The sad thing about this is that people look to detect 'fakes' in order that they can be weeded out and derided as "not good". But if an artist can paint as well as one of the Masters, shouldn't we be excited to find a 'fake' because it means that there is another great painter out there who we know nothing about - and who paints so well that even an art expert can't point out why that person is a worse painter than Rembrandt?
We should be looking for other masterpieces by the same guy and hanging those up next to the Rembrandts too.
That is true - some news does end up in Wikipedia.
However, that "problem" (well, perhaps not problem...but "issue") is being actively dealt with in the slowly growing WikiNews spin-off. WikiNews isn't yet strong enough to replace the kinds of information that Wikipedia has on natural disasters - but that's largely a matter of education - both of authors and readers.
It would (IMHO) be very wrong to exclude the newest 1% of members from posting to WikiNews. But when news turns into history - we can take a more leisurely approach to fixing problems.
Aside from the principles of the thing - that's three articles out of 800,000 that can't be edited by *everyone* - and with the number of members growing at a rate of 5 to 10% per month - anyone who has been a member for a week or so will out of the 'newest 1%' catagory. Sure, more articles will inevitably get added to that list over time - but it's never going to be more than a vanishingly small percentage of articles.
In terms of practical limitations, that's pretty minor - and if it keeps the site maintainable and useful - it gets my vote.
As a matter of principle - well, Wiki isn't about giving people the right to free speech - it's about getting facts into an encyclopedia.
It is believed that the encyclopedia will be better if everyone can edit any article at any time because 'Many eyes make all bugs shallow'. Even as an uninformed layperson looking up Aardvarks, I can spot a spelling mistake in an article and fix it right then and there...but in the case of the kinds of articles being restricted here, there are already PLENTY of eyes on them and adding more won't improve the encyclopedia.
From that perspective, how likely is it that someone who has authoritative knowledge about those few articles will know something that is verifiable that can't wait one week to be posted?
You might argue that (say) some insider in the pay of George Bush needs to be able to post especially incriminating evidence that he/she just discovered onto the Wiki page - and might need to do so either urgently - or anonymously. But that kind of information is unverifiable and falls under the 'no original research' criteria which would eliminate it from Wiki anyway. Wiki isn't a news site - information of that kind should be posted elsewhere first - and only end up in the encyclopedia when it's been verified, understood, etc.
People who visit the Wiki and search on 'George Bush' should not expect to find the latest, juicy tidbits about him there. It's an encyclopedia - they should expect to find historical information that's reasonably well established. It should contain information ABOUT any controversy without actually being controversial itself.
The VAST amount of work that goes on in the Wiki is far more mundane. The other day, I looked up Red Squirrels - found that a sentance about the number of young they bear was incomprehensible - so I looked the information up on half a dozen web pages about squirrels to find out the truth - and corrected the sentence right then and there.
Red Squirrels - not reigning US monarchs^H^H^H^H^H^H^H^Hpresidents.
Cool! Language Wars. Let loose the flaming trolls!!
If you are a Linux nut (as well you should be) then it's gotta be Java since C# is a work of the devil.
In the end, once you know one OOP language, you know 95% of what you need to work in any OOP language - so if you learned Java - but needed to pick up C# or C++ or something in the future, it wouldn't be that hard.
I guess you could do the course in C# and teach yourself Java in parallel on your Linux box...but that's more work.
The issue is that of the tradeoff between the cost of launching a base camp and the cost of maintaining it.
The cost of putting a simple air-filled cylinder into orbit - with enough consumables to keep a handful of people alive for a week or two is MUCH smaller than the cost of getting the ISS up there. But the trouble with simple approach that is that every time you need a base station, you have to launch a new one. So over N missions that need such a base station, you pay N x LaunchCost.
However, with the ISS, having launched it, you have to maintain it. The original plan for ISS was that it would pay for it's maintenance with science results - but because the crew have to spend 100% of their time maintaining the place - zero science ever comes out of it. So we are paying solely for maintenance. That wouldn't be so bad if we had a heck of a lot of missions using it as a stepping stone to wherever - but we aren't. We have not had a single mission actually use the ISS - and we're unlikely to have one for another 10 years.
Even after 10 years, we'll probably only fly new missions that rely on a base camp every 5 years or so (and that's pretty darned optimistic).
So we're paying decades worth of maintenance for just one or two uses (or maybe none at all).
The cost of repeatedly launching cheap, disposable stations (such as SkyLab) is much less.
