Thanks. Let me try to write what you said, sticking to facts and neutral POV. I tried to remove those of your answers which didn't seem to be factual statements of agreement/disagreement with the literal text that was written, but might have made some mistakes.
* We believe that interpersonal skills are at least as important as technical skills. - You believe instead that technical skills are more important than technical skills (either in software in general, or OSS in general, or Linux in particular? unclear).
* We can add the most value as professionals by drawing on the diversity... - About race/sex diversity, you believe that the quantity/quality of a team's technical contributions is unaffected by this. About background/international diversity, you believe that the quantity/quality of a team's technical contributions is increased a little by this in some areas. (I'm reluctant to introduce the word "team" here, but it's the cleanest way I can see to define the question and then reconcile your answer. I could have talked about "contributor-pool" but then we'd get into subtle definitions.)
* We have the obligation to use our positions of privilege, however tenuous, to improve the lives of others. - You believe instead that there is no such obligation. (or, I guess, you believe it's an acceptable exercise of free speech for a project to adopt this as a rule, but you believe the quality of the project will suffer for it).
* We must make room for people who are not like us to enter our field and succeed there. - You believe instead that there's no obligation on anyone to make room.
* We have an ethical responsibility to refuse to work on software that will negatively impact the well-being of other people. - You instead believe there is no such ethical responsibility, maybe because every action has some positive and some negative impact. (It was unclear to me how this relates - is there a logical step which says "anything which can be used for good and bad obviates the developer of evaluating the good/bad likely effects because that's solely down to the users"?)
* We acknowledge the value of non-technical contributors as equal to the value of technical contributors. - You believe instead that that value of technical contributors is higher.
* We understand that working in our field is a privilege, not a right. - You instead believe that working in OSS is a right, indeed just a special case of the freedom of speech+expression. (I don't think that you and the manifesto have the same definition of "field" and "work" on this point?)
* The negative impact of toxic people in the workplace or the larger community is not offset by their technical contributions. - You instead believe that "toxic person" is too loaded/dangerous a term to ever be used (in this context? or in general? unclear).
(I'm not trying to argue anything here nor disagree with you. I'm trying to neutrally characterize+understand people's positions. The manifesto had seemed mild and inoffensive to me on first reading; in the light of other people's reactions I need to dig deeper).
Tell me: which specific bullet-points or sentences in that manifesto did you disagree with? Or did you disagree with implications or consequences of the manifesto rather than something written in it?
We've read similar wording a thousand times, be it in academia, in schools or in corporate codes of conduct and the purpose is always the same...
Should I understand from the way you chose to answer that you don't disagree with any individual bullet-points or sentences in the manifesto, but you do disagree with the implications and consequences (and also motives)?
Actually, this is generally what happens to a project. It adopts a CoC or enough anonymous/SJW complaints are created, big contributors leave or are forced out, the project dies or slows because it is spending inane amounts of energy on political discussions rather than technical. It happened with NodeJS, Kubernetes, LLVM, Tor, Debian all of which are decent projects but kind of 'stuck' now that boards and bug lists are overrun by "complaints" rather than technical discussions.
I wondered if what you said is true? Here's a graph of LLVM contributions. I pulled it from github's llvm-mirror/llvm repo, and used github's "Insights > Contributions" tab to display the graph. Github claimed that the graph shows up to September 2018 -- that's contradicted by the labels and tickmarks. I wonder if github's graph-plotting routines don't get the label+tickmark positions quite right? I aimed to include roughly equal spans before and after the code of conduct to see if there's a clear effect.
My conclusion? I don't see a clear signal. The graph seems to have roughly the same noise level and averages before and after, and a better narrative would be "there are sometimes huge sustained spikes of activity for reasons presumably related to the content of the work". (This is just from eyeballing it; I can't be bothered to put the numbers into a spreadsheet and do the calculations more carefully). If you say that LLVM is now dead, that's not supported by the data. If you say it's slowed, that's plausible (we'd need numbers), but the data says there hasn't been a major slowdown.
If you believe a different numerical analysis would be better, one that takes into account the quality of the contributions, then yes please I'd love for you to provide that! Or for you to provide equivalent graphs for the other projects you listed.
I'd say the title doesn't say it all, nor even much of it -- the title seems to be intentionally written to sound paradoxical and so entice you to read further. Based on the first four paragraphs, I think the title would be more accurate (but more wordy) if it said "Meritocracy is obviously a good principle in theory -- but in practice the actual measurement of merit has been so inaccurate and biased that we could collectively be more productive if we think through what we're trying to achieve and figure out other principles that are more practically effective."
Tell me: which specific bullet-points or sentences in that manifesto did you disagree with? Or did you disagree with implications or consequences of the manifesto rather than something written in it?
What's a nice way to say that the code is a total disaster area and that it has no business being near the kernel?
