...arrested for trespass and for blocking traffic, or worse, not speech. If the administration were busy silencing critics, Bush would have had Bob Herbert and Paul Krugman executed already...
Perhaps my English keyword filters get most of the US-based spam, but until I added rather draconian filters to auto-reject Korean and Chinese e-mail (with certain exceptions, but not many) I got a *lot* of spam from those areas, in their respective languages, for web sites in those countries.
If memory serves, my filters now frown upon just about any e-mail that passes through either country, unless it comes from certain blessed domains or uses similarly blessed keywords.
This bothered me about the MechWarrior games since the mechs have the technology to perfectly detect, locate and identify enemies but lack the obvious hookups to actually *target* the enemies. In that respect, they're far inferior technology than today's fighter aircraft, I think.
However, letting them aim on their own would change it from being an action game to a tactical game, and it would have been more difficult to implement a "reasonable" challenging AI (e.g. one that didn't cheat blatantly by having infinite jumpjets or anything like that, and one where your superiors weren't stupid enough to send a single unsupported lance with no MFBs against two hundred Timberwolves armed with X-Pulses and Thunderbolts defending a ruined city).
The Germans should also realize that even if you don't want it, a war may come to you anyway -- and not on your own terms. Poland didn't want the Heer to come marching in, anymore than the Kuwaitis wanted to be annexed or the Israelis appreciate being blown up.
One could argue that the United States and terrorists have been at war for many years -- as a victim, such as in the bombing of the Marine embassy in Beirut; and as a participant (if usually indirect one), by backing states (notably Israel) and organizations (intelligence organizations of other states) whose interests frequently involve opposing terrorists more directly. From that point of view, attacks on the United States and its interests are likely inevitable unless it were to completely withdraw from world affairs, which is not going to happen.
Depending on who's doing the planning, both "acceptable collateral damage" and attacking UN convoys strike me as realistic, not cynical. Brutal, yes... but it may be a good thing to remind people how vicious it can be.
The former certainly was common in the Second World War; in fact, both the Allies and the Axis deliberately targeted civillians in their bombing / rocket campaigns. It still occurs today, to a certain degree. For instance, consider the apparent errant missile strike that killed perhaps a few dozen Iraqi citizens in a market lately. From what I've heard, the objective may have been a mobile missile launcher that was in the vicinity. Since the US knows full well that while its Tomahawks are pretty damned accurate, they're not perfect, it implies some non-zero tolerance for civilian casualties in order to hit valid military targets.
The whole Palestinian-Israeli conflict also involves frequent targeting of civilians by both sides -- with Palestinians attacking Israeli communities, and Israelis retalliating against just about everybody. In both of their viewpoints, the niceties of "civilized" warfare lose to their objectives of ejecting either the Israeli settlers or the Palestinian militants.
The UN has also been considered fair game by at least some combatants in recent conflicts, such as the whole mess in Somalia. UN troops have been intimidated, robbed, shot, and probably kidnapped and whatever else.
If memory serves, you can take a (fully automatic) AK-47, bury it in mud for a while, dig it back up, and it'll still work. They can be/very/ robust devices.
Most people, for instance, would probably talk if the alternative was seeing acid injected into the eyeballs of their coworkers, or being forced to watch the slow execution of villagers they're supposed to be helping and then to eat their remains.
If you take that attitude, to be fair you have to boycott just about every country in the entire world. Not even the United States -- which is willing to sell weapons systems to Taiwan -- recognizes them as independent.
The original author was Sima Qian; in my edition, the translator was Burton Watson, and he chose the title "Records of the Grand Historian". Technically, this particular volume was only a part of the original history; apparently, there were additional sections detailing laws, philosophies, et al, and also additions by later historians. I don't know whether Mr. Watson translated the rest.
Tracking rating per user provides some possibilities, if you don't mind the forfeiture of privacy and additional drain on hardware resources.
For instance, my reading habits aren't even close to uniform across all books. A browsing of my collection would suggest a fairly strong interest in certain types of novels and authors (for instance, quite a few Russian authors show up; likewise, 'dark' literature such as works by Camus, Kafka, and Huxley also appear, and so does some "hard" science fiction and certain histories) and sparse or complete absence in other genres (biographies, for instance; no Westerns; no non-English books; no poetry; few serials).
