Yes. And "autopilot" in boats and planes means you can take your hands off the controls for significant periods of time (hours in some case) and actually take your close attention off the forward progress of the thing you are piloting for substantial periods of time (tens of seconds or more) as your radar and/or visual scans tell you that there's nothing too close on a heading to worry about.
In a Tesla with autopilot you can never safely take your full attention from the road or your hands off the controls. Musk seems to agree with this. Several accidents also seem to confirm this -- the Tesla crashing into the gore point at the 85 HOV offramp from the 101; the Tesla crashing (apparently at full speed) into a big red fire engine at an accident scene on 405 in Culver City; and this accident where the Tesla ran right into a stationary vehicle.
So, why keep calling it "autopilot" when it is NOTHING like autopilot? Musk will eventually listen to his lawyers and change the name, but he's just too stubborn to do it until he sees multi-million dollar settlements and (more importantly to him) more and more bad PR from his bone headed decision.
The Post Office is one of the few things specifically mentioned in the Constitution as a service the Government is expected to provide.
That is incorrect. In Article I, Section 8 the Constitution grants Congress the power to "establish Post Offices and post Roads". If they don't believe Post Offices or post Roads are needed, they are free not to exercise that power and to dismantle said system. In fact, right now, are are there many (any?) active "Post roads" of note in the United States? It seems likely that the postal service will be dismantled in the next 50 years as there will just be insufficient need for it.
Is it also your belief that because Congress has the power to declare war that they should do so even if they don't think it's necessary?
Even if Teslas are safer than a comparable priced new BMW or Volvo or Mercedes because they have better survivability (due, for example, to a superior front end crumple zone because there's no need to accommodate an engine/transmission in that area in a Tesla), that's irrelevant to any judgement on Autopilot (for example, if it's technically appropriate or if its name is misleading).
Suppose a Tesla model had 10% the fatality rate per passenger mile of the "average" car. Musk then figured out they could decrease the weight of the model to improve its acceleration performance by reducing strength of crash safety members and eliminating "extra" battery shielding and sold this option as "Insane Ludicrous Mode" -- but elimination of those safety features would make the car 3X more lethal to occupants in crashes. The car would still have 30% the fatality rate per passenger mile of the "average" car. Would Tesla fans here say "Oh, that's fine -- it's still safer than the 'average' car and the fine print mentions that this performance improvement was accomplished by reducing the safety so drivers should drive more carefully" (and, would they say that if a family member was killed as a passenger in someone else's "Insane Ludicrous Mode" Tesla when it was t-boned by a red light runner and it was apparent that the fiery death would likely not have occurred in the "standard" model)?
Another thing to consider is "passenger miles" vs. "vehicle miles" (the alleged stats quoted above are not clear, but the stats appear to be vehicle, not passenger, mile based). I rarely see a Tesla with more than one person in it. I don't know if that's typical, but it's my observation. With the exception of the recent "Tesla crematorium" case, I believe all of the publicized fatal and non-fatal Tesla crashes had only one person in the Tesla - the driver. If the number of passenger miles per vehicle mile are lower in Teslas than, say, Dodge minivians, of course the fatalities per vehicle mile will be less for Teslas because there are fewer people IN the average Tesla subjected to the force of the crash. It's quite common for cars with multiple occupants to crash and result in fatalities to some, not all, of the occupants (sometimes at least one occupant is killed and at least one literally walks away).
The demographics of the drivers are also important. Generally, due to the cost and interest in technology, Teslas are owned and driven by those in the age "sweet spot" of safe driving -- old enough not to be really stupid (i.e., 19 years old males prefacing their driving antics with "here, hold my beer") and young enough not to have impaired night vision, reaction time, and cognitive functions. The younger demographic usually just can't afford a Tesla and the older demographic (even if they feel they can afford one even though they are living on retirement savings) is generally less inclined to be drawn to newer technology.
Also, remember that (unless their batteries give out which is a distinct possibility -- but it's likely by then that there will be rebuilders who will take worn out battery packs and rebuild them with slightly used batteries from electric cars which have been totalled), the "safe" Teslas of today will be being compared in 15 years to the "average" car which by then will have more safety features and Teslas won't look as good when the $3000 used Tesla with 2016 safety levels are included in the mix. If Tesla wanted to avoid that, they could have by only leasing the cars and recycliing them after five years but they didn't choose to do that.
