Taubes points out that the current medical orthodoxy -- that consuming fat makes you fat and exercise makes you thin -- has no basis in research.
This is, AFAICT, not the current "medical orthodoxy", but a strawman. Sure, most medical practitioners, dieticians, etc., will advise people who want to lose weight to cut down on fat (its the easiest place for most people to cut down without losing necessary nutrition, and has other, non-weight health benefits when cut) and to exercise more (because it keeps your body from slowing down when you cut calories, making it more practical to maintain a calorie deficit, and because it has other, non-weight health benefits) if they want to lose weight, but I've never heard one say either that "eating fat makes you fat" or "exercise makes you thin".
Most seem to say the same thing: calorie surpluses make you gain weight, calorie deficits make you lose weight. The advice they give is on how to maintain good health and a calorie deficit simultaneous, so that you can lose weight and improve health. And while less fat and more exercise are common parts of that (since they are appropriate for most people), they aren't the only advice given, nor are they given to everyone (because what is appropriate does, in fact, vary based on what the person is doing already, their other health conditions, etc.)
Public domain software - no copyright, no nothing. Rare and not very useful, but it does exist. Well, it did exist until universities wised up to what some of the faculty were doing.
SQLite still exists and, as best I can tell, is moderately popular.
If this is the case, why don't FOSS licenses specifically disavow that right.
Such a disavowal (which is what the claim to be "irrevocable" is) generally, as I understand, has no more durability than the license itself, as a promise unsupported by consideration. A gratuitous license (with or without a claim of irrevocability) may provide a bar against the licensor pursuing infringement actions after a sudden revocation under an estoppel theory, and that certainly is more likely if a promise was given that the license would be valid for a certain period of time or irrevocable, but estoppel will generally be extended only so far as the court feels is necessary to prevent the former licensee from suffering a substantial injustice from the extent to which they had already relied on the license, it won't make the license really irrevocable. It will just give the licensee the opportunity to argue for some consideration to the resources they've already expended under the expectation that the license would not be revoked.
The GPL, and all other licenses based in copyrights, only kick in when you perform the actual copying.
Rather, gratuitous copyright licenses like the GPL only restrict your use of rights that are exclusive to the copyright holder under the law applicable to your activity. This includes most copying, but it also includes things that are not copying, and excludes copying under circumstances where that copying is either not within the basic scope of the exclusive rights under copyright or is within the scope of a specific exception to copyright (e.g., in the U.S. at least, "fair use".)
Copyright law allows the copyright holder control over public performance of their work.
U.S. copyright law provides the copyright owner an exclusive right of public performance only for "literary, musical, dramatic, and choreographic works, pantomimes, and motion pictures and other audiovisual works" (17 USC 106). Some computer software (many games) I suppose might qualify as either "motion pictures" or "other audiovisual work", but an open-source programming language interpreter, database access layer, or web application framework aren't going to qualify as anyo of those things.
AGPL is no more "viral" than GPL is, it's just appropriate for "free as in free speech" web-apps development.
In that both are completely inappropriate for that, this is correct. The GPL is not "free as in speech", however much its proponents want to claim that way. Its a conditional license offered by the owner of a property (a copyright), and is, as such, "free as in beer".
Public domain software is "free as in free speech" -- you may use it freely without restrictions or conditions imposed by any private party so long as you don't break generally applicable laws in what you do with it. GPL software is a "gift" given to you in the hope that you will serve the interest of the property-holder granting you the license, and with terms that restrict your ability to use it in other ways. "Free as in free speech" for that is a clever sales tactic, but not even a remotely accurate parallel.
The only restriction on me is that if someone already has a BSD licensed copy now I can't say "Your copy is now AGPL licensed"
Maybe so, maybe not. Gratuitous licenses (i.e., ones that are not contracts, which would include the BSDL and GPL) are generally, in US law at least, revocable at will by the licensor (whether or not their terms say they are "irrevocable" or something similar.) This is, of course, not something the FSF ever mentions in their attempts to argue that the GPL is a license on which both developers and users of software can depend and feel secure in their consistent, unambiguously-defined rights.
Bet you 10 slashbucks if you do some research behind where this study came from, it is companies who claim to have the fix for this.
Its from an lobbying group whose pushing a tiered internet and other telecom-friendly government policy as the solution; so its not the "we have a product that is a solution" type of thing, but essentially the political equivalent.
