Imagine that we re-brand the 'expensive robot' as a 'turnkey managed surgery solution' or a 'patient outcome appliance' and ask your IT minions about whether HQ likes capital costs or expensive human resources less...
Have you tried cutting it open and suturing it shut again? If not, I'll hang up to keep my metrics high enough to avoid getting reamed out by my supervisor in order to allow you to try that. If so, please bleed out on hold while I stall on escalating you to a tier 2 support rep, much less to engineering...
And to get that we'd need to replace all that laggy legacy fiber with a series of tubes pumped down as close to vacuum as possible, run perfectly straight between nodes. That might get costly.
The interesting question is whether allowing surgeons to telecommute makes you more or less likely to be chopped open by some hack.
On the minus side, we've certainly all experienced the fact that when the call center gets off shored it is because nobody gives a damn and a cheaper labor force can do a bad job for less. On the other hand, there's the old quip about 'what do you call a med school grad in the bottom ten percent of his class? Doctor.' and the fact that only having access to on-site talent means that you are substantially at the mercy of the quality of whatever medical experts happen to be where you need treatment, which varies enormously between wealthy medical hubs and low density and/or really poor backwaters.
Are you better off with whatever surgeon lives within commuting distance, because they can't hire any cheap 'n cheerful bottom feeder within your light cone? Or are you better off with any surgeon in your light cone; because they aren't limited to recruiting only the second-stringers willing to settle for the lousy location/lower salary/inferior institutional prestige?
Aside from its puff-piece value(here we are talking about it, despite the fact that academics and DARPA have been puttering around with speech recognition since the 70s if not earlier, and a number of companies-that-aren't-facebook are already aggressively attempting to give away voice-assistant products in order to score that juicy data, and facebook has nothing except its CEO's new year's resolution...) it may also be a matter of cultural fit.
I'd be pretty surprised if he does his own housecleaning, when their are contractors for that; but having 'the help' actually embedded in your home is a bit weird and intrusive unless they are amazingly skilled at being deferential and you are somewhat used to the idea(all the subtle joy of walking back into your hotel room and noticing that the cleaners moved things around a bit while you were out, except all the time and at home). There's also the probable level of trust that someone who appears to dislike intrusions on his own privacy and built a fortune by intruding on that of others would have for comparatively low-paid lackeys with privileged physical access to his house.
Some dodgy speech-to-text system bolted onto a pile of scripts is certainly going to do a vastly worse job than a human would; but it won't be nearly as noticeable or likely to sell pictures of you snorting lines off your iphone to the tabloids.
The trouble with athelete-research is that it is nominally wicked and immoral, so everybody carefully avoids keeping useful records about the experiments.
It's a real pity, there would probably be some very interesting stuff there if the dataset were available. At least the Purity of Sport remains unsullied, so there's that.
So long as it isn't a germline modification there will probably be some enthusiasm for even high risk treatments. DMD reliably cripples and then kills you in a time long enough to be an agonizing ordeal and short enough that you can be said to have died tragically young. When Plan A sucks just that much, even serious side effects start to look pretty good.
If it is a germline modification, or if somebody can't keep their genetic engineering virus from hopping around, a lot more caution will be warranted because removing a change committed to the wild will be a real problem; but this is a nasty disease with a lousy prognosis so if the risks are confined to the patients they will likely be tolerated pretty well.
That's actually what ended up happening to thalidomide. The FDA's refusal to approve a zesty teratogen for marketing to pregnant women was roundly vindicated in hindsight; but it has some niche applications in situations where the risks are lower and the diseases more serious.
Any situation where you end up having to trust the clients is more or less entirely hopeless. Team DRM has spent years and a great deal of money and effort demonstrating this; and they are working from a much stronger legal position that twitter is; the twits can deny API access but have effectively zero leverage over basic web scraping tools, where the DRM people can usually at least ensure that noncompliant clients are theoretically illegal.
Perhaps more dangerously(for the politicians, not that this is a bad thing) it's a largely false hope. If some pesky bunch of idealists is running a collection of deleted tweets just because it warms their hearts can you really believe that your opponent's campaign isn't going to be poring over your history with substantially greater motivation?