But all, when engaged, are expected to quite safely move the vehicle through the space it's traveling in and allow the pilot to take their hands off the controls and, for modestly short times (ten seconds or more), divert their attention to doing other things (such as looking at charts) with only occasional scans of the area (either visual or via radar etc). Tesla autopilot seems to require the driver's constant attention to avoid running into stationary objects that are routinely encountered on roads (gore points, fire trucks, etc...). The "auto" pilot in "autopilot" implies that the feature can be safely used for significant portions of your trip without close second by second monitoring and that's simply not the case with Tesla's feature so the name is very misleading.
Yes, drivers should heed all the warnings, but it's a little like selling "8 ton automotive jackstands" with a sticker on them saying "Do not get any part of body under a vehicle elevated on these jackstands" and blaming the user because the "jackstands" collapsed when their 1.5 ton car fell on them during brake job.
Interestingly, you used the vague terms "perhaps" and "likely" and "often" as well as selected a rather arbitrary "5 year" relevance window. These are your political opinions and mine may (and I suspect do) differ greatly. A ten year old video of a politician engaging in sexual harassment video IS politically relevant to some voters. If you don't think it is, fine -- just ignore the video. However, those that do think it's relevant should be able to see it and judge for themselves.
The whole nature of the Free Speech and Free Press clauses of the First Amendment is that there is virtually no control over such things because by necessity the inverse that means that the very government that controls speech and press is also the government that has a vested interest in suppressing relevant information. China would be a good example here. And, if it wasn't for the First Amendment, imagine what Trump and Sessions and the newly formed Ministry of Truth cabinet department would consider relevant -- they might decide that ten minutes is the window of relevance and ban all discussion of anything Trump did more than ten minutes ago.
I'm skeptical that is a "debt" legally. Assuming the policy of "No Cash" is posted, you agreed to those terms when you ordered.
If you are in a Walmart and eat a candy bar while in the store and intentionally walk out without paying for it, what is that? It sounds like theft to me. If you think it's just a "debt", what about if you put the candy bar in your pocket and intentionally walk out without paying for it, what is that? It sounds like theft to me rather than owing a "debt". What if you are in a Walmart and intentionally walk out with a television set without paying for it, what is that? It sounds like theft to me rather than owing a debt. Intent may matter here though -- if you were unaware of the contract terms (didn't notice the sign), it's probably not a criminal matter.
I would think in the case where the diner has eaten the meal and then doesn't have a credit or debit card to pay for it or simply refuses to use a card, it would be a civil issue that the restaurant could pursue in court and it might become a "debt" when a judgement against the diner is entered. In practice, they will probably take the cash and advise the diner that next time they can't pay via cash. If the diner repeats this behavior, the restaurant might again take cash but then inform the diner they are no longer welcome at the establishment and that they will be trespassing and the police will be called to arrest them if they return and don't leave upon request.
The key word is "debt". Barring a contract clause to the contrary, a debtor in the US must accept USD cash to satisfy a financial debt.
However, when you buy a meal at a restaurant, you are not entering into a debtor:creditor relationship (of course assuming you didn't enter into some bizarre contract with the restaurateur indicating that they had loaned you the money for the meal and you, then, were to repay that loan before leaving the restaurant).
As the Federal Reserve site explains, there is "no Federal statute mandating that a private business, a person, or an organization must accept currency or coins as payment for goods or services."
Your tests are not as simple as you seem to think.
First, the First Amendment does not protect just "political speech". With narrow exceptions, it protects all forms of speech for all purposes - including artistic, educational, religious, and entertainment. So, why have an exception to data retention restrictions for just data that might be useful in supporting political speech.
Second, identifying that something will, someday, support political speech is impossible. For example, if there is a video of random unknown John Doe grabbing a coworker's ass at a company event, does retaining that serve to support political speech? Well, generally no of course. However if twenty years later, John Doe ends up running for President on a platform including "Eliminate Sexual Harassment in the Workplace", the video will have turned out to be useful in supporting political speech -- but if Google, Bing, Facebook, Wayback Machine, and the local newspaper et al had been forced to delete and deindex it at Doe's "forget this" request a year after it was made public, voters and journalists would never know to even inquire as to the apparent hypocrisy on the part of Candidate Doe.
