The real power is in the corporations. They buy the politicians (oops I meant "lobby") and get their way every single time.
I agree that corporate influence over politics is a problem.
We erode the Bill of Rights and nobody bats an eye at it. We can now indefinitely imprison anyone that might be a terrorist. So there goes Habeas Corpus. First amendment is shot to hell. What do you think the founding fathers would think of "free speech zones"? I have a feeling they'd be loading muskets. Second amendment? Also boned. Nagin after Katrina went through the gun licenses and ordered the national guard to confiscate every gun in the city. And they did. It goes on and on. Warantless wiretaps, GPS tracking devices without a court order, Carnivore...you no longer have hardly any rights at all, but you don't even notice it. They really did a number on us. A real pro job.
However, I can't see how the above can be attributed to corporate influence. There are some areas... big content has way too much influence over copyright law, and corporations are far too often shielded from the effects of their own bad acts -- and even bailed out by the government if they're too big to fail.
But what's the corporate interest in strip-searching airline passengers, or confining political speech to designated zones away from politicians, or getting the government to confiscate guns, or to install GPS tracking devices on people or read their e-mail, etc.?
I assert that none of those things are the result of corporate influence, because big business couldn't care less about them. Some of them, such as the warrantless wiretaps, actually bring heat down on the corps, and others, like all of the interference with business travel, make their lives harder.
No, I don't think you can lay any of that at the feet of corporate lobbyists. All that other stuff is just politicians trying to do what they think the people want them to do.
I seriously doubt the "electronic junk" adds much at all, not compared to the weight expended on improved crash protection. The bodies of modern cars, before any of that other junk is added, are 50% heavier than those of similar cars 30 years ago.
Your telling me that winter tires are not mandatory in wintery states in the US?
Not only are they not mandatory, but in many states they are, in fact, illegal.
Cite?
Studded tires are illegal in many states, but not snow tires. The main difference between snow tires and all-weather tires is that the snow tires are made of softer rubber. The harder rubber of all-weather tires is too stiff in cold weather to grip the ground well.
You're information is only somewhat accurate. A typical interview path for Google consists of about 7 interviews during which you're interviewed by at least a couple different people and although there is no official stance on asking such questions there are most definitely interviewers that do.
That information is also somewhat dated.
There has been an effort to standardize the Google interview process, and except in unusual circumstances, here's how it works:
The first step is a phone screen. This is a 45-minute phone call, during which the interviewer poses a series of hypothetical programming problems. Many phone screeners will set up a shared Writer document in Google Docs and ask the candidate to write some code, but many don't.
The purpose of the phone screen is purely to evaluate whether or not the candidate should be brought in for an on-site interview. The screener doesn't participate further in the hiring process, and the results of the screen don't go to the hiring committee, other than in the sense that if the candidate doesn't pass the screen the committee will never hear of the candidate.
The next step is the on-site interview. This consists of five one-hour interviews with five different people. The third "interview" however, is just lunch. The Google employee assigned for that hour takes the candidate to lunch and answers questions about the company, their job, etc. and just generally socializes. That employee doesn't provide any hiring feedback.
The other four interviews are what really matters. HR defines the standard set of off-limits questions -- anything relevant to gender, religion, ethnicity, economic background, etc. There is also a list of engineering questions which are banned, some because they're not very useful, most because they're too well-known, published on the web, etc. Outside of that the interviewer can do what they like, with the goal of evaluating the candidate's problem-solving, coding, design and architecture abilities, as well as "Googliness" (culture fit), but the approach should be centered around solving problems.
The interviewers do not compare notes or consult with one another. There is a piece of paper on which each jots down the technical problems presented and the areas tested (e.g. coding, design, etc.). They don't write down how the candidate performed; the purpose of the paper is just to ensure that they all use different problems and adequately cover all of the areas of ability that need to be evaluated.
After the interviews, the interviewers still don't consult with one another, in order to avoid one influencing the opinion of another. Each writes up a report with a description of the problems presented, how the candidate responded (including the code the candidate wrote) and their overall evaluation. The report culminates in a hire/no-hire recommendation and a numeric score between 1 (I'll quit if you hire this person) and 4 (I'll quit if you don't hire this person). Each interviewer's scores are tracked for calibration purposes.
