Roundabouts are very simple. They are not a single intersection with a concrete circle in the middle, they are a series of T intersections on a one-way ring road.
Oh, I know how simple they are, conceptually. It's a flawed idea because a "series of T intersections on a one-way road" is only safe if (1) they're spaced far enough apart that drivers can focus on each of them properly and (2) if the traffic on both of the roads is relatively light. If traffic is heavy on one of the roads, you need a traffic signal to regulate the flow of traffic, or you get DDoS conditions where it's difficult for anyone to turn onto the heavier-traveled road (the ring, since that carries traffic from all of the roads that enter it). Unless you have a light somewhere breaking up the flow of traffic on the heavier-traveled road.
Lights, properly timed and networked (manually, or with some sort of electronic interconnectivity), are self-regulating in that they break traffic into groups of heavy traffic followed by groups of very light traffic. This allows traffic at intersections not controlled by a light (such as the ones in a roundabout) to get across or into the flow where the traffic is minimal due to the red light. Roundabouts are not self-regulating in this manner, just the opposite in fact; they slow everyone down and the result is a steady slow stream of traffic, which exacerbates the DDoS situation I described above.
Traffic on the stem of the T must give way to traffic on the bar of the T. In the case of the roundabout, this becomes "Traffic entering a roundabout must give way to traffic on the roundabout."
(1) That's the opposite of everything drivers here know, because here, drivers are taught to yield to traffic coming from their right side, which would mean traffic inside the roundabout yielding to traffic entering the roundabout. Roundabouts would be an exception to the rule.
This yield-to-right rule has two good things going for it: drivers are on the left side of the vehicle (in this country) and can't see how much clearance their right side has as well, and secondly, the left side has a decent-sized metal frame passing across the driver's vision and they can't see vehicles coming from that side as well as they can vehicles coming from their right. It makes sense that they should yield to the cars that they can see well, but whose positions relative to their own are less well known.
(2) You obviously didn't read my post fully, because I specifically mentioned that (in my country, at least) that is not any sort of universal standard. The "yield-to-right" rule was applied to roundabouts and only recently have people started thinking that a specific exception to that rule should be made for roundabouts.
Yes. Same as any other site that is being scraped by a bot that isn't respecting robots.txt.
Detect its user agent and feed it error pages (or even fake data).
If the bot starts spoofing a different user agent, detect its IP address.
And if the bot is operating off a botnet and IP detection fails, you'll have to use heuristic-style methods if you want to detect it. You might just plain be SOL.
Yes it did. Right there, 2nd comment from the top of the page. I suggest you use Ctrl-F to find it. Google indexed the comment but it doesn't hold your hand all the way.
There's an HTML link to it right there. You don't have to enable or disable anything to use it. Middle-click it, or right click and select "Open link in new tab". It's exactly like the "Reply to This" link on Slashdot. Javascript, but with an HTML link for fall-back.
Ever spend 2-3 minutes at a red light waiting for it to turn when there is no cross traffic? Roundabouts are specifically designed to alleviate that
I'd generally try to avoid those lights, too. They're a problem that shouldn't exist. The roundabout is still the solution which is simple, elegant, and wrong. Two-way stop intersections alleviate that problem just fine. If you're on the busier street, you don't even have to slow down. So do sensors in the road, or cameras. So do lights which are programmed to go onto long-green, short-red cycles when there's statistically-low traffic on the side-street. It can even vary depending on the time of day.
the rules are as simple as knowing wtf yield actually means in Webster terms
Oh, sure. It's so simple. So does traffic outside the roundabout yield to traffic inside it, or does traffic inside it yield to traffic merging in? And don't say it's obvious. It's not a universal standard. People think one is "better", but changing it in areas where it's "wrong" would probably be even more dangerous than leaving it as is. Here, here's Wikipedia's explanation of the clusterfuck which is traffic circles or roundabouts:
"Traffic circle" is a term mainly used in the United States to describe a junction which in other countries would be called a roundabout. Although a traffic circle is sometimes called a roundabout even in the U.S., U.S. traffic engineers make the distinction that in a roundabout entering traffic must always yield to traffic already in the circle, whereas in a traffic circle entering traffic is controlled by stop signs, traffic signals, or is not formally controlled.