Third, a LOT of things are reported that are not "public record" - investigative reporters, for example, sometimes get data that is not "public" (for example, from an employee of a company they are investigating). Surely if such an investigation results in the discovery that General Electric knowingly cut corners in the design of a nuclear reactor that then caused a radiation release that killed thousands of people, the reporter should be free to reveal the information they discovered and not be required to "take it down" when GE requested that it be "forgotten".
The courts have long upheld that what happens in public is, just that, public. The police can observe it (and record it) just as you can observe (and record) the police. If you don't want pictures of yourself urinating in public, don't do it. If you don't want pictures of yourself pouring a mug of beer over your own head, don't do that in a public place.
Since the dawn of civilization, people have known what other people did in public. Although, now, with larger more anonymous communities coupled with the ease of mobility allowing one to easily move hundreds or thousands of miles away every few years, you have more privacy from those around than you once did (unless you do something so notorious and stupid that it goes viral).
The interpretation required would be so extreme as to render the Free Speech and Free Press clause of the First Amendment to be nearly impotent. It would require judicial handstands almost as extreme as some want the courts to do in an attempt to invalidate the Second Amendment because they know there isn't public support for repealing it.
So, maybe the Ninth Circuit Court of Appeals would uphold a "right to be forgotten" law, but neither the "liberal" or "conservative" wing of the Supreme Court would let their ruling stand.
If you are letting people take pictures of you and they are posting those pictures, that's not Facebook's fault. That same person can take that same picture and post it on their own web site on their own server. If you don't want your picture taken and posted, avoid places where people take pictures and avoid interacting with people who take pictures and post them.
"Right to be forgotten" would require amending the Constitution to eliminate that pesky Free Speech Clause in the United States Constitution.
(What good is a 'Preview' button when it doesn't display the comment as it will be seen by others? Sorry about the unsuccessful attempt to format a list.)
The "Irish slave" claim is really a myth. There were many Irish indentured servants but indentured servitude is not the same as slavery. Being an indentured servant often was not a good life, but there were very critical differences between being an indentured servant and a slave.
For example:
Indentured servants were considered a full "person". Slaves were not.
Indentured servants entered into a contract which required, typically in exchange for passage to the Colonies/US, them to provide services for a fixed period of time (often seven years). Being a slave was "for life".
Indentured servants didn't pass their status on to their offspring. Slaves did.
Many, perhaps most, indentured servants willingly entered in to the deal (albeit, maybe not completely aware of what they were getting into). Slaves did not.
Indentured servants (and their offspring) were not the "property" of the person they were serving. Slaves were.
The article includes quotes from Richard Stallman, arguing that data privacy isn't the problem. "The problem is that these companies are collecting data about you, period. We shouldn't let them do that. The data that is collected will be abused..." He later adds that "We need a law that requires every system to be designed in a way that achieves its basic goal with the least possible collection of data... No company is so important that its existence justifies setting up a police state."
Stallman is quite confused. A "police state" is one that imposes its will on the people via laws. If someone wants to start a company like Facebook and consumers decide to use that company's services in exchange for giving up personal information and being tracked, only a police state would prevent that voluntary arrangement. I have virtually no footprint on Facebook because my reluctance to let them track me exceeds my desire to use their services. However, why should the government pass a law that prevents those that feel otherwise from acting on their desires?
A contractor's rate is not fixed per item of work. A contractor sets their price and a client either agrees or rejects. On the next item of work, the process happens over again.
That's not always the case. Companies will sometimes enter into an agreement with a contractor for the contractor to do Task A for $X, Task B for $Y, and Task C for $Z for the next 12 months -- the number and mix of A, B, and C that the company actually requests the contractor to do can vary widely based on the day-to-day needs of the company that hired the contractor. There may be guarantees of at least some number of each type of task per year and there may be a maximum number of each type of task that the contractor will promise to do for the negotiated rates. Each job, however, is not specifically negotiated.
For example, the facilities department may contract with a painting contractor to do small paint jobs around the facility (where people put their feet on the wall or whatever) where the contract sets a "per request" charge and a "per square foot" charge on top of that
In the case of Tesla "autopilot" miles (which isn't "self driving"), one needs to consider the driver and vehicle demographics.