The next step is that the reports are forwarded to the local hiring committee. This committee typically meets bi-weekly, so this often introduces a significant delay into the decision process. The committee evaluates the candidates based on their resume and on the reports and makes a hire/no-hire decision. The candidate's packet is also forwarded to another, corporate, hiring committee. This committee normally follows the recommendation of the lower committee but in some cases does reverse it.
That's the process, and the technical questions definitely are not brain-teaser type problems. They're abstract and generally somewhat contrived, because that's all there's time for. But questions that rely on some flash of insight are not appropriate, nor are questions that depend on knowledge of trivia, and interviewers that use them regularly will be taken to task by the hiring committee. Good questions permit ma
My response to questions like that improved dramatically when I read an article that explained questions way out of left field like that are intended to test your problem solving ability, so do your best to estimate an answer and explain your thought process. Reading that article didn't make me better at debugging hard technical issues, but made me dramatically better at handling off-the-wall interview questions nimbly. You're not measuring what you think you're measuring.
When I interviewed at Google, every one of the five people who interviewed me explained that they were interested in understanding how I approached and solved problems, and that it mattered less that I succeeded at solving the problem than that I demonstrated how I went about trying to solve it. They all asked me to please explain my thought processes and not to worry about making sure everything I said was perfectly correct.
Does that include the right to hear what you don't want to hear or be subject to advertising?
I'm not sure what "right to hear what you don't want to hear" means. You have the freedom to ignore what you want to ignore. You can plug your ears, close your eyes, ban speech you don't like from your property, etc.
The right for idiots worldwide to make shallow judgments about you that you can't ignore?
Why could you not ignore them? But, in general, no you have no right to control what others think or believe.
Compare it to the Second Amendment which basically says there is some sort of right to bear arms, but doesn't say much about how that is to be implemented other than saying there is a "militia" that is "regulated" which should be free to bear arms (which are not specified).
Actually, SCOTUS decided in Heller that the militia portion was just a motivational clause and not descriptive of the nature of the right. So the operative part of the second amendment simply says "The right to keep and bear arms shall not be infringed." You are, of course, correct that it doesn't define the scope of the right, nor even the meaning of "arms". All of that has to be worked out by the courts. So far all we know is that possession of "common" types of firearms may not be banned, but that the right is subject to "reasonable restrictions" (as are all the fundamental rights). No word on what the right to "bear" arms means at all, and nothing about what level of scrutiny must be applied to determine what restrictions are "reasonable".
For example, a country could honor this "right" by creating a radio station out in the middle of nowhere that people could drive to and say anything they want. You probably would be arrested afterwards for saying the wrong thing, and nobody in the country would be allowed to listen, but it'd honor this right enough to count.
No, that would violate the right in multiple ways. Allowing people to speak their minds only via this radio station would clearly "abridge the freedom of speech" by restricting where it can be done. Preventing the radio station from being re-broadcast elsewhere falls into the same category. And arresting people for saying the wrong thing would be a restriction on what can be said. Trying to argue otherwise is equivalent to arguing that murder isn't a crime because you're free to kill whomever you like, but will be arrested afterward. If behavior is penalized, then that behavior is not allowed.
That's not to say that location and content can never be restricted in any way. All of the fundamental rights are subject to reasonable restrictions when there is a clear and overriding public need. For example, deliberately giving military secrets to the enemy in wartime is treason, punishable by death. The classic example, of course, is shouting "FIRE!" in a crowded theater (in practice these days people would just look at you funny, but in the days before modern fire codes it was a deadly dangerous thing to do). The courts have accordingly established a process for evaluating any such restrictions, a procedure which requires the government to prove that any restriction is truly necessary, that the restriction is narrowly tailored and that it is the least restrictive possible means of achieving the goal. If the government cannot demonstrate all three of those, then the restriction is unconstitutional.
When did I ever propose any specific solution? There are many that would work. The point is that we have a real problem, that under the current system there IS much of value that is being lost, and we need to find some solution to address it. The purpose of copyright (the US form, anyway) was originally to increase the flow of material into the public domain, but pervasive, perpetual copyright is choking the public domain and causing some material to be lost forever. But unless there is a sea change in Congress' approach, that status quo is just going to continue and, indeed, to become more severe.