NASA had all the right technologies in their hands with the Saturn V. It could launch moon missions, short-lived space stations, it was capable of reaching the Lagrange points - the works. The shuttle on the other hand can't even reach synchronous orbit where the big money is at for commercial satellite recovery and repairs! It's orbital payload is pathetic in comparison to the Saturn V and it's cost per mission is more - despite it's much vaunted reusablility!
If the ISS and Shuttle debacles have taught us anything, it's that disposable spacecraft and disposable space stations are the way to go. IMHO, we knew that just a few years into shuttle operations. The ISS taught us nothing.
Whilst it's true that most reputable media wouldn't go with something as extreme as this - they do get successfully sued for libel all the time.
The ability of the victim (or anyone else for that matter) to correct the problem at source the very moment it's detected is entirely new - and offers a more effective and more amicable way to sort out small mistakes, larger errors and malicious bullshit.
In most cases when someone says or prints something nasty and incorrect about you, you have a major struggle on your hand to get it retracted.
This is the one - possibly the *only* - place where you can simply get in there and fix it yourself. Yes, someone can then go back and trash you again - but there are Wiki mechanisms to get that fixed.
If someone had said this stuff about the guy on Slashdot - or in the New York Times - or on radio or TV - he'd have had an enormous fight on his hands to get his good name cleared - and in all likelyhood, never have gotten clear retractions. A retraction in a newspaper doesn't retract all of the copies already in print - an erratum or even a full apology is going to go unread by the vast percentage of readers and would possibly occur weeks or months after the damage was done.
In this case, a dozen keystrokes would have fixed the problem within minutes of the problem being discovered - and REPLACED the offending material burying the original maligning text where most people will never look - and those who do will understand clearly what happened from the document history. Furthermore, the fact that nobody noticed the problem means that almost certainly nobody read the darned article in the first place.
This should never have happened to Wikipedia - it's the one place where this kind of thing isn't a real problem.
Since anonymous users can still edit articles, it is perfectly possible to log in using a legitimate account, create a new page with minimal content then log out and put whatever crap you want into it using an anonymous account.
Perhaps a better approach is to somehow disallow access to disconnected pages. When the last link to the page goes away, the article is put into hibernation until someone again links to it.
In the Mythbusters book, there is a list of "The 12 myths that you'll probably never see broken on Mythbusters". Some of them are obviously untestable ("Does your soul weigh 21 grams?").
Number six on the list says: "Flat VW: Is it possible that a flattened VW bug was found between a head-on collision of two massive semi trucks."
Well, that seems *AWFULLY* similar to the show I saw this evening. I'd like to hear the process by which this changed status from impossible to actually happening? Why did you change the myth to require that the trucks would be fused together - and most important: Why didn't you use a VW bug? How can we expect any scientific validity without a squished bug?
I *really* want to see you do "Birds in a Truck" though - "Can a truckload of 1000 pigeons cause a truck to levitate if it hits a bump?"....well, how about "Does the truck weigh less if you bang on the sides to make them all fly around inside?" (Yeah - I know the answer).
Great show though - the world needs more entertaining science on TV.
Any serious or even semi-serious Lego builder will tell you that MegaBloks are *AWFUL* compared to the genuine article.
They are made of a much harder plastic and after being snapped together and pulled apart just a couple of times, they wear out to the point where they hardly stick together at all.
Lego parts are of a slightly more 'rubbery' plastic - they feel almost oily to the touch. I have Lego bricks from 40 years ago that still work just as well as they day they were first used.
When my son was given a bunch of MegaBloks as a present, they 'polluted' our vast Lego collection. Every time I find one, I toss it straight into the trash.
About the only use for MegaBloks is in making large sculptural pieces that you want to glue together to make permenant. The hard polystyrene in MegaBloks can be glued together with polystyrene cement - and the issue of wear becomes irrelevent!
Yet other Lego clones exist - but they tend to have poorer tolerances than either Lego or MegaBloks and can actually damage your real Lego if you mix them.
Oxford Dictionary of Quotations is illegal?!
on
Reining in Google
·
· Score: 1
Pick up a good book of quotations.
You'll find a bunch of quotations - many from copyrighted works.
They are there in part so you can find something witty to say at parties (plagiarism in it's worst form) - or you can use them to find the book from which some famous quotation comes.
This seems to be no different from what Google is doing - except that they are able to produce even the most obscure quotations as well as the famous ones (that's just a much *better* dictionary of quotations - it's not logically different)...and Google has copied the books in their entierity in order to do that.
Copying those books is only illegal if:
1) The book is still in copyright...and...
2a) They don't own a copy of the book...or...