The most glaring one I can remember was on the morning of September 11, 2001, but its not the only one that has occurred, and undoubtedly won't be the last. Also, the same thing happens with any other limited communications service (POTS systems can be -- and have been -- overloaded during major events!), and with (and where we get the name) electrical grids.
So, yeah, by 2010, internet brownouts "might" happen. They already do happen. And we all survive.
Aside from pushing a meaningles scary buzzword ("exaflood"), this is an unsurprising study by a largely telecom-industry-funded lobbying group favoring tiered internet services and other telecom-friendly policy that, surprise of surprises, finds that with the current, mostly-neutral internet, the whole system is about to collapse, and it will be used to sell the idea that we have to abandon that model, let telecoms charge additional fees to get data delivered even though they already charge each end for every byte transferred, etc.
I am no a mathematician but surely if you're going to submit a computer aided proof you must submit a full copy of the program.
Why? If you do a proof without a program, you don't have to submit a full copy of your brain, do you?
The "proof" is the thing that proves the theorem (hence the name), and it must do so in adequate detail showing all necessary steps and, as necessary, justifying them. If it does so, no additional information about how it was decided to take particular steps is necessary. If it does not, it fails regardless of what additional information provided.
He thinks that it goes against the Japanese culture to use a technology without paying for it, that it shows disrespect to not pay for software licenses. He is not even allowed to consider using Linux or any other OSS for that matter.
Many companies I've known won't use software if they aren't paying someone for support and a license (and, often, particular support guarantees and/or performance warranties.) I don't know anything about Japanese culture, but in business the need to have some else that's feet can be held to the fire if something goes wrong is a big deal.
Of course, you can get paid support (and sometimes licensing, when the software is under a dual OSS/commercial licensing model) for most OSS you might want to adopt in a business environment, so neither cultural nor business-based reluctance to use software without paying for it should be a major barrier to OSS adoption.
How do they know they're accurately converting the signals to sound, if they're basing this off a man who has no ability to speak?
Many people who are unable to speak are able to communicate in some other way (usually, some form of gesture, whether sign language, nodding, blinking, whatever.) It doesn't take a much to be able to indicate "right" or "wrong".
No disrespect to Lajos, but have we really fallen so far in programming standards that it's considered "amazing" to disassemble a 1024 byte program?
Good question. Lets look at the excerpt from TFA included in TFS:
'He is an amazing reverse-engineer,' recounts team leader Tim McNerney, 'We understood the disassembled calculator code well enough to simulate it, but Lajos really turned it into "source code" of the highest standards.'
Sure looks to me that what Lajos is being credited with isn't the disassembly, at all.
Would you rather the MS guys spend time seeing if they can force their 114k application down into 10k, or perhaps writing an operating system that doesn't suck?
None of this addresses the, original point of this fork, fact that you cannot trust the client.
Yeah, it does. That's not a fact. You can trust the client, and in fact most systems do. Some forms of cheating that would not be possible without that are possible (and, given enough users, certain) if you do, but that doesn't mean you can't do it. Game companies aren't generally interested in 100% security against cheating, they are interested in maximimizing profitability, which is a very different thing.
If you trust the client you will have cheating.
You can (and will, given a big enough user base) have bots and many forms of cheating even if the client isn't trusted to push state changes back to the server.
It's not free but that doesn't detract from it's validity as an idea.
It does detract from the validity of the assertion that the game would have been more profitable if that architecture had been chosen, which is what I was responding to.
You can never trust the client. It is in the hands of the enemy.
This assumes that the fundamental goal of a gaming company is to defeat cheaters. This is not the case. Their fundamental goal is to make a profit. Making cheating difficult is useful only insofar as it helps acheive that goal; if doing it makes it impossible (or even just too much more costly) to develop a game with adequate performance to attract players, its not worth it, however ideal it might be in some abstract anti-cheating analysis.
If you start trusting the client because of the warden then someone is going to cheat by figuring out how to spoof the warden.
Blizzard isn't trusting the client because of the warden; they are trusting the client, most likely, because its the only way (in their view) to make a game with adequate performance. The warden is added to the mix because it mitigates (not eliminates) the cheating problems that come along with trusting the client, on the belief that it does so well enough to keep them at a level where the game will not be overwhelmed by cheaters so as to destroy its userbase and, hence, profitability. And that's ultimately all they are concerned with.
They'd have made a more profitable game, yes, unless you think that the guys maintaining Warden work for free.
I think you've left out the lost revenue from having more synchronization artifacts, higher client bandwidth requirements, and also the additional cost from having more server CPU requirements to meet a given level of player load. Moving processing from the client CPUs to the server CPUs isn't free.
