Oh, sorry, I thought it was Google that detected the problem and acted (though in keeping with/. tradition, I didn't RTFA). If Amazon *told* Google that it was infringing, then I agree that Amazon should have known better.
Invisible? No. Vanishingly rare, yes. Among people who pay attention to such things, CC is widely known precisely because it's a radical departure from the norm. A good one, but not a common one by any means. In fact, I'd call it an obscure corner case, though I'd love for that to change.
Nah, you just need to postulate technologies that allow preposterous accelerations. Check out David Weber's Honor Harrington books. The ships have accelerations measured in hundreds of gravities, and their missiles accelerate at tens of thousands of gravities. All of this is made possible by technology that allows gravity and inertia to be manipulated (within limits).
The neatest thing I found about the series (other than all the geometry), is that Black Jack -- a war hero revived after 100 years in stasis -- is the reverse-type to the trope of Ancient Badass Warrior Travels to Future and Kicks Soft Pasty-White Butts Living in Luxury Too Long. In Geary's "relative future," he is the Enlightened Man, using sophisticated naval tactics no longer taught at Academy because the peeps of that relative future are so angry and beaten down by a century of war that all they want to do is just ram their ships into the enemy and rip out their opponents' lungs with their teeth.
Heh. That was one of the most annoying things to me. Societies at war tend to get better at it, not worse. Much better, and very quickly. The notion that a peacetime navy could have better strategists and tacticians than a wartime navy just makes no sense to me at all. I had a hard time suspending disbelief on that issue.
If you haven't read David Weber's Honor Harrington books, you should. They're somewhat similar, but much, much better in almost every way, including the realism of space combat. The big thing Campbell gets very wrong in the Lost Fleet books is that ships in space can't "turn"! They can rotate and apply vector changes but they can't bend their course around in big loops. Well, they could with a carefully-planned series of vector changes, but it would be inefficient and pointless.
The biggest problem with Campbell's approach, which annoyed me to no end while reading them, was the way his ships could turn. They had to accelerate and decelerate reasonably when adding or losing velocity, but they could maneuver in curves, including big looping turns that reversed their direction, and it seemed they could do so much faster than by simply flipping end over end and applying maximum acceleration.
Campbell also has a hard time creating really good, believable characters. That annoyed me, but it drove my wife bonkers.
Ships in combat rarely move at anything other than maximum speed. Even ground forces tend to move at the maximum speed they can, consistent with not exhausting themselves. This is because maneuvering for advantage means getting to the critical position before your opponent, who is trying to get there before you do.
In the case of Weber's stories there's also the fact that missile drives "burn out" after a period of time and that a missile that reaches attack range with a dead drive is ineffective, so drive burnout imposes an upper limit on the range at which you can attack -- but since missiles begin with an initial velocity based on your ship's velocity (plus some velocity imparted by the launcher), moving faster means that your missiles can travel farther before their drives burn out. So in addition to maneuver advantages, accelerating at maximum will give your force greater range when it comes time to launch.
Consequently, ships in Weber's stories do tend to operate at maximum "safe" acceleration at all times in combat (they can actually accelerate faster, at the expense of some risk of blowing themselves up -- and sometimes they make that choice). Unless they're using a lower speed in order to deceive the enemy in some fashion. And if they are looking for an engagement that doesn't just involve blowing past one another at significant fractions of light speed, they do end up decelerating into contact -- though it's more common that they just choose courses that converge to the degree they want, in order to arrange to be in the range they want for the amount of time they want (and of course the enemy is adjusting course to try to shape the engagement the way they want, too).
The combat tactics in Weber's books are very well thought-out; almost excessively so.
David Weber invents whole swaths of physics, notably detailed artificial gravity control (without the use of large masses) and inertial dampening, both of which are completely fiction and likely to stay that way.
Meh. If you're going to have stories set in space you have to invent technology that makes it reasonable. In the case of the Honorverse, Weber clearly wanted to be able to draw a lot of inspiration from great naval battles, where maneuvering for position is essential, which implies a need for extremely high acceleration rates, unless you're going to abandon physics entirely or ignore issues of scale (the approach taken by most space combat).
Finding a way to make space combat both plausible and exciting, and similar enough to be able to draw on naval history, with only one or two small postulated advances in fundamental technology -- and advances that are pretty broadly accepted by sci-fi readers already -- is an outstanding achievement
It's great that he then sticks to his own framework assiduously, but it's a completely artificial framework, and the whole thing is based on historical naval warfare.
Duh. Honor Harrington is Horatio Hornblower in space. Weber would be the first to tell you that... note the similarity of their names!
However, it doesn't help that the political situation is a not even thinly veiled rehash of the French Revolution, right down to the name of the People's Republic leader, Rob S. Pierre. I'm bad at thinking up names, but come on.
Given the literally hundreds of characters in the books who have good names, it's quite obvious that Weber is good at thinking up names (I particularly like the Chinese/German names of the Andermanis). Given that, it's obvious that Rob S. Pierre's name was chosen on purpose. Not only that, the Havenite empire is populated by people with French names... Weber clearly decided that if he was going to draw on a historical parallel, he might as well be obvious about it.