2b) They distribute copies of the book beyond what is allowed under fair use.
But Google aren't distributing copies - so as far as I can tell, so long as they buy copies of any copyrighted books they might index, they are on solid ground.
There is definitely an argument that Google should purchase one copy of every still-in-copyright book they index (for all I know, they may actually be doing that - does anyone know for sure?)
Whilst that would represent a significant cash outlay for them, it's not going to assuage the fears of all the authors and publishers out there who stand to sell exactly one additional copy of their precious works.
However, I don't see how Google could be accused of breaking any law by doing that - any more than the Oxford Dictionary of Quotations is breaking the law by printing quotations from the works of famous authors.
> That said, with a decent OCR program, it would be pretty trivial to > write a script that just dumps a book from the snippets provided by > Google.
As a practical matter, I'm not sure it's even possible.
If you searched on the title of the book and (say) the last sentance of the last piece of text you got from the Google Index quotation, you *might* get the next couple of sentances - but you might also get a quotation from a review of the book. Your search mechanism would also tend to get stuck in a loop if the book used the exact same sentance in two or more places.
As a practical method for stealing content, it's really not gonna fly.
However, one must be concerned with theoretical issues - so I guess your point is taken.
> The big difference between a "book data" cache and a web content cache is > that web data is already freely available on the web.
The big difference between a "web data" cache and a book content cache is that book data is already freely available in public libraries.
Same difference.
You are allowed to read library books freely - but you aren't allowed to make permenant copies of them. You are allowed to read web sites freely - but you aren't allowed to make permenant copies of them.
There is complete symmetry here - both legally speaking - and in practice.
There isn't a different copyright law for the Internet.
Re:Indexing or Caching?
on
Reining in Google
·
· Score: 2, Insightful
> Indexing a printed work in no leads to the user actually doing anything that will > make money for the person(s) responsible for that work.
How not?
How is this different from the list of citations at the end of a scientific paper?
How else do you lead someone to something that'll make money for the author?
1) You quote the title of the work, the name of the author and the name of the publisher. I'm pretty sure Google will do that.
2) You might provide a link to a website where you can purchase the book. How long do you think it'll be before Google are making money by pointing you to Amazon and taking a click-though fee for that?
I'm 100% certain that indexing printed works will be a win for the authors and publishers.
What's much less certain is how much of the work Google could quote without destroying sales - their usual two lines of contextual quotation is *occasionally* enough to prevent me from clicking through to the underlying web site. If you were looking for the atomic weight of cadmium and enough of the target were quoted to tell you the answer - why would you click through? (Well, you might...but you wouldn't have to). In that case, the original web page author might be denied some advertising revenue or other valuable statistical data that he was seeking when putting up that information for free access.
I think a couple of sentances of content is enough to verify that the web site is talking about the chemistry of Cadmium - and is not an artist supplier selling Cadmium Yellow paint. But it's a thin line.
But what about books? I think the threshold for a book is much larger than for a web site. I would be buy the book just in order to read the rest of that paragraph? I think not. I think quoting *more* of the book is useful to both the author and to the person doing the searching.
The cache is a different matter...there are many good reasons (listed ably by others earlier in this thread) why I might want my copy of the page to be the only one people access. I let them read it - but I might prefer that copying were not allowed.
There is also a distinction here between what IS legal and what OUGHT TO BE legal. Both are worthy of discussion in the context of "Should the copyright law be changed in order to *clearly* allow indexing?" - which is obviously a huge public good.
I don't see any difference between what Google are doing here and what they do to index web sites.
The roam the web - they take local copies of every web page - they index those pages - then they display a 'snippet' of the page in response to a search query.
Same deal with the books. Scan them into a private archive, index the archive - display the title and a sentence or two of content to provide context. I see no problem with that.
What is problematic (both with the Web indexing and Book indexing) is the Google 'cache' - where you can get the content of the web page from Googles cache if the original web page is missing or slow. That is (in my opinion) a breach of the Web page owner's copyright - and would be a breech of the book's copyright too.
However, the indexing service that Google (and others) provide for the Web is the only thing that makes the Internet useful. Doing that for books would be of HUGE benefit to mankind and absolutely must be allowed - even if copyright law has to be changed to make it happen.
Let's think carefully about the 'Google cache' thing though - that's dubious because it allows people access to content without going through the content provider's access mechanisms. That's the thing that deprives the author of value. Indexing actually increases the value of a work because it allows people to find it - and therefore increases the pool of potential purchasers by an enormous factor.
Google indexing should be the savior of printed media and authors should support it.