The client cannot be trusted. Clients request, servers decide and dictate. Let the client anticipate and drift its local world state all you like, but the server must never, ever, accept a state change from the client, only requests.
The problem with this, in practice, is while it kills cheating, it can increase the server-client bandwidth and the server CPU load. Both increase the expense of hosting the server, and the first also potentially increases the cost of maintaining a performing client setup (by potentially requiring a faster connection.)
If you can explain how to prove a hidden volume exists within a Truecrypt container, you shouldn't be on Slashdot, you should be out raking in a million-dollar salary.
Technically-focussed people tend to forget what "proof" means in a court of law. It does not mean mathematical proof. It means evidence from which a person (judge or juror) could determine that a given proposition is likely to be true to given degree of certainty (which varies from jurisdiction to jurisdiction and between different types of cases and different propositions.)
The fact that it is mathematically impossible to demonstrate the existence of a second encrypted volume within the first by examining the contents of the partition does not mean that it is impossible to show, to any legally-required degree of certainty that such a volume exists by other evidence, such as:
1) Witness testimony about the computer's owners activities, including their use of the computer, 2) What is and isn't found on the open and first encrpyted partitions on the computer, 3) The fact that the software used to encrypt the disk supports a second hidden partition.
"Social engineering" is (or at least used to be) widely recognized as being as important, often more important, than pure technical attacks in hacking. The same thing, essentially, applies here.
I'm not saying you're wrong, but wouldn't this make every piece of literary criticism and analysis of a copyrighted work infringing?
No, it wouldn't.
What is the difference between this case and a doctoral student exploring the themes of e.g. a collection of poems by E.E. Cummings in a thesis?
The difference is that purpose is a consideration in Fair Use analysis, and that "criticism", "scholarship", and "non-commercial educational purposes" are explicitly held out as purposes that weigh in favor of a finding of Fair Use. Note that Fair Use is a limitation on the basic exclusive rights under copyright, so before Fair Use analysis is necessary, the use must be one that would otherwise be infringing.
A few would be willing to give up the right for the rest of their lives for one million dollars
$1 million dollars invested to produce even a rather miserable 2.5% return would give you $25,000 a year. Assuming that was entirely devoted to political activism (campaign donations, independent expenditures, etc.), I would think it'd produce substantially more effect than the one average voter. So, really, even looked at from an "concerned citizen" standpoint, it seems rational under the status quo system.
If the DB (schema) will be used by multiple different projects, it should be accesed over SOAP/XML/Rest/WhatEverProtocolDuJour _only_ Otherwise schema changes to the DB will affect more and more projects directly.
If you can provide a consistent SOAP/XML/REST/etc. interface, you can provide a consistent interface at the database level, using views and related DB features. Ideally, at least; some db servers may be deficient in this regard, and of course you can work around this by using a web application to paper over server deficiencies just as you can with security deficiencies, but in either case you ideally shouldn't need to do that for technical reasons.
Seams a lot more useful in a war zone. At a roadblock in Iraq i think people would appreciate their engine getting shut off a little more the getting shot at.
OTOH, its as likely that military/security forces enforcing roadblocks in a war/occupation zone (or totalitarian state) would use it to disable a car and still shoot at it rather than as an alternative to shooting at the vehicle.
What you said is true, but then why do people who even request fixes and changes get a ton of +5 insightful 'if you don't like it, for it' replies here on Slashdot?
(1) Because slashdot moderation is fairly meaningless. There are lots of + reasons, and very few - reasons, and concepts that are 180-degrees opposed, on the same thread, will get modded up to +5 because different segments of the community approve for different reasons.
(2) Because Slashdot isn't the place to request a fix, and "if you don't like it, fork it" is about the best that can be reasonably expected of Slashdot (except where there is an existing fix, in which case, "if you don't like it, use foo that fixes the problem" is the best response.)
Is that a reasonable response to someone who's asking to fix bugs and leaks?
If they are asking on Slashdot (and not in a thread gathering questions for, say, a lead Mozilla developer for a Slashdot interviewer or an Ask Slashdot piece looking for existing patches, add-ons, etc., that address a need), yes.
If they are asking, say, on Bugzilla, maybe or maybe not, that depends what the specific "bug and leaks" in question are. In many cases, "we don't view this as a high priority fix; you are welcome to either fork Firefox or submit a patch on your own" would be a reasonable response.