It also wouldn't be a huge surprise if experience suggested that the bad-PR tweets that actually gain traction are typically the ones that people capture and distribute in awful-mobile-app-screenshot form, rather than the systematic but mostly uninteresting automated collections of every tweet some twit attempted to untweet.
If, in practice, the juiciest accidental honesty is already being captured manually you just end up looking like you have something to hide by selectively denying API access. Plus, in absence of a suitably robust search tool, a flood of mostly uninteresting noise is a fair deterrent to all but the most enthusiastic investigators(who are probably already operating less visible equivalents of the same tool for 'opposition research' purposes tied to various campaigns).
If, god help us, twitter is still relevant in a couple of decades this may come back to bite them, since the automatic logs of what future presidential candidates said while campaigning for their entry-level positions in local markets today will probably be available; but in terms of immediate PR handling I suspect that a lot of the good stuff gets scraped up by humans without automated assistance.
Organizationally, the school district might actually be pretty logical.
These sorts of things aren't supposed to be the school's problem; but it seems hard to argue with the idea that they have become the school's problem anyway, since the readiness of the students they have to work with is being seriously affected. When something becomes your problem, having the option to deal with it is ugly; but often much less painful than waiting hopefully for whoever is supposed to fix the root cause of the problem to sort things out so that you can do your work.
They are using the NW.js javascript environment, packaged in their executable, to provide javascript interpretation without the browser limitations; but according to the article it is just being used in social engineering attacks at present, not coupled with an exploit.
Presumably having the guts of the application in javascript will make the developer's life easier if he wants to produce a version for another platform and nothing prevents this being used as a payload for some other exploit that allows the attacker to execute something for you; but the current version appears to be all payload and not much delivery mechanism.
And Java's predecessors were old enough to drink when Java was laid out. In fact, given that computers used to be a great deal rarer than mathematicians, it may well be argued that we've had architecture-independent programs longer than we've had architectures and certainly longer than we've had OSes.
Being capable of more dynamic range is certainly good; my concern is just what they'll do to source material that doesn't have it. Since the limitations in dynamic range usually show as uniform blown-out or shadowed areas, with no detail to play with, I can only guess that they'll be taking the areas of the image that do have detail to work with(but were also likely carefully tweaked to look good within the constrained range expected) and smear them out over the wider range in order to look more-contrasty-and-vibrant. That just isn't going to end well.
The HDR capabilities themselves are all well and good, just as greater color depth is a good thing; but I'd be similarly concerned if somebody' s TV purported to 'improve' 8 and 16 bit color into 24/32 bit color.
I'm always up for more resolution; but my eyes are bleeding just imagining what fake-HDR effects are going to look like. More dynamic range is useful; but what horrors are they going to invoke when faking data not specified by the source material to get a 'Best Buy Brite(tm)!' effect that 8 out of 10 focus group participants agreed looked brighter and more vivid than the competitor's image under retail conditions?
The major trick, in humans, is that (so far as we know) there aren't a bunch of conveniently defined castes to work with. With these ants there ended up being a single fairly neat switch that the researchers could apply to get an entire cascade of behavioral changes because that entire cascade of changes is already one that the ants have evolved to employ. We know of similar things in humans(the most dramatic probably being the developmental trajectory into either male or female development from an originally very similar state); but we don't seem to have a 'service-sector worker', or 'pre-med' caste, despite the frantic efforts of parents to the contrary.
This isn't to say that applications in humans would be impossible; but there is a reason why we've been medicating for generations(arguably, at least since we developed brewing, possibly earlier in populations whose hunter-gathering regions had decent psychoactives naturally available): because we rarely even know what we want, except in the vaguest terms, much less how to get there.
Since most of these distractions are probably smartphone related, and contemporary smartphones tend to have pretty substantial activity-logging capabilities and sensor packages, it would be technically feasible to get an 'aircraft blackbox' style snapshot of what was going on at the time of the injury.
Anyone know if you could get away with an insurance policy that demands such data, in the event of a claim, and then attempts to deny coverage for any costs deemed to have been incurred because of negligent distraction? Just think of how many claims you could deny with the right data...