As a consequence, my recommendations would vary in value based on the reader who's interested. If, for instance, you're the sort of reader who's interested mostly in/recent/ books, my opinions won't have that much to offer since by the time I buy a book, the author's probably been dead for quite some time. If you're interested in knowing how the latest _Dune_ or _Star Wars_ book turns out, well, I've never read/any/ SW book and I stopped reading _Dune_ books after _Crackhouse: Dune_, so again I'd be the wrong chap to ask. But if just finished, say, "The Trial" by Kafka and liked it, you might be wanting to hear from others -- like myself -- who found it interesting and see what we would recommend.
That requires associating reviews with each other, with the most obvious but most invasive approach being the storage of all recommendations that each person has made, on a person-rates-book basis. Then, to derive additional "if you liked this you may also like" recommendations, a database could search for other individuals whose ratings profile was similar to yours, and identify books you haven't rated but were liked by those others.
Of course, that hinges on being able to define similar in a VERY sparse, extremely high-dimensional space -- given the large number of books, nobody has the TIME to have read even a miniscule fraction of a percent of them all. In addition, some books may exist in multiple versions (different translations or editions, for instance) and those ratings may need to be collated. And so forth. It also requires a large enough user base to build a reasonable predictor.
If you don't want to have reviews tracked per user, they could still be tracked in limited batches -- provide a form in which many ratings can be set at once. That way, they could all be associated with each other without sacrificing anonymity -- give 'em some primary key to distinguish that batch of votes from other batches, but don't tie the key to the user. The number of batches might be much larger than the number of users, however, and it'll lose some value because of batches that should have been tied together but weren't.
Re:You know you are dealing with nerds...
on
An IMDb for Books
·
· Score: 1
Any half-decent editor would have told the authors of the Bible to tone down the bloody "Alpha lived for 120 years and begat Bar who lived for..." mess in, hm, Chronicles unless it was _meant_ to be a soporific. Gott im Himmel...;)
Collaborative filtering has possibilities, at least for the more narrowly-focused ones, or those who stick to popular domains. If you pick works unpopular due to age and perhaps cultural obscurity, then (a) there won't be that many others to collaborate with, and (b) each vote will have a correspondingly high impact on that item.
Of course, that also requires at a minimum per-session profiling, if not per-user -- because multiple ratings have to be associated with each other somehow.
Re:Asking slashdotters...
on
An IMDb for Books
·
· Score: 2, Insightful
You might find more variety among the Slashdot audience than you'd think. Not all of us constantly wallow in technical manuals from O'Reilly or _Star Wars_ / _Star Trek_ fiction.
Hell, my most recent readings include a history of the Qin dynasty as written by a historian of the Han; a short novel about a family of Tsarist aristocrats set in Kiev shortly after the Bolshevik revolution in Moscow; the second half of the "Hyperion" series; and a "novel of ideas" (nihilism, socialism, atheism, et al) of intrigue and conspiracy set in various places in Russia in the late 1800s. Not caring too much as to the age of a text frees one to pick and choose.
Well, no kidding. Axe-carrying dwarves, woods-dwelling elves, elderly staff-bearing wizards, and foul orcs were brought to the masses in LOTR.
Plenty of subsequent fantasy books, games and movies probably draw upon that, the same way many draw upon fantasy-medieval-European themes -- it saves work, and makes the material more accessible to readers who are already familiar with the rehashed ideas.
If it allows Microsoft to disavow any support for the modified versions (big if; I don't know the terms of such an arrangement) it could save them a considerable amount of money while getting sales they might not otherwise get.
Shareware gets pirated en masse as well. How many people actually register their copy of WinZip? The reason software copyrights get infringed is because it's trivial to do, usually, and the likelihood of getting caught is extremely low. Luxury cars, Armani suits and "collectable" card games are all vastly overpriced, but most people don't bother stealing them because they're intrinsically harder to steal without consequence.
Wrong. You also have to take into account economic impact, which can be substantial, and how much of the work was used, which in the case of software is usually everything.
Universities frequently buy site licenses. This is for legal reasons, and not simply to be nice to software publishers, since they're usually not rolling around in excess positive cash flow.