The USA can only exist by the strength of it's economy.
It's economy can only exist if it's industry can continue to be profitable.
But every industry the USA develops is eventually understood well enough by other societies with lower wage costs - so to preserve it's high cost of living, outsourcing is inevitable.
The only way to survive waves of outsourcing is to develop new industries that are not yet accessible to low-wage countries. This is a never ending cycle - invent, exploit, outsource, abandon.
New Industries are driven by new technology.
New Technology is driven by new science.
And new science is driven by high standards of education coupled to the kinds of blue-sky research that pretty much only comes from government and university programmes.
Pull away the rug at the bottom - and the whole edifice comes tumbling down within maybe one or two human generations.
> Or we could remove any CSS mandate from the DVD spec. > Release an open DVD, get distribution protection., > Release encrypted DVD, you're protecting yourself.
This goes beyond the DVD standard...which carries no legal force anyway. This problem exists in all digital media.
> Obviously the better choice is copyright protection since > no encryption works anyway.
That's only true as long as future computers are much faster than present day machines. It typically takes a vastly more powerful computer to break a code than the computer that does the encoding - or the one that decodes using the official key. However, when Moores' law runs out of steam, computers will no longer be getting faster and the most sophisticated encryption scheme that can be devised at that point in time will be well-neigh impossible to crack unless there is some horrible loophole in the encryption math. With all the people studying encryption these days, that prospect vanishes too.
Remember, nobody 'cracked' the CSS scheme - the original author was given the key-code by somewhat nefarious means. If the CSS key had remained secret, DeCSS would not exist. Notice for example that the X-box encryption scheme has not been broken - despite an awful lot of effort. In that case, the key is only required by one manufacturer - so it's remained a close-kept secret. The CSS key was required by dozens of manufacturers of DVD players - so it was eventually leaked.
> As a result this will never happen because our government likes > its corporate sugar daddies.
Indeed.
The copyright office will probably do due diligence, collect public comments, pass them on to the law makers - who will ignore them and extend the life of copyrights again and again until they are effectively infinite.
Try to imagine a day when Mickey Mouse becomes public domain. Now try to imagine a world in which he never does. Which seems more plausible?
That strikes to the heart of what we're going to call 'The Work'. If the encrypted digital version is identical to the analog version then I guess the fact that the analog version is freely available means that the public has access to the work in the future so copyright protection is a fair trade. Of course the temptation would be to make and sell just one copy of the VHS in order to meet my hypothetical copyright law - then ship the remainder as encrypted digital copies - so I guess that wouldn't work. Even if you required an unencrypted version to be lodged at the library of congress or some such repository, you could make that copy be so blurry and noisey as to be essentially unwatchable.
However, if the digital version is significantly different (in the way that you can improve on an existing patent and patent the result for example) - then it must be treated as a different work.
A third option would be to say that the work is conferred copyright privilages only so long as it's not shipped in an encrypted form...so having released the VHS version under copyright law would effectively prevent you from ever releasing an encrypted digital version without losing your rights to the analog version. I don't like that option at all.
I think my preference would be to say that the digital version is treated as a separate work. So the analog version is protected by copyright law - but if you, the owner of the copyright, subsequently decided that you'd prefer to rely on encryption to getting copyright law protection - then you could release an encrypted version which (as you say) could be legally copied if the encryption was broken.
Copyrights and patents exist in order that the holder may gain some period of commercial value - protected by the state, in return for which, the work returns to the public domain once that period has expired.
It's a reasonably fair trade.
The DMCA (and encryption in general) overturns that trade. It allows the owner the protection of the law - but it doesn't give the public their rights to eventual open use.
That's not fair.
IMHO, people who encrypt their products should be denied copyright protection - in just the way that people with technical innovation can choose to keep that innovation as a Trade Secret instead of getting Patent coverage.
In the absence of this, historians of the future are going to have a very hard time finding out about our society. If most or even all media of the next century ends up locked away behind almost unbreakable encryption - with DRM hardware locking that encryption to a particular computer - it may become impossible for someone a few hundred years from now to read our newspapers, books or watch our movies. That would be a terrible thing.
It's bad enough that we've been primarily responsible for screwing over the planet for our descendents - but now we're working hard towards denying them access to our greatest musicians, authors and other media?
Right now, it seems like a computer from a few hundred years into the future would be able to break any of our codes quite easily - but when Moores' Law runs out of steam, the most powerful encryption systems we have on that day will probably be *FOREVER* uncrackable.
OpenGL has an analog of 'Retained mode' - it's called a 'Display List' and the games are already using that. