It goes way back: Rowan Atkinson (9th) -> Richard E Grant (10th) -> Hugh Grant (11th) -> Jonathan Pryce (12th) -> Joanna Lumley (13th). The writer (Steven Moffat) in The Curse of Fatal Death (1999) foretold that the 13th doctor will be a blond woman.
how can you call a language complete if it does not have a GOTO statement ? I'm not saying that I use it a lot, but it is useful for handling errors & similar
The valid use-case of GOTO I have seen is recourse cleanup. For that rust has RAII approach. Breaking out of multiple for loops can be a bit more tricky, but those can still be done rather painlessly and if not, use of GOTO would not help with understanding what the code actually does.
While mathematically interesting, in real life people will figure out you are using (c) option. If the checks were any deterrent at all, less offending folks will start to offend more.
Your basic AI "expert system" outputs a result, a confidence level, and self-adjusts when it's fed training data (including being re-fed data it gave the wrong result for along with the desired result/correction).
If updated data is contains systemic errors, statistical models can't deal with that. No confidence levels can help you if the input is systematically biased. Remember those “faster than light” neutrinos? Statistical models showed they were FTL and that error was not fixed until the measurement bias was removed.
Because idiots need to be told they're idiots.
Just a friendly reminder — each and every one is an idiot by someone else's standard. It seems that you needed to be told that.
The training data doesn't have to include a “race” field, for the algorithm to figure out the race of a person. If other fields (the more the better) correlate with race, the algo can figure out that there exist a distinct group of people, that share common properties. The algo then can look up the arrest / conviction / parole / whatever information on this “group of people” and come to conclusion that this group is “risky”. Problems raises when the training data contains racially biased arrests / convictions and so on.
Another side effect of the incomprehensible risk assessments is that it doesn't help with rooting out causes of criminal behaviour. So far the AI can say “this person is risk to the society”, but it can't say that “this person is risk because they are economically desperate” or “this person is unlikely to comply with parole because they debilitating mental disorder”. Sure, if society believes that isolation from society is the last and the first resort in dealing with harmful individuals, causes are not important, but I believe that this approach is both inhumane and uneconomical.
incarceration rates for different races is different; differences between races are insignificant; therefore AI is biased
then your rebuttal would have been valid. However, the issue here is bias in training data. If AI learns from what the law enforcement / judiciary feed them, then the AI will reflect the biases of said institutions.
You seem to suggest that all the biases the justice system has is well founded and reflecting the reality. I find it hard to believe.
To assist reading, I would prefer using keywords. For example, 'not A and B' rather than '!A && B'. It seems to improve readability, less error prone (no accidental 'A & B', which is 'A bitand B' with keywords).
I don't know your teaching methodologies, but please watch the Stop Teaching C presentation by Kate Gregory, where she shares her experience on teaching C++ and how to make it less confusing for novices.
In that case it shows how far the Linux ecosystem has matured. I have been using Fedora for couple of years and latest Fedora Alpha had no issues (for me, at least). Ten years ago even LTS Ubuntu was bag of bugs in comparison.
Hmm, “doing well” is a vague term. I would describe those as having future prospects open for the inhabitants, which roughly means education, nature protection, reasonably stable political environment and wealth. On these counts Northern Europe (Iceland, Norway, Denmark, Sweden, Finland and Estonia) are going pretty well. If Russia invaded, I would certainly also consider Netherlands, Ireland, UK, France, Germany, Switzerland and Austria as go to places.
Orthogonal or not, if they get into government, they will have to vote for not just their pet issues. I suppose that it just means that they would support whatever economic system they have at the time.
From the top of my head I can't remember what kind of economic system they support. Skimming the wikipedia page gives me an impression that it is regulated capitalism.
The Soviet system is corrupt, but they couldn't imagine any other way of living.
Who are “they” exactly? Even on paper the nomenklatura was working towards communism (they just had no idea how to get there). Even those not in power had in their living memories times of democracies and nationalisms.
Capitalism produces bad results, but communism is apparently the only other option and is reportedly terrible
Then be so kind and tell us about those other options. Seriously. We have had feudalism, mercantilism, autarky, collectivism, communism, wild capitalism, state capitalism, regulated capitalism, libertarianism. None of these seems to “work”.
Do cars really get scrapped under 150k miles now adays?
Unlikely. They tend to get exported to “poorer” countries. For example, I drive a 15yo car with 300Gm on the road and I gather I could easily sell it for 3000 euros.
You talk as if it is a bad thing. Let's take banal an example. Some 12-year-old is smoking cigarettes at school and a teacher has spotted him doing so. Let's say the teacher has 4 options: 1) do nothing, 2) tell the parents and the principal, 3) threaten to do do the (2) option, if is caught smoking again 4) threaten to do do the (2) option, unless money/favours are given.