The usual objective behind various rights and protections afforded to renters is addressing the fact that settling a contractual dispute while unexpectedly homeless is a...rather lopsided...process.
It's not news that contract law just doesn't work as smoothly when the parties involved are of vastly different power; and resolving matters through a civil proceeding can be problematic when there is particularly urgent time pressure. In situations involving people who rent their sole residence; both of these conditions tend to hold.
It's one thing to, say, go through a hassle with Dell because your computer was DOA. It's a bit more problematic to have a landlord stalling you while the heating system isn't working; or a sewage leak is making it impossible to use anything involving water(not my most entertaining college experience...)
Once you have basic logistics covered it becomes much easier to take up a property issue as just another civil dispute; but the thing about being a renter is that most of your disputes tend to start because certain basic logistics are suddenly not covered.
I wouldn't have modded it 'flamebait'; but it does fall into the relatively unhelpful category of being overbroad(if you go state-by-state, the degree to which US education is totally fucked varies quite widely); and it also ignores the important fact that the DOE isn't actively changing the state of CS education here; but merely signalling an unwillingness to get tough on trying to improve it.
They aren't saying "Yeah, you can fire those uppity CS people and replace them with guidance counselors who will teach the students how to empathize with the computes"; as much as they are saying "At present, effectively zero high schoolers are taking CS courses in some entire states; this is such a disaster that we will look the other way if you need to use dubiously suitable teachers to make this number better."
That's still not a good thing, since it's likely to result in a number of students getting CS-in-name-only rather than no CS, which will take more time but be about as useful; but it's not as though the DOE is saying "Yeah, things are great, go ahead and cut costs!"; but looking at the utter clusterfuck that exists in some parts of the country and informing them that they'd rather see the appearance of improvement, even if the schools don't have the faculty resources to do it properly, than more of the same.
Civics might be a better fit for software project management or software engineering, if only in the weak sense that someone with knowledge of how the noble theory behind various forms of government goes through a long, byzantine, often actively counterproductive, development process and finally ends up as a sprawling, sclerotic, morass of scope creep and dirty hacks should have an advantage in understanding how software projects tend to do the same.
Not much related to CS, of course; but that's the best I can think of.
Probably not from one of the well-reputed research universities where those who can do both; and even those who would prefer to be doing pretend to teach; but I'm not sure offhand.
Out of curiosity, do you know if your school was just caught flat-footed(none of their existing recruiting channels/methods of evaluating applicants or recruiting people supplied teachers with the necessary skills), or did it fall into the ugly morass where people can't quite decide what "IT" or "CS" or "Computer Literacy" is supposed to involve; and so predictably fail to find the right person for the job because they don't actually know what the job is?
It's not necessarily an easy position to get teachers for even if you do know exactly what you are doing, since it's a skillset with strong demand, and often more competitive salary, outside education; but you really don't have a prayer of getting the right person for the job if you aren't sure whether you are teaching 'button clicking in MS Office', 'Java 101', 'Math, with an emphasis on algorithms and computational complexity', or 'how to IT the computers into submission'. All of these are useful things to learn, at least for some people; but they are also all entirely different; and unless you realize that your ability to obtain the right teacher is going to be tragic at best.
If they recruited an actual chemical engineer, it sounds like they had options, chemical engineering is also a skill that probably doesn't pay best in education; but also like they had no clue what they were looking for, since that's even further from logical than accidentally hiring an IT guy to handle CS.
There are some subject-matter-experts who are a genuine menace in the classroom(I suspect that most of us probably took at least one course in college from the TAs they shoved all the actual work of teaching the course onto...); but it is pretty sad how the K-12 level, both in terms of incentives for people teaching there, and in terms of applicants, skews heavy on people with qualifications 'in teaching'; but somewhat light on people with serious grounding in what they are supposed to be teaching.