"Thanks for your contribution. This diff you sent touches a lot of areas of code, which means we'll need a more careful test matrix to make sure we've shaken out every possible consequence. Also, past experience has shown that changes to files XYZ have been correlated with a high rate of undiscovered bugs. I agree that this is an important area, but let's step back and see if we can figure out a different plan of attack."
(You just said "total disaster area", so I had to fill in one possible scenario for why it's a total disaster area, so that I could give a constructive reason. One of the key parts about being nice/respectful/mentoring is to give constructive feedback.)
What if "code improve" means that it needs to rewritten from scratch by someone else who actually understands?
"Let's step back and see if we can figure out a different plan of attack."
Plenty of people simply have the wrong idea about how to solve a problem... In order words, you just want to fire them from the project, because they aren't good enough. What's a good inclusive and welcoming way to fire someone?
That's a bad reason to fire someone. You should fire someone either because they lack POTENTIAL to get up to par, or because you see that they do have the potential but you lack the resources to help them get up to par (i.e. you're willing to let a future good coder go because you're currently resource-constrained, or you think it will be cheaper to have the training done elsewhere and pay for the trained person).
It would be better if the environment was not particularly welcoming and friendly to those people who can't produce the highest quality code. If you can't keep up, get out of the way.
Citation needed.
I've found it better to reject code that isn't high enough quality in a friendly and welcoming manner -- from my personal experience from my 15 years in the industry, I've found that a quicker and more successful way to motivate substandard coders to become good and productive coders.
It reminds me of the Aesop's Fable about when the sun and the wind made a bet to see who could get someone to take off my cloak -- so I assume my experience reflects a more or less universal human experience.
But I'm not aware of studies one way or the other in the specific area of OSS code contributions.
Is he fucking serious? This is a world class governing body passing laws that affect people literally all over the globe... and their excuse was "we didn't have time to sufficiently scrutinize these before voting for them."? This is... fucking insane to be light about it.
Isn't this standard practice for governing bodies? Rand Paul in the US recently complained about a 700 page spending bill, complained that neither he nor his colleagues were given time to read it. I've heard the same from the UK parliament.
It sounds good, but who defines "extremist content" or worse "incites acts of terrorism". It might be clear to you and me what this would be, but how about if you don't agree with what people say who are in control of this definition?
Just to note, you don't need a definition if instead you have a decision procedure.
That's basically true of much of common law, and why a lawyer will so often tell you "I can advise you on how to mitigate risk based on precedent but the only way to get a definitive answer for your question is to take it to court".
I think DMCA is a good example of this. It talks about "infringing material" but the DMCA law as written is actually independent of the details of what counts as infringing. Everything is expressed in terms of the process of sending a takedown letter, then responding, then going to court should there be disagreement. (It didn't touch upon how to deal with DOS takedown-letter attacks, nor how an SLA for responses, but in the light of DMCA then I bet the EU will at least think about these).
So what it would ultimately boil down to is this: the people who control the definition would be (1) the nation's supreme court, albeit with the narrow remit of having to stick with the intent of the vague words in the statute rather than complete freedom to define it any way they want, and with the various institutional checks and balances that countries have built up over the centuries to stop the supreme courts getting far out of line, and with the ultimate sanction of popular revolt if they do, (2) the practical business considerations that encourage companies to use caution rather than pushing at the edge of the definition, (3) the practical business desires to make money even when that does push at the edge of the definition.
If the bottom line is you think we shouldn't have any laws unless there's a 100% objective unarguable measurement to determine things -- that would be an interesting thought experiment, but it's far removed from how things are today.
...I'm sure all of Trump's detractors will agree that he has done the right thing and it's about time. Won't they?
I hope you didn't just anticipate a negative response from a group of people and use your anticipation feel critical about them!
I hope you won't expect ALL members of a group to do something right (since that's not how groups of humans ever work). And I hope you won't criticize the group as a whole based on just some of them failing to do something right; only if you have evidence that a significant portion of group members fail to do it.
I think it's very possible that you won't do either of those things. (but your rhetorical flourish has me worried...)
It's like all of the healthcare and insurance websites that say you can only use 8-16 characters and provide a tiny whitelist of special characters. Why are those even allowed to be online if that's all they can handle? That's all but an admission you're too cheap to update anything to make it really secure.
I think "restrict to tiny whitelist of special characters" sounds like a really good idea. As soon as we delve into variable-length characters, I worry that some (any) part of the stack might not handle them right. Maybe there's code which has the wrong computation about how many bytes a given character takes. Maybe some code is vulnerable to a string which terminates half way through a multibyte sequence. Maybe some code indexes into a string, which you can't do with variable-length.
Yeah it is a matter of cheapness. "Do I want to spend my security $$ auditing every single codepath and library for being safe with respect to variable-length characters, or do I think that a whitelist is good enough so I can spend my $$ on other security matters?"
Even just dealing with capitalization is confusing, in case your business needs involve case-insensitivity. In Turkish there are two forms of the letter "i" and turning something uppercase then lowercase is no longer a straightforward reversible operation -- and if your business needs require case insensitivity (e.g. for names) then the normal programmer techniques for generating a canonical capitalization won't work.