Add to that his total inability to craft believable mannerisms for humans. That really irritated me. He would pick one mannerism for his characters, like rubbing their temple, and then every single character in the novel would use exactly that mannerism when he wanted to inject a bit of humanity into the situation.
Nonsense. Pick any major character in the series and I can tell you what their major mannerisms are, and they're different. Many of Honor's subordinates pick up some of her mannerisms, but that's also deliberate, and not unreasonable.
There would be some potential issues with copyleft licenses being unenforceable,
Some? SOME?
It would be entirely unenforceable.
Obviously it would be unenforceable. The GPs point was that this could create some potential issues. The success of permissive, non-copyleft licenses like BSD, MIT, Apache, etc., however, show that this isn't necessarily a huge problem.
I doubt the model is able to discern between initial versions and revisions. It's not the sort of thing that would occur to me when creating such a model, anyway. I suspect that the fact that the bot didn't trigger until the update was a coincidence.
I've enjoyed a lot of Weber books, but I feel he gets too bogged down in the statistics of battle. Fleet A has 1000 missile launchers with a 22 second cycle time and a missile defense system that is 75% effective, fleet B has 2000 missile launchers with a 28 second cycle time and a missile defense system that is 55% effective. Do the math, figure out who has the highest DPS and you have a winner.
I really like that aspect of his Honorverse battles, and also his attention to detail with respect to distances, acceleration rates, light-speed propagation delays, etc. It gives the battles a much greater feeling of realism, and allows him (with some careful pre-planning) to set up battle scenarios that are both tailored to the needs of the plot and seem completely uncontrived. I think he also does a proper job of balancing when the statistics dominate the engagements and when they don't. In individual ship-on-ship engagements the individual heroism of crewmembers means as much or more than the numbers (think about HMS Fearless' "death ride" against the Peep Q-ship), but as the battles scale up to squadron and then fleet levels the statistics utterly dominate the engagement outcomes.
Of course, I'm the sort of person who sometimes double-checks his distance and acceleration figures to make sure they actually make sense, so I may have a higher tolerance
And of course there is the inevitable fighting with a damaged ship. I know it makes for boring books, but I don't see how realistically any future space combat would be anything other than one hit one kill.
I think Weber addresses that quite effectively as well. With few exceptions his combatants are killed by a single hit, if it comes in through the open aspects of the drive wedge. But sidewalls mitigate the damage and they also mean that contact weapons are completely ineffective unless you can get them through those same open aspects, because physical missiles can't get through the sidewalls at all. The fact that effective weapons must be standoff weapons which detonate at ranges of tens of thousands of kilometers from the sidewall-protected ship means that the warheads can't just be omnidirectional bombs. Even shaped charges wouldn't work.
So to get hits, the warheads are bomb-pumped X-ray lasers which have tremendous energy but merely poke holes through the ships rather than blowing them apart. The effect is (by design, I'm sure -- but it's a reasonable one, given the assumed wedges and sidewalls) more like a wooden ship being struck by a cannonball than a steel ship being struck by an anti-ship missile.
There are a few times in the books when one side uses "old-fashioned nukes" against ships that don't have wedges and sidewalls up, or which get through the open ends of the wedge, and the result is devastating. Even at some distance unprotected ships' hulls are "slagged" by being near a massive explosion.
But given that most combat damage is done by lasers (or in close-in combat, grasers), I think it's entirely reasonable that ships, especially the enormous ships of the wall, can take many hits and keep on fighting.
The default assumption of these automated checkers is that anything shared is infringing.
Is that assumption not generally correct? If I were building a bayesian network to estimate the probability that a given piece of shared content is infringing, I'd have to estimate the prior probability that a given torrent is infringing, assuming nothing else is known -- and any statistical analysis of the torrents on the web would put that prior probability at around 0.9999, I'm sure.
The fact is that any model that is remotely close to accurate will end up being structured so that the default assumption is "infringing", in the absence of sufficiently strong indicators to the contrary.
IMO, the failure here wasn't that the bots considered him to be an infringer, the failure was that it was so hard for him to get the decision re-evaluated. I also understand Google's point of view that since infringers will also try to use whatever process is available to get accounts restored, that process has to include some barriers to screen most of them out. But it sounds like the barriers were too high in this case.
If Google sees more profit in "being a victim" to laws they could very well change, then they will do precisely dick to change them. Period.
Over the last few years Google has established a major lobbying operation in Washington, and has been spending significant amounts of money at trying to change the laws. The fact is that individual corporations, however wealthy or influential, don't actually have the ability to rewrite the laws, and certainly not in a timeframe less than decades.
Note that I work for Google, but I'm not involved in any of what's being discussed here. I did, however, have a chance to discuss the issues with an attorney Google hired to lobby against software patents. She was hired the same day I was and while going through our orientation I had a chance to chat with her over lunch. She had previously worked for other lobbying firms and had a deep understanding of how the system works, and her comment was that it can be changed, but at a rate of inches per year, not miles per hour -- and that for every lobbyist working to push the law one way there is another lobbyist pushing back the other way.