Especially at lower grade levels, you'd be a fool to deny that 'teaching' is, in itself, a valuable and necessary skill that some people really, really, don't have; but merely being necessary doesn't make it a substitute for actually knowing what you are teaching and how to teach it. There are some who have excellent knowledge of a field; but simply either need to shape up or head over to a research/professional track and stay away from students(at least until the students are old enough to learn in a capacity as 'junior colleague' rather than 'student'); but exposure to a classroom has a way of weeding out those who can't teach rather brutally; while it is less effective at weeding out people with at least adequate 'teaching' skills; but subject matter knowledge that is a little too shaky to allow them to teach it with full confidence and facility.
Computer Science would look pretty weird if it tried to step away from the math department. That said, it would be nice if the twin evils of "if it involves a computer, we can call it CS, right?" and "Well, this isn't CS, so computers are good for nothing but typing papers" were dealt with.
Every time some 'button pressing in MS Office' tutorial gets a 'CS' course code, a turing machine's tape snaps; but it is also the case that there are many useful things to be done with computers, some requiring nontrivial skill, that can't honestly be called 'CS'; but are of great use to subjects not typically associated with computers.
I had a buddy in college who was nominally studying 'history'; but the substance of the work was heavily GIS-driven analysis of existing archeological sites and(using what was known about trade, cultural contact, etc. from those sites; along with information about the capabilities and needs of the available travel options of the day and enough geology and biology to advance reasonably educated hypotheses about past landscapes from present-day GIS data) using aerial and satellite imagery to identify likely trade routes and intermediate settlements in areas where little or no ground-level archeology had been done, to prioritize future on-the-ground digging. It wasn't 'CS'; but it was a flavor of 'history' where use of computer tools well beyond the point-n'-click level was simply non-optional. Pretending that that makes it 'CS' would be dishonest and misleading; but (given that not everyone is actually going to become a programmer and get rich selling mobile apps to everyone else's little programmers), there is something to be said for K-12 level introduction to the fact that even if you don't "do CS" knowing at least enough to be slightly dangerous and write awful little scripts that make real programmers cry is quite valuable in all sorts of areas.
Slashdot Mirror
Imagine that we re-brand the 'expensive robot' as a 'turnkey managed surgery solution' or a 'patient outcome appliance' and ask your IT minions about whether HQ likes capital costs or expensive human resources less...
Have you tried cutting it open and suturing it shut again? If not, I'll hang up to keep my metrics high enough to avoid getting reamed out by my supervisor in order to allow you to try that. If so, please bleed out on hold while I stall on escalating you to a tier 2 support rep, much less to engineering...
And to get that we'd need to replace all that laggy legacy fiber with a series of tubes pumped down as close to vacuum as possible, run perfectly straight between nodes. That might get costly.
The interesting question is whether allowing surgeons to telecommute makes you more or less likely to be chopped open by some hack.
On the minus side, we've certainly all experienced the fact that when the call center gets off shored it is because nobody gives a damn and a cheaper labor force can do a bad job for less. On the other hand, there's the old quip about 'what do you call a med school grad in the bottom ten percent of his class? Doctor.' and the fact that only having access to on-site talent means that you are substantially at the mercy of the quality of whatever medical experts happen to be where you need treatment, which varies enormously between wealthy medical hubs and low density and/or really poor backwaters.
Are you better off with whatever surgeon lives within commuting distance, because they can't hire any cheap 'n cheerful bottom feeder within your light cone? Or are you better off with any surgeon in your light cone; because they aren't limited to recruiting only the second-stringers willing to settle for the lousy location/lower salary/inferior institutional prestige?
Aside from its puff-piece value(here we are talking about it, despite the fact that academics and DARPA have been puttering around with speech recognition since the 70s if not earlier, and a number of companies-that-aren't-facebook are already aggressively attempting to give away voice-assistant products in order to score that juicy data, and facebook has nothing except its CEO's new year's resolution...) it may also be a matter of cultural fit.
I'd be pretty surprised if he does his own housecleaning, when their are contractors for that; but having 'the help' actually embedded in your home is a bit weird and intrusive unless they are amazingly skilled at being deferential and you are somewhat used to the idea(all the subtle joy of walking back into your hotel room and noticing that the cleaners moved things around a bit while you were out, except all the time and at home). There's also the probable level of trust that someone who appears to dislike intrusions on his own privacy and built a fortune by intruding on that of others would have for comparatively low-paid lackeys with privileged physical access to his house.