Now if all the stack is written in a language like C# (all characters are 16 bits wide -- unless you're using the newer.NETCore which has UTF8) or a language like Swift (all characters are variable-length but the string library is locked down enough that you won't accidentally do something wrong) then I'd feel mostly safe, so long as I didn't need case-insensitivity. Otherwise, I wouldn't.
You might say "well, coders are paid to sort out these problems". Completely true. And a well-paid programmer will be well-paid because of their experience, and their experience will say "be very careful about allowing arbitrary characters, and only go there if there's a compelling business need that outweighs the disadvantages".
Mindfulness is the new religion of modern medicine. What exactly is it? No one can clearly explain. How is it achieved? Well, opinions differ. How can it be objectively measured?
The definition used in this study is "mindfulness is the self-reported score on the Frieburg Mindfulness Inventory". That's exactly what it is, and how it's objectively measured (by asking people to self-report it). Yes that is an objective measure even though it's an objective report of a self-reported subjective thing, just like "QRDeNameland likes vanilla ice-cream" is an objective report of your self-reported subjective preference.
Anyway, the news in this study is that there's a statistical correlation between how people self-report on the Inventory, and how their brain responds to pain.
Yes it would totally be an interesting study that gets widely reported if scores on a self-reported prayer inventory also correlated to brain MRI scans (or indeed most other machine-measured phenomena). It'd be fascinating, say, to ask questions like "when you pray do you hear Jesus" or "when you pray do you lose track of where you are", and then correlate with MRI scan data. The Frieburg Mindfulness Inventory has the advantage of being an existing standard to use; I'm not aware of existing standard Prayer Indexes, so it'd be important to pre-specify such a prayer inventory to avoid falling into the trap of data-mining correlations.
However, it's insulting when a practitioner of supposedly science-based medicine starts touting ill-defined magical solutions as if they were science.
To be clear, this was a correlation study, not causation. It therefore wasn't touting solutions. The correlations were very well-defined and precise. I don't see any magic in either Freiburg Inventory nor in MRI scan data.
Uh huh. Try getting a severe burn and see how much that helps. I know everyone is a special snowflake, but pain is pain. Insect bites are hardly real pain.
I'm not sure what point you're trying to make. Is it that there exists a pain for which the technique doesn't apply, and therefore you say the technique doesn't apply to any cases of pain (a logical fallacy)? Or are you trying to define a concept of "true pain" for which the concept doesn't apply and then maybe attempt to correlate your definition of "true pain" with some other observed MRI response?
I mentioned insect bites because (1) the itch sensation is conveyed by C fibers, the same ones that carry pain, so it's a difference in magnitude not in kind; (2) I use this mindfulness technique to avoid scratching my insect bites -- I observe most people do scratch their bites even when it makes them worse.
In this context, Mindfulness is an oxymoron, where to become mindful is to remove oneself from one's own mind. It's the application of adding a level of indirection (*) to every variable.
Mindfulness in this context is just a measure of how people score on a straightforward practical questionnaire. The questions are practical, so there's no oxymoron. The only oxymoron comes from your own putative definition "remove oneself from one's own mind" which isn't what was used in the study. Here are some of the questions in the questionnaire, on a scale from Rarely to Almost Always:
Q. I sense my body, whether eating, cooking, cleaning or talking. Q. I am able to appreciate myself. Q. I pay attention to what’s behind my actions. Q. I am friendly to myself when things go wrong. Q. I am impatient with myself and with others.
Note that the "mindfulness" used in this context doesn't require you to semantically analyze the questions for what you believe to be contradictions or oddities or the logic of the questions; it's merely an observation of how people respond to it, correlated with population observations of how other people respond to it.
I think your analogy of "adding a level of indirection" isn't the right one. These questions make it sound like mindfulness is more like executing the code under a debugger, and being in the habit of pausing it in tough situations or just periodically, and inspecting the values of local variables so as to have a better idea of what's going on. The alternative would be to only figure out what's going on by looking at the output values of all functions, or what's printed to stdout.
This study resonates with my personal experience. Say I'm in pain at the dentist's, or an insect bite, or fatigue from endurance exercise. I could drop straight into the normal instinctual fight-or-flight emotional response to the pain. But instead I get my mind to observe the pain as a detached analytical observer -- to try to document the sensations of the pain in all their aspects, like a scientist would. I pretend there's no axiom that says "this sensorial experience implies that emotional response". And, hey presto, there the emotional response just doesn't happen.
We know what AI means in the industry today. It means machine-learning of some sorts - putting together a parameterized classifier, training those parameters against a training data-set, and then running it against real-world input. Yes there is considerable work needed to figure out what precise things to classify, what kind of parameterized classifier to use, and to gather the training data-set. Yes it is indeed very different from coding a classification algorithm directly. Yes it will be used in impactful ways. No it's not what we colloquially refer to as "intelligence". Yes it will have blind spots that can be gamed.
