Come on. This kind of front page publicity is very, very rare. Two days, no less. And this research has been done before. And in general, one or two experiments are not going to reveal the ultimate truth, so why the sudden interest in this?
> Because this research is interesting for humans who reproduce and attempt to teach their offspring to act like adult human beings in 18 years or so.
So because the staff is interested in it personally? Then just write so upfront. "I'm personally interested in this."
Why is Slashdot posting this run-of-the-mill type of research that cannot get normal funding on its front page, two days in a row? Cui bono? To whose benefit?
Don't crowdfund research. It's bad for publicly funded research. As soon as the government, the universities and the grant organizations find out that crowdfunding works, they'll cut budgets or make their funding dependent on privately acquired money. That means only public darlings will get funded, or projects with corporate backing, and that the research results can end up in a drawer if there are larger backers that have made such provisions. In short, by funding research, you're killing it. Instead, write your congressman or member of parliament.
You're right. Number of files? Comment language? Size? History length? Just a few parameters that seem to influence the likelihood with which "Ugly hack" appears. Nope, first a good model, then the analysis.
I don't understand it either, but people seem to be voluntarily sucking up to Über, probably blinded by the thought that anything big must be good. They are just a taxi service, dammit. Oh yeah, they have an app. That must be it.
Ah, you a worthy of front page attention by being the visionless CEO of some unimportant and technologically totally uninteresting company as long as enough drool comes from the mouth of Goldman-Sachs analysts?
I'm an Apple fan (well, 80%), and I would mod it up. It's important information. It's at +5 now, but I can't understand why anyone would want to mod it down, except for a malicious hacker.
Wow, I did not believe this. I skimmed the article, and that's quite impressive. I don't see how to could do anything more nefarious than some observations on user activity with just Javascript, but thanks for pointing it out.
Your view on this is quite absurd. No-one should get fired for occasionally "pissing off" a manager or trying an ambitious project. If you get fired over this, and find that acceptable, I suggest you look up the definition of Stockholm Syndrome.
> If she doesn't understand the internet, that's a serious problem.
You're basically saying that every member of congress in the oversight committee should have a deep understanding of how internet works. Well, that's absurd, for two reasons.
1. They share tasks. Not everyone can overlook everything. And they surely will have staff for these matters, too. Otherwise there wouldn't be any oversight at all. 2. There is a lot more in this world to worry about than Internet.
> there are many very good reasons people are upset with her.
Possibly, but lack of knowledge on one of many possible topics shouldn't be one.
Sure. I only wanted to point out that Rust doesn't seem any worse than other languages in that particular respect.
BTW, I looked at Ruby and thought: what an awful mess. It's the kind of language that allows you to do things too cleverly. Redefining what a declaration means by overriding functions, was that really necessary?
I'm not too positive about Rust, but I admire the attempt.
That is by itself not an argument. The same obscure errors can be present in code written in another language. Rust does not seem to induce code with subtle bugs.
The comments in this thread are all made from a reasonable understanding of how the Internet works. Sen. Feinstein doesn't seem to share it. Ok, you can laugh about that once. But the contempt for her is beyond normal. It is at troll level. How much do her critics know about politics and running large public organizations? Next to nothing, I bet. And they have even less experience. In line with the season, let me say: think, before you throw the first stone
Seeing the world in a bad light, doesn't make you superior. (source forgotten)
I also have a hard time believing a request like this makes it to the/. frontpage. It is too much to ask. First, license plate data is protected. No organization in their right mind will give you access to it. Second, no-one has the ability to recognize car models, and pictures are not necessarily stored, and certainly not pictures that allow recognition of the car. LPR cameras focus on the license plate alone. The error rate is somewhere between 1% and 5% for that, but at least the data is highly specific. Imagine there were cameras that recognized models. Error rates would be staggering, and what would you do with that data? Query how many Chevrolets Maribu have passed a certain camera? What good does that do? Nothing, and that's why there is only LPR on a large scale (and bluetooth, by the way). Third, the information and amount of work you request is in excess of the value of the car, and certainly more than your job's worth. It's not economic.
Did these "many people" ever look at the offerings of Khan academy? That's not academic stuff. And Coursera lacks serious cohesion and supervision. Those are two necessary (but not sufficient) conditions.
But university is about more than learning some formula by heart or reading a book. You need to get an understanding of the context of the theories, the process of discovery, and be guided through the history and current practices. It's not for everyone, but it's certainly not something an online course can provide.