Even better, replace the comment with code that tests for the criteria.
In other words, write overcomplicated code that ships late?
Why would it be overcomplicated? If it is, that just means you've failed in another way.
As to shipping late... that's a decision that needs to be made jointly by business and engineering. It's called "technical debt", and the sort of comment you mention is technical debt, even if it's less bad than the same code without the comment.
In any case, that's only one example of commenting "why". Many of them cannot be tested in code at all.
Those that can't, can't. But many can. The key is to pause and think hard about whether there's any way to make the comment unnecessary. Comments are not a good, they're a necessary evil.
I suspect you're another one whose code isn't as self-documenting as you think it is.
I'm definitely still learning the art of comment-free programming, so my code has many more comments in it than I'd like. I work with some seriously awesome engineers, though, and I'm learning quickly.
One gentleman in particular really amazes me. His functions read like narratives, without any need for comments, and he always seems to find the perfect way to structure the solution to minimize both the lines of code and the effort required to understand them. The core of his talent is a true gift for finding the perfect name to capture the meaning of a class, function or variable -- or perhaps it's being able to precisely express the meaning of the name in code, no more and no less.
Someday I hope to be like him. But he's been writing code for 20 years longer than I have, and is smarter than me, to boot. So I'll probably never reach that level. But it's certainly a goal worth striving for.
The why matters a lot. Good comments should be things like/* I chose this algorithm because I expect the data to meet these criteria. */
Even better, replace the comment with code that tests for the criteria.
There are some cases where comments add value, but I'm increasingly of the opinion that they're few and far between (document-generating comments are an exception, but even they should avoid being redundant). Any time you feel the need to write an explanatory comment, step back and think hard about whether you can instead fix the code to make the comment unnecessary.
Why? What is it about GPLv3 that would have dissuaded Google from using Linux?
The ability to license Android to control-freak companies that don't want you to be able to root your phone? Because the GPLv3 requires that it includes all the information to install your own modified version, no more "our signed binaries only".
Google would actually prefer that control-freak companies not be able to lock users out. In that respect, GPLv3 would have worked better.
If "whole lot of people" cannot come up with enough money to get into agreement with copyright holders, then the crap is not worth it.
The problem is that no one knows who the copyright holders *are*. That's what is meant by "dead" or "orphaned" works. You can't pay someone if you don't know who to pay.
What about all of the old celluloid films which are disintegrating but can't be copied to preserve them because their copyright ownership is cloudy?
Nothing is stopping anyone from copying an old celluloid film (I'm assuming someone actually holds the old celluloid film in their hands) They just can't turn around and start distributing that copy. Of course if you don't have the film in your hand you can't copy it no matter what the law is
See Eldred. I'm not proposing a theoretical problem.
Copying old movies isn't cheap or easy. Without the ability to redistribute them, even people who are motivated out of pure altruism can't really afford to do it.
The fee is at the least needed to cover the cost of this system.
That's reasonable.
A second reason is that there won't be a "Copyright Reregistration Service, Inc." where for a one-time low-low-fee the service will automatically take care of all the needed registrations for you!
That's not actually a problem, as long as the registration service also keeps track of who the current and correct owner(s) are so that it can file correct registrations
The third reason is that it needs to stop "frivolous" registrations - there needs to be an expected net economic benefit to the copyright and the fee needs to be at a level where the expected benefit is sufficient to warrant payment of this fee.
Maybe. I'm not so sure that really matters. As long as there's some cost and some effort required, truly "dead" copyrights will be allowed to expire.
My perception is that copyrighted works fall into one of two categories: Works which someone actively cares about, whether or not they're actively exploiting them at the present, and works which are basically forgotten. IMO, one of the biggest problems we have right now is the non-emptiness of the second set, and I think setting even a very, very low barrier for copyright retention would fix that.
Note that I don't have any objection to schemes which dramatically shorten all copyrights, or that make it prohibitively expensive to maintain them long-term. I think both of those make perfect sense. But I think the problems they solve are much less important than the problem of dead works, and they'd obviously be harder to implement.
The fee doesn't really even need to be large. You could probably achieve the same results with no fee at all, actually, just a periodic re-registration requirement. The key is to ensure that the copyright holder still knows and cares about the work, and that others who are interested in doing something with the work can identify the owner.