Please, I'm so tired of reading posts saying "this isn't really artificial intelligence" when we're all aware of what the industry jargon means, and all aware of the utility and limitations of the current approaches.
So, Russia placed misinformation on social networks, and who do we believe that swayed? I've yet to see a single study, or even anyone claiming, those ads and fake news reports actually had an effect on the election, i.e., convinced voters to choose one candidate over another.
Why is that your definition of an effect? Another effect, indeed the one mentioned in the summary, would be to sow discord. Do you consider "foreign power sowing discord" a possible concerning effect? How would you measure it?
My impression from http://www.people-press.org/20... is that discord has been growing steadily for decades. If I wanted to weaken America, I reckon that my starting point would be contributing to this tread. Personally I want to strengthen America, but I'm just an individual not a nation, so my approach is to express moderation and balance to everyone.
What exactly is the experiment meant to prove? Moving little devices over a 10 metre taut wire isn't exactly pushing the boundaries of science.
From the article:
"It's going to be the world's first experiment to test elevator movement in space," a university spokesman told AFP on Tuesday.
I'm not sure what you're saying. Do you think they should push a larger device over a longer distance for their first attempt? Or do you think they should skip this test entirely and just assume it will work fine?
1. No targeted harassment (i.e. repeated behavior that causes alarm/annoyance/distress). I think there have been a total of two tweets from Elon Musk in this case, so it hardly seems repeated.
2. No unwanted sexual advances. This clearly wasn't.
3. No promoting violence on the basis of race, ethnicity, sexual orientation, gender, religion, age, disability. Musk wasn't promoting violence nor was this one of the protected categories.
4. No hateful display names or profile images.
Elon Musk's tweets clearly haven't broken the Twitter rules. It should be stressed that "11001000100 think the tweets shouldn't have been made" and indeed "most people think the tweets shouldn't have been made" are both very different from "violate the standard"
If you're right, that merely elevates the article from "crackpot claptrap" to "shoddily written clickbait". The mystery is less what happens than why it happens only to some ships, and why ship owners don't take the safety measures that are described in other comments here.
The article and accompanying discussion addresses these questions and was technically interesting and high quality. "A lot is known about the physics of the liquefaction [...] Yet despite our understanding of this phenomenon (and the guidelines in place to prevent it occurring), it is still causing ships to sink and take their crew with them."
The technical answer is that the existing guidance on stowing and shipping solid bulk cargoes is too simplistic. Liquefaction potential depends not just on how much moisture is in a bulk cargo but also other material characteristics, such as the particle size distribution, the ratio of the volume of solid particles to water and the relative density of the cargo, as well as the method of loading and the motions of the vessel during the voyage.
The economics make it clear that it's not worth spending huge amounts of money (that also make ships more awkward to load) for an event that's comparatively rare. It's doubly not worth spending money on refits when we don't even have a good physics model of it, one that's backed by data+observations. And if we kitted out some experimental ships but they proved to be in the 99.9% of ships that aren't affected by the phenomenon, then we won't have gathered any data.
The main idea in this thread was lengthwise bulkheads. I've never seen a ship designed that way. Not sure why. Maybe because it's a bulkhead that would need to be seriously strong (to stop the millions of tonnes) and would contribute nothing to the structure of the ship; only weight. Another idea was pumps to remove the water. Those would have to be exceptionally robust to handle being hammered by lumps of bauxite, and not get clogged by the finer bauxite. Maybe you could put a pump behind a mesh, if you could invent a mesh that would stand up to the millions of tons. The article suggests that another cause might actually just be the *speed* of loading. If that's true, how would we measure+test that? Could we invent a different loading technique that's mostly as fast? It'd be massively cheaper than lengthwise bulkheads.
In the video, the Australian Maritime Safety Organizations suggests a different preventative measure which is - captains should be aware that this is a phenomenon, aware of what are the warning signs, should pay attention to those first signs, and should consider seeking a Port of Refuge.
In short - excellent article, interesting phenomenon that even though we know about liquefaction and know about ocean shipping most of us still wouldn't have thought of, is deeper than "just add bulkheads".
So long as the Constitution Stands, the notion of hate speech is nonsense. To quote an American President:...
That's a blisteringly US-centric response. Here was a paper authored by some Fins and an Italian. From the introduction of the paper, hate speech is defined in both US law and EU law. The paper looked at datasets from Wikipedia and Twitter, both places with a hugely international corpus.
The paper in any case is independent of what you or the law defines as hate speech. In the paper, they judged whether a variety of published algorithms could come up with the same "hatefulness scoring" as a crowd-sourced collection of people did or a group of researchers. They found the algorithms couldn't do that in general, and in the limited cases where the algorithms could do it then the algorithms could be easily defeated.
It is basically irrelevant to the results what definition they used for hate speech. They could have scored them on some other criteria, such as how funny they are or how frequently they implicitly allude to hamsters, or anything where the human taggers had the same level of agreement. The research and results would still be the same.