Some dodgy speech-to-text system bolted onto a pile of scripts is certainly going to do a vastly worse job than a human would; but it won't be nearly as noticeable or likely to sell pictures of you snorting lines off your iphone to the tabloids.
The trouble with athelete-research is that it is nominally wicked and immoral, so everybody carefully avoids keeping useful records about the experiments.
It's a real pity, there would probably be some very interesting stuff there if the dataset were available. At least the Purity of Sport remains unsullied, so there's that.
So long as it isn't a germline modification there will probably be some enthusiasm for even high risk treatments. DMD reliably cripples and then kills you in a time long enough to be an agonizing ordeal and short enough that you can be said to have died tragically young. When Plan A sucks just that much, even serious side effects start to look pretty good.
If it is a germline modification, or if somebody can't keep their genetic engineering virus from hopping around, a lot more caution will be warranted because removing a change committed to the wild will be a real problem; but this is a nasty disease with a lousy prognosis so if the risks are confined to the patients they will likely be tolerated pretty well.
That's actually what ended up happening to thalidomide. The FDA's refusal to approve a zesty teratogen for marketing to pregnant women was roundly vindicated in hindsight; but it has some niche applications in situations where the risks are lower and the diseases more serious.
This naming convention is mandatory, on pain of having to beat a nighttime crysalid terror mission with a rookie squad that lacks plasma weapons.
Mandatory Permissive Action Links for all providers of idiotic 'social' platforms. Surely this is an easy question?
Any situation where you end up having to trust the clients is more or less entirely hopeless. Team DRM has spent years and a great deal of money and effort demonstrating this; and they are working from a much stronger legal position that twitter is; the twits can deny API access but have effectively zero leverage over basic web scraping tools, where the DRM people can usually at least ensure that noncompliant clients are theoretically illegal.
Perhaps more dangerously(for the politicians, not that this is a bad thing) it's a largely false hope. If some pesky bunch of idealists is running a collection of deleted tweets just because it warms their hearts can you really believe that your opponent's campaign isn't going to be poring over your history with substantially greater motivation?
It also wouldn't be a huge surprise if experience suggested that the bad-PR tweets that actually gain traction are typically the ones that people capture and distribute in awful-mobile-app-screenshot form, rather than the systematic but mostly uninteresting automated collections of every tweet some twit attempted to untweet.
If, in practice, the juiciest accidental honesty is already being captured manually you just end up looking like you have something to hide by selectively denying API access. Plus, in absence of a suitably robust search tool, a flood of mostly uninteresting noise is a fair deterrent to all but the most enthusiastic investigators(who are probably already operating less visible equivalents of the same tool for 'opposition research' purposes tied to various campaigns).
If, god help us, twitter is still relevant in a couple of decades this may come back to bite them, since the automatic logs of what future presidential candidates said while campaigning for their entry-level positions in local markets today will probably be available; but in terms of immediate PR handling I suspect that a lot of the good stuff gets scraped up by humans without automated assistance.
Organizationally, the school district might actually be pretty logical.
These sorts of things aren't supposed to be the school's problem; but it seems hard to argue with the idea that they have become the school's problem anyway, since the readiness of the students they have to work with is being seriously affected. When something becomes your problem, having the option to deal with it is ugly; but often much less painful than waiting hopefully for whoever is supposed to fix the root cause of the problem to sort things out so that you can do your work.
They are using the NW.js javascript environment, packaged in their executable, to provide javascript interpretation without the browser limitations; but according to the article it is just being used in social engineering attacks at present, not coupled with an exploit.
Presumably having the guts of the application in javascript will make the developer's life easier if he wants to produce a version for another platform and nothing prevents this being used as a payload for some other exploit that allows the attacker to execute something for you; but the current version appears to be all payload and not much delivery mechanism.
And Java's predecessors were old enough to drink when Java was laid out. In fact, given that computers used to be a great deal rarer than mathematicians, it may well be argued that we've had architecture-independent programs longer than we've had architectures and certainly longer than we've had OSes.