Look at some of the changes Google would clearly like to make: Software patents are nothing but pain for Google. The company has had to cave in and start acquiring and creating patents, but only because to do otherwise would result in being destroyed by them. Copyright law causes huge pain for Google. Not just because it ends up having to try to enforce copyrights -- even though it really couldn't care less; almost none of Google's business relies in any way on copyrights -- and not just because it ends up taking crap over its inevitable enforcement errors, but because copyright law as-is actively impedes much of what Google wants to do. It appears that the system has decided that Google's caching of web pages is probably okay, but there has been a lot of question and controversy which has cost a lot of money. There's the Google Books thing which still hasn't been settled.
An area in which Google is finally starting to make a little progress is the law around self-driving cars... but even there Google has really only gotten a couple of states to agree that it's okay for a car to drive itself as long as there's a human driver ready to take over at any instant and who takes full responsibility for anything the car does, which is only a small first step to what Google really wanted.
And what about SOPA? If Google is so all-powerful politically, why was the whole Internet blackout day even necessary? Why didn't Google just pay off the lawmakers in question and shut the whole thing down? Because it doesn't work that way. Many people criticized Google's post-SOPA efforts, saying that Google's efforts to help craft compromise legislation proved that Google didn't really care about the fundamental principle -- but influencing compromises is how you make progress in Washington and Google simply doesn't have the power to take a hard line and be successful.
There's no doubt that corporate lobbying does influence our laws, substantially, and probably excessively. But that's a far cry from saying that any corporation can just buy whatever legislation it likes -- and that is especially untrue for a new participant like Google. Google has only existed for 14 years, and has only gotten involved in politics in the last four or five years. Give them another 20 years of lobbying and they'll probably have built the sort of influence that may allow them to affect the laws in significant ways -- but it still won't be dramatic, or the changes very fast.
Honestly, with respect to copyright law, I think the first thing Google and other proponents of a more rational copyright world need to do is not lobby for changes in laws, but to educate the public. The public needs to understand that copyright is itself a compromise, where society grants a temporary ownership to a creator in order to increase the flow of works into t
Not only should all states be funding open-source textbooks, they should really start with the K-12 space.
Schools in our country spend insane amounts of money buying very expensive textbooks again and again, in spite of the fact that the topics they cover haven't changed significantly centuries. Most elementary schools save money by not allowing the books to leave the classroom, ever, and even secondary schools have to put a lot of effort into trying to reuse books year over year, which changes how students use the books. In addition to all of that, when errors are discovered or (rare) changes in subject matter crop up, there is no way to correct them except by shelling out another $50-75 per text, per student.
The content for all of this basic education material should be in the public domain, freely available for anyone to use in whatever way they like. If hardcover, bound copies are useful then schools (or states) should hire publishers to print them, on a competitive basis. If Kinko's can provide a comparable product at a lower price than Houghton-Mifflin, then Kinko's gets the contract. This would reduce the per-text price from $50+ to around $5. At that price, every student can have their own copy to use and abuse and keep or recycle, as they like.
Or, if teachers find it more useful to have their students use the material in electronic form, or to print off selected subsets of the material to hand out on a periodic basis, or... whatever, it can be done in whatever way makes the most sense for the context.
And all of this could be accomplished if states only diverted a small fraction of the money they currently spend on buying expensive textbooks to hiring good education authors to write them on a for-hire basis. This has been obvious for at least 20 years.
Why hasn't it happened? Actually, it has... but not in the official education establishment. The home-schooling community has developed tons of great, low-cost educational materials. Much of it is open source, and much of the rest includes specific permission to make copies as needed for educational purposes. Why hasn't it happened in public school bureaucracies? IMO, it's because their centralization and subservience to political structures has left them open to manipulation by people with vested interest in the status quo.
Weber is the first to point out that Honor Harrington is a deliberate re-imagining of Horatio Hornblower. Her arc has continued beyond the point where Hornblower died, but that was actually a change in plans caused by angst among fans who didn't want the story to end. The stories, however, definitely go well beyond Forester's books in the complexity of the plots and on the number and choice of themes. And I agree with the GP that Weber does a great job of inventing a form of space combat that is plausible and realistic (within the bounds of his postulated physics around the possibility of gravity manipulation) -- and it actually does include carriers in space for reasons that make sense.
Safehold doesn't follow any Forester story line (or even themes) as far as I can tell, though Weber is obviously taking the opportunity to give full expression to his love for and deep knowledge of sailing -- to a degree that actually bothers many readers without nautical experience because they find his lovingly-detailed descriptions of rigging plans and sailing evolutions baffling or boring or both (Personally, I really like that aspect of the books.)
I haven't read Out of the Dark (yet).
Weber has also published plenty of books that have no relation to Forester's books, and don't involve naval warfare. The Dahak series is good, and the War God series is great fun.
Weber is one of my favorite contemporary authors. And I really like the fact that he manages to sustain a pace of three to four novels per year. That's not quite an Asimovian output (Weber would have to more than double his pace), but it's getting up there.
You may have read articles in the press over the last couple of weeks about YouTube videos in Brazil.