I don't watch TV much. But I watched the World Cup earlier this year and was *dismayed* at the poor picture quality. There weren't enough pixels to see which player had the ball when the camera was zoomed out. When I sat up close then I saw a load of compression artifacts from transmission. The whole picture also lost clarity when panning - which happens all the time.
I would love to watch an ultra-high bandwidth transmission of 8K soccer in future. I'd happily go to whichever pub or IMAX could broadcast it, and I'd pay a decent amount for it.
Slashdot Mirror
This is so dumb. A tracker would show me as terribly inactive because...
It's only dumb if a financial majority of the insurer's customers are similar to you.
Thanks. Let me try to write what you said, sticking to facts and neutral POV. I tried to remove those of your answers which didn't seem to be factual statements of agreement/disagreement with the literal text that was written, but might have made some mistakes.
(I'm not trying to argue anything here nor disagree with you. I'm trying to neutrally characterize+understand people's positions. The manifesto had seemed mild and inoffensive to me on first reading; in the light of other people's reactions I need to dig deeper).
Tell me: which specific bullet-points or sentences in that manifesto did you disagree with? Or did you disagree with implications or consequences of the manifesto rather than something written in it?
We've read similar wording a thousand times, be it in academia, in schools or in corporate codes of conduct and the purpose is always the same ...
Should I understand from the way you chose to answer that you don't disagree with any individual bullet-points or sentences in the manifesto, but you do disagree with the implications and consequences (and also motives)?
Actually, this is generally what happens to a project. It adopts a CoC or enough anonymous/SJW complaints are created, big contributors leave or are forced out, the project dies or slows because it is spending inane amounts of energy on political discussions rather than technical. It happened with NodeJS, Kubernetes, LLVM, Tor, Debian all of which are decent projects but kind of 'stuck' now that boards and bug lists are overrun by "complaints" rather than technical discussions.
I wondered if what you said is true? Here's a graph of LLVM contributions. I pulled it from github's llvm-mirror/llvm repo, and used github's "Insights > Contributions" tab to display the graph. Github claimed that the graph shows up to September 2018 -- that's contradicted by the labels and tickmarks. I wonder if github's graph-plotting routines don't get the label+tickmark positions quite right? I aimed to include roughly equal spans before and after the code of conduct to see if there's a clear effect.
https://i.imgur.com/fXorXyo.pn...
My conclusion? I don't see a clear signal. The graph seems to have roughly the same noise level and averages before and after, and a better narrative would be "there are sometimes huge sustained spikes of activity for reasons presumably related to the content of the work". (This is just from eyeballing it; I can't be bothered to put the numbers into a spreadsheet and do the calculations more carefully). If you say that LLVM is now dead, that's not supported by the data. If you say it's slowed, that's plausible (we'd need numbers), but the data says there hasn't been a major slowdown.
If you believe a different numerical analysis would be better, one that takes into account the quality of the contributions, then yes please I'd love for you to provide that! Or for you to provide equivalent graphs for the other projects you listed.
If there is any doubt on their agenda, read "The Post-Meritocracy Manifesto", but in fact, the title says it all.
As you suggested, I did read the manifesto https://postmeritocracy.org/.
I'd say the title doesn't say it all, nor even much of it -- the title seems to be intentionally written to sound paradoxical and so entice you to read further. Based on the first four paragraphs, I think the title would be more accurate (but more wordy) if it said "Meritocracy is obviously a good principle in theory -- but in practice the actual measurement of merit has been so inaccurate and biased that we could collectively be more productive if we think through what we're trying to achieve and figure out other principles that are more practically effective."
Tell me: which specific bullet-points or sentences in that manifesto did you disagree with? Or did you disagree with implications or consequences of the manifesto rather than something written in it?
What's a nice way to say that the code is a total disaster area and that it has no business being near the kernel?
"Thanks for your contribution. This diff you sent touches a lot of areas of code, which means we'll need a more careful test matrix to make sure we've shaken out every possible consequence. Also, past experience has shown that changes to files XYZ have been correlated with a high rate of undiscovered bugs. I agree that this is an important area, but let's step back and see if we can figure out a different plan of attack."
(You just said "total disaster area", so I had to fill in one possible scenario for why it's a total disaster area, so that I could give a constructive reason. One of the key parts about being nice/respectful/mentoring is to give constructive feedback.)
What if "code improve" means that it needs to rewritten from scratch by someone else who actually understands?
"Let's step back and see if we can figure out a different plan of attack."
Plenty of people simply have the wrong idea about how to solve a problem... In order words, you just want to fire them from the project, because they aren't good enough. What's a good inclusive and welcoming way to fire someone?
That's a bad reason to fire someone. You should fire someone either because they lack POTENTIAL to get up to par, or because you see that they do have the potential but you lack the resources to help them get up to par (i.e. you're willing to let a future good coder go because you're currently resource-constrained, or you think it will be cheaper to have the training done elsewhere and pay for the trained person).