Are people being jealous of fat useless leach members of "estates"? Let them have their dubious castles and kitschy art collections at the expense of fools who still pay for this old crap.
Nothing of value was lost?
What about all of the old celluloid films which are disintegrating but can't be copied to preserve them because their copyright ownership is cloudy?
The problem isn't people who are actively profiting from old works. The problem is old works that are locked up to the benefit of none and the detriment of all.
Numbers like these don't bode well for PC gamers and will only serve to encourage even more draconian DRM measures than we've seen in the past.
Thus only punishing customers who paid, not the people downloading the game illegally and applying a crack.
Makes perfect sense
And thus making the cracked version more useful than the retail version. It's not uncommon for people who buy a legitimate copy to download a cracked copy as well, because the cracked copy is less annoying. Increasing the DRM will increase piracy.
Barring a sea change in Congress' perception of copyright, we'll get another Mickey Mouse Protection Act by 2051. Remember, the Supreme Court already gave Congress permission to continue extending copyright indefinitely.
I'm sympathetic to mass suffering, and I think the world can be improved in all sorts of ways, but I still don't "get" Occupy. It seemed to be sort of a mass whine-in, without any clear goals. I'm all for protesting what's wrong with the world, and with trying to make it better, but you have to actually say what's wrong, and ideally propose how it can be fixed. Or if you can't actually articulate a solution, you ought to at least be able to describe the state you'd like to get to.
I do take issue with your characterization of Pascal's wager as a form of innumeracy. I would agree that the assumption that life extension has infinite value is a fallacy of some sort, which might be characterized as innumeracy, if you squint.
The point is that when you ascribe infinite value to a particular thing, then there should be no consideration of anything of finite value. For example, if you consider going to heaven of infinite value, then the obvious strategy is to kill yourself in an approved manner (say martyrdom) so that you don't risk backsliding at a future date.
Ah, but suicide is a sin, so that option's not available. Deliberately seeking out martyrdom is still suicide.
First, while people can easily remember a half-dozen common words, they're going to have a much harder time remembering a selection of words they've never heard of before. The xkcd suggestion of choosing from a restricted dictionary is more practical, but it drops the entropy from your suggested 115 bits to 66 bits (which is still slightly better than the 10-character password I suggested, but not hugely so).
Second, what you're talking about is passphrases that are 30-40 characters long. Half the web sites I use -- especially the financial ones -- won't accept more than 12 characters, and a good number won't take more than 8.
Third, even if people can remember the words, and how to spell them, and web sites will allow them, how many people can quickly and accurately type them, especially when they can't see what they're typing? I couldn't.
Selecting six words from an extremely large set would provide a great deal of entropy, but it's not very practical.
However, I certainly do concede that it is possible to choose passwords/passphrases that provide long-term resistance against brute force attacks. But few people will do it -- and many web sites won't even allow it. Given the other avenues of attack (shoulder surfing, mistakenly typing a password the wrong place, unsrupulous web admins), the most practical method, at present, is to use unique per-site passwords that are moderately long and complex, unique per site, and change them periodically.
I'm a big fan of OpenID for this reason. It allows me to have one fairly strong password that my fingers can type quickly (because I use it a lot), plus a second authentication factor (OTP generator on my phone), and to use that same login credential at a lot of web sites. But just try to convince your bank that they should trust Google, or Blizzard, to handle their client authentication for them -- in spite of the fact that they do a far better job. Even if Verisign or some similar "trusted" company were to offer strong OpenIDs with multi-factor authentication, it'd still be tough to get the banks and other important sites to trust them.
Something like that is where we've got to go, though. Password-only authentication isn't a viable long-term strategy, and it's not going to be practical to have a different second factor token for every site you use.
Slashdot Mirror
The real power is in the corporations. They buy the politicians (oops I meant "lobby") and get their way every single time.
I agree that corporate influence over politics is a problem.
We erode the Bill of Rights and nobody bats an eye at it. We can now indefinitely imprison anyone that might be a terrorist. So there goes Habeas Corpus. First amendment is shot to hell. What do you think the founding fathers would think of "free speech zones"? I have a feeling they'd be loading muskets. Second amendment? Also boned. Nagin after Katrina went through the gun licenses and ordered the national guard to confiscate every gun in the city. And they did. It goes on and on. Warantless wiretaps, GPS tracking devices without a court order, Carnivore...you no longer have hardly any rights at all, but you don't even notice it. They really did a number on us. A real pro job.