Given all the interest, we wanted to explain what has happened, and why. First of all some basic
principles about the service. Our goal is for YouTube to be a community that everyone can enjoy, as well
as a platform for free speech around the world. This can cause real challenges, because what is OK in
one country may be offensive or even illegal in another.
So we have clear community guidelines about the kind of videos that are unacceptable--and when they
are flagged, we review and if necessary remove them. If a video is illegal in a particular country--and we
have a local version of the service there, as in Brazil--we will restrict access to it, after receiving a valid
court order or government complaint. Because we are deeply committed to free expression, we often
push back on requests that we do not believe are valid. For example, we were recently in court in the US
arguing that videos were perfectly legitimate and should stay on YouTube.
Now for what’s happened in Brazil. As usual during an election season, we have had a lot of court
orders to remove videos that are critical of political candidates. As always, we have reviewed them all--
and pushed back on the many legal complaints that we believe are invalid. For example, last week, we
appealed a court order to remove videos from YouTube. While we were waiting for that appeal to be
heard, an arrest warrant was issued for me as country director of Google Brazil.
Late last night, we learned that our final legal appeal has been denied and so now we have no choice but
to block the video in Brazil. We are deeply disappointed that we have never had the full opportunity to
argue in court that these were legitimate free speech videos and should remain available in Brazil.
Despite all this, we will continue to campaign for free expression globally—not just because it’s a key
tenet of free societies, but also because more information generally means more choice, more power,
more economic opportunity and more freedom for people. As Article 19 of the United Nations’ Universal
Declaration of Human Rights states, “Everyone has the right to freedom of opinion and expression; this
right includes freedom to hold opinions without interference and to seek, receive, and impart information
and ideas through any media and regardless of frontiers.”
Ironically, the user who published one of the videos has now removed it and closed their account--
showing just what a chilling effect these episodes can have on free speech.
If congress really cared to stop a military action authorized by the president they have the power to withdraw funding for any and all foreign adventures.
They could do more than that: They could simply direct the president to stop. If he failed to do so, they could impeach him and then have the courts try him. No need to act indirectly via funding.
But the truth is that both the Repubicrats and the Democans are in general agreement about the US' right and need to lash out at random.
Well, arguably we could have expected that a constitutional lawyer would follow the Constitution. But that would have been stupid, of course, since Obama is from the humpty-dumpty branch of Constitutional law ("when I read the Constitution it means just what I choose it to mean -- neither more nor less").
I don't know what that single instruction would be (I am not an assembler expert), or how likely it is that a compiler would recognize it.
Followup: Just for fun I decided to test it. I compiled the code with -O1 on my handy compiler (g++ 4.6.3) and what it produced was:
imulq %rdi, %rax
shrq $32, %rax
So, two instructions. However, it occurred to me that perhaps the code in question was to be run on a 32-bit processor, and my compiler is compiling for 64 bits. So I changed the problem a bit, to the analogous one on a 64-bit CPU:
uint64_t((__uint128_t(a) * b) >> 64)
and what the compiler produced was:
mulq %rdi
So, it looks like gcc 4.6.3 does, in fact, recognize how to optimize this particular code. No need for inline assembler here.
overloaded operators are functions, so the function call will already lose more time
Functions, including in-line assembler implementations, can be inlined, so there isn't necessarily any functional call overhead whatsoever.
pretty sure gp was joking
No, I don't think he was, although he perhaps somewhat missed the point. He just pointed out a convenient way to do the escape to assembler with minimal impact on the remainder of the code, but the core point was that you need assembler to code this operation efficiently -- the argument is that few, if any, C or C++ compilers would recognize that something like:
uint32((uint64(a) * b) >> 32);
can be reduced to a single instruction. I don't know what that single instruction would be (I am not an assembler expert), or how likely it is that a compiler would recognize it.
Nope. This is (yet again) about a US company trying to pretend that only US law applies as soon as they enter another country.
I don't think so.
I think this is Google arguing that Google shouldn't be responsible for policing what people choose to post on their service. Instead, Google appears to believe that the person who posted the video in violation of the law should be held accountable -- and that's not an unreasonable position, particularly since the question goes far beyond just this case, or election laws, or even YouTube.
My perception is that Google is quite rational about these things, and does abide by local laws. In fact, Google specifically is very careful in general to abide by legal takedown requests. If they're willing to brave the court's wrath, it must be because they see tremendous value in establishing the precedent that users are responsible for what they post.
(Disclaimer: I work for Google. I do not, however, speak for Google, and Google doesn't speak for me. Also, I have no insider knowledge about this issue.)
Nothing worse than everyone trying to check out the latest version on a slow network attached to a memory starved server.
Git is another solution to this problem. Cloning the repository is harder on the network because you're downloading a complete revision history (though represented in a very compact way) but requires virtually no CPU or RAM on the server, and once you've got your clone all of your subsequent work is very low-impact on the server in all respects: network, RAM, CPU.
Slashdot Mirror
Oh, sorry, I thought it was Google that detected the problem and acted (though in keeping with /. tradition, I didn't RTFA). If Amazon *told* Google that it was infringing, then I agree that Amazon should have known better.