It would be better if the environment was not particularly welcoming and friendly to those people who can't produce the highest quality code. If you can't keep up, get out of the way.
Citation needed.
I've found it better to reject code that isn't high enough quality in a friendly and welcoming manner -- from my personal experience from my 15 years in the industry, I've found that a quicker and more successful way to motivate substandard coders to become good and productive coders.
It reminds me of the Aesop's Fable about when the sun and the wind made a bet to see who could get someone to take off my cloak -- so I assume my experience reflects a more or less universal human experience.
But I'm not aware of studies one way or the other in the specific area of OSS code contributions.
Is he fucking serious? This is a world class governing body passing laws that affect people literally all over the globe... and their excuse was "we didn't have time to sufficiently scrutinize these before voting for them."? This is... fucking insane to be light about it.
Isn't this standard practice for governing bodies? Rand Paul in the US recently complained about a 700 page spending bill, complained that neither he nor his colleagues were given time to read it. I've heard the same from the UK parliament.
It sounds good, but who defines "extremist content" or worse "incites acts of terrorism". It might be clear to you and me what this would be, but how about if you don't agree with what people say who are in control of this definition?
Just to note, you don't need a definition if instead you have a decision procedure.
That's basically true of much of common law, and why a lawyer will so often tell you "I can advise you on how to mitigate risk based on precedent but the only way to get a definitive answer for your question is to take it to court".
I think DMCA is a good example of this. It talks about "infringing material" but the DMCA law as written is actually independent of the details of what counts as infringing. Everything is expressed in terms of the process of sending a takedown letter, then responding, then going to court should there be disagreement. (It didn't touch upon how to deal with DOS takedown-letter attacks, nor how an SLA for responses, but in the light of DMCA then I bet the EU will at least think about these).
So what it would ultimately boil down to is this: the people who control the definition would be (1) the nation's supreme court, albeit with the narrow remit of having to stick with the intent of the vague words in the statute rather than complete freedom to define it any way they want, and with the various institutional checks and balances that countries have built up over the centuries to stop the supreme courts getting far out of line, and with the ultimate sanction of popular revolt if they do, (2) the practical business considerations that encourage companies to use caution rather than pushing at the edge of the definition, (3) the practical business desires to make money even when that does push at the edge of the definition.
If the bottom line is you think we shouldn't have any laws unless there's a 100% objective unarguable measurement to determine things -- that would be an interesting thought experiment, but it's far removed from how things are today.
...I'm sure all of Trump's detractors will agree that he has done the right thing and it's about time. Won't they?
I hope you didn't just anticipate a negative response from a group of people and use your anticipation feel critical about them!
I hope you won't expect ALL members of a group to do something right (since that's not how groups of humans ever work). And I hope you won't criticize the group as a whole based on just some of them failing to do something right; only if you have evidence that a significant portion of group members fail to do it.
I think it's very possible that you won't do either of those things. (but your rhetorical flourish has me worried...)
I've found the quality/ranking of amazon results to be TERRIBLE. I always do a google search when I want to find products on amazon.
It's like all of the healthcare and insurance websites that say you can only use 8-16 characters and provide a tiny whitelist of special characters. Why are those even allowed to be online if that's all they can handle? That's all but an admission you're too cheap to update anything to make it really secure.
I think "restrict to tiny whitelist of special characters" sounds like a really good idea. As soon as we delve into variable-length characters, I worry that some (any) part of the stack might not handle them right. Maybe there's code which has the wrong computation about how many bytes a given character takes. Maybe some code is vulnerable to a string which terminates half way through a multibyte sequence. Maybe some code indexes into a string, which you can't do with variable-length.
Yeah it is a matter of cheapness. "Do I want to spend my security $$ auditing every single codepath and library for being safe with respect to variable-length characters, or do I think that a whitelist is good enough so I can spend my $$ on other security matters?"
Even just dealing with capitalization is confusing, in case your business needs involve case-insensitivity. In Turkish there are two forms of the letter "i" and turning something uppercase then lowercase is no longer a straightforward reversible operation -- and if your business needs require case insensitivity (e.g. for names) then the normal programmer techniques for generating a canonical capitalization won't work.
Now if all the stack is written in a language like C# (all characters are 16 bits wide -- unless you're using the newer .NETCore which has UTF8) or a language like Swift (all characters are variable-length but the string library is locked down enough that you won't accidentally do something wrong) then I'd feel mostly safe, so long as I didn't need case-insensitivity. Otherwise, I wouldn't.
You might say "well, coders are paid to sort out these problems". Completely true. And a well-paid programmer will be well-paid because of their experience, and their experience will say "be very careful about allowing arbitrary characters, and only go there if there's a compelling business need that outweighs the disadvantages".
Mindfulness is the new religion of modern medicine. What exactly is it? No one can clearly explain. How is it achieved? Well, opinions differ. How can it be objectively measured?