I doubt the EFF and their ilk do all that much on the firearms front, either. For both groups, their message is much clearer due to focus on a single issue... and their membership, probably, that much larger.
However, for people that care about/both/ issues, there's absolutely nothing stopping them from participating in BOTH groups unless they're so impoverished they can't afford to contribute to two causes.
You see, once you've made that MP3 and given it away, you've made and redistributed a derivative work. The CD-R isn't the original medium, so first-sale doctrine DOES NOT APPLY.
If you GAVE UP your original CD by passing it along, THEN you would be within your rights (assuming there's no legally valid EULA prohibiting otherwise, and that seems unlikely). However, that's different from burning a CD-R and giving it away... or, for that matter, recording an audio cassette of the music and giving that away (due to AHRA).
Likewise, for P2P, people aren't sending their original medium; they're making and republishing versions of the original work. That's unauthorized if done without explicit permission from the copyright holders, unless the copyrights have already expired.
(1) P2P is only really useful when there are enough users providing material to make it interesting. Remember, one of the selling points of Napster was the sheer number of people using it.
(2) Regardless of encryption, requests still have to be sent. If they're to be understood, then the encryption keys must be available to everybody who might want to make a request.
The only way to make the requests unreadable by the general population (e.g. your local enforcer) would be to use public-key encryption with one key pair per peer, which means you can no longer broadcast search requests -- you instead have to issue one custom-encrypted request to each and every host you query. That won't scale, either computationally (have to use large keys to make them computationally unfeasible to break) or in terms of bandwidth.
Bandwidth costs money; in addition, it's a shared resource, so rampant file trading cuts into everybody else's right to use the networks for their intended purpose.
On the other hand, sitting in your own room getting boozed up won't affect that many people so long as you stay there and don't harrass people by, say, throwing stuff out your window.
Different-quality versions of the same content -- for instance, the "unlimited download" versions could be content with advertising, pauses, more compression artifacts, or other deliberate defects. You might be able to burn a CD full of music with ads and other junk in it, but would you want to?
In addition, one could implement a proprietary format and player, although that still leaves certain hardware hacks possible.
Slashdot Mirror
...arrested for trespass and for blocking traffic, or worse, not speech. If the administration were busy silencing critics, Bush would have had Bob Herbert and Paul Krugman executed already...
Perhaps my English keyword filters get most of the US-based spam, but until I added rather draconian filters to auto-reject Korean and Chinese e-mail (with certain exceptions, but not many) I got a *lot* of spam from those areas, in their respective languages, for web sites in those countries.
If memory serves, my filters now frown upon just about any e-mail that passes through either country, unless it comes from certain blessed domains or uses similarly blessed keywords.
This bothered me about the MechWarrior games since the mechs have the technology to perfectly detect, locate and identify enemies but lack the obvious hookups to actually *target* the enemies. In that respect, they're far inferior technology than today's fighter aircraft, I think.
However, letting them aim on their own would change it from being an action game to a tactical game, and it would have been more difficult to implement a "reasonable" challenging AI (e.g. one that didn't cheat blatantly by having infinite jumpjets or anything like that, and one where your superiors weren't stupid enough to send a single unsupported lance with no MFBs against two hundred Timberwolves armed with X-Pulses and Thunderbolts defending a ruined city).
You're comparing government-imposed restrictions with voluntary, industry-imposed labeling? Mmm hmmm.
The Germans should also realize that even if you don't want it, a war may come to you anyway -- and not on your own terms. Poland didn't want the Heer to come marching in, anymore than the Kuwaitis wanted to be annexed or the Israelis appreciate being blown up.
One could argue that the United States and terrorists have been at war for many years -- as a victim, such as in the bombing of the Marine embassy in Beirut; and as a participant (if usually indirect one), by backing states (notably Israel) and organizations (intelligence organizations of other states) whose interests frequently involve opposing terrorists more directly. From that point of view, attacks on the United States and its interests are likely inevitable unless it were to completely withdraw from world affairs, which is not going to happen.
Depending on who's doing the planning, both "acceptable collateral damage" and attacking UN convoys strike me as realistic, not cynical. Brutal, yes... but it may be a good thing to remind people how vicious it can be.