However, I can't see how the above can be attributed to corporate influence. There are some areas... big content has way too much influence over copyright law, and corporations are far too often shielded from the effects of their own bad acts -- and even bailed out by the government if they're too big to fail.
But what's the corporate interest in strip-searching airline passengers, or confining political speech to designated zones away from politicians, or getting the government to confiscate guns, or to install GPS tracking devices on people or read their e-mail, etc.?
I assert that none of those things are the result of corporate influence, because big business couldn't care less about them. Some of them, such as the warrantless wiretaps, actually bring heat down on the corps, and others, like all of the interference with business travel, make their lives harder.
No, I don't think you can lay any of that at the feet of corporate lobbyists. All that other stuff is just politicians trying to do what they think the people want them to do.
I seriously doubt the "electronic junk" adds much at all, not compared to the weight expended on improved crash protection. The bodies of modern cars, before any of that other junk is added, are 50% heavier than those of similar cars 30 years ago.
Your telling me that winter tires are not mandatory in wintery states in the US? Not only are they not mandatory, but in many states they are, in fact, illegal.
Cite?
Studded tires are illegal in many states, but not snow tires. The main difference between snow tires and all-weather tires is that the snow tires are made of softer rubber. The harder rubber of all-weather tires is too stiff in cold weather to grip the ground well.
You're information is only somewhat accurate. A typical interview path for Google consists of about 7 interviews during which you're interviewed by at least a couple different people and although there is no official stance on asking such questions there are most definitely interviewers that do.
That information is also somewhat dated.
There has been an effort to standardize the Google interview process, and except in unusual circumstances, here's how it works:
The first step is a phone screen. This is a 45-minute phone call, during which the interviewer poses a series of hypothetical programming problems. Many phone screeners will set up a shared Writer document in Google Docs and ask the candidate to write some code, but many don't.
The purpose of the phone screen is purely to evaluate whether or not the candidate should be brought in for an on-site interview. The screener doesn't participate further in the hiring process, and the results of the screen don't go to the hiring committee, other than in the sense that if the candidate doesn't pass the screen the committee will never hear of the candidate.
The next step is the on-site interview. This consists of five one-hour interviews with five different people. The third "interview" however, is just lunch. The Google employee assigned for that hour takes the candidate to lunch and answers questions about the company, their job, etc. and just generally socializes. That employee doesn't provide any hiring feedback.
The other four interviews are what really matters. HR defines the standard set of off-limits questions -- anything relevant to gender, religion, ethnicity, economic background, etc. There is also a list of engineering questions which are banned, some because they're not very useful, most because they're too well-known, published on the web, etc. Outside of that the interviewer can do what they like, with the goal of evaluating the candidate's problem-solving, coding, design and architecture abilities, as well as "Googliness" (culture fit), but the approach should be centered around solving problems.
The interviewers do not compare notes or consult with one another. There is a piece of paper on which each jots down the technical problems presented and the areas tested (e.g. coding, design, etc.). They don't write down how the candidate performed; the purpose of the paper is just to ensure that they all use different problems and adequately cover all of the areas of ability that need to be evaluated.
After the interviews, the interviewers still don't consult with one another, in order to avoid one influencing the opinion of another. Each writes up a report with a description of the problems presented, how the candidate responded (including the code the candidate wrote) and their overall evaluation. The report culminates in a hire/no-hire recommendation and a numeric score between 1 (I'll quit if you hire this person) and 4 (I'll quit if you don't hire this person). Each interviewer's scores are tracked for calibration purposes.
The next step is that the reports are forwarded to the local hiring committee. This committee typically meets bi-weekly, so this often introduces a significant delay into the decision process. The committee evaluates the candidates based on their resume and on the reports and makes a hire/no-hire decision. The candidate's packet is also forwarded to another, corporate, hiring committee. This committee normally follows the recommendation of the lower committee but in some cases does reverse it.