The cost of a Tesla eliminates a huge number of the riskiest drivers - such as young males and old folks. Mostly drivers of Teslas are middle age wealthy folks -- and many of them are tech oriented (telling since some insurance companies give a discount to engineers since they seem to be safer drivers). Few of them are in the "Here, hold my beer while I show you how to do a four wheel drift around the corner of 1st and Main St" demographics.
As well, fatality numbers for autopilot miles need to be adjusted to take into account the safety and survivablity features of the base car independent of autopilot. Teslas have the modern crash prevention features (such as ESC) and crash survival capabilities (such as side air bags) that many older cars on the road don't have. As well, Teslas seem have good crash surviablity.
And, of course, "autopilot" isn't widely used in some of the more dangerous scenarios (black ice, snow, roads with poor lane markings, etc).
The purpose is to determine if they can complete a programming task in the real world with the skills they have gained. Using the internet and bringing in materials is expected everywhere else except an exam at a university. The only cheating that can be done is asking others for help and that is easily monitored.
Wrong (usually), the purpose of an exam in school (unless perhaps the course is Computer Plagiarism 101A) is to measure your understanding of the material, not "to get it done" -- after all, your degree is telling the world that you understood the curriculum and, if you have the degree and didn't get the curriculum, you are sullying the reputation of the University and reducing the value of the degrees of the other students around you. That is not the purpose of programming in the "real world" - the goal is to get it done properly (including not compromising IP by copying code that you don't have rights to use in the way you are using it). Hence, except what the professor authorizes (as in an open book test where she states that you can use the textbook or some other specific reference material), everything is off-limits -- and that usually will be "the whole internet".
This pretty much mandates locked down dedicated "workstations" w/o internet access and without networked access to other students or anything but the "authorized open books" unless the University has an honor code that is taken seriously (likely requiring severe consequences for a violation -- such as expulsion with no opportunity to return and permanent refusal to release transcripts on a student's first violation).
Let them use whatever they normally use.
Unfortunately, the need for dedicated locked down systems makes this impractical -- although, there's no reason (except administrative costs) several popular environments shouldn't be available.
It's hard to say that one has a well rounded CS degree if one does not know at least one language and preferably more. I probably used at least 15 different languages in the 3-4 years I spent getting my CS BSc decades ago from a fairly well respected school. Of course, only one of languages I used in school did I ever use professionally and that was quite a different dialect of the language. However, this exposure to multiple ways of expressing algorithms and working at different levels was quite helpful - Simula, LISP, APL, Assembly (various), Microcode (Varian), Algol, and SNOBOL come to mind as being among the useful ones. Some were similar enough to others (or just annoying) to not be as useful (FORTRAN, Basic, Pascal come to mind in this category - although spending way too many recreational hours reverse engineering the intermediate "compiled" file format of BASIC on a PDP with not one word of documentation was fun and there were still a few bytes I never figured out - for some reason my girlfriend didn't seem nearly as interested in this as I was!).
Fortunately, I never had to, or chose to, actually write COBOL - in my spare time I read a book about it and swore I would never ever:
PERFORM VARYING Pain FROM Low BY TooMuch UNTIL Pain GREATER THAN Tolerance
Frankly though, even just writing by hand for two hours is painful, frustrating, and slow if the rest of the time you type almost everything. If you take notes in class on a laptop, type all your papers, and (of course) do your programming assignments on a (guess what) computer, the muscles in your fingers are probably just are not used to writing longhand for very long.
A CS test should not be a test of how diligent you are at doing hand exercises during the semester in order to prepare your hands for a longhand test.
I'm LONG out of school but if I try to write more than a few lines of English text, my fingers get tired and it becomes almost illegible as I've been typing 99.9+% of my written work (including "out of class" written assignments when I was in school) since the mid 70's.
If dedicated computers are configured for "test taking", there shouldn't be any room for academic dishonesty. You probably can't let students use their own laptops unless you really think your honor code is taken seriously by all (which, I think, is rare).
Any "skilled programmer" should know the specifics of most common algorithms and datatypes - although if they haven't actually used a particular algorithm in years, I don't see why they can't refresh their brain by spending a couple minutes online (i.e., I would count that as "know" as it was "once knew but have forgotten some of the specifics").