The former certainly was common in the Second World War; in fact, both the Allies and the Axis deliberately targeted civillians in their bombing / rocket campaigns. It still occurs today, to a certain degree. For instance, consider the apparent errant missile strike that killed perhaps a few dozen Iraqi citizens in a market lately. From what I've heard, the objective may have been a mobile missile launcher that was in the vicinity. Since the US knows full well that while its Tomahawks are pretty damned accurate, they're not perfect, it implies some non-zero tolerance for civilian casualties in order to hit valid military targets.
The whole Palestinian-Israeli conflict also involves frequent targeting of civilians by both sides -- with Palestinians attacking Israeli communities, and Israelis retalliating against just about everybody. In both of their viewpoints, the niceties of "civilized" warfare lose to their objectives of ejecting either the Israeli settlers or the Palestinian militants.
The UN has also been considered fair game by at least some combatants in recent conflicts, such as the whole mess in Somalia. UN troops have been intimidated, robbed, shot, and probably kidnapped and whatever else.
...or suicidal, but looking for a way in which their insurance company might still pay off.
If memory serves, you can take a (fully automatic) AK-47, bury it in mud for a while, dig it back up, and it'll still work. They can be /very/ robust devices.
...except that Robert Scheer was spewing bullshit, not truth.
Everybody has a breaking point.
Most people, for instance, would probably talk if the alternative was seeing acid injected into the eyeballs of their coworkers, or being forced to watch the slow execution of villagers they're supposed to be helping and then to eat their remains.
If you take that attitude, to be fair you have to boycott just about every country in the entire world. Not even the United States -- which is willing to sell weapons systems to Taiwan -- recognizes them as independent.
The original author was Sima Qian; in my edition, the translator was Burton Watson, and he chose the title "Records of the Grand Historian". Technically, this particular volume was only a part of the original history; apparently, there were additional sections detailing laws, philosophies, et al, and also additions by later historians. I don't know whether Mr. Watson translated the rest.
Tracking rating per user provides some possibilities, if you don't mind the forfeiture of privacy and additional drain on hardware resources.
/recent/ books, my opinions won't have that much to offer since by the time I buy a book, the author's probably been dead for quite some time. If you're interested in knowing how the latest _Dune_ or _Star Wars_ book turns out, well, I've never read /any/ SW book and I stopped reading _Dune_ books after _Crackhouse: Dune_, so again I'd be the wrong chap to ask. But if just finished, say, "The Trial" by Kafka and liked it, you might be wanting to hear from others -- like myself -- who found it interesting and see what we would recommend.
For instance, my reading habits aren't even close to uniform across all books. A browsing of my collection would suggest a fairly strong interest in certain types of novels and authors (for instance, quite a few Russian authors show up; likewise, 'dark' literature such as works by Camus, Kafka, and Huxley also appear, and so does some "hard" science fiction and certain histories) and sparse or complete absence in other genres (biographies, for instance; no Westerns; no non-English books; no poetry; few serials).
As a consequence, my recommendations would vary in value based on the reader who's interested. If, for instance, you're the sort of reader who's interested mostly in
That requires associating reviews with each other, with the most obvious but most invasive approach being the storage of all recommendations that each person has made, on a person-rates-book basis. Then, to derive additional "if you liked this you may also like" recommendations, a database could search for other individuals whose ratings profile was similar to yours, and identify books you haven't rated but were liked by those others.
Of course, that hinges on being able to define similar in a VERY sparse, extremely high-dimensional space -- given the large number of books, nobody has the TIME to have read even a miniscule fraction of a percent of them all. In addition, some books may exist in multiple versions (different translations or editions, for instance) and those ratings may need to be collated. And so forth. It also requires a large enough user base to build a reasonable predictor.
If you don't want to have reviews tracked per user, they could still be tracked in limited batches -- provide a form in which many ratings can be set at once. That way, they could all be associated with each other without sacrificing anonymity -- give 'em some primary key to distinguish that batch of votes from other batches, but don't tie the key to the user. The number of batches might be much larger than the number of users, however, and it'll lose some value because of batches that should have been tied together but weren't.