I think we all can agree that the option (4) is morally wrong thing to do. Whether option (3) is morally acceptable is depending on morality of option (2). I believe that if (2) is ok, (3) is arguably ok as well.
In this particular case I think the CNN case is more like the option (3). CNN has full rights to lift anonymity regardless of the subject wishes. If you think that news organizations should be morally compelled to not to reveal the name of someone using pseudonyms against their will, what is the justification?
Saying you don't want to answer is far better than some sort of ridiculous attack.
I did answer explicitly, and the answer was, and I quote, Nope. Then I went on explaining that this position was not set in stone and if there is good evidence to do so, this position should be reconsidered.
You are the one that attacked an entire culture instead of discussing.
Why shouldn't I attack an entire culture? And why would you take offence if you are not part of it?
Politeness [...] doesn't really work if the misinformation is persistent [...]
I do understand the sentiment, but in my experience being rude doesn't do the trick either. I am not a psychologist, but my understanding is that when people get attacked, they begin to stick to their misconceptions even harder. After that there is even more incentive to defend the indefensible, because now it is personal. I guess I have just answered the “Why shouldn't I attack an entire culture?” question.
I dwell on the opposite side of the world to where I do.
Ok, so where do you live and what do you believe in? I assumed you are form USA because/. is US centric and my first example was from US media landscape, which you had no issues with.
you've gone and attacked a strawman
Ok, let me attack you personally. Why are you afraid of discussing issues that might possibly lead to banning stuff?
Surely you can do much better than such stupidity?
I took the liberty to take a peek in your comment history and it seems you tent do insult in a condescending tone. Has this helped you to have an intelligent conversation where you have gained new insights?
This method would help to determine if the “therapy” changes the perception of children (which is certainly useful), but I don't think that it would establish if the person is more or less likely to directly or indirectly harm children. In this particular case, if arousal is brought to nil, it would make sense that the person is less of a threat, but if the levels stay the same or are increased, it tells very little about harmful behaviour.
Slashdot Mirror
It goes way back: Rowan Atkinson (9th) -> Richard E Grant (10th) -> Hugh Grant (11th) -> Jonathan Pryce (12th) -> Joanna Lumley (13th). The writer (Steven Moffat) in The Curse of Fatal Death (1999) foretold that the 13th doctor will be a blond woman.
British TV shows tend to be written by the show creators, a relatively small team
I should add that roughly half of Doctor Who episodes are written by “guest writers”, but Doctor Who is not a typical British TV example.
The toll bot was misconfigured. The stdout and stderr were pointing to the same object.
Oh, you cheeky bastard! Mod this up!
how can you call a language complete if it does not have a GOTO statement ? I'm not saying that I use it a lot, but it is useful for handling errors & similar
The valid use-case of GOTO I have seen is recourse cleanup. For that rust has RAII approach. Breaking out of multiple for loops can be a bit more tricky, but those can still be done rather painlessly and if not, use of GOTO would not help with understanding what the code actually does.
You should examine your own weasel words (namely, “competent”) before casting stones.
Honestly, I want to see YouTube just outright removing all music videos licensed under big labels that lobby under RIAA just to see what would happen
So, let me get this straight. You want Youtube to demonstrate it's dominant position in the market, by abusing it... in the EU? Are you new here?
What success rate do we pick?
While mathematically interesting, in real life people will figure out you are using (c) option. If the checks were any deterrent at all, less offending folks will start to offend more.
Your basic AI "expert system" outputs a result, a confidence level, and self-adjusts when it's fed training data (including being re-fed data it gave the wrong result for along with the desired result/correction).
If updated data is contains systemic errors, statistical models can't deal with that. No confidence levels can help you if the input is systematically biased. Remember those “faster than light” neutrinos? Statistical models showed they were FTL and that error was not fixed until the measurement bias was removed.
Because idiots need to be told they're idiots.
Just a friendly reminder — each and every one is an idiot by someone else's standard. It seems that you needed to be told that.
The training data doesn't have to include a “race” field, for the algorithm to figure out the race of a person. If other fields (the more the better) correlate with race, the algo can figure out that there exist a distinct group of people, that share common properties. The algo then can look up the arrest / conviction / parole / whatever information on this “group of people” and come to conclusion that this group is “risky”. Problems raises when the training data contains racially biased arrests / convictions and so on.
Another side effect of the incomprehensible risk assessments is that it doesn't help with rooting out causes of criminal behaviour. So far the AI can say “this person is risk to the society”, but it can't say that “this person is risk because they are economically desperate” or “this person is unlikely to comply with parole because they debilitating mental disorder”. Sure, if society believes that isolation from society is the last and the first resort in dealing with harmful individuals, causes are not important, but I believe that this approach is both inhumane and uneconomical.
If the argument would have been
then your rebuttal would have been valid. However, the issue here is bias in training data. If AI learns from what the law enforcement / judiciary feed them, then the AI will reflect the biases of said institutions.