Being capable of more dynamic range is certainly good; my concern is just what they'll do to source material that doesn't have it. Since the limitations in dynamic range usually show as uniform blown-out or shadowed areas, with no detail to play with, I can only guess that they'll be taking the areas of the image that do have detail to work with(but were also likely carefully tweaked to look good within the constrained range expected) and smear them out over the wider range in order to look more-contrasty-and-vibrant. That just isn't going to end well.
The HDR capabilities themselves are all well and good, just as greater color depth is a good thing; but I'd be similarly concerned if somebody' s TV purported to 'improve' 8 and 16 bit color into 24/32 bit color.
I'm always up for more resolution; but my eyes are bleeding just imagining what fake-HDR effects are going to look like. More dynamic range is useful; but what horrors are they going to invoke when faking data not specified by the source material to get a 'Best Buy Brite(tm)!' effect that 8 out of 10 focus group participants agreed looked brighter and more vivid than the competitor's image under retail conditions?
The major trick, in humans, is that (so far as we know) there aren't a bunch of conveniently defined castes to work with. With these ants there ended up being a single fairly neat switch that the researchers could apply to get an entire cascade of behavioral changes because that entire cascade of changes is already one that the ants have evolved to employ. We know of similar things in humans(the most dramatic probably being the developmental trajectory into either male or female development from an originally very similar state); but we don't seem to have a 'service-sector worker', or 'pre-med' caste, despite the frantic efforts of parents to the contrary.
This isn't to say that applications in humans would be impossible; but there is a reason why we've been medicating for generations(arguably, at least since we developed brewing, possibly earlier in populations whose hunter-gathering regions had decent psychoactives naturally available): because we rarely even know what we want, except in the vaguest terms, much less how to get there.
Since most of these distractions are probably smartphone related, and contemporary smartphones tend to have pretty substantial activity-logging capabilities and sensor packages, it would be technically feasible to get an 'aircraft blackbox' style snapshot of what was going on at the time of the injury.
Anyone know if you could get away with an insurance policy that demands such data, in the event of a claim, and then attempts to deny coverage for any costs deemed to have been incurred because of negligent distraction? Just think of how many claims you could deny with the right data...
The usual objective behind various rights and protections afforded to renters is addressing the fact that settling a contractual dispute while unexpectedly homeless is a...rather lopsided...process.
It's not news that contract law just doesn't work as smoothly when the parties involved are of vastly different power; and resolving matters through a civil proceeding can be problematic when there is particularly urgent time pressure. In situations involving people who rent their sole residence; both of these conditions tend to hold.
It's one thing to, say, go through a hassle with Dell because your computer was DOA. It's a bit more problematic to have a landlord stalling you while the heating system isn't working; or a sewage leak is making it impossible to use anything involving water(not my most entertaining college experience...)
Once you have basic logistics covered it becomes much easier to take up a property issue as just another civil dispute; but the thing about being a renter is that most of your disputes tend to start because certain basic logistics are suddenly not covered.
I wouldn't have modded it 'flamebait'; but it does fall into the relatively unhelpful category of being overbroad(if you go state-by-state, the degree to which US education is totally fucked varies quite widely); and it also ignores the important fact that the DOE isn't actively changing the state of CS education here; but merely signalling an unwillingness to get tough on trying to improve it.
They aren't saying "Yeah, you can fire those uppity CS people and replace them with guidance counselors who will teach the students how to empathize with the computes"; as much as they are saying "At present, effectively zero high schoolers are taking CS courses in some entire states; this is such a disaster that we will look the other way if you need to use dubiously suitable teachers to make this number better."
That's still not a good thing, since it's likely to result in a number of students getting CS-in-name-only rather than no CS, which will take more time but be about as useful; but it's not as though the DOE is saying "Yeah, things are great, go ahead and cut costs!"; but looking at the utter clusterfuck that exists in some parts of the country and informing them that they'd rather see the appearance of improvement, even if the schools don't have the faculty resources to do it properly, than more of the same.