Slashdot Mirror
This is, AFAICT, not the current "medical orthodoxy", but a strawman. Sure, most medical practitioners, dieticians, etc., will advise people who want to lose weight to cut down on fat (its the easiest place for most people to cut down without losing necessary nutrition, and has other, non-weight health benefits when cut) and to exercise more (because it keeps your body from slowing down when you cut calories, making it more practical to maintain a calorie deficit, and because it has other, non-weight health benefits) if they want to lose weight, but I've never heard one say either that "eating fat makes you fat" or "exercise makes you thin".
Most seem to say the same thing: calorie surpluses make you gain weight, calorie deficits make you lose weight. The advice they give is on how to maintain good health and a calorie deficit simultaneous, so that you can lose weight and improve health. And while less fat and more exercise are common parts of that (since they are appropriate for most people), they aren't the only advice given, nor are they given to everyone (because what is appropriate does, in fact, vary based on what the person is doing already, their other health conditions, etc.)
SQLite still exists and, as best I can tell, is moderately popular.
Rather, gratuitous copyright licenses like the GPL only restrict your use of rights that are exclusive to the copyright holder under the law applicable to your activity. This includes most copying, but it also includes things that are not copying, and excludes copying under circumstances where that copying is either not within the basic scope of the exclusive rights under copyright or is within the scope of a specific exception to copyright (e.g., in the U.S. at least, "fair use".)
U.S. copyright law provides the copyright owner an exclusive right of public performance only for "literary, musical, dramatic, and choreographic works, pantomimes, and motion pictures and other audiovisual works" (17 USC 106). Some computer software (many games) I suppose might qualify as either "motion pictures" or "other audiovisual work", but an open-source programming language interpreter, database access layer, or web application framework aren't going to qualify as anyo of those things.
In that both are completely inappropriate for that, this is correct. The GPL is not "free as in speech", however much its proponents want to claim that way. Its a conditional license offered by the owner of a property (a copyright), and is, as such, "free as in beer".
Public domain software is "free as in free speech" -- you may use it freely without restrictions or conditions imposed by any private party so long as you don't break generally applicable laws in what you do with it. GPL software is a "gift" given to you in the hope that you will serve the interest of the property-holder granting you the license, and with terms that restrict your ability to use it in other ways. "Free as in free speech" for that is a clever sales tactic, but not even a remotely accurate parallel.
Maybe so, maybe not. Gratuitous licenses (i.e., ones that are not contracts, which would include the BSDL and GPL) are generally, in US law at least, revocable at will by the licensor (whether or not their terms say they are "irrevocable" or something similar.) This is, of course, not something the FSF ever mentions in their attempts to argue that the GPL is a license on which both developers and users of software can depend and feel secure in their consistent, unambiguously-defined rights.
Its from an lobbying group whose pushing a tiered internet and other telecom-friendly government policy as the solution; so its not the "we have a product that is a solution" type of thing, but essentially the political equivalent.
The most glaring one I can remember was on the morning of September 11, 2001, but its not the only one that has occurred, and undoubtedly won't be the last. Also, the same thing happens with any other limited communications service (POTS systems can be -- and have been -- overloaded during major events!), and with (and where we get the name) electrical grids.
So, yeah, by 2010, internet brownouts "might" happen. They already do happen. And we all survive.
Aside from pushing a meaningles scary buzzword ("exaflood"), this is an unsurprising study by a largely telecom-industry-funded lobbying group favoring tiered internet services and other telecom-friendly policy that, surprise of surprises, finds that with the current, mostly-neutral internet, the whole system is about to collapse, and it will be used to sell the idea that we have to abandon that model, let telecoms charge additional fees to get data delivered even though they already charge each end for every byte transferred, etc.
Why? If you do a proof without a program, you don't have to submit a full copy of your brain, do you?
The "proof" is the thing that proves the theorem (hence the name), and it must do so in adequate detail showing all necessary steps and, as necessary, justifying them. If it does so, no additional information about how it was decided to take particular steps is necessary. If it does not, it fails regardless of what additional information provided.
Many companies I've known won't use software if they aren't paying someone for support and a license (and, often, particular support guarantees and/or performance warranties.) I don't know anything about Japanese culture, but in business the need to have some else that's feet can be held to the fire if something goes wrong is a big deal.
Of course, you can get paid support (and sometimes licensing, when the software is under a dual OSS/commercial licensing model) for most OSS you might want to adopt in a business environment, so neither cultural nor business-based reluctance to use software without paying for it should be a major barrier to OSS adoption.
Its odd that he would say "disassembled code" if he meant "machine language".
