators needs to check and decide, which in some cases requires knowledge? Is that a deepfake first-person-shooter video game? Is it real? Is it a movie scence? Does our policy care about these distinctions?
That's what takes nearly an hour to get reliable blocking going. We should not be surprised.
Or maybe I'm wrong. Put yourself in Zuckerberg's shoes and tell me with some technical detail how YOU would solve this problem. I'm not being sarcastic. Realistic ideas well described welcome here. Ideas with gaping holes in feasibility with current tech not useful though.
what category of product or service or influence campaign etc the ad is about. That's some pretty heavy AI, unless they ask advertisers to self-disclose in some official form what kind of ad this is, which seems impractical to enforce.
by positive targeting on stereotyped "white people" interests, I suppose. I guess you have to ban all culture-indicating tags from housing and similar opportunity (e,g, employment) ad targeting.
How do you allow advertisers to target, say, latin-dancing party ads to a market segment "interested in Hispanic culture" without permitting the kind of discrimination talked about? Do you just not allow the NOT operator in the targeting? or don't allow NOT in front of a set of protected categories?
Maybe it's just that simple. Thoughts, all you clever data types?
Wait, did I say per working capita? No, I believe I didn't. I said per capita, which means per person with a head. USA 833 moter vehicles per 1000 people. China 173 motor vehicles per 1000 people
I know how numbers work pretty well.
Per capita is the only fair way to assess GHG emissions policy and progress.
Conventional diesel buses are already much more efficient than cars. A bus emits approximately 1/6 the CO2 per passenger kilometer as does a single-occupant ICE car.
From that perspective, electrifying cars should give more ghg reductions, in a population where most people are using cars.
In China, the car use is probably not that high yet, as a proportion of population, so it makes sense that busses are having a bigger oil displacement there. That is not replicable to other car-centric places like USA.
"One current FAA safety engineer said that every time the pilots on the Lion Air flight reset the (trim) switches on their control columns to pull the nose back up, MCAS would (reset its 0 degree reference and) have kicked in again and “allowed new increments of 2.5 degrees.”
“So once they pushed a couple of times, they were at full stop,” meaning at the full extent of the tail swivel, he said.
So in summary a system FAA-certified on the basis of being able to adjust nose-down trim by 0.6 degrees could actually, (after a few cycles of the pilot correcting it a little bit with trim up), command full nose-down trim, about 5 or 6 degrees tailplane tilt.
All of this relying on input from a single angle-of-attack sensor. Get this, the plane has two such sensors, one on each side, but the MCAS only uses input from one of them!!! ! !! ! ! ! ! What the hell? If you use two of them, then your software can check if they diverge, and disable systems relying on the input, and warn the pilots. That is some criminally bad development cost saving judgement there.
IMHO there is a structural problem in many "hybrid" (non developer-run) companies where non-technical, unskilled-technical (migrated to mgmt), or vaguely historically technical management need to see the technical staff "productively mirroring" their own highly social and highly multi-tasking management jobs.
The incentives are mostly aligned with willingness to butterfly from one urgent task/project/proposal to another. Aligned with making consistent, predicted, scheduled, budget (no surprises, no innovation) progress in one week intervals, even if one is only making low-hanging-fruit progress on easy problems, and avoiding the hard problems because they would get you in trouble with management quickly.
In these organizations, trying to get into and maintain flow state to work with sustained focus on actually hard technical problems (be it architecture, thoughtful requirements analysis, thorough survey of prior work, good design, good quality well-factored coding, etc) is usually punished. Would-be practitioners of focus are thought of as loners, mavericks, not communicating, missing administrative tasks, etc etc.
Even though those people, if given a conducive work environment, would almost always be the most productive contributors to sustainable and innovative s/w development. And if they had enough freedom from interruption, gratuitous extra meetings on random side topics, etc, then they could also choose their own natural focus-break times.
Are there any software workplaces like those described in my last paragraph? Somewhere over across the wide river?
Slashdot Mirror
Interesting idea. Keep going with your prototyping.
But it's addressed at quite a different problem than "desktop" user interface. A mass-adoption user interface has to be a solution to how a lot of people can understand what things they can do and how to find those things and how to get them to work.
You can offer a lot of functions, but it has to be possible for people to have a mental model of what's possible and how to do it.
For that, often uniformity and simplicity (and lack of choice) are the keys to success. Or choice as an option (hidden away somewhere for the curious to discover), but not required to for the average user to succeed.
Things like - common language, common visual language, common process-design language, and other UX principles are determinants of success.
You can offer a lot of things, but offer them through a narrow, one-size-fits-all pipe. And make the things just work, with very little specialized knowledge needed.