That's the process, and the technical questions definitely are not brain-teaser type problems. They're abstract and generally somewhat contrived, because that's all there's time for. But questions that rely on some flash of insight are not appropriate, nor are questions that depend on knowledge of trivia, and interviewers that use them regularly will be taken to task by the hiring committee. Good questions permit ma
My response to questions like that improved dramatically when I read an article that explained questions way out of left field like that are intended to test your problem solving ability, so do your best to estimate an answer and explain your thought process. Reading that article didn't make me better at debugging hard technical issues, but made me dramatically better at handling off-the-wall interview questions nimbly. You're not measuring what you think you're measuring.
When I interviewed at Google, every one of the five people who interviewed me explained that they were interested in understanding how I approached and solved problems, and that it mattered less that I succeeded at solving the problem than that I demonstrated how I went about trying to solve it. They all asked me to please explain my thought processes and not to worry about making sure everything I said was perfectly correct.
Does that include the right to hear what you don't want to hear or be subject to advertising?
I'm not sure what "right to hear what you don't want to hear" means. You have the freedom to ignore what you want to ignore. You can plug your ears, close your eyes, ban speech you don't like from your property, etc.
The right for idiots worldwide to make shallow judgments about you that you can't ignore?
Why could you not ignore them? But, in general, no you have no right to control what others think or believe.
Compare it to the Second Amendment which basically says there is some sort of right to bear arms, but doesn't say much about how that is to be implemented other than saying there is a "militia" that is "regulated" which should be free to bear arms (which are not specified).
Actually, SCOTUS decided in Heller that the militia portion was just a motivational clause and not descriptive of the nature of the right. So the operative part of the second amendment simply says "The right to keep and bear arms shall not be infringed." You are, of course, correct that it doesn't define the scope of the right, nor even the meaning of "arms". All of that has to be worked out by the courts. So far all we know is that possession of "common" types of firearms may not be banned, but that the right is subject to "reasonable restrictions" (as are all the fundamental rights). No word on what the right to "bear" arms means at all, and nothing about what level of scrutiny must be applied to determine what restrictions are "reasonable".
For example, a country could honor this "right" by creating a radio station out in the middle of nowhere that people could drive to and say anything they want. You probably would be arrested afterwards for saying the wrong thing, and nobody in the country would be allowed to listen, but it'd honor this right enough to count.
No, that would violate the right in multiple ways. Allowing people to speak their minds only via this radio station would clearly "abridge the freedom of speech" by restricting where it can be done. Preventing the radio station from being re-broadcast elsewhere falls into the same category. And arresting people for saying the wrong thing would be a restriction on what can be said. Trying to argue otherwise is equivalent to arguing that murder isn't a crime because you're free to kill whomever you like, but will be arrested afterward. If behavior is penalized, then that behavior is not allowed.
That's not to say that location and content can never be restricted in any way. All of the fundamental rights are subject to reasonable restrictions when there is a clear and overriding public need. For example, deliberately giving military secrets to the enemy in wartime is treason, punishable by death. The classic example, of course, is shouting "FIRE!" in a crowded theater (in practice these days people would just look at you funny, but in the days before modern fire codes it was a deadly dangerous thing to do). The courts have accordingly established a process for evaluating any such restrictions, a procedure which requires the government to prove that any restriction is truly necessary, that the restriction is narrowly tailored and that it is the least restrictive possible means of achieving the goal. If the government cannot demonstrate all three of those, then the restriction is unconstitutional.
When did I ever propose any specific solution? There are many that would work. The point is that we have a real problem, that under the current system there IS much of value that is being lost, and we need to find some solution to address it. The purpose of copyright (the US form, anyway) was originally to increase the flow of material into the public domain, but pervasive, perpetual copyright is choking the public domain and causing some material to be lost forever. But unless there is a sea change in Congress' approach, that status quo is just going to continue and, indeed, to become more severe.
In other words, write overcomplicated code that ships late?
Why would it be overcomplicated? If it is, that just means you've failed in another way.
As to shipping late... that's a decision that needs to be made jointly by business and engineering. It's called "technical debt", and the sort of comment you mention is technical debt, even if it's less bad than the same code without the comment.
In any case, that's only one example of commenting "why". Many of them cannot be tested in code at all.