This repetition suppression (as it's called) is normal in BOLD responses (the thing fMRI measures). It happens for every stimulus. It also happens when someone reads a word for the second time, and guess what: when reading it for the second time, processing is faster and less error prone. This is called the priming effect. It's hypothesized that it actually shows an accumulation of neural activity. So a "precipitous drop" is nothing to worry about: it's a symptom of the underlying processes, and moreover: it's the wrong thing to look at when you're concerned with traffic safety.
I'm half sympathetic towards your remark. But. The first exploits were apparently against Flash. Now, that is something everyone expects, but you cannot accuse the makers of Flash of "constantly fucking about with GUI and javascript interpreter". The product seems to be "feature stable" since a long time, and rather simple in concept and implementation, in comparison to a browser. So, according to me, that indicates that even sticking to a simple code base does not guarantee it to be bug free. Look at OpenSSL: that is even smaller, and still had a nasty bug.
But in the end, it must be easier to improve a small code base, I agree. Then the issue becomes: how to convince people that we should get rid of javascript and css, to mention two of the more complex components of a browser? Or to accept a much worse performance (the JS engine is large because of optimization, not because interpreting Javascript is super complex)?
You don't know what you're talking about. I've been running simulations in C++ all my life, and I couldn't be happier with OSX: it runs on top of BSD (a Unix flavor, like Linux), which makes scripting a doddle, and has a free IDE with all the graphical dev tools to add interaction and visualization to the simulation program. Mind you, that part doesn't come for free. The availability of other, commercial software puts OSX a notch above Linux. Nowadays, you can install Cygwin and get the free Visual Studio edition for Windows, but for years, OSX has been the best dev platform around.
> Any entry level laptop will have more CPU and GPU capability to do whatever she's gonna be asked.
Simulations require so much CPU, you've got no idea. I had one of my first simulation programs run for two weeks over the Christmas holidays to get a decent set of parameters. More power is simply better in this case. Remember that optimizing simulation programs is far from trivial and will take a significant part of the PhD.
Slashdot Mirror
Come on. This kind of front page publicity is very, very rare. Two days, no less. And this research has been done before. And in general, one or two experiments are not going to reveal the ultimate truth, so why the sudden interest in this?
> Because this research is interesting for humans who reproduce and attempt to teach their offspring to act like adult human beings in 18 years or so.
So because the staff is interested in it personally? Then just write so upfront. "I'm personally interested in this."
Why is Slashdot posting this run-of-the-mill type of research that cannot get normal funding on its front page, two days in a row? Cui bono? To whose benefit?
Don't crowdfund research. It's bad for publicly funded research. As soon as the government, the universities and the grant organizations find out that crowdfunding works, they'll cut budgets or make their funding dependent on privately acquired money. That means only public darlings will get funded, or projects with corporate backing, and that the research results can end up in a drawer if there are larger backers that have made such provisions. In short, by funding research, you're killing it. Instead, write your congressman or member of parliament.
You're right. Number of files? Comment language? Size? History length? Just a few parameters that seem to influence the likelihood with which "Ugly hack" appears. Nope, first a good model, then the analysis.
Of course, that hadn't crossed my mind. My enemy's enemy, that sort of stuff.
I don't understand it either, but people seem to be voluntarily sucking up to Über, probably blinded by the thought that anything big must be good. They are just a taxi service, dammit. Oh yeah, they have an app. That must be it.
Closer to 300,000, I think.
> as far as valuation.
Ah, you a worthy of front page attention by being the visionless CEO of some unimportant and technologically totally uninteresting company as long as enough drool comes from the mouth of Goldman-Sachs analysts?
At least you're honest, I suppose.
I'm an Apple fan (well, 80%), and I would mod it up. It's important information. It's at +5 now, but I can't understand why anyone would want to mod it down, except for a malicious hacker.
Wow, I did not believe this. I skimmed the article, and that's quite impressive. I don't see how to could do anything more nefarious than some observations on user activity with just Javascript, but thanks for pointing it out.
But Javascript?
Your view on this is quite absurd. No-one should get fired for occasionally "pissing off" a manager or trying an ambitious project. If you get fired over this, and find that acceptable, I suggest you look up the definition of Stockholm Syndrome.
You mean some bystander picked it up and made it public? Possible. The story itself doesn't tell.
Anyway, it seems the ministry offers to send it well after the exams, and he would have to pay for printing and shipping.
So he made this request, haha, but who informed all the numerous reporters, and to what end?
> If she doesn't understand the internet, that's a serious problem.
You're basically saying that every member of congress in the oversight committee should have a deep understanding of how internet works. Well, that's absurd, for two reasons.