And once you get the rules for traffic set firmly so we can all get through roundabouts in one piece, care to explain how pedestrians should attempt get across a roundabout safely? Answer: can't be done. Get a safe distance away from the roundabout and cross the road, preferably at an intersection with a stop light.
I like them, but on major intersections... they can get a little bit crazy
On major intersections, they're not just crazy, they're insane. On small intersections, they're annoying and unnecessarily slow down traffic on the higher-traveled of the two roads, where a simple two-way stop on the less-traveled one would be smarter.
I think it's the new trendy thing to drive around with high beams on all the time.
Or maybe I'm just noticing it more because it's dark when I drive to work now. I'll be glad for the brief reprieve when DST ends next Monday. I'll be driving to work while it's light out for a while.
Actually, I'd make that knowing the whereabouts of the roundabouts, and avoiding them if at all possible. Doubly so if they're busy or have more than one lane.
Roundabout: an elegant device to squish all of the traffic passing through an intersection into a single or few lanes where it converges, twists around, and exits onto its respective roads. It's a fucking dance and hardly anyone knows the rules to it well enough to attempt it safely. The roundabout was the solution which was simple, elegant, and wrong.
Virtually every other traffic sign is already adequately understood. Except maybe the distinction between "this lane is ending, merge now" and "this lane is not ending, so please don't try to merge unless/until you have to because you'll just block traffic".
CCDs are often placed behind IR cutoff filters. You phone probably does this, your digital camera as well
Cutoff filters may block some IR but they do not block it entirely. At least, the ones on phones or digital cameras don't. Try pointing a phone or digital camera at the IR LED in a TV remote while you press a button on the remote and see what happens.
They basically just have to be good enough to avoid normal levels of IR blinding the sensor. Unless they're expecting people to be actively trying to blind sensors with intense IR, they don't need to filter it that well.
More than likely, though, you'd get pulled over by the first cop who noticed that your license plate frame blinded his dash cam.
you know who the driver is [or] you are driving it
Hypothetically:
If I have a wife and a teenager who're both allowed to use the family car (fill it up before you bring it back) and I'm not at home 24 hours a day, no, I don't know who the driver is when your fancy camera took my car's picture. Too bad that picture doesn't show who the driver was, hmm?
And I won't do your detective work for you, officer (a.k.a. go asking everyone to try to find out who was driving it).
And you can't charge me for a crime I didn't commit, and if I did commit it, you'll have to have some evidence because I'm not testifying against myself.
And I very politely just told you to go fuck yourself, and there's not a damn thing you can do about it. Have a nice day, officer.
Come back with a picture showing who the driver was.
You can make "clean-room" reverse engineering while (somebody is) looking at the code. The only requirement is that the person writting your software dosen't touch the foreign code.
AFAIK he was doing it all himself.
You are probably confusing it with "black box" reverse engenireeng.
Actually, I was repeating what TFA said:
For reference, it does appear that this researcher is not doing "clean-room" reverse engineering. One of the comments he writes on his blog reads, "It is because I have only de-obfuscted 3.8 and 4.1(BETA) versions of skype binary." (In response to why he isn't targeting Skype 5 support at this time.)
Moot point, because if he's one person looking at Skype's code and then writing code to interface with it using its protocols, that's not black-box or clean-room reverse engineering.
Coal is mostly carbon. Carbon weighs about 12 g/mol *.
To burn, coal requires oxygen. Oxygen is found in the air. Oxygen has an atomic weight of about 16 g/mol and is found in the form of O2, which weighs twice as much, 32 g/mol.
1 mol C + 1 mol O2 => 1 mol CO2 *
CO2 weighs about 44 g/mol, or about 3.66 times the weight of carbon.
How could burning 1 pound of coal result in 3 pounds of CO2? Well, apparently the coal was only about 82% pure carbon.
You might have to have pure water to begin with for this catalyst to give you pure hydrogen and pure oxygen. They probably started with DI water from their RO unit, before they added whatever alkali they needed to activate the catalyst.
Water + Energy (+ Catalyst) => Hydrogen + Oxygen (+ Catalyst) Hydrogen + Oxygen + Energy => Water + Energy
Notice anything funny about that? I do... if the energies were just right (namely, less in than out), it'd be a perpetual motion machine.
Hint: the energies aren't right, and it's not. You'll have to put more energy into it than you'll get back in a useful form. No catalyst will ever get you to the point where it takes less energy to split the water than you'd get back by recombining the H and O.