Of course, a keypunch operator working off of coding forms doesn't need to understand common algorithms and datatypes.
Why not allow testing out of a class that is core to the curriculum? Perhaps the "test out" test should be harder (deeper and broader and probably longer) than the tests that would be taken by someone taking the class and/or perhaps the "passing score" should be an "A" just to make sure that the person really wouldn't benefit from taking the class because they would likely learn something from it.
Slashdot Mirror
http://dilbert.com/strip/2018-01-30
Yes. And "autopilot" in boats and planes means you can take your hands off the controls for significant periods of time (hours in some case) and actually take your close attention off the forward progress of the thing you are piloting for substantial periods of time (tens of seconds or more) as your radar and/or visual scans tell you that there's nothing too close on a heading to worry about.
In a Tesla with autopilot you can never safely take your full attention from the road or your hands off the controls. Musk seems to agree with this. Several accidents also seem to confirm this -- the Tesla crashing into the gore point at the 85 HOV offramp from the 101; the Tesla crashing (apparently at full speed) into a big red fire engine at an accident scene on 405 in Culver City; and this accident where the Tesla ran right into a stationary vehicle.
So, why keep calling it "autopilot" when it is NOTHING like autopilot? Musk will eventually listen to his lawyers and change the name, but he's just too stubborn to do it until he sees multi-million dollar settlements and (more importantly to him) more and more bad PR from his bone headed decision.
That is incorrect. In Article I, Section 8 the Constitution grants Congress the power to "establish Post Offices and post Roads". If they don't believe Post Offices or post Roads are needed, they are free not to exercise that power and to dismantle said system. In fact, right now, are are there many (any?) active "Post roads" of note in the United States? It seems likely that the postal service will be dismantled in the next 50 years as there will just be insufficient need for it.
Is it also your belief that because Congress has the power to declare war that they should do so even if they don't think it's necessary?
Indeed. If you're pulling a 747 out of a ditch with a plane tug, at least two serious errors in judgement are in evidence.
Even if Teslas are safer than a comparable priced new BMW or Volvo or Mercedes because they have better survivability (due, for example, to a superior front end crumple zone because there's no need to accommodate an engine/transmission in that area in a Tesla), that's irrelevant to any judgement on Autopilot (for example, if it's technically appropriate or if its name is misleading).
Suppose a Tesla model had 10% the fatality rate per passenger mile of the "average" car. Musk then figured out they could decrease the weight of the model to improve its acceleration performance by reducing strength of crash safety members and eliminating "extra" battery shielding and sold this option as "Insane Ludicrous Mode" -- but elimination of those safety features would make the car 3X more lethal to occupants in crashes. The car would still have 30% the fatality rate per passenger mile of the "average" car. Would Tesla fans here say "Oh, that's fine -- it's still safer than the 'average' car and the fine print mentions that this performance improvement was accomplished by reducing the safety so drivers should drive more carefully" (and, would they say that if a family member was killed as a passenger in someone else's "Insane Ludicrous Mode" Tesla when it was t-boned by a red light runner and it was apparent that the fiery death would likely not have occurred in the "standard" model)?
Another thing to consider is "passenger miles" vs. "vehicle miles" (the alleged stats quoted above are not clear, but the stats appear to be vehicle, not passenger, mile based). I rarely see a Tesla with more than one person in it. I don't know if that's typical, but it's my observation. With the exception of the recent "Tesla crematorium" case, I believe all of the publicized fatal and non-fatal Tesla crashes had only one person in the Tesla - the driver. If the number of passenger miles per vehicle mile are lower in Teslas than, say, Dodge minivians, of course the fatalities per vehicle mile will be less for Teslas because there are fewer people IN the average Tesla subjected to the force of the crash. It's quite common for cars with multiple occupants to crash and result in fatalities to some, not all, of the occupants (sometimes at least one occupant is killed and at least one literally walks away).
The demographics of the drivers are also important. Generally, due to the cost and interest in technology, Teslas are owned and driven by those in the age "sweet spot" of safe driving -- old enough not to be really stupid (i.e., 19 years old males prefacing their driving antics with "here, hold my beer") and young enough not to have impaired night vision, reaction time, and cognitive functions. The younger demographic usually just can't afford a Tesla and the older demographic (even if they feel they can afford one even though they are living on retirement savings) is generally less inclined to be drawn to newer technology.