Any half-decent editor would have told the authors of the Bible to tone down the bloody "Alpha lived for 120 years and begat Bar who lived for..." mess in, hm, Chronicles unless it was _meant_ to be a soporific. Gott im Himmel... ;)
Collaborative filtering has possibilities, at least for the more narrowly-focused ones, or those who stick to popular domains. If you pick works unpopular due to age and perhaps cultural obscurity, then (a) there won't be that many others to collaborate with, and (b) each vote will have a correspondingly high impact on that item.
Of course, that also requires at a minimum per-session profiling, if not per-user -- because multiple ratings have to be associated with each other somehow.
You might find more variety among the Slashdot audience than you'd think. Not all of us constantly wallow in technical manuals from O'Reilly or _Star Wars_ / _Star Trek_ fiction.
Hell, my most recent readings include a history of the Qin dynasty as written by a historian of the Han; a short novel about a family of Tsarist aristocrats set in Kiev shortly after the Bolshevik revolution in Moscow; the second half of the "Hyperion" series; and a "novel of ideas" (nihilism, socialism, atheism, et al) of intrigue and conspiracy set in various places in Russia in the late 1800s. Not caring too much as to the age of a text frees one to pick and choose.
Well, no kidding. Axe-carrying dwarves, woods-dwelling elves, elderly staff-bearing wizards, and foul orcs were brought to the masses in LOTR.
Plenty of subsequent fantasy books, games and movies probably draw upon that, the same way many draw upon fantasy-medieval-European themes -- it saves work, and makes the material more accessible to readers who are already familiar with the rehashed ideas.
If it allows Microsoft to disavow any support for the modified versions (big if; I don't know the terms of such an arrangement) it could save them a considerable amount of money while getting sales they might not otherwise get.
Shareware gets pirated en masse as well. How many people actually register their copy of WinZip? The reason software copyrights get infringed is because it's trivial to do, usually, and the likelihood of getting caught is extremely low. Luxury cars, Armani suits and "collectable" card games are all vastly overpriced, but most people don't bother stealing them because they're intrinsically harder to steal without consequence.
Wrong. You also have to take into account economic impact, which can be substantial, and how much of the work was used, which in the case of software is usually everything.
Universities frequently buy site licenses. This is for legal reasons, and not simply to be nice to software publishers, since they're usually not rolling around in excess positive cash flow.
I doubt the EFF and their ilk do all that much on the firearms front, either. For both groups, their message is much clearer due to focus on a single issue... and their membership, probably, that much larger.
/both/ issues, there's absolutely nothing stopping them from participating in BOTH groups unless they're so impoverished they can't afford to contribute to two causes.
However, for people that care about
Actually, no.
You see, once you've made that MP3 and given it away, you've made and redistributed a derivative work. The CD-R isn't the original medium, so first-sale doctrine DOES NOT APPLY.
If you GAVE UP your original CD by passing it along, THEN you would be within your rights (assuming there's no legally valid EULA prohibiting otherwise, and that seems unlikely). However, that's different from burning a CD-R and giving it away... or, for that matter, recording an audio cassette of the music and giving that away (due to AHRA).
Likewise, for P2P, people aren't sending their original medium; they're making and republishing versions of the original work. That's unauthorized if done without explicit permission from the copyright holders, unless the copyrights have already expired.
Unlikely.
(1) P2P is only really useful when there are enough users providing material to make it interesting. Remember, one of the selling points of Napster was the sheer number of people using it.
(2) Regardless of encryption, requests still have to be sent. If they're to be understood, then the encryption keys must be available to everybody who might want to make a request.
The only way to make the requests unreadable by the general population (e.g. your local enforcer) would be to use public-key encryption with one key pair per peer, which means you can no longer broadcast search requests -- you instead have to issue one custom-encrypted request to each and every host you query. That won't scale, either computationally (have to use large keys to make them computationally unfeasible to break) or in terms of bandwidth.
Bandwidth costs money; in addition, it's a shared resource, so rampant file trading cuts into everybody else's right to use the networks for their intended purpose.
On the other hand, sitting in your own room getting boozed up won't affect that many people so long as you stay there and don't harrass people by, say, throwing stuff out your window.
Different-quality versions of the same content -- for instance, the "unlimited download" versions could be content with advertising, pauses, more compression artifacts, or other deliberate defects. You might be able to burn a CD full of music with ads and other junk in it, but would you want to?
In addition, one could implement a proprietary format and player, although that still leaves certain hardware hacks possible.