5 REM Print a message onto the user's screen 10 PRINT "They did, however, get in a tiff when a judge ruled jail-breaking was not illegal nor did it violate the DMCA" 20 END
The only browser that I know probably won't support it is IE (even the newer versions, I think - support for data URIs is limited to certain parameters, e.g. backgrounds), but I was also unsure of the mouse events since clientX, clientY are not computed the same over all browsers. Although since it doesn't care about the actual x,y of the click, just the relative distance between two clicks, that shouldn't cause a problem.
It is handy, though, to have a free host for a few k worth of HTML to do a demo like that. It was pretty cool when I discovered tinyurl allows data URIs. E.g. here's another I did to demonstrate multiple background images (transparent PNG) with prime tile sizes to create a background pattern with a long repeat interval.
I'm not saying it's impossible, but I haven't seen multi-column layout with images spanning columns done in JavaScript. You reach some walls in JavaScript.
Can anybody figure out what he's trying to say there? You wouldn't even need Javascript - you'd do that with some very basic CSS. I don't see the problem he's trying to point out.
\Net neutrality doesn't return to reply squat, because it doesn't care to protect illegal content. Net neutrality exists to ensure equal transmission for content that may be undesirable
Net neutrality means that if you request a packet from IP:port to IP:port, I don't ask questions.
Getting an illegal IP:port off the internet... well, that's someone else's problem.
Slashdot Mirror
Actually, I didn't and still don't. There was no "post that was in response to", it was the root post in its comment thread.
Unless that was the joke, in which case, no, it wasn't funny.
Roundabouts are very simple. They are not a single intersection with a concrete circle in the middle, they are a series of T intersections on a one-way ring road.
Oh, I know how simple they are, conceptually. It's a flawed idea because a "series of T intersections on a one-way road" is only safe if (1) they're spaced far enough apart that drivers can focus on each of them properly and (2) if the traffic on both of the roads is relatively light. If traffic is heavy on one of the roads, you need a traffic signal to regulate the flow of traffic, or you get DDoS conditions where it's difficult for anyone to turn onto the heavier-traveled road (the ring, since that carries traffic from all of the roads that enter it). Unless you have a light somewhere breaking up the flow of traffic on the heavier-traveled road.
Lights, properly timed and networked (manually, or with some sort of electronic interconnectivity), are self-regulating in that they break traffic into groups of heavy traffic followed by groups of very light traffic. This allows traffic at intersections not controlled by a light (such as the ones in a roundabout) to get across or into the flow where the traffic is minimal due to the red light. Roundabouts are not self-regulating in this manner, just the opposite in fact; they slow everyone down and the result is a steady slow stream of traffic, which exacerbates the DDoS situation I described above.
Traffic on the stem of the T must give way to traffic on the bar of the T. In the case of the roundabout, this becomes "Traffic entering a roundabout must give way to traffic on the roundabout."
(1) That's the opposite of everything drivers here know, because here, drivers are taught to yield to traffic coming from their right side, which would mean traffic inside the roundabout yielding to traffic entering the roundabout. Roundabouts would be an exception to the rule.
This yield-to-right rule has two good things going for it: drivers are on the left side of the vehicle (in this country) and can't see how much clearance their right side has as well, and secondly, the left side has a decent-sized metal frame passing across the driver's vision and they can't see vehicles coming from that side as well as they can vehicles coming from their right. It makes sense that they should yield to the cars that they can see well, but whose positions relative to their own are less well known.
(2) You obviously didn't read my post fully, because I specifically mentioned that (in my country, at least) that is not any sort of universal standard. The "yield-to-right" rule was applied to roundabouts and only recently have people started thinking that a specific exception to that rule should be made for roundabouts.
Yes. Same as any other site that is being scraped by a bot that isn't respecting robots.txt.
Detect its user agent and feed it error pages (or even fake data).
If the bot starts spoofing a different user agent, detect its IP address.
And if the bot is operating off a botnet and IP detection fails, you'll have to use heuristic-style methods if you want to detect it. You might just plain be SOL.