Also, remember that (unless their batteries give out which is a distinct possibility -- but it's likely by then that there will be rebuilders who will take worn out battery packs and rebuild them with slightly used batteries from electric cars which have been totalled), the "safe" Teslas of today will be being compared in 15 years to the "average" car which by then will have more safety features and Teslas won't look as good when the $3000 used Tesla with 2016 safety levels are included in the mix. If Tesla wanted to avoid that, they could have by only leasing the cars and recycliing them after five years but they didn't choose to do that.
But all, when engaged, are expected to quite safely move the vehicle through the space it's traveling in and allow the pilot to take their hands off the controls and, for modestly short times (ten seconds or more), divert their attention to doing other things (such as looking at charts) with only occasional scans of the area (either visual or via radar etc). Tesla autopilot seems to require the driver's constant attention to avoid running into stationary objects that are routinely encountered on roads (gore points, fire trucks, etc...). The "auto" pilot in "autopilot" implies that the feature can be safely used for significant portions of your trip without close second by second monitoring and that's simply not the case with Tesla's feature so the name is very misleading.
Yes, drivers should heed all the warnings, but it's a little like selling "8 ton automotive jackstands" with a sticker on them saying "Do not get any part of body under a vehicle elevated on these jackstands" and blaming the user because the "jackstands" collapsed when their 1.5 ton car fell on them during brake job.
Interestingly, you used the vague terms "perhaps" and "likely" and "often" as well as selected a rather arbitrary "5 year" relevance window. These are your political opinions and mine may (and I suspect do) differ greatly. A ten year old video of a politician engaging in sexual harassment video IS politically relevant to some voters. If you don't think it is, fine -- just ignore the video. However, those that do think it's relevant should be able to see it and judge for themselves.
The whole nature of the Free Speech and Free Press clauses of the First Amendment is that there is virtually no control over such things because by necessity the inverse that means that the very government that controls speech and press is also the government that has a vested interest in suppressing relevant information. China would be a good example here. And, if it wasn't for the First Amendment, imagine what Trump and Sessions and the newly formed Ministry of Truth cabinet department would consider relevant -- they might decide that ten minutes is the window of relevance and ban all discussion of anything Trump did more than ten minutes ago.
I'm skeptical that is a "debt" legally. Assuming the policy of "No Cash" is posted, you agreed to those terms when you ordered.
If you are in a Walmart and eat a candy bar while in the store and intentionally walk out without paying for it, what is that? It sounds like theft to me. If you think it's just a "debt", what about if you put the candy bar in your pocket and intentionally walk out without paying for it, what is that? It sounds like theft to me rather than owing a "debt". What if you are in a Walmart and intentionally walk out with a television set without paying for it, what is that? It sounds like theft to me rather than owing a debt. Intent may matter here though -- if you were unaware of the contract terms (didn't notice the sign), it's probably not a criminal matter.
I would think in the case where the diner has eaten the meal and then doesn't have a credit or debit card to pay for it or simply refuses to use a card, it would be a civil issue that the restaurant could pursue in court and it might become a "debt" when a judgement against the diner is entered. In practice, they will probably take the cash and advise the diner that next time they can't pay via cash. If the diner repeats this behavior, the restaurant might again take cash but then inform the diner they are no longer welcome at the establishment and that they will be trespassing and the police will be called to arrest them if they return and don't leave upon request.
The key word is "debt". Barring a contract clause to the contrary, a debtor in the US must accept USD cash to satisfy a financial debt.
However, when you buy a meal at a restaurant, you are not entering into a debtor:creditor relationship (of course assuming you didn't enter into some bizarre contract with the restaurateur indicating that they had loaned you the money for the meal and you, then, were to repay that loan before leaving the restaurant).
As the Federal Reserve site explains, there is "no Federal statute mandating that a private business, a person, or an organization must accept currency or coins as payment for goods or services."
Your tests are not as simple as you seem to think.
First, the First Amendment does not protect just "political speech". With narrow exceptions, it protects all forms of speech for all purposes - including artistic, educational, religious, and entertainment. So, why have an exception to data retention restrictions for just data that might be useful in supporting political speech.