The definition used in this study is "mindfulness is the self-reported score on the Frieburg Mindfulness Inventory". That's exactly what it is, and how it's objectively measured (by asking people to self-report it). Yes that is an objective measure even though it's an objective report of a self-reported subjective thing, just like "QRDeNameland likes vanilla ice-cream" is an objective report of your self-reported subjective preference.
Anyway, the news in this study is that there's a statistical correlation between how people self-report on the Inventory, and how their brain responds to pain.
Yes it would totally be an interesting study that gets widely reported if scores on a self-reported prayer inventory also correlated to brain MRI scans (or indeed most other machine-measured phenomena). It'd be fascinating, say, to ask questions like "when you pray do you hear Jesus" or "when you pray do you lose track of where you are", and then correlate with MRI scan data. The Frieburg Mindfulness Inventory has the advantage of being an existing standard to use; I'm not aware of existing standard Prayer Indexes, so it'd be important to pre-specify such a prayer inventory to avoid falling into the trap of data-mining correlations.
However, it's insulting when a practitioner of supposedly science-based medicine starts touting ill-defined magical solutions as if they were science.
To be clear, this was a correlation study, not causation. It therefore wasn't touting solutions. The correlations were very well-defined and precise. I don't see any magic in either Freiburg Inventory nor in MRI scan data.
Uh huh. Try getting a severe burn and see how much that helps. I know everyone is a special snowflake, but pain is pain. Insect bites are hardly real pain.
I'm not sure what point you're trying to make. Is it that there exists a pain for which the technique doesn't apply, and therefore you say the technique doesn't apply to any cases of pain (a logical fallacy)? Or are you trying to define a concept of "true pain" for which the concept doesn't apply and then maybe attempt to correlate your definition of "true pain" with some other observed MRI response?
I mentioned insect bites because (1) the itch sensation is conveyed by C fibers, the same ones that carry pain, so it's a difference in magnitude not in kind; (2) I use this mindfulness technique to avoid scratching my insect bites -- I observe most people do scratch their bites even when it makes them worse.
In this context, Mindfulness is an oxymoron, where to become mindful is to remove oneself from one's own mind. It's the application of adding a level of indirection (*) to every variable.
Mindfulness in this context is just a measure of how people score on a straightforward practical questionnaire. The questions are practical, so there's no oxymoron. The only oxymoron comes from your own putative definition "remove oneself from one's own mind" which isn't what was used in the study. Here are some of the questions in the questionnaire, on a scale from Rarely to Almost Always:
Q. I sense my body, whether eating, cooking, cleaning or talking.
Q. I am able to appreciate myself.
Q. I pay attention to what’s behind my actions.
Q. I am friendly to myself when things go wrong.
Q. I am impatient with myself and with others.
Note that the "mindfulness" used in this context doesn't require you to semantically analyze the questions for what you believe to be contradictions or oddities or the logic of the questions; it's merely an observation of how people respond to it, correlated with population observations of how other people respond to it.
I think your analogy of "adding a level of indirection" isn't the right one. These questions make it sound like mindfulness is more like executing the code under a debugger, and being in the habit of pausing it in tough situations or just periodically, and inspecting the values of local variables so as to have a better idea of what's going on. The alternative would be to only figure out what's going on by looking at the output values of all functions, or what's printed to stdout.
This study resonates with my personal experience. Say I'm in pain at the dentist's, or an insect bite, or fatigue from endurance exercise. I could drop straight into the normal instinctual fight-or-flight emotional response to the pain. But instead I get my mind to observe the pain as a detached analytical observer -- to try to document the sensations of the pain in all their aspects, like a scientist would. I pretend there's no axiom that says "this sensorial experience implies that emotional response". And, hey presto, there the emotional response just doesn't happen.
From the same article - Visual Basic overtakes C#, PHP and Javascript...
We know what AI means in the industry today. It means machine-learning of some sorts - putting together a parameterized classifier, training those parameters against a training data-set, and then running it against real-world input. Yes there is considerable work needed to figure out what precise things to classify, what kind of parameterized classifier to use, and to gather the training data-set. Yes it is indeed very different from coding a classification algorithm directly. Yes it will be used in impactful ways. No it's not what we colloquially refer to as "intelligence". Yes it will have blind spots that can be gamed.
Please, I'm so tired of reading posts saying "this isn't really artificial intelligence" when we're all aware of what the industry jargon means, and all aware of the utility and limitations of the current approaches.
Why would you advise people on natural sources of probiotics - in response to an article which said that they're either ineffectual or actually bad?
So, Russia placed misinformation on social networks, and who do we believe that swayed? I've yet to see a single study, or even anyone claiming, those ads and fake news reports actually had an effect on the election, i.e., convinced voters to choose one candidate over another.
Why is that your definition of an effect? Another effect, indeed the one mentioned in the summary, would be to sow discord. Do you consider "foreign power sowing discord" a possible concerning effect? How would you measure it?