Yes it did. Right there, 2nd comment from the top of the page. I suggest you use Ctrl-F to find it. Google indexed the comment but it doesn't hold your hand all the way.
http://lmgtfy.com/?q=%22Hey+guys%2C+where+else+can+we+find+more+drivel+to+dilute+our+search+results%22
There's an HTML link to it right there. You don't have to enable or disable anything to use it. Middle-click it, or right click and select "Open link in new tab". It's exactly like the "Reply to This" link on Slashdot. Javascript, but with an HTML link for fall-back.
Ever spend 2-3 minutes at a red light waiting for it to turn when there is no cross traffic? Roundabouts are specifically designed to alleviate that
I'd generally try to avoid those lights, too. They're a problem that shouldn't exist. The roundabout is still the solution which is simple, elegant, and wrong. Two-way stop intersections alleviate that problem just fine. If you're on the busier street, you don't even have to slow down. So do sensors in the road, or cameras. So do lights which are programmed to go onto long-green, short-red cycles when there's statistically-low traffic on the side-street. It can even vary depending on the time of day.
the rules are as simple as knowing wtf yield actually means in Webster terms
Oh, sure. It's so simple. So does traffic outside the roundabout yield to traffic inside it, or does traffic inside it yield to traffic merging in? And don't say it's obvious. It's not a universal standard. People think one is "better", but changing it in areas where it's "wrong" would probably be even more dangerous than leaving it as is. Here, here's Wikipedia's explanation of the clusterfuck which is traffic circles or roundabouts:
"Traffic circle" is a term mainly used in the United States to describe a junction which in other countries would be called a roundabout. Although a traffic circle is sometimes called a roundabout even in the U.S., U.S. traffic engineers make the distinction that in a roundabout entering traffic must always yield to traffic already in the circle, whereas in a traffic circle entering traffic is controlled by stop signs, traffic signals, or is not formally controlled.
And once you get the rules for traffic set firmly so we can all get through roundabouts in one piece, care to explain how pedestrians should attempt get across a roundabout safely? Answer: can't be done. Get a safe distance away from the roundabout and cross the road, preferably at an intersection with a stop light.
I like them, but on major intersections... they can get a little bit crazy
On major intersections, they're not just crazy, they're insane. On small intersections, they're annoying and unnecessarily slow down traffic on the higher-traveled of the two roads, where a simple two-way stop on the less-traveled one would be smarter.
I think it's the new trendy thing to drive around with high beams on all the time.
Or maybe I'm just noticing it more because it's dark when I drive to work now. I'll be glad for the brief reprieve when DST ends next Monday. I'll be driving to work while it's light out for a while.
Actually, I'd make that knowing the whereabouts of the roundabouts, and avoiding them if at all possible. Doubly so if they're busy or have more than one lane.
Roundabout: an elegant device to squish all of the traffic passing through an intersection into a single or few lanes where it converges, twists around, and exits onto its respective roads. It's a fucking dance and hardly anyone knows the rules to it well enough to attempt it safely. The roundabout was the solution which was simple, elegant, and wrong.
Virtually every other traffic sign is already adequately understood. Except maybe the distinction between "this lane is ending, merge now" and "this lane is not ending, so please don't try to merge unless/until you have to because you'll just block traffic".
CCDs are often placed behind IR cutoff filters. You phone probably does this, your digital camera as well
Cutoff filters may block some IR but they do not block it entirely. At least, the ones on phones or digital cameras don't. Try pointing a phone or digital camera at the IR LED in a TV remote while you press a button on the remote and see what happens.
They basically just have to be good enough to avoid normal levels of IR blinding the sensor. Unless they're expecting people to be actively trying to blind sensors with intense IR, they don't need to filter it that well.
More than likely, though, you'd get pulled over by the first cop who noticed that your license plate frame blinded his dash cam.
you know who the driver is [or] you are driving it
Hypothetically:
If I have a wife and a teenager who're both allowed to use the family car (fill it up before you bring it back) and I'm not at home 24 hours a day, no, I don't know who the driver is when your fancy camera took my car's picture. Too bad that picture doesn't show who the driver was, hmm?
And I won't do your detective work for you, officer (a.k.a. go asking everyone to try to find out who was driving it).
And you can't charge me for a crime I didn't commit, and if I did commit it, you'll have to have some evidence because I'm not testifying against myself.
And I very politely just told you to go fuck yourself, and there's not a damn thing you can do about it. Have a nice day, officer.
Come back with a picture showing who the driver was.
You can make "clean-room" reverse engineering while (somebody is) looking at the code. The only requirement is that the person writting your software dosen't touch the foreign code.
AFAIK he was doing it all himself.
You are probably confusing it with "black box" reverse engenireeng.
Actually, I was repeating what TFA said:
For reference, it does appear that this researcher is not doing "clean-room" reverse engineering. One of the comments he writes on his blog reads, "It is because I have only de-obfuscted 3.8 and 4.1(BETA) versions of skype binary." (In response to why he isn't targeting Skype 5 support at this time.)
Moot point, because if he's one person looking at Skype's code and then writing code to interface with it using its protocols, that's not black-box or clean-room reverse engineering.
I thought about calling him on that but I figured it smelled too much like a troll.
*sigh* and then I said molecules when I meant atoms. Silly me. There are the same number of atoms on both sides of the balanced equation.
He wasn't doing "clean-room" reverse engineering, though; TFA mentions that he de-obfuscated a few different versions of the Skype binary.
Coal is mostly carbon. Carbon weighs about 12 g/mol *.
To burn, coal requires oxygen. Oxygen is found in the air. Oxygen has an atomic weight of about 16 g/mol and is found in the form of O2, which weighs twice as much, 32 g/mol.
1 mol C + 1 mol O2 => 1 mol CO2 *
CO2 weighs about 44 g/mol, or about 3.66 times the weight of carbon.
How could burning 1 pound of coal result in 3 pounds of CO2? Well, apparently the coal was only about 82% pure carbon.
* The mole, abbreviated "mol", is just a number of atoms or molecules. A very large number. It's a constant. As a matter of fact, it's exactly defined to be the number of atoms of C-12 in 12 grams of pure carbon-12. So that equation is perfectly balanced; there are equal numbers of molecules of C and O on both sides.
You might have to have pure water to begin with for this catalyst to give you pure hydrogen and pure oxygen. They probably started with DI water from their RO unit, before they added whatever alkali they needed to activate the catalyst.
If it's coated in this stuff, we probably won't know how close we are until we've hit it.
Water + Energy (+ Catalyst) => Hydrogen + Oxygen (+ Catalyst)
Hydrogen + Oxygen + Energy => Water + Energy
Notice anything funny about that? I do... if the energies were just right (namely, less in than out), it'd be a perpetual motion machine.
Hint: the energies aren't right, and it's not. You'll have to put more energy into it than you'll get back in a useful form. No catalyst will ever get you to the point where it takes less energy to split the water than you'd get back by recombining the H and O.
You forgot the two most important parts!
5 REM Print a message onto the user's screen
10 PRINT "They did, however, get in a tiff when a judge ruled jail-breaking was not illegal nor did it violate the DMCA"
20 END
The only browser that I know probably won't support it is IE (even the newer versions, I think - support for data URIs is limited to certain parameters, e.g. backgrounds), but I was also unsure of the mouse events since clientX, clientY are not computed the same over all browsers. Although since it doesn't care about the actual x,y of the click, just the relative distance between two clicks, that shouldn't cause a problem.
It is handy, though, to have a free host for a few k worth of HTML to do a demo like that. It was pretty cool when I discovered tinyurl allows data URIs. E.g. here's another I did to demonstrate multiple background images (transparent PNG) with prime tile sizes to create a background pattern with a long repeat interval.
Yes you can do it in Javascript, but the code would be convoluted.
Proof-of-concept demo. Click and drag the empty image frame to dynamically relocate it.
(Only works in browsers which recognize the data: URI scheme. Tested in FF and Opera.)
Guess that trip to Bolivia I had planned just went out the window.
In all seriousness, though... will this affect me? If at all, how?
I'm not saying it's impossible, but I haven't seen multi-column layout with images spanning columns done in JavaScript. You reach some walls in JavaScript.
Can anybody figure out what he's trying to say there? You wouldn't even need Javascript - you'd do that with some very basic CSS. I don't see the problem he's trying to point out.
\Net neutrality doesn't return to reply squat, because it doesn't care to protect illegal content. Net neutrality exists to ensure equal transmission for content that may be undesirable
Net neutrality means that if you request a packet from IP:port to IP:port, I don't ask questions.
Getting an illegal IP:port off the internet... well, that's someone else's problem.