Second, identifying that something will, someday, support political speech is impossible. For example, if there is a video of random unknown John Doe grabbing a coworker's ass at a company event, does retaining that serve to support political speech? Well, generally no of course. However if twenty years later, John Doe ends up running for President on a platform including "Eliminate Sexual Harassment in the Workplace", the video will have turned out to be useful in supporting political speech -- but if Google, Bing, Facebook, Wayback Machine, and the local newspaper et al had been forced to delete and deindex it at Doe's "forget this" request a year after it was made public, voters and journalists would never know to even inquire as to the apparent hypocrisy on the part of Candidate Doe.
Third, a LOT of things are reported that are not "public record" - investigative reporters, for example, sometimes get data that is not "public" (for example, from an employee of a company they are investigating). Surely if such an investigation results in the discovery that General Electric knowingly cut corners in the design of a nuclear reactor that then caused a radiation release that killed thousands of people, the reporter should be free to reveal the information they discovered and not be required to "take it down" when GE requested that it be "forgotten".
The courts have long upheld that what happens in public is, just that, public. The police can observe it (and record it) just as you can observe (and record) the police. If you don't want pictures of yourself urinating in public, don't do it. If you don't want pictures of yourself pouring a mug of beer over your own head, don't do that in a public place.
Since the dawn of civilization, people have known what other people did in public. Although, now, with larger more anonymous communities coupled with the ease of mobility allowing one to easily move hundreds or thousands of miles away every few years, you have more privacy from those around than you once did (unless you do something so notorious and stupid that it goes viral).
The interpretation required would be so extreme as to render the Free Speech and Free Press clause of the First Amendment to be nearly impotent. It would require judicial handstands almost as extreme as some want the courts to do in an attempt to invalidate the Second Amendment because they know there isn't public support for repealing it.
So, maybe the Ninth Circuit Court of Appeals would uphold a "right to be forgotten" law, but neither the "liberal" or "conservative" wing of the Supreme Court would let their ruling stand.
If you are letting people take pictures of you and they are posting those pictures, that's not Facebook's fault. That same person can take that same picture and post it on their own web site on their own server. If you don't want your picture taken and posted, avoid places where people take pictures and avoid interacting with people who take pictures and post them.
"Right to be forgotten" would require amending the Constitution to eliminate that pesky Free Speech Clause in the United States Constitution.
(What good is a 'Preview' button when it doesn't display the comment as it will be seen by others? Sorry about the unsuccessful attempt to format a list.)
The "Irish slave" claim is really a myth. There were many Irish indentured servants but indentured servitude is not the same as slavery. Being an indentured servant often was not a good life, but there were very critical differences between being an indentured servant and a slave.
For example:
More information is here, here, here, and here.
Stallman is quite confused. A "police state" is one that imposes its will on the people via laws. If someone wants to start a company like Facebook and consumers decide to use that company's services in exchange for giving up personal information and being tracked, only a police state would prevent that voluntary arrangement. I have virtually no footprint on Facebook because my reluctance to let them track me exceeds my desire to use their services. However, why should the government pass a law that prevents those that feel otherwise from acting on their desires?
That's not always the case. Companies will sometimes enter into an agreement with a contractor for the contractor to do Task A for $X, Task B for $Y, and Task C for $Z for the next 12 months -- the number and mix of A, B, and C that the company actually requests the contractor to do can vary widely based on the day-to-day needs of the company that hired the contractor. There may be guarantees of at least some number of each type of task per year and there may be a maximum number of each type of task that the contractor will promise to do for the negotiated rates. Each job, however, is not specifically negotiated.
For example, the facilities department may contract with a painting contractor to do small paint jobs around the facility (where people put their feet on the wall or whatever) where the contract sets a "per request" charge and a "per square foot" charge on top of that
In the case of Tesla "autopilot" miles (which isn't "self driving"), one needs to consider the driver and vehicle demographics.
The cost of a Tesla eliminates a huge number of the riskiest drivers - such as young males and old folks. Mostly drivers of Teslas are middle age wealthy folks -- and many of them are tech oriented (telling since some insurance companies give a discount to engineers since they seem to be safer drivers). Few of them are in the "Here, hold my beer while I show you how to do a four wheel drift around the corner of 1st and Main St" demographics.