Those that can't, can't. But many can. The key is to pause and think hard about whether there's any way to make the comment unnecessary. Comments are not a good, they're a necessary evil.
I suspect you're another one whose code isn't as self-documenting as you think it is.
I'm definitely still learning the art of comment-free programming, so my code has many more comments in it than I'd like. I work with some seriously awesome engineers, though, and I'm learning quickly.
One gentleman in particular really amazes me. His functions read like narratives, without any need for comments, and he always seems to find the perfect way to structure the solution to minimize both the lines of code and the effort required to understand them. The core of his talent is a true gift for finding the perfect name to capture the meaning of a class, function or variable -- or perhaps it's being able to precisely express the meaning of the name in code, no more and no less.
Someday I hope to be like him. But he's been writing code for 20 years longer than I have, and is smarter than me, to boot. So I'll probably never reach that level. But it's certainly a goal worth striving for.
The why matters a lot. Good comments should be things like /* I chose this algorithm because I expect the data to meet these criteria. */
Even better, replace the comment with code that tests for the criteria.
There are some cases where comments add value, but I'm increasingly of the opinion that they're few and far between (document-generating comments are an exception, but even they should avoid being redundant). Any time you feel the need to write an explanatory comment, step back and think hard about whether you can instead fix the code to make the comment unnecessary.
Why? What is it about GPLv3 that would have dissuaded Google from using Linux?
The ability to license Android to control-freak companies that don't want you to be able to root your phone? Because the GPLv3 requires that it includes all the information to install your own modified version, no more "our signed binaries only".
Google would actually prefer that control-freak companies not be able to lock users out. In that respect, GPLv3 would have worked better.
If "whole lot of people" cannot come up with enough money to get into agreement with copyright holders, then the crap is not worth it.
The problem is that no one knows who the copyright holders *are*. That's what is meant by "dead" or "orphaned" works. You can't pay someone if you don't know who to pay.
Seriously, Google Eldred.
GPL3 licensed code in the Linux kernel would have made a huge difference to people building their own versions of android to install on phones.
While I wish we had that - a GPL3 licensed Linux kernel would not have been used in android. It probably would have been a BSD derivative.
Why? What is it about GPLv3 that would have dissuaded Google from using Linux?
I believe I answered that in the subject and in the expression "old crap".
One man's "old crap" is "irreplaceable film history" to a whole lot of people. Seriously, did you google Eldred?
What about all of the old celluloid films which are disintegrating but can't be copied to preserve them because their copyright ownership is cloudy?
Nothing is stopping anyone from copying an old celluloid film (I'm assuming someone actually holds the old celluloid film in their hands) They just can't turn around and start distributing that copy. Of course if you don't have the film in your hand you can't copy it no matter what the law is
See Eldred. I'm not proposing a theoretical problem.
Copying old movies isn't cheap or easy. Without the ability to redistribute them, even people who are motivated out of pure altruism can't really afford to do it.
The fee is at the least needed to cover the cost of this system.
That's reasonable.
A second reason is that there won't be a "Copyright Reregistration Service, Inc." where for a one-time low-low-fee the service will automatically take care of all the needed registrations for you!
That's not actually a problem, as long as the registration service also keeps track of who the current and correct owner(s) are so that it can file correct registrations
The third reason is that it needs to stop "frivolous" registrations - there needs to be an expected net economic benefit to the copyright and the fee needs to be at a level where the expected benefit is sufficient to warrant payment of this fee.
Maybe. I'm not so sure that really matters. As long as there's some cost and some effort required, truly "dead" copyrights will be allowed to expire.
My perception is that copyrighted works fall into one of two categories: Works which someone actively cares about, whether or not they're actively exploiting them at the present, and works which are basically forgotten. IMO, one of the biggest problems we have right now is the non-emptiness of the second set, and I think setting even a very, very low barrier for copyright retention would fix that.
Note that I don't have any objection to schemes which dramatically shorten all copyrights, or that make it prohibitively expensive to maintain them long-term. I think both of those make perfect sense. But I think the problems they solve are much less important than the problem of dead works, and they'd obviously be harder to implement.
The fee should have an N^2 or 2^N or N! factor, where N is the number of years since expiration.
Meh. That might be satisfying, but it's unlikely to be enacted, and it's not necessary to solve the dead works problem.