1. They share tasks. Not everyone can overlook everything. And they surely will have staff for these matters, too. Otherwise there wouldn't be any oversight at all.
2. There is a lot more in this world to worry about than Internet.
> there are many very good reasons people are upset with her.
Possibly, but lack of knowledge on one of many possible topics shouldn't be one.
Sure. I only wanted to point out that Rust doesn't seem any worse than other languages in that particular respect.
BTW, I looked at Ruby and thought: what an awful mess. It's the kind of language that allows you to do things too cleverly. Redefining what a declaration means by overriding functions, was that really necessary?
I'm not too positive about Rust, but I admire the attempt.
Yes, but you'll need 5 to 10 years of Rust experience. In a lead role.
> the remaining ones will be much more obscure
That is by itself not an argument. The same obscure errors can be present in code written in another language. Rust does not seem to induce code with subtle bugs.
The comments in this thread are all made from a reasonable understanding of how the Internet works. Sen. Feinstein doesn't seem to share it. Ok, you can laugh about that once. But the contempt for her is beyond normal. It is at troll level. How much do her critics know about politics and running large public organizations? Next to nothing, I bet. And they have even less experience. In line with the season, let me say: think, before you throw the first stone
Seeing the world in a bad light, doesn't make you superior.
(source forgotten)
I also have a hard time believing a request like this makes it to the /. frontpage. It is too much to ask. First, license plate data is protected. No organization in their right mind will give you access to it. Second, no-one has the ability to recognize car models, and pictures are not necessarily stored, and certainly not pictures that allow recognition of the car. LPR cameras focus on the license plate alone. The error rate is somewhere between 1% and 5% for that, but at least the data is highly specific. Imagine there were cameras that recognized models. Error rates would be staggering, and what would you do with that data? Query how many Chevrolets Maribu have passed a certain camera? What good does that do? Nothing, and that's why there is only LPR on a large scale (and bluetooth, by the way). Third, the information and amount of work you request is in excess of the value of the car, and certainly more than your job's worth. It's not economic.
Did these "many people" ever look at the offerings of Khan academy? That's not academic stuff. And Coursera lacks serious cohesion and supervision. Those are two necessary (but not sufficient) conditions.
But university is about more than learning some formula by heart or reading a book. You need to get an understanding of the context of the theories, the process of discovery, and be guided through the history and current practices. It's not for everyone, but it's certainly not something an online course can provide.
Who writes these free courses anyway?
And you live in Indiana? Because you just sound like my colleague from Terre Haute.
This repetition suppression (as it's called) is normal in BOLD responses (the thing fMRI measures). It happens for every stimulus. It also happens when someone reads a word for the second time, and guess what: when reading it for the second time, processing is faster and less error prone. This is called the priming effect. It's hypothesized that it actually shows an accumulation of neural activity. So a "precipitous drop" is nothing to worry about: it's a symptom of the underlying processes, and moreover: it's the wrong thing to look at when you're concerned with traffic safety.
I'm half sympathetic towards your remark. But. The first exploits were apparently against Flash. Now, that is something everyone expects, but you cannot accuse the makers of Flash of "constantly fucking about with GUI and javascript interpreter". The product seems to be "feature stable" since a long time, and rather simple in concept and implementation, in comparison to a browser. So, according to me, that indicates that even sticking to a simple code base does not guarantee it to be bug free. Look at OpenSSL: that is even smaller, and still had a nasty bug.
But in the end, it must be easier to improve a small code base, I agree. Then the issue becomes: how to convince people that we should get rid of javascript and css, to mention two of the more complex components of a browser? Or to accept a much worse performance (the JS engine is large because of optimization, not because interpreting Javascript is super complex)?
You don't know what you're talking about. I've been running simulations in C++ all my life, and I couldn't be happier with OSX: it runs on top of BSD (a Unix flavor, like Linux), which makes scripting a doddle, and has a free IDE with all the graphical dev tools to add interaction and visualization to the simulation program. Mind you, that part doesn't come for free. The availability of other, commercial software puts OSX a notch above Linux. Nowadays, you can install Cygwin and get the free Visual Studio edition for Windows, but for years, OSX has been the best dev platform around.
> Any entry level laptop will have more CPU and GPU capability to do whatever she's gonna be asked.
Simulations require so much CPU, you've got no idea. I had one of my first simulation programs run for two weeks over the Christmas holidays to get a decent set of parameters. More power is simply better in this case. Remember that optimizing simulation programs is far from trivial and will take a significant part of the PhD.