You seem to suggest that all the biases the justice system has is well founded and reflecting the reality. I find it hard to believe.
To assist reading, I would prefer using keywords. For example, 'not A and B' rather than '!A && B'. It seems to improve readability, less error prone (no accidental 'A & B', which is 'A bitand B' with keywords).
I don't know your teaching methodologies, but please watch the Stop Teaching C presentation by Kate Gregory, where she shares her experience on teaching C++ and how to make it less confusing for novices.
Fedora is the alpha test release for RHEL
In that case it shows how far the Linux ecosystem has matured. I have been using Fedora for couple of years and latest Fedora Alpha had no issues (for me, at least). Ten years ago even LTS Ubuntu was bag of bugs in comparison.
Was that a description of Apple walled garden or spiralling debt?
I can get pretty much any phone for a dollar if I get a new mobile contract.
How does contract costs change if you get it without phone? Isn't it just a hidden leasing?
Hmm, “doing well” is a vague term. I would describe those as having future prospects open for the inhabitants, which roughly means education, nature protection, reasonably stable political environment and wealth. On these counts Northern Europe (Iceland, Norway, Denmark, Sweden, Finland and Estonia) are going pretty well. If Russia invaded, I would certainly also consider Netherlands, Ireland, UK, France, Germany, Switzerland and Austria as go to places.
Orthogonal or not, if they get into government, they will have to vote for not just their pet issues. I suppose that it just means that they would support whatever economic system they have at the time.
Socialism and regulated capitalism seems to work reasonably well.
Sorry, my bad. When I hear “the west”, I assume it's Scandinavia, UK, France, Germany, which is more of a mixed bad.
This is why I vote for the pirate party.
From the top of my head I can't remember what kind of economic system they support. Skimming the wikipedia page gives me an impression that it is regulated capitalism.
The Soviet system is corrupt, but they couldn't imagine any other way of living.
Who are “they” exactly? Even on paper the nomenklatura was working towards communism (they just had no idea how to get there). Even those not in power had in their living memories times of democracies and nationalisms.
Capitalism produces bad results, but communism is apparently the only other option and is reportedly terrible
Then be so kind and tell us about those other options. Seriously. We have had feudalism, mercantilism, autarky, collectivism, communism, wild capitalism, state capitalism, regulated capitalism, libertarianism. None of these seems to “work”.
Do cars really get scrapped under 150k miles now adays?
Unlikely. They tend to get exported to “poorer” countries. For example, I drive a 15yo car with 300Gm on the road and I gather I could easily sell it for 3000 euros.
It's coercion, plain and simple
You talk as if it is a bad thing. Let's take banal an example. Some 12-year-old is smoking cigarettes at school and a teacher has spotted him doing so. Let's say the teacher has 4 options: 1) do nothing, 2) tell the parents and the principal, 3) threaten to do do the (2) option, if is caught smoking again 4) threaten to do do the (2) option, unless money/favours are given.
I think we all can agree that the option (4) is morally wrong thing to do. Whether option (3) is morally acceptable is depending on morality of option (2). I believe that if (2) is ok, (3) is arguably ok as well.
In this particular case I think the CNN case is more like the option (3). CNN has full rights to lift anonymity regardless of the subject wishes. If you think that news organizations should be morally compelled to not to reveal the name of someone using pseudonyms against their will, what is the justification?
Saying you don't want to answer is far better than some sort of ridiculous attack.
I did answer explicitly, and the answer was, and I quote, Nope. Then I went on explaining that this position was not set in stone and if there is good evidence to do so, this position should be reconsidered.
You are the one that attacked an entire culture instead of discussing.
Why shouldn't I attack an entire culture? And why would you take offence if you are not part of it?
Politeness [...] doesn't really work if the misinformation is persistent [...]
I do understand the sentiment, but in my experience being rude doesn't do the trick either. I am not a psychologist, but my understanding is that when people get attacked, they begin to stick to their misconceptions even harder. After that there is even more incentive to defend the indefensible, because now it is personal. I guess I have just answered the “Why shouldn't I attack an entire culture?” question.
I dwell on the opposite side of the world to where I do.
Ok, so where do you live and what do you believe in? I assumed you are form USA because /. is US centric and my first example was from US media landscape, which you had no issues with.
you've gone and attacked a strawman
Ok, let me attack you personally. Why are you afraid of discussing issues that might possibly lead to banning stuff?
Surely you can do much better than such stupidity?
I took the liberty to take a peek in your comment history and it seems you tent do insult in a condescending tone. Has this helped you to have an intelligent conversation where you have gained new insights?
This method would help to determine if the “therapy” changes the perception of children (which is certainly useful), but I don't think that it would establish if the person is more or less likely to directly or indirectly harm children. In this particular case, if arousal is brought to nil, it would make sense that the person is less of a threat, but if the levels stay the same or are increased, it tells very little about harmful behaviour.