Invisible? No. Vanishingly rare, yes. Among people who pay attention to such things, CC is widely known precisely because it's a radical departure from the norm. A good one, but not a common one by any means. In fact, I'd call it an obscure corner case, though I'd love for that to change.
"Stalked". I don't think that word means what you think it means.
If I have to see advertising, I'd rather it be for stuff I'm actually interested in.
Nah, you just need to postulate technologies that allow preposterous accelerations. Check out David Weber's Honor Harrington books. The ships have accelerations measured in hundreds of gravities, and their missiles accelerate at tens of thousands of gravities. All of this is made possible by technology that allows gravity and inertia to be manipulated (within limits).
Space isn't too big if you can go fast enough.
The neatest thing I found about the series (other than all the geometry), is that Black Jack -- a war hero revived after 100 years in stasis -- is the reverse-type to the trope of Ancient Badass Warrior Travels to Future and Kicks Soft Pasty-White Butts Living in Luxury Too Long. In Geary's "relative future," he is the Enlightened Man, using sophisticated naval tactics no longer taught at Academy because the peeps of that relative future are so angry and beaten down by a century of war that all they want to do is just ram their ships into the enemy and rip out their opponents' lungs with their teeth.
Heh. That was one of the most annoying things to me. Societies at war tend to get better at it, not worse. Much better, and very quickly. The notion that a peacetime navy could have better strategists and tacticians than a wartime navy just makes no sense to me at all. I had a hard time suspending disbelief on that issue.
If you haven't read David Weber's Honor Harrington books, you should. They're somewhat similar, but much, much better in almost every way, including the realism of space combat. The big thing Campbell gets very wrong in the Lost Fleet books is that ships in space can't "turn"! They can rotate and apply vector changes but they can't bend their course around in big loops. Well, they could with a carefully-planned series of vector changes, but it would be inefficient and pointless.
The biggest problem with Campbell's approach, which annoyed me to no end while reading them, was the way his ships could turn. They had to accelerate and decelerate reasonably when adding or losing velocity, but they could maneuver in curves, including big looping turns that reversed their direction, and it seemed they could do so much faster than by simply flipping end over end and applying maximum acceleration.
Campbell also has a hard time creating really good, believable characters. That annoyed me, but it drove my wife bonkers.
Only if you accelerate and decelerate at maximum.
Ships in combat rarely move at anything other than maximum speed. Even ground forces tend to move at the maximum speed they can, consistent with not exhausting themselves. This is because maneuvering for advantage means getting to the critical position before your opponent, who is trying to get there before you do.
In the case of Weber's stories there's also the fact that missile drives "burn out" after a period of time and that a missile that reaches attack range with a dead drive is ineffective, so drive burnout imposes an upper limit on the range at which you can attack -- but since missiles begin with an initial velocity based on your ship's velocity (plus some velocity imparted by the launcher), moving faster means that your missiles can travel farther before their drives burn out. So in addition to maneuver advantages, accelerating at maximum will give your force greater range when it comes time to launch.
Consequently, ships in Weber's stories do tend to operate at maximum "safe" acceleration at all times in combat (they can actually accelerate faster, at the expense of some risk of blowing themselves up -- and sometimes they make that choice). Unless they're using a lower speed in order to deceive the enemy in some fashion. And if they are looking for an engagement that doesn't just involve blowing past one another at significant fractions of light speed, they do end up decelerating into contact -- though it's more common that they just choose courses that converge to the degree they want, in order to arrange to be in the range they want for the amount of time they want (and of course the enemy is adjusting course to try to shape the engagement the way they want, too).
The combat tactics in Weber's books are very well thought-out; almost excessively so.
David Weber invents whole swaths of physics, notably detailed artificial gravity control (without the use of large masses) and inertial dampening, both of which are completely fiction and likely to stay that way.
Meh. If you're going to have stories set in space you have to invent technology that makes it reasonable. In the case of the Honorverse, Weber clearly wanted to be able to draw a lot of inspiration from great naval battles, where maneuvering for position is essential, which implies a need for extremely high acceleration rates, unless you're going to abandon physics entirely or ignore issues of scale (the approach taken by most space combat).
Finding a way to make space combat both plausible and exciting, and similar enough to be able to draw on naval history, with only one or two small postulated advances in fundamental technology -- and advances that are pretty broadly accepted by sci-fi readers already -- is an outstanding achievement
It's great that he then sticks to his own framework assiduously, but it's a completely artificial framework, and the whole thing is based on historical naval warfare.
Duh. Honor Harrington is Horatio Hornblower in space. Weber would be the first to tell you that... note the similarity of their names!
However, it doesn't help that the political situation is a not even thinly veiled rehash of the French Revolution, right down to the name of the People's Republic leader, Rob S. Pierre. I'm bad at thinking up names, but come on.
Given the literally hundreds of characters in the books who have good names, it's quite obvious that Weber is good at thinking up names (I particularly like the Chinese/German names of the Andermanis). Given that, it's obvious that Rob S. Pierre's name was chosen on purpose. Not only that, the Havenite empire is populated by people with French names... Weber clearly decided that if he was going to draw on a historical parallel, he might as well be obvious about it.