As well, fatality numbers for autopilot miles need to be adjusted to take into account the safety and survivablity features of the base car independent of autopilot. Teslas have the modern crash prevention features (such as ESC) and crash survival capabilities (such as side air bags) that many older cars on the road don't have. As well, Teslas seem have good crash surviablity.
And, of course, "autopilot" isn't widely used in some of the more dangerous scenarios (black ice, snow, roads with poor lane markings, etc).
Wrong (usually), the purpose of an exam in school (unless perhaps the course is Computer Plagiarism 101A) is to measure your understanding of the material, not "to get it done" -- after all, your degree is telling the world that you understood the curriculum and, if you have the degree and didn't get the curriculum, you are sullying the reputation of the University and reducing the value of the degrees of the other students around you. That is not the purpose of programming in the "real world" - the goal is to get it done properly (including not compromising IP by copying code that you don't have rights to use in the way you are using it). Hence, except what the professor authorizes (as in an open book test where she states that you can use the textbook or some other specific reference material), everything is off-limits -- and that usually will be "the whole internet".
This pretty much mandates locked down dedicated "workstations" w/o internet access and without networked access to other students or anything but the "authorized open books" unless the University has an honor code that is taken seriously (likely requiring severe consequences for a violation -- such as expulsion with no opportunity to return and permanent refusal to release transcripts on a student's first violation).
Unfortunately, the need for dedicated locked down systems makes this impractical -- although, there's no reason (except administrative costs) several popular environments shouldn't be available.
It's hard to say that one has a well rounded CS degree if one does not know at least one language and preferably more. I probably used at least 15 different languages in the 3-4 years I spent getting my CS BSc decades ago from a fairly well respected school. Of course, only one of languages I used in school did I ever use professionally and that was quite a different dialect of the language. However, this exposure to multiple ways of expressing algorithms and working at different levels was quite helpful - Simula, LISP, APL, Assembly (various), Microcode (Varian), Algol, and SNOBOL come to mind as being among the useful ones. Some were similar enough to others (or just annoying) to not be as useful (FORTRAN, Basic, Pascal come to mind in this category - although spending way too many recreational hours reverse engineering the intermediate "compiled" file format of BASIC on a PDP with not one word of documentation was fun and there were still a few bytes I never figured out - for some reason my girlfriend didn't seem nearly as interested in this as I was!).
Fortunately, I never had to, or chose to, actually write COBOL - in my spare time I read a book about it and swore I would never ever:
PERFORM VARYING Pain FROM Low BY TooMuch UNTIL Pain GREATER THAN Tolerance
Frankly though, even just writing by hand for two hours is painful, frustrating, and slow if the rest of the time you type almost everything. If you take notes in class on a laptop, type all your papers, and (of course) do your programming assignments on a (guess what) computer, the muscles in your fingers are probably just are not used to writing longhand for very long.
A CS test should not be a test of how diligent you are at doing hand exercises during the semester in order to prepare your hands for a longhand test.
I'm LONG out of school but if I try to write more than a few lines of English text, my fingers get tired and it becomes almost illegible as I've been typing 99.9+% of my written work (including "out of class" written assignments when I was in school) since the mid 70's.
If dedicated computers are configured for "test taking", there shouldn't be any room for academic dishonesty. You probably can't let students use their own laptops unless you really think your honor code is taken seriously by all (which, I think, is rare).
Any "skilled programmer" should know the specifics of most common algorithms and datatypes - although if they haven't actually used a particular algorithm in years, I don't see why they can't refresh their brain by spending a couple minutes online (i.e., I would count that as "know" as it was "once knew but have forgotten some of the specifics").
Of course, a keypunch operator working off of coding forms doesn't need to understand common algorithms and datatypes.
That does teach one to be careful. But, what's this "three compiles a day"? Did you bribe the RJE operator?
Why not allow testing out of a class that is core to the curriculum? Perhaps the "test out" test should be harder (deeper and broader and probably longer) than the tests that would be taken by someone taking the class and/or perhaps the "passing score" should be an "A" just to make sure that the person really wouldn't benefit from taking the class because they would likely learn something from it.