+1
The fee doesn't really even need to be large. You could probably achieve the same results with no fee at all, actually, just a periodic re-registration requirement. The key is to ensure that the copyright holder still knows and cares about the work, and that others who are interested in doing something with the work can identify the owner.
Are people being jealous of fat useless leach members of "estates"? Let them have their dubious castles and kitschy art collections at the expense of fools who still pay for this old crap.
Nothing of value was lost?
What about all of the old celluloid films which are disintegrating but can't be copied to preserve them because their copyright ownership is cloudy?
The problem isn't people who are actively profiting from old works. The problem is old works that are locked up to the benefit of none and the detriment of all.
Numbers like these don't bode well for PC gamers and will only serve to encourage even more draconian DRM measures than we've seen in the past.
Thus only punishing customers who paid, not the people downloading the game illegally and applying a crack.
Makes perfect sense
And thus making the cracked version more useful than the retail version. It's not uncommon for people who buy a legitimate copy to download a cracked copy as well, because the cracked copy is less annoying. Increasing the DRM will increase piracy.
Barring a sea change in Congress' perception of copyright, we'll get another Mickey Mouse Protection Act by 2051. Remember, the Supreme Court already gave Congress permission to continue extending copyright indefinitely.
But we're talking about 17th-century French Catholicism, not the whole wide world of religious ideas and motives.
Aha!
I think I finally get Occupy! It wasn't intended to change anything, it was just people wanting to vent.
I'm sympathetic to mass suffering, and I think the world can be improved in all sorts of ways, but I still don't "get" Occupy. It seemed to be sort of a mass whine-in, without any clear goals. I'm all for protesting what's wrong with the world, and with trying to make it better, but you have to actually say what's wrong, and ideally propose how it can be fixed. Or if you can't actually articulate a solution, you ought to at least be able to describe the state you'd like to get to.
I do take issue with your characterization of Pascal's wager as a form of innumeracy. I would agree that the assumption that life extension has infinite value is a fallacy of some sort, which might be characterized as innumeracy, if you squint.
The point is that when you ascribe infinite value to a particular thing, then there should be no consideration of anything of finite value. For example, if you consider going to heaven of infinite value, then the obvious strategy is to kill yourself in an approved manner (say martyrdom) so that you don't risk backsliding at a future date.
Ah, but suicide is a sin, so that option's not available. Deliberately seeking out martyrdom is still suicide.
Blaise Pascal was not innumerate.
There are multiple problems with your proposal.
First, while people can easily remember a half-dozen common words, they're going to have a much harder time remembering a selection of words they've never heard of before. The xkcd suggestion of choosing from a restricted dictionary is more practical, but it drops the entropy from your suggested 115 bits to 66 bits (which is still slightly better than the 10-character password I suggested, but not hugely so).
Second, what you're talking about is passphrases that are 30-40 characters long. Half the web sites I use -- especially the financial ones -- won't accept more than 12 characters, and a good number won't take more than 8.
Third, even if people can remember the words, and how to spell them, and web sites will allow them, how many people can quickly and accurately type them, especially when they can't see what they're typing? I couldn't.
Selecting six words from an extremely large set would provide a great deal of entropy, but it's not very practical.
However, I certainly do concede that it is possible to choose passwords/passphrases that provide long-term resistance against brute force attacks. But few people will do it -- and many web sites won't even allow it. Given the other avenues of attack (shoulder surfing, mistakenly typing a password the wrong place, unsrupulous web admins), the most practical method, at present, is to use unique per-site passwords that are moderately long and complex, unique per site, and change them periodically.
I'm a big fan of OpenID for this reason. It allows me to have one fairly strong password that my fingers can type quickly (because I use it a lot), plus a second authentication factor (OTP generator on my phone), and to use that same login credential at a lot of web sites. But just try to convince your bank that they should trust Google, or Blizzard, to handle their client authentication for them -- in spite of the fact that they do a far better job. Even if Verisign or some similar "trusted" company were to offer strong OpenIDs with multi-factor authentication, it'd still be tough to get the banks and other important sites to trust them.
Something like that is where we've got to go, though. Password-only authentication isn't a viable long-term strategy, and it's not going to be practical to have a different second factor token for every site you use.