Add to that his total inability to craft believable mannerisms for humans. That really irritated me. He would pick one mannerism for his characters, like rubbing their temple, and then every single character in the novel would use exactly that mannerism when he wanted to inject a bit of humanity into the situation.
Nonsense. Pick any major character in the series and I can tell you what their major mannerisms are, and they're different. Many of Honor's subordinates pick up some of her mannerisms, but that's also deliberate, and not unreasonable.
There would be some potential issues with copyleft licenses being unenforceable,
Some? SOME? It would be entirely unenforceable.
Obviously it would be unenforceable. The GPs point was that this could create some potential issues. The success of permissive, non-copyleft licenses like BSD, MIT, Apache, etc., however, show that this isn't necessarily a huge problem.
I doubt the model is able to discern between initial versions and revisions. It's not the sort of thing that would occur to me when creating such a model, anyway. I suspect that the fact that the bot didn't trigger until the update was a coincidence.
I've enjoyed a lot of Weber books, but I feel he gets too bogged down in the statistics of battle. Fleet A has 1000 missile launchers with a 22 second cycle time and a missile defense system that is 75% effective, fleet B has 2000 missile launchers with a 28 second cycle time and a missile defense system that is 55% effective. Do the math, figure out who has the highest DPS and you have a winner.
I really like that aspect of his Honorverse battles, and also his attention to detail with respect to distances, acceleration rates, light-speed propagation delays, etc. It gives the battles a much greater feeling of realism, and allows him (with some careful pre-planning) to set up battle scenarios that are both tailored to the needs of the plot and seem completely uncontrived. I think he also does a proper job of balancing when the statistics dominate the engagements and when they don't. In individual ship-on-ship engagements the individual heroism of crewmembers means as much or more than the numbers (think about HMS Fearless' "death ride" against the Peep Q-ship), but as the battles scale up to squadron and then fleet levels the statistics utterly dominate the engagement outcomes.
Of course, I'm the sort of person who sometimes double-checks his distance and acceleration figures to make sure they actually make sense, so I may have a higher tolerance
And of course there is the inevitable fighting with a damaged ship. I know it makes for boring books, but I don't see how realistically any future space combat would be anything other than one hit one kill.
I think Weber addresses that quite effectively as well. With few exceptions his combatants are killed by a single hit, if it comes in through the open aspects of the drive wedge. But sidewalls mitigate the damage and they also mean that contact weapons are completely ineffective unless you can get them through those same open aspects, because physical missiles can't get through the sidewalls at all. The fact that effective weapons must be standoff weapons which detonate at ranges of tens of thousands of kilometers from the sidewall-protected ship means that the warheads can't just be omnidirectional bombs. Even shaped charges wouldn't work.
So to get hits, the warheads are bomb-pumped X-ray lasers which have tremendous energy but merely poke holes through the ships rather than blowing them apart. The effect is (by design, I'm sure -- but it's a reasonable one, given the assumed wedges and sidewalls) more like a wooden ship being struck by a cannonball than a steel ship being struck by an anti-ship missile.
There are a few times in the books when one side uses "old-fashioned nukes" against ships that don't have wedges and sidewalls up, or which get through the open ends of the wedge, and the result is devastating. Even at some distance unprotected ships' hulls are "slagged" by being near a massive explosion.
But given that most combat damage is done by lasers (or in close-in combat, grasers), I think it's entirely reasonable that ships, especially the enormous ships of the wall, can take many hits and keep on fighting.
The default assumption of these automated checkers is that anything shared is infringing.
Is that assumption not generally correct? If I were building a bayesian network to estimate the probability that a given piece of shared content is infringing, I'd have to estimate the prior probability that a given torrent is infringing, assuming nothing else is known -- and any statistical analysis of the torrents on the web would put that prior probability at around 0.9999, I'm sure.
The fact is that any model that is remotely close to accurate will end up being structured so that the default assumption is "infringing", in the absence of sufficiently strong indicators to the contrary.
IMO, the failure here wasn't that the bots considered him to be an infringer, the failure was that it was so hard for him to get the decision re-evaluated. I also understand Google's point of view that since infringers will also try to use whatever process is available to get accounts restored, that process has to include some barriers to screen most of them out. But it sounds like the barriers were too high in this case.
If Google sees more profit in "being a victim" to laws they could very well change, then they will do precisely dick to change them. Period.
Over the last few years Google has established a major lobbying operation in Washington, and has been spending significant amounts of money at trying to change the laws. The fact is that individual corporations, however wealthy or influential, don't actually have the ability to rewrite the laws, and certainly not in a timeframe less than decades.
Note that I work for Google, but I'm not involved in any of what's being discussed here. I did, however, have a chance to discuss the issues with an attorney Google hired to lobby against software patents. She was hired the same day I was and while going through our orientation I had a chance to chat with her over lunch. She had previously worked for other lobbying firms and had a deep understanding of how the system works, and her comment was that it can be changed, but at a rate of inches per year, not miles per hour -- and that for every lobbyist working to push the law one way there is another lobbyist pushing back the other way.
