"It doesn't matter what a feminist says; critics will always not like it."
Unless properly measured and sufficient evidence thrown, any "women this while men that" *is* sexist by its very definition.
Even more: even if properly measured and sufficient evidence thrown, any "women this while men that, so let's be it" without an analysis on why it is so and why it shouldn't need to be changed is still sexist.
"The less successful ones had more of a concept of forcing a new paradigm on the workplace, with a sort of 40 hours only outlook, no lost vacation, sick leave balance of 0."
Oh! but, but... what a f* bastards! 40 hours/week, and stay by the hiring contract they signed... how they dared!!!???
"I think that in the end they might make it harder for themselves to recruit talent."
Yes, you're rationale is quite convincing. But don't fool yourself thinking this haven't been already thought of by them. Two possible outcomes I see:
1) They plan on reaching non-poaching/salary rates' collusion agreements with their competitors (not as if it was the first time).
2) For the ones they really like, they won't negotiate, but they'll throw "clues": -I won't negotiate but, of course, before I make you an offer you please tell me what would you accept -Let's see... 150K -OK. Now, as per my company policies I'll "come up with an offer that we think is fair". Please remember this is our one and only offer: 150K.
"If the ability to negotiate aggressively is not a talent required for the job, there is no reason why someone who negotiates well should get a higher salary."
Maybe you are right.
But how funny that the "solution" this CEO proposes to avoid negotiation is "I'll make an offer and you'll take it" instead of, say, "you'll make an offer and I'll take it".
"What does a negotiation accomplish?"
Last I reviewed, a hiring contract is still a contract. You know, that stuff about "meeting of the minds", "consensus ad idem", "mutual assent"...
And this specific kind of contracts are basically about exchanging labour for *money*. It's difficult to reach that "meeting of the minds", "consensus ad idem", "mutual assent"... about the exchange of labour for money when one party is void to bring the issue about money onto the table. Oh! and how convenient for the hirer while, at the same time, inconvenient for the hiree.
"I'm not THAT smart, but even I know that the auto-flown plane would need to be programmed to ditch into the least populated areas possible."
No, certainly you are not THAT smart if you think that this is the only needed input and that you can set its value well in advance.
"Without some data to consider, it could very well be both. If the 80% pilot error can be brought down to a 70% automation error"
That's right. But if you don't consider the data it might be very well that a party will push you to a no-return path for their own profit. What if the promised 70% automation error ends up being 85% (oh, who could have expected that!) but the companies still save the half a million in salaries and the extra cargo space?
"That's the interesting angle. Which would you rather have on the stick? A computer or an inexperienced human?"
A very interesting one, in fact, because as you automate more and more and the pilot is seen less and less the mighty hero he used to be (along with the pay range he used to have), the less experienced humans you'll have on the stick so you are risking reaching full automation in a self-fulfilling prophecy kind of way.
In the end, you'll have accidents due to automation as you have accidents by human error. The (difficult) point is to set which of both scenarios render a better overall outcome.
Readings don't make sense only when they come from defective equipment. These readings made perfect sense since they were demonstrated to be correct: they were faithfully transmiting what was really happening.
They were only percieved as nonsensical by the guy that, thousands of miles away, wasn't able to think on that specific scenario where they made perfect sense coupled with a cabin crew that was trained to be overly confident on the ability of that distant guy to properly assess real time flight environment.
"The pilot in a modern plane like the A320 doesn't manage, and isn't even allowed to manage, the finer details of flying. The plane does that. The pilot is only there for general overarching strategy. The plane makes it happen."
These all mean that in this case the system (composed both of the machine and the pilot) worked exactly and beatifully as designed, with the plane flying and the pilot thinking for the best result. Now, if you take away the pilot, who's going to do the thinking?
"Wouldn't an automated system have "ditch" sites programmed in?"
Yes, if it's programmed to do so.
"It wouldn't need to be smart enough to identify a river - just know that the river is there."
No, the one that needed to be smart enough is the programer, and if the programer is that smart he'd immediately ditch it as an option since he wouldn't now in advance if the river is frozen, deep enough, sustained traffic at the moment of the accident, etc.
The fact is that the automated system would need to be as fast as the human to evaluate the situation (very easy), had the ability to process as much environmental information as the human to input the evaluation process (very hard) and as clever as the human to think out of the box (impossible in the next few decades).
No. Taking off the pilot from the cockpit is as of now not in the slightest about increasing security but about how higher a dead toll looks acceptable for better financial results.
"Why can't a computer figure out if there are no runways to fall back on emergency options like roads or rivers?"
It can, of course. But then, you need real time awareness of *all* your environment, not only the track you are expected to follow, and the ability to discern which paths are unavoidingly unpracticable and those where an emergency call can clear the path, and 100% perfect instrumentation so the system doesn't take a frozen pitot tube for a stalled plane.
And a programing team with enough experience to know in advance about all the weird things that can happen during a flight... in a time were there's no access to experienced pilots because you retired all them long ago (you don't think you'll make the new system 100% fool proof overnight, do you?).
And all that just to cope with a single non-standard situation, now you still to develop the system to cope with the other hundreds ones.
"Hard, I'm sure, but far from the hardest part of flying a plane."
Are you sure? What you are really asking for is for the planes to achieve orders of magnitude higher level of environment awareness coupled to orders of magnitud harder artificial intelligence.
"The simple facts are, accident rates on planes have gone way down as the amount of automation has gone way up"
Correction: accident rates of planes have gone way down as the amount of *pilot-supporting* automation and instrumentation has gone way up.
"A computer will always follow the checklist for the right response to a given problem"
Yes, of course, because the last, say, 40 years have teach us that software is always bug-free and that blind confidence on instrumentation about flight status hasn't ever lead to bad assessment which, in turn, would lead to the wrong checklist to be followed.
"80% figure above, that means 20% of accidents are still due to weather, equipment failure, etc. Those same conditions exist even if pilots are no longer flying."
Humm, no. At the very least, taking out the pilot from the cockpit means more equipment so, at current equipment failure rates, will mean more accidents due to that reason. And this at current failure rates. One can guess that as of currently, the things the pilot is still doing are the ones that are difficult to take out of her and thus, more prone to higher failure rates -and once you start thinking on the economic incentive to outsource those most dificult parts to cheap third world companies the 80% figure above starts looking too much an apples to oranges comparation case.
Yes. Now, what of that percentage do you think is due to unproper instruments' ergonomy? And then, why do you think automation to substitute the pilot handling, or better yet, the programer doing it whose life is not at stake, will do any better?
"A better way to say this might be: the effort the manufacturer puts into security will be equal to the perceived risk. Since my garage door is already easy to open with a crowbar, the manufacturer might perceive that the risk of some wireless vulnerability is no worse than the risk I am already accepting"
Half true.
The effort a manufacturer puts into security will be equal to the percieved risk... to them, not to you.
People buy because of features, not because security and there's basically no liability for the vendor in case of (even hugh) security problems.
So, imagine for a minute you are a vendor: each day it passes without puting your product in the market means expenses you have to affront without sales to compensate for them and risk of a competitor to reach the market before you do.
Naturally you focus in the feature set and launch as soon as it looks like more or less being there, everything else be damned. After all, the worst that can come from a security problem with my gadget will be bad PR *IF AND ONLY IF* my gadget successes in the market, while the risk of reaching the market too late or at too a high cost, is bankrupcy, so let's focus on reaching the market first, which is already difficult enough.
"Linux and FreeBSD grew at the same exponential rate for a decade+, after which Linux network effects took over. There's a bunch of advantages in terms of code contribution to using the BSD license, but there's a bunch of advantages in terms of marketing to use the GPL"
No, I don't think it's a matter of code contribution vs marketing, but code contribution vs code contribution and what you say, fits my hypothesis:
At the very begining, GPL and BSD licenses look very much the same in terms of code contribution to the "free" branch of the project at hand in that, no matter what, there will not be external contributions. You see, most "minor" GPL projects have just one/two main developers and that's it.
But when a project success, the incentive for external code grows and here comes the difference: companies (and some individual developers) will always prefer BSD licensing in the hope of taking the cake and eating it too by means of closed sourced, let's say, extensions. This naturally leads to a more or less stable "core" and, if lucky, some closed sourced projects taking advantage of it.
GPL, on the other hand, may have a higher "point of no return" in terms of companies wanting to produce code to the project but once reached, the incentive is just too big to ignore and then, GPL will make sure that all those contributions add exponentially to the base code value.
This neatly explains the development of BSD's vs Linux as well as other projects.
"If you get hardware that supports FreeBSD there's a good chance the driver was written by the hardware company themselves. Everything 'just works' (if it's supported). ZFS is a great file system and having ZFS on Root with each update a true 'snapshot' is also great."
And still, hasn't grown to critical mass in, what? almost 20 years? Maybe a licensing decision has something to do with it.
"Where I work, we're trying to spin up amazon infrastructure because our IT department can't deliver reliable servers"
You most probably won't sucess becose your problem seems not to be your IT guys but lousy management and you won't get rid of them moving to the cloud.
"The average reviewer has nary a clue about cryptography"
The average user's review is of no value anywhere.
Just go Amazon and check the reviews of any item at hand, say, a french skillet. They are "physical" items, the kind of things people have been using for ages and still, half of the reviews will be more or less like "the item arrived this morning; I haven't used it yet, it's still in its package so I didn't even see it, but I'll give it five stars because, hey, I'm happy".
"while cockpit video cameras may help determine the cause of some crashes, there are plenty of recent examples where it wouldn't help."
While logging data from left engine working regime may help determine the cause of some crashes, there are plenty of recent examples where it wouldn't help.
"I thought we got rid of slavery a century or two ago."
Well, there are in fact professions that already have special labour conditions that would be considered border to slavery were not for the fact that it is socially accepted that those constrains should be in place. Police and military come immediately to mind.
AFAIK, at least in EU, flight controllers already come under this designation due to national security considerations and maybe pilots are also of this kind. If they are not, they could be in the future.
"Why exactly are data recorders antiquated? I mean the concept, not a specific device."
They aren't, in the slightest.
"Pilots hate this idea because it will show they are human."
Exactly that. You, of course, already know it. You want non-instrumental data recorders because not everything that happens in a flight is measured by instruments. You already have voice recordings and, as you say, they are triggered in specific moments and events. Video could be the same as long as it offers more relevant information than other techniques -and it sure can do, or else we wouldn't have developed TV, radio being enough.
"They make jokes, complain about work, talk about their weekends, etc."
Exactly. They *already* do this despite having their conversations saved so, what's the added problem about video imaginery?
If it can be used to increase flight safety it must be put in place. Full stop.
It's only that it have to cope with the fact that part of the "machinery" is human and it has, therefore, to cope with "human envelope conditions" which include there must be moments for relax and moments for intimacy, just like is already happening with the currently in place data logging facilities.
Slashdot Mirror
"It doesn't matter what a feminist says; critics will always not like it."
Unless properly measured and sufficient evidence thrown, any "women this while men that" *is* sexist by its very definition.
Even more: even if properly measured and sufficient evidence thrown, any "women this while men that, so let's be it" without an analysis on why it is so and why it shouldn't need to be changed is still sexist.
"The less successful ones had more of a concept of forcing a new paradigm on the workplace, with a sort of 40 hours only outlook, no lost vacation, sick leave balance of 0."
Oh! but, but... what a f* bastards! 40 hours/week, and stay by the hiring contract they signed... how they dared!!!???
What will come next? Abolishing slavery!!!???
"I think that in the end they might make it harder for themselves to recruit talent."
Yes, you're rationale is quite convincing. But don't fool yourself thinking this haven't been already thought of by them. Two possible outcomes I see:
1) They plan on reaching non-poaching/salary rates' collusion agreements with their competitors (not as if it was the first time).
2) For the ones they really like, they won't negotiate, but they'll throw "clues":
-I won't negotiate but, of course, before I make you an offer you please tell me what would you accept
-Let's see... 150K
-OK. Now, as per my company policies I'll "come up with an offer that we think is fair". Please remember this is our one and only offer: 150K.
"If the ability to negotiate aggressively is not a talent required for the job, there is no reason why someone who negotiates well should get a higher salary."
Maybe you are right.
But how funny that the "solution" this CEO proposes to avoid negotiation is "I'll make an offer and you'll take it" instead of, say, "you'll make an offer and I'll take it".
"What does a negotiation accomplish?"
Last I reviewed, a hiring contract is still a contract. You know, that stuff about "meeting of the minds", "consensus ad idem", "mutual assent"...
And this specific kind of contracts are basically about exchanging labour for *money*. It's difficult to reach that "meeting of the minds", "consensus ad idem", "mutual assent"... about the exchange of labour for money when one party is void to bring the issue about money onto the table. Oh! and how convenient for the hirer while, at the same time, inconvenient for the hiree.
In Soviet Russia!
"I'm not THAT smart, but even I know that the auto-flown plane would need to be programmed to ditch into the least populated areas possible."
No, certainly you are not THAT smart if you think that this is the only needed input and that you can set its value well in advance.
"Without some data to consider, it could very well be both. If the 80% pilot error can be brought down to a 70% automation error"
That's right. But if you don't consider the data it might be very well that a party will push you to a no-return path for their own profit. What if the promised 70% automation error ends up being 85% (oh, who could have expected that!) but the companies still save the half a million in salaries and the extra cargo space?
"That's the interesting angle. Which would you rather have on the stick? A computer or an inexperienced human?"
A very interesting one, in fact, because as you automate more and more and the pilot is seen less and less the mighty hero he used to be (along with the pay range he used to have), the less experienced humans you'll have on the stick so you are risking reaching full automation in a self-fulfilling prophecy kind of way.
In the end, you'll have accidents due to automation as you have accidents by human error. The (difficult) point is to set which of both scenarios render a better overall outcome.
"I can't even imagine how it must feel (not feel?) to hand-fly something as fast and heavy as an airliner *with no stick feedback!*"
No, you can't. You visually fly a Cessna.
"the readings didnt make any sense."
Wrong, wrong, utterly wrong.
Readings don't make sense only when they come from defective equipment. These readings made perfect sense since they were demonstrated to be correct: they were faithfully transmiting what was really happening.
They were only percieved as nonsensical by the guy that, thousands of miles away, wasn't able to think on that specific scenario where they made perfect sense coupled with a cabin crew that was trained to be overly confident on the ability of that distant guy to properly assess real time flight environment.
"The pilot in a modern plane like the A320 doesn't manage, and isn't even allowed to manage, the finer details of flying. The plane does that. The pilot is only there for general overarching strategy. The plane makes it happen."
These all mean that in this case the system (composed both of the machine and the pilot) worked exactly and beatifully as designed, with the plane flying and the pilot thinking for the best result. Now, if you take away the pilot, who's going to do the thinking?
"Wouldn't an automated system have "ditch" sites programmed in?"
Yes, if it's programmed to do so.
"It wouldn't need to be smart enough to identify a river - just know that the river is there."
No, the one that needed to be smart enough is the programer, and if the programer is that smart he'd immediately ditch it as an option since he wouldn't now in advance if the river is frozen, deep enough, sustained traffic at the moment of the accident, etc.
The fact is that the automated system would need to be as fast as the human to evaluate the situation (very easy), had the ability to process as much environmental information as the human to input the evaluation process (very hard) and as clever as the human to think out of the box (impossible in the next few decades).
No. Taking off the pilot from the cockpit is as of now not in the slightest about increasing security but about how higher a dead toll looks acceptable for better financial results.
"Why can't a computer figure out if there are no runways to fall back on emergency options like roads or rivers?"
It can, of course. But then, you need real time awareness of *all* your environment, not only the track you are expected to follow, and the ability to discern which paths are unavoidingly unpracticable and those where an emergency call can clear the path, and 100% perfect instrumentation so the system doesn't take a frozen pitot tube for a stalled plane.
And a programing team with enough experience to know in advance about all the weird things that can happen during a flight... in a time were there's no access to experienced pilots because you retired all them long ago (you don't think you'll make the new system 100% fool proof overnight, do you?).
And all that just to cope with a single non-standard situation, now you still to develop the system to cope with the other hundreds ones.
"Hard, I'm sure, but far from the hardest part of flying a plane."
Are you sure? What you are really asking for is for the planes to achieve orders of magnitude higher level of environment awareness coupled to orders of magnitud harder artificial intelligence.
It doesn't look like peanuts to me.
"The simple facts are, accident rates on planes have gone way down as the amount of automation has gone way up"
Correction: accident rates of planes have gone way down as the amount of *pilot-supporting* automation and instrumentation has gone way up.
"A computer will always follow the checklist for the right response to a given problem"
Yes, of course, because the last, say, 40 years have teach us that software is always bug-free and that blind confidence on instrumentation about flight status hasn't ever lead to bad assessment which, in turn, would lead to the wrong checklist to be followed.
Or are they?
"80% figure above, that means 20% of accidents are still due to weather, equipment failure, etc. Those same conditions exist even if pilots are no longer flying."
Humm, no. At the very least, taking out the pilot from the cockpit means more equipment so, at current equipment failure rates, will mean more accidents due to that reason. And this at current failure rates. One can guess that as of currently, the things the pilot is still doing are the ones that are difficult to take out of her and thus, more prone to higher failure rates -and once you start thinking on the economic incentive to outsource those most dificult parts to cheap third world companies the 80% figure above starts looking too much an apples to oranges comparation case.
"80% of accidents are pilot error."
Yes. Now, what of that percentage do you think is due to unproper instruments' ergonomy? And then, why do you think automation to substitute the pilot handling, or better yet, the programer doing it whose life is not at stake, will do any better?
"A better way to say this might be: the effort the manufacturer puts into security will be equal to the perceived risk. Since my garage door is already easy to open with a crowbar, the manufacturer might perceive that the risk of some wireless vulnerability is no worse than the risk I am already accepting"
Half true.
The effort a manufacturer puts into security will be equal to the percieved risk... to them, not to you.
People buy because of features, not because security and there's basically no liability for the vendor in case of (even hugh) security problems.
So, imagine for a minute you are a vendor: each day it passes without puting your product in the market means expenses you have to affront without sales to compensate for them and risk of a competitor to reach the market before you do.
Naturally you focus in the feature set and launch as soon as it looks like more or less being there, everything else be damned. After all, the worst that can come from a security problem with my gadget will be bad PR *IF AND ONLY IF* my gadget successes in the market, while the risk of reaching the market too late or at too a high cost, is bankrupcy, so let's focus on reaching the market first, which is already difficult enough.
"Linux and FreeBSD grew at the same exponential rate for a decade+, after which Linux network effects took over. There's a bunch of advantages in terms of code contribution to using the BSD license, but there's a bunch of advantages in terms of marketing to use the GPL"
No, I don't think it's a matter of code contribution vs marketing, but code contribution vs code contribution and what you say, fits my hypothesis:
At the very begining, GPL and BSD licenses look very much the same in terms of code contribution to the "free" branch of the project at hand in that, no matter what, there will not be external contributions. You see, most "minor" GPL projects have just one/two main developers and that's it.
But when a project success, the incentive for external code grows and here comes the difference: companies (and some individual developers) will always prefer BSD licensing in the hope of taking the cake and eating it too by means of closed sourced, let's say, extensions. This naturally leads to a more or less stable "core" and, if lucky, some closed sourced projects taking advantage of it.
GPL, on the other hand, may have a higher "point of no return" in terms of companies wanting to produce code to the project but once reached, the incentive is just too big to ignore and then, GPL will make sure that all those contributions add exponentially to the base code value.
This neatly explains the development of BSD's vs Linux as well as other projects.
"If you get hardware that supports FreeBSD there's a good chance the driver was written by the hardware company themselves. Everything 'just works' (if it's supported). ZFS is a great file system and having ZFS on Root with each update a true 'snapshot' is also great."
And still, hasn't grown to critical mass in, what? almost 20 years? Maybe a licensing decision has something to do with it.
"Where I work, we're trying to spin up amazon infrastructure because our IT department can't deliver reliable servers"
You most probably won't sucess becose your problem seems not to be your IT guys but lousy management and you won't get rid of them moving to the cloud.
"The average reviewer has nary a clue about cryptography"
The average user's review is of no value anywhere.
Just go Amazon and check the reviews of any item at hand, say, a french skillet. They are "physical" items, the kind of things people have been using for ages and still, half of the reviews will be more or less like "the item arrived this morning; I haven't used it yet, it's still in its package so I didn't even see it, but I'll give it five stars because, hey, I'm happy".
"You can quit both the police and the military"
As you say, you can quit... except when you can't. They are also restricted rights (at least) on strike, unionize and free speech.
"The draft IS slavery, IMHO."
Well, that's exactly what I said, isn't it? It's only it is a kind of socially accepted one.
"What's the point of fighting to defend "freedom" when you don't have it?"
I think Gene Hackman's role on Crimson Tide already gave the perfect answer: "We're here to preserve democracy -not to practice it."
"while cockpit video cameras may help determine the cause of some crashes, there are plenty of recent examples where it wouldn't help."
While logging data from left engine working regime may help determine the cause of some crashes, there are plenty of recent examples where it wouldn't help.
And your point is?
"I thought we got rid of slavery a century or two ago."
Well, there are in fact professions that already have special labour conditions that would be considered border to slavery were not for the fact that it is socially accepted that those constrains should be in place. Police and military come immediately to mind.
AFAIK, at least in EU, flight controllers already come under this designation due to national security considerations and maybe pilots are also of this kind. If they are not, they could be in the future.
"Why exactly are data recorders antiquated? I mean the concept, not a specific device."
They aren't, in the slightest.
"Pilots hate this idea because it will show they are human."
Exactly that. You, of course, already know it. You want non-instrumental data recorders because not everything that happens in a flight is measured by instruments. You already have voice recordings and, as you say, they are triggered in specific moments and events. Video could be the same as long as it offers more relevant information than other techniques -and it sure can do, or else we wouldn't have developed TV, radio being enough.
"They make jokes, complain about work, talk about their weekends, etc."
Exactly. They *already* do this despite having their conversations saved so, what's the added problem about video imaginery?
If it can be used to increase flight safety it must be put in place. Full stop.
It's only that it have to cope with the fact that part of the "machinery" is human and it has, therefore, to cope with "human envelope conditions" which include there must be moments for relax and moments for intimacy, just like is already happening with the currently in place data logging facilities.