If you have lots of microservices, what is the common ontology of process concepts and entities that allow their APIs to "just work" together, to form a coherent whole? You can't rely on there being smart programmers in the middle to glue A to B to Z.
If you're offering complicated stuff, then in today's world, you probably need a very good AI to "explain it like I'm 5" to every new user, and to remove unnecessary choices (that each require knowledge) from the end users.
End users want something they already understand, and they want it to just work.
in the market (by adoption numbers) is Android.
If wide adoption is what you want, you need a SINGLE, UNIFORM thing with a strong "design opinion", user experience simplicity, aimed at mass market user, yet evolving, but with principled change management, as lessons are learned etc. etc.
Mass-market user interfaces are about mass-market users. Their purpose is not to serve the tinkering needs of geeks. The geeks can keep playing to their heart's content on alt-UIs, but there needs to be a benevelent-dictator-controlled single-themed mass-market UI, if mass-market desktop UI is one of the goals.
If politicians could demonstrate an accurate understanding of science and technology, perhaps we could trust them to regulate it....
But, um, no.
If you have a completely open-entry apps market, it will turn into a cesspool of malware, misleading fraudware. i.e. like the total set of websites on WWW.
If user choice of what app does every key function is not only allowed but required by regulators, the complexity of the user experience will rise beyond many peoples' comprehension. The utility they get out of the ecosystem will decline.
Quality-control (and avoidance of technical debt on a platform) is a legitimate GOOD that should be seriously weighed by market regulators.
User-experience complexity-creep prevention (and prevention of increase of choice-knowledge requirements of users) is a legitimate GOOD that should be seriously weighed by market regulators.
If I had any faith that market regulators could be that enlightened when balancing their decisions, I would say, sure, regulate to ensure competition in that market. Right now, I have no faith that anyone other than software architects would know what the hell I'm talking about.
When iPhone first came out in 2007, it was a completely closed computing eco-system with nothing but a handful of Apple apps.
It wasn't til more than 3 years later that Apple announced availability of 3rd party developer apps, tightly quality controlled by Apple.
Now Apple is going to be regulated in this market that they are entirely responsible for creating in the first place.
Not sure what I think of this. Is it regulatory overreach?
In theory, there's nothing except network effect stopping new company X from creating new mobile device line and OS X1 with completely new market for apps Y with arbitrarily different rules for participation in that market. Oh wait, that already happened. X1 = Android. X1b = Windows Phone (oh never mind),....
So how is this abuse of a monopoly.
When the first real AR platform comes out, it might eventually overtake screen-phones. And it will have its own controlled app market. Isn't that a new competitor in the wider definition of the ubiquitous mobile apps market? Innovative disruption defeats potential monopoly, as far as I know. Isn't it fine to leave these markets to do what they will?
Ok so you too know how to critique. Really hard to tell where you are coming from politically, other than you have something against socialism (and capitalism? but it's hard to tell)
:-)
Perhaps you would be so kind as to tell what kind of political solutions you think we do need? Give me a target
I guess this will have to come in 5 different levels of explanation of the AI decision:
... ...
Level 1: Explain it to me like I'm 5 or wear a MAGA hat.
Level 2: Explain it to me like I'm the average regulatory enforcement bureacrat
Level 5: Explain it to me like I'm Geoffrey Hinton
I did say that those who publish illegal material (that is, the poster onto the public network) should be held accountable.
Although it is currently unclear as to in which country's jurisdiction and standards - is it determined by:
a) where you post from (which could be random),
b) where the poster is a citizen of,
c) where the filmed act took place, or
d) where it was copied to by web browsers i.e. everywhere in the world?)
If you're going to hold middleware responsible, which of the above a) b) c) or d) should determine whether it should be filtered and who should pay what penalty, to which jurisdictional authority?
Is the principle, for example, that a citizen of any country, and publishing from anywhere (else), should be charged with an offensive/illegal material publication offense separately by each jurisdiction in which at least one user of the Internet / middleware was able to see the content? So charged in some countries and not others, according to different standards and laws? If we do it that way, the Internet itself would become a de-facto illegal concept, and would have to be divided into siloed national intranets instead.
If middleware is to be held accountable for user-posted content, what are the standards of reasonableness of response-time? Given that there is as yet no AI that can reliably diagnose the really bad stuff, and that a human complaint-then-review process takes time and imperfect and somewhat arbitrary, and jurisdiction-relative decision-making, what is reasonable there. Given that content is edited / morphed specifically to avoid detection while being replicated around the web or re-posted on the middleware platforms, to what degree should the middleware owner be held accountable for reliably keeping the essence of the content off their platform at all times?
Re: capitalists
/deal makers / aggregators; look, it's more stable and economically efficient if I'm in charge without quarrel for a while, they've basically re-invented hierarchical governance by "elite-at-organizing" organizers.
Things like capitalism, hierarchical governance, taxes etc just natural emerge, as they are evolutionarily stable, and not co-incidentally thermodynamically efficient, solutions in social game theory.
As soon as one person says: I have an idea, why don't you let me organize some collective activity so we can do it better with less wasted effort, and here, chip in some of your collected resources / effort-payment tokens to help with it, and get some benefit out of it, they've basically re-invented capitalism.
As soon as some good organizer / deal maker / resource aggregator says to the next rung of not quite as excellent organizers
And as soon as the central organizer(s) tip of the pyramid realizes it can only do its organized-initiatives thing by collecting a percentage of the resources / effort that the rest of the population is gathering / putting out, then they've basically re-invented taxation.
All of these things are inevitable in human society, or, indeed, probably, in any society of resource-constrained, environment-threatened semi-autonomous intelligent agents trying to optimize their energy-expended-per-survival-probability-increment, and able to communicate with each other and influence each others' behaviour.
Re: populism - One problem with it is that that populist leaders tend to exploit their superior grasp of psychological manipulation to amp up the anger by whatever rhetorical tricks will do the trick. This often means cynically riling people up about the wrong issues (spurious issues, or entirely mischaracterized or fictional issues) with (then, obviously) wrong diagnoses for a fix.
Rationality and good information seem to be in extremely short supply in populist movements and leadership, as does general good will toward all of humanity, never mind the ecosystems / biosphere.
So yes, I'm opposed to that kind of irrational, mean-spirited, xenophobic populism that is all the rage these days.
Populism today has a clear connotation of manipulation and amplification of peoples' worst instincts (misdirected fear, anger, willful ignorance, denialism, inappropriate blaming, emphasis of insider/outsider division based on some arbitrary criterion, fanning flames of conflict etc etc).
---------------
We don't need: When the going gets tough, the bullies get going. That's chimpanzee-pack level. We don't need MMA/WWE-style politics where the loudest roared bullshit wins support. That's not going to help anything. That's just a symptom of the shitstorm, not a cure.
When the going gets tough, we need knowledgeable and compassionate leadership to implement rational, fair changes and solutions that have a chance of effectively improving the situation. We need eloquence and charisma in service of informed, decent, intelligent ideas that could actually work in our complex world.
1) Well there are many countries in the world where the top 10% or so live a "western standards" wealthy or upper middle class lifestyle and 90% essentially rot in a subsistence black/grey market economy. That pattern is already well established. And the top 10% there do fine, (as long as they can afford adequate security measures to fend off the moderately embittered opportunistic underclass.)
I'm not saying that's good. I'm just saying it's a historically well established pattern that still persists in various places. Not implausible at all.
4) The voting power of the disenfranchised will eventually trump (no pun intended heh heh) that of the shrinking number of beneficiaries of the automation. The results of this are unpredictable but could polarize to: a) Uber-Trump Moron Populist Fascism lays waste to pretty much everything, or b) Sensible leader puts in good universal basic income program (and then is probably assassinated, unfortunately - see a)
5) Or of course, there WAS the French Revolution kind of scenario. Of course, after having decapitated the problem, you have to figure out what to do next.
The web as a whole is event larger, and with only a tweak to search tech, an unmoderated "virtual, distributed" facebook-like browsing pattern could be supported.
So how are you going to police the whole web? And why would you even try?
I really think people need to get thicker skins, and accept that the full, ugly range of human behaviour exists. People who don't want to be able to see it all should start becoming patrons of some new filtered-search service that works FOR THEM, according to their own specified standards.
The onus should be on the viewer not to look.
And of course, posters of illegal material should also be identified and prosecuted.
Middleware is just middleware, a platform, and new ones can spring up like hydras heads in the Internet infrastructure as a whole. Middleware is not the problem. It's a mistake to focus on that. That leads to totalitarian censorship.
by a billion or so users.
Yes. AI is being used to try to police/filter it quickly these days (because with a billion possible contributors, how else could you do it fast?) but the challenge for that AI to recognize really bad content (the only kind that should be filtered, right?) is very complex. Human moderators often can't even do it reliably.
So realistically, it has to be a system that relies partly on human users to flag bad content for immediate review. Then a large team of moderators needs to check and decide, which in some cases requires knowledge? Is that a deepfake first-person-shooter video game? Is it real? Is it a movie scence? Does our policy care about these distinctions?
That's what takes nearly an hour to get reliable blocking going.
We should not be surprised.
Or maybe I'm wrong. Put yourself in Zuckerberg's shoes and tell me with some technical detail how YOU would solve this problem. I'm not being sarcastic. Realistic ideas well described welcome here. Ideas with gaping holes in feasibility with current tech not useful though.
That we have to protect all technology against psychopathic super-assholes.
what category of product or service or influence campaign etc the ad is about.
That's some pretty heavy AI, unless they ask advertisers to self-disclose in some official form what kind of ad this is, which seems impractical to enforce.
by positive targeting on stereotyped "white people" interests, I suppose.
I guess you have to ban all culture-indicating tags from housing and similar opportunity (e,g, employment) ad targeting.
How do you allow advertisers to target, say, latin-dancing party ads to a market segment "interested in Hispanic culture" without permitting the kind of discrimination talked about? Do you just not allow the NOT operator in the targeting? or don't allow NOT in front of a set of protected categories?
Maybe it's just that simple. Thoughts, all you clever data types?
and taking "back" control of the mobile platform.
Maybe it would have had to rebrand it. e.g. Orafix, Mava (mobile java),..., what have you.
Nothing stopped Oracle from pulling a CyanogenMod move, but backed by Oracle's money.
Nothing except lack of willpower, imagination, and skill.
Wait, did I say per working capita?
No, I believe I didn't.
I said per capita, which means per person with a head.
USA 833 moter vehicles per 1000 people.
China 173 motor vehicles per 1000 people
I know how numbers work pretty well.
Per capita is the only fair way to assess GHG emissions policy and progress.
Conventional diesel buses are already much more efficient than cars.
A bus emits approximately 1/6 the CO2 per passenger kilometer as does a single-occupant ICE car.
From that perspective, electrifying cars should give more ghg reductions, in a population where most people are using cars.
In China, the car use is probably not that high yet, as a proportion of population, so it makes sense that busses are having a bigger oil displacement there. That is not replicable to other car-centric places like USA.
really important point.
As is common, this accident seems to have been caused by a conjunction of a design flaw and various human process flaws.
"One current FAA safety engineer said that every time the pilots on the Lion Air flight reset the (trim) switches on their control columns to pull the nose back up, MCAS would (reset its 0 degree reference and) have kicked in again and “allowed new increments of 2.5 degrees.”
“So once they pushed a couple of times, they were at full stop,” meaning at the full extent of the tail swivel, he said.
So in summary a system FAA-certified on the basis of being able to adjust nose-down trim by 0.6 degrees could actually, (after a few cycles of the pilot correcting it a little bit with trim up), command full nose-down trim, about 5 or 6 degrees tailplane tilt.
All of this relying on input from a single angle-of-attack sensor. Get this, the plane has two such sensors, one on each side, but the MCAS only uses input from one of them!!! ! !! ! ! ! ! What the hell? If you use two of them, then your software can check if they diverge, and disable systems relying on the input, and warn the pilots. That is some criminally bad development cost saving judgement there.
you just weeded out the ones that are prone to fail early.
You're now flying in one that's statistically more prone to failing slightly later. And it's now slightly later.
from the FAA, right now, before the black box analysis, would be to say that:
The Boeing 737 Max 8 is "probably, most of the time" still airworthy, as far as we know.
[cynicism]But that doesn't keep the sales flowing in, so what's the harm in a little time-shifting of the truth? [end cynicism]
IMHO there is a structural problem in many "hybrid" (non developer-run) companies where non-technical, unskilled-technical (migrated to mgmt), or vaguely historically technical management need to see the technical staff "productively mirroring" their own highly social and highly multi-tasking management jobs.
The incentives are mostly aligned with willingness to butterfly from one urgent task/project/proposal to another. Aligned with making consistent, predicted, scheduled, budget (no surprises, no innovation) progress in one week intervals, even if one is only making low-hanging-fruit progress on easy problems, and avoiding the hard problems because they would get you in trouble with management quickly.
In these organizations, trying to get into and maintain flow state to work with sustained focus on actually hard technical problems (be it architecture, thoughtful requirements analysis, thorough survey of prior work, good design, good quality well-factored coding, etc) is usually punished. Would-be practitioners of focus are thought of as loners, mavericks, not communicating, missing administrative tasks, etc etc.
Even though those people, if given a conducive work environment, would almost always be the most productive contributors to sustainable and innovative s/w development. And if they had enough freedom from interruption, gratuitous extra meetings on random side topics, etc, then they could also choose their own natural focus-break times.
Are there any software workplaces like those described in my last paragraph? Somewhere over across the wide river?