Look at some of the changes Google would clearly like to make: Software patents are nothing but pain for Google. The company has had to cave in and start acquiring and creating patents, but only because to do otherwise would result in being destroyed by them. Copyright law causes huge pain for Google. Not just because it ends up having to try to enforce copyrights -- even though it really couldn't care less; almost none of Google's business relies in any way on copyrights -- and not just because it ends up taking crap over its inevitable enforcement errors, but because copyright law as-is actively impedes much of what Google wants to do. It appears that the system has decided that Google's caching of web pages is probably okay, but there has been a lot of question and controversy which has cost a lot of money. There's the Google Books thing which still hasn't been settled.
An area in which Google is finally starting to make a little progress is the law around self-driving cars... but even there Google has really only gotten a couple of states to agree that it's okay for a car to drive itself as long as there's a human driver ready to take over at any instant and who takes full responsibility for anything the car does, which is only a small first step to what Google really wanted.
And what about SOPA? If Google is so all-powerful politically, why was the whole Internet blackout day even necessary? Why didn't Google just pay off the lawmakers in question and shut the whole thing down? Because it doesn't work that way. Many people criticized Google's post-SOPA efforts, saying that Google's efforts to help craft compromise legislation proved that Google didn't really care about the fundamental principle -- but influencing compromises is how you make progress in Washington and Google simply doesn't have the power to take a hard line and be successful.
There's no doubt that corporate lobbying does influence our laws, substantially, and probably excessively. But that's a far cry from saying that any corporation can just buy whatever legislation it likes -- and that is especially untrue for a new participant like Google. Google has only existed for 14 years, and has only gotten involved in politics in the last four or five years. Give them another 20 years of lobbying and they'll probably have built the sort of influence that may allow them to affect the laws in significant ways -- but it still won't be dramatic, or the changes very fast.
Honestly, with respect to copyright law, I think the first thing Google and other proponents of a more rational copyright world need to do is not lobby for changes in laws, but to educate the public. The public needs to understand that copyright is itself a compromise, where society grants a temporary ownership to a creator in order to increase the flow of works into t
Not only should all states be funding open-source textbooks, they should really start with the K-12 space.
Schools in our country spend insane amounts of money buying very expensive textbooks again and again, in spite of the fact that the topics they cover haven't changed significantly centuries. Most elementary schools save money by not allowing the books to leave the classroom, ever, and even secondary schools have to put a lot of effort into trying to reuse books year over year, which changes how students use the books. In addition to all of that, when errors are discovered or (rare) changes in subject matter crop up, there is no way to correct them except by shelling out another $50-75 per text, per student.
The content for all of this basic education material should be in the public domain, freely available for anyone to use in whatever way they like. If hardcover, bound copies are useful then schools (or states) should hire publishers to print them, on a competitive basis. If Kinko's can provide a comparable product at a lower price than Houghton-Mifflin, then Kinko's gets the contract. This would reduce the per-text price from $50+ to around $5. At that price, every student can have their own copy to use and abuse and keep or recycle, as they like.
Or, if teachers find it more useful to have their students use the material in electronic form, or to print off selected subsets of the material to hand out on a periodic basis, or... whatever, it can be done in whatever way makes the most sense for the context.
And all of this could be accomplished if states only diverted a small fraction of the money they currently spend on buying expensive textbooks to hiring good education authors to write them on a for-hire basis. This has been obvious for at least 20 years.
Why hasn't it happened? Actually, it has... but not in the official education establishment. The home-schooling community has developed tons of great, low-cost educational materials. Much of it is open source, and much of the rest includes specific permission to make copies as needed for educational purposes. Why hasn't it happened in public school bureaucracies? IMO, it's because their centralization and subservience to political structures has left them open to manipulation by people with vested interest in the status quo.
Weber is the first to point out that Honor Harrington is a deliberate re-imagining of Horatio Hornblower. Her arc has continued beyond the point where Hornblower died, but that was actually a change in plans caused by angst among fans who didn't want the story to end. The stories, however, definitely go well beyond Forester's books in the complexity of the plots and on the number and choice of themes. And I agree with the GP that Weber does a great job of inventing a form of space combat that is plausible and realistic (within the bounds of his postulated physics around the possibility of gravity manipulation) -- and it actually does include carriers in space for reasons that make sense.
Safehold doesn't follow any Forester story line (or even themes) as far as I can tell, though Weber is obviously taking the opportunity to give full expression to his love for and deep knowledge of sailing -- to a degree that actually bothers many readers without nautical experience because they find his lovingly-detailed descriptions of rigging plans and sailing evolutions baffling or boring or both (Personally, I really like that aspect of the books.)
I haven't read Out of the Dark (yet).
Weber has also published plenty of books that have no relation to Forester's books, and don't involve naval warfare. The Dahak series is good, and the War God series is great fun.
Weber is one of my favorite contemporary authors. And I really like the fact that he manages to sustain a pace of three to four novels per year. That's not quite an Asimovian output (Weber would have to more than double his pace), but it's getting up there.
It's still vastly better than the "it's just a goddamned piece of paper" approach.
I fail to see an appreciable difference. It's a difference of style, not substance.