Good question. Lets look at the excerpt from TFA included in TFS:
Sure looks to me that what Lajos is being credited with isn't the disassembly, at all.
It'd be an improvement if MS did either.
It does detract from the validity of the assertion that the game would have been more profitable if that architecture had been chosen, which is what I was responding to.
This assumes that the fundamental goal of a gaming company is to defeat cheaters. This is not the case. Their fundamental goal is to make a profit. Making cheating difficult is useful only insofar as it helps acheive that goal; if doing it makes it impossible (or even just too much more costly) to develop a game with adequate performance to attract players, its not worth it, however ideal it might be in some abstract anti-cheating analysis.
Blizzard isn't trusting the client because of the warden; they are trusting the client, most likely, because its the only way (in their view) to make a game with adequate performance. The warden is added to the mix because it mitigates (not eliminates) the cheating problems that come along with trusting the client, on the belief that it does so well enough to keep them at a level where the game will not be overwhelmed by cheaters so as to destroy its userbase and, hence, profitability. And that's ultimately all they are concerned with.
I think you've left out the lost revenue from having more synchronization artifacts, higher client bandwidth requirements, and also the additional cost from having more server CPU requirements to meet a given level of player load. Moving processing from the client CPUs to the server CPUs isn't free.
The problem with this, in practice, is while it kills cheating, it can increase the server-client bandwidth and the server CPU load. Both increase the expense of hosting the server, and the first also potentially increases the cost of maintaining a performing client setup (by potentially requiring a faster connection.)
Technically-focussed people tend to forget what "proof" means in a court of law. It does not mean mathematical proof. It means evidence from which a person (judge or juror) could determine that a given proposition is likely to be true to given degree of certainty (which varies from jurisdiction to jurisdiction and between different types of cases and different propositions.)
The fact that it is mathematically impossible to demonstrate the existence of a second encrypted volume within the first by examining the contents of the partition does not mean that it is impossible to show, to any legally-required degree of certainty that such a volume exists by other evidence, such as:
1) Witness testimony about the computer's owners activities, including their use of the computer,
2) What is and isn't found on the open and first encrpyted partitions on the computer,
3) The fact that the software used to encrypt the disk supports a second hidden partition.
"Social engineering" is (or at least used to be) widely recognized as being as important, often more important, than pure technical attacks in hacking. The same thing, essentially, applies here.
No, it wouldn't.
The difference is that purpose is a consideration in Fair Use analysis, and that "criticism", "scholarship", and "non-commercial educational purposes" are explicitly held out as purposes that weigh in favor of a finding of Fair Use. Note that Fair Use is a limitation on the basic exclusive rights under copyright, so before Fair Use analysis is necessary, the use must be one that would otherwise be infringing.
$1 million dollars invested to produce even a rather miserable 2.5% return would give you $25,000 a year. Assuming that was entirely devoted to political activism (campaign donations, independent expenditures, etc.), I would think it'd produce substantially more effect than the one average voter. So, really, even looked at from an "concerned citizen" standpoint, it seems rational under the status quo system.
If you can provide a consistent SOAP/XML/REST/etc. interface, you can provide a consistent interface at the database level, using views and related DB features. Ideally, at least; some db servers may be deficient in this regard, and of course you can work around this by using a web application to paper over server deficiencies just as you can with security deficiencies, but in either case you ideally shouldn't need to do that for technical reasons.
OTOH, its as likely that military/security forces enforcing roadblocks in a war/occupation zone (or totalitarian state) would use it to disable a car and still shoot at it rather than as an alternative to shooting at the vehicle.
(1) Because slashdot moderation is fairly meaningless. There are lots of + reasons, and very few - reasons, and concepts that are 180-degrees opposed, on the same thread, will get modded up to +5 because different segments of the community approve for different reasons.
(2) Because Slashdot isn't the place to request a fix, and "if you don't like it, fork it" is about the best that can be reasonably expected of Slashdot (except where there is an existing fix, in which case, "if you don't like it, use foo that fixes the problem" is the best response.)
If they are asking on Slashdot (and not in a thread gathering questions for, say, a lead Mozilla developer for a Slashdot interviewer or an Ask Slashdot piece looking for existing patches, add-ons, etc., that address a need), yes.
If they are asking, say, on Bugzilla, maybe or maybe not, that depends what the specific "bug and leaks" in question are. In many cases, "we don't view this as a high priority fix; you are welcome to either fork Firefox or submit a patch on your own" would be a reasonable response.