Civics might be a better fit for software project management or software engineering, if only in the weak sense that someone with knowledge of how the noble theory behind various forms of government goes through a long, byzantine, often actively counterproductive, development process and finally ends up as a sprawling, sclerotic, morass of scope creep and dirty hacks should have an advantage in understanding how software projects tend to do the same.
Not much related to CS, of course; but that's the best I can think of.
Probably not from one of the well-reputed research universities where those who can do both; and even those who would prefer to be doing pretend to teach; but I'm not sure offhand.
Out of curiosity, do you know if your school was just caught flat-footed(none of their existing recruiting channels/methods of evaluating applicants or recruiting people supplied teachers with the necessary skills), or did it fall into the ugly morass where people can't quite decide what "IT" or "CS" or "Computer Literacy" is supposed to involve; and so predictably fail to find the right person for the job because they don't actually know what the job is?
It's not necessarily an easy position to get teachers for even if you do know exactly what you are doing, since it's a skillset with strong demand, and often more competitive salary, outside education; but you really don't have a prayer of getting the right person for the job if you aren't sure whether you are teaching 'button clicking in MS Office', 'Java 101', 'Math, with an emphasis on algorithms and computational complexity', or 'how to IT the computers into submission'. All of these are useful things to learn, at least for some people; but they are also all entirely different; and unless you realize that your ability to obtain the right teacher is going to be tragic at best.
If they recruited an actual chemical engineer, it sounds like they had options, chemical engineering is also a skill that probably doesn't pay best in education; but also like they had no clue what they were looking for, since that's even further from logical than accidentally hiring an IT guy to handle CS.
There are some subject-matter-experts who are a genuine menace in the classroom(I suspect that most of us probably took at least one course in college from the TAs they shoved all the actual work of teaching the course onto...); but it is pretty sad how the K-12 level, both in terms of incentives for people teaching there, and in terms of applicants, skews heavy on people with qualifications 'in teaching'; but somewhat light on people with serious grounding in what they are supposed to be teaching.
Especially at lower grade levels, you'd be a fool to deny that 'teaching' is, in itself, a valuable and necessary skill that some people really, really, don't have; but merely being necessary doesn't make it a substitute for actually knowing what you are teaching and how to teach it. There are some who have excellent knowledge of a field; but simply either need to shape up or head over to a research/professional track and stay away from students(at least until the students are old enough to learn in a capacity as 'junior colleague' rather than 'student'); but exposure to a classroom has a way of weeding out those who can't teach rather brutally; while it is less effective at weeding out people with at least adequate 'teaching' skills; but subject matter knowledge that is a little too shaky to allow them to teach it with full confidence and facility.
Computer Science would look pretty weird if it tried to step away from the math department. That said, it would be nice if the twin evils of "if it involves a computer, we can call it CS, right?" and "Well, this isn't CS, so computers are good for nothing but typing papers" were dealt with.
Every time some 'button pressing in MS Office' tutorial gets a 'CS' course code, a turing machine's tape snaps; but it is also the case that there are many useful things to be done with computers, some requiring nontrivial skill, that can't honestly be called 'CS'; but are of great use to subjects not typically associated with computers.
I had a buddy in college who was nominally studying 'history'; but the substance of the work was heavily GIS-driven analysis of existing archeological sites and(using what was known about trade, cultural contact, etc. from those sites; along with information about the capabilities and needs of the available travel options of the day and enough geology and biology to advance reasonably educated hypotheses about past landscapes from present-day GIS data) using aerial and satellite imagery to identify likely trade routes and intermediate settlements in areas where little or no ground-level archeology had been done, to prioritize future on-the-ground digging. It wasn't 'CS'; but it was a flavor of 'history' where use of computer tools well beyond the point-n'-click level was simply non-optional. Pretending that that makes it 'CS' would be dishonest and misleading; but (given that not everyone is actually going to become a programmer and get rich selling mobile apps to everyone else's little programmers), there is something to be said for K-12 level introduction to the fact that even if you don't "do CS" knowing at least enough to be slightly dangerous and write awful little scripts that make real programmers cry is quite valuable in all sorts of areas.