You may have read articles in the press over the last couple of weeks about YouTube videos in Brazil. Given all the interest, we wanted to explain what has happened, and why. First of all some basic principles about the service. Our goal is for YouTube to be a community that everyone can enjoy, as well as a platform for free speech around the world. This can cause real challenges, because what is OK in one country may be offensive or even illegal in another.
So we have clear community guidelines about the kind of videos that are unacceptable--and when they are flagged, we review and if necessary remove them. If a video is illegal in a particular country--and we have a local version of the service there, as in Brazil--we will restrict access to it, after receiving a valid court order or government complaint. Because we are deeply committed to free expression, we often push back on requests that we do not believe are valid. For example, we were recently in court in the US arguing that videos were perfectly legitimate and should stay on YouTube.
Now for what’s happened in Brazil. As usual during an election season, we have had a lot of court orders to remove videos that are critical of political candidates. As always, we have reviewed them all-- and pushed back on the many legal complaints that we believe are invalid. For example, last week, we appealed a court order to remove videos from YouTube. While we were waiting for that appeal to be heard, an arrest warrant was issued for me as country director of Google Brazil.
Late last night, we learned that our final legal appeal has been denied and so now we have no choice but to block the video in Brazil. We are deeply disappointed that we have never had the full opportunity to argue in court that these were legitimate free speech videos and should remain available in Brazil.
Despite all this, we will continue to campaign for free expression globally—not just because it’s a key tenet of free societies, but also because more information generally means more choice, more power, more economic opportunity and more freedom for people. As Article 19 of the United Nations’ Universal Declaration of Human Rights states, “Everyone has the right to freedom of opinion and expression; this right includes freedom to hold opinions without interference and to seek, receive, and impart information and ideas through any media and regardless of frontiers.”
Ironically, the user who published one of the videos has now removed it and closed their account-- showing just what a chilling effect these episodes can have on free speech.
Source: http://googlebrasilblog.blogspot.com/2012/09/youtube-no-brasil.html
If congress really cared to stop a military action authorized by the president they have the power to withdraw funding for any and all foreign adventures.
They could do more than that: They could simply direct the president to stop. If he failed to do so, they could impeach him and then have the courts try him. No need to act indirectly via funding.
But the truth is that both the Repubicrats and the Democans are in general agreement about the US' right and need to lash out at random.
Well, arguably we could have expected that a constitutional lawyer would follow the Constitution. But that would have been stupid, of course, since Obama is from the humpty-dumpty branch of Constitutional law ("when I read the Constitution it means just what I choose it to mean -- neither more nor less").
I don't know what that single instruction would be (I am not an assembler expert), or how likely it is that a compiler would recognize it.
Followup: Just for fun I decided to test it. I compiled the code with -O1 on my handy compiler (g++ 4.6.3) and what it produced was:
imulq %rdi, %rax
shrq $32, %rax
So, two instructions. However, it occurred to me that perhaps the code in question was to be run on a 32-bit processor, and my compiler is compiling for 64 bits. So I changed the problem a bit, to the analogous one on a 64-bit CPU:
uint64_t((__uint128_t(a) * b) >> 64)
and what the compiler produced was:
mulq %rdi
So, it looks like gcc 4.6.3 does, in fact, recognize how to optimize this particular code. No need for inline assembler here.
overloaded operators are functions, so the function call will already lose more time
Functions, including in-line assembler implementations, can be inlined, so there isn't necessarily any functional call overhead whatsoever.
pretty sure gp was joking
No, I don't think he was, although he perhaps somewhat missed the point. He just pointed out a convenient way to do the escape to assembler with minimal impact on the remainder of the code, but the core point was that you need assembler to code this operation efficiently -- the argument is that few, if any, C or C++ compilers would recognize that something like:
uint32((uint64(a) * b) >> 32);
can be reduced to a single instruction. I don't know what that single instruction would be (I am not an assembler expert), or how likely it is that a compiler would recognize it.
Nope. This is (yet again) about a US company trying to pretend that only US law applies as soon as they enter another country.
I don't think so.
I think this is Google arguing that Google shouldn't be responsible for policing what people choose to post on their service. Instead, Google appears to believe that the person who posted the video in violation of the law should be held accountable -- and that's not an unreasonable position, particularly since the question goes far beyond just this case, or election laws, or even YouTube.
My perception is that Google is quite rational about these things, and does abide by local laws. In fact, Google specifically is very careful in general to abide by legal takedown requests. If they're willing to brave the court's wrath, it must be because they see tremendous value in establishing the precedent that users are responsible for what they post.
(Disclaimer: I work for Google. I do not, however, speak for Google, and Google doesn't speak for me. Also, I have no insider knowledge about this issue.)
Nothing worse than everyone trying to check out the latest version on a slow network attached to a memory starved server.
Git is another solution to this problem. Cloning the repository is harder on the network because you're downloading a complete revision history (though represented in a very compact way) but requires virtually no CPU or RAM on the server, and once you've got your clone all of your subsequent work is very low-impact on the server in all respects: network, RAM, CPU.
There is no problem that cannot be solved, or created, by adding another layer of regulation.