My impression from http://www.people-press.org/20... is that discord has been growing steadily for decades. If I wanted to weaken America, I reckon that my starting point would be contributing to this tread. Personally I want to strengthen America, but I'm just an individual not a nation, so my approach is to express moderation and balance to everyone.
What exactly is the experiment meant to prove? Moving little devices over a 10 metre taut wire isn't exactly pushing the boundaries of science.
From the article:
"It's going to be the world's first experiment to test elevator movement in space," a university spokesman told AFP on Tuesday.
I'm not sure what you're saying. Do you think they should push a larger device over a longer distance for their first attempt? Or do you think they should skip this test entirely and just assume it will work fine?
How does Elon Musk not get kicked off for calling a guy a pedophile and a "child rapist"? What is the standard? Why isn't it being enforced?
The standard? https://help.twitter.com/en/ru...
1. No targeted harassment (i.e. repeated behavior that causes alarm/annoyance/distress). I think there have been a total of two tweets from Elon Musk in this case, so it hardly seems repeated.
2. No unwanted sexual advances. This clearly wasn't.
3. No promoting violence on the basis of race, ethnicity, sexual orientation, gender, religion, age, disability. Musk wasn't promoting violence nor was this one of the protected categories.
4. No hateful display names or profile images.
Elon Musk's tweets clearly haven't broken the Twitter rules. It should be stressed that "11001000100 think the tweets shouldn't have been made" and indeed "most people think the tweets shouldn't have been made" are both very different from "violate the standard"
If you're right, that merely elevates the article from "crackpot claptrap" to "shoddily written clickbait". The mystery is less what happens than why it happens only to some ships, and why ship owners don't take the safety measures that are described in other comments here.
The article and accompanying discussion addresses these questions and was technically interesting and high quality. "A lot is known about the physics of the liquefaction [...] Yet despite our understanding of this phenomenon (and the guidelines in place to prevent it occurring), it is still causing ships to sink and take their crew with them."
The technical answer is that the existing guidance on stowing and shipping solid bulk cargoes is too simplistic. Liquefaction potential depends not just on how much moisture is in a bulk cargo but also other material characteristics, such as the particle size distribution, the ratio of the volume of solid particles to water and the relative density of the cargo, as well as the method of loading and the motions of the vessel during the voyage.
The economics make it clear that it's not worth spending huge amounts of money (that also make ships more awkward to load) for an event that's comparatively rare. It's doubly not worth spending money on refits when we don't even have a good physics model of it, one that's backed by data+observations. And if we kitted out some experimental ships but they proved to be in the 99.9% of ships that aren't affected by the phenomenon, then we won't have gathered any data.
The main idea in this thread was lengthwise bulkheads. I've never seen a ship designed that way. Not sure why. Maybe because it's a bulkhead that would need to be seriously strong (to stop the millions of tonnes) and would contribute nothing to the structure of the ship; only weight. Another idea was pumps to remove the water. Those would have to be exceptionally robust to handle being hammered by lumps of bauxite, and not get clogged by the finer bauxite. Maybe you could put a pump behind a mesh, if you could invent a mesh that would stand up to the millions of tons. The article suggests that another cause might actually just be the *speed* of loading. If that's true, how would we measure+test that? Could we invent a different loading technique that's mostly as fast? It'd be massively cheaper than lengthwise bulkheads.
There's a good video too: https://www.youtube.com/watch?...
In the video, the Australian Maritime Safety Organizations suggests a different preventative measure which is - captains should be aware that this is a phenomenon, aware of what are the warning signs, should pay attention to those first signs, and should consider seeking a Port of Refuge.
In short - excellent article, interesting phenomenon that even though we know about liquefaction and know about ocean shipping most of us still wouldn't have thought of, is deeper than "just add bulkheads".
So long as the Constitution Stands, the notion of hate speech is nonsense. To quote an American President: ...
That's a blisteringly US-centric response. Here was a paper authored by some Fins and an Italian. From the introduction of the paper, hate speech is defined in both US law and EU law. The paper looked at datasets from Wikipedia and Twitter, both places with a hugely international corpus.
The paper in any case is independent of what you or the law defines as hate speech. In the paper, they judged whether a variety of published algorithms could come up with the same "hatefulness scoring" as a crowd-sourced collection of people did or a group of researchers. They found the algorithms couldn't do that in general, and in the limited cases where the algorithms could do it then the algorithms could be easily defeated.
It is basically irrelevant to the results what definition they used for hate speech. They could have scored them on some other criteria, such as how funny they are or how frequently they implicitly allude to hamsters, or anything where the human taggers had the same level of agreement. The research and results would still be the same.
I don't watch TV much. But I watched the World Cup earlier this year and was *dismayed* at the poor picture quality. There weren't enough pixels to see which player had the ball when the camera was zoomed out. When I sat up close then I saw a load of compression artifacts from transmission. The whole picture also lost clarity when panning - which happens all the time.
I would love to watch an ultra-high bandwidth transmission of 8K soccer in future. I'd happily go to whichever pub or IMAX could broadcast it, and I'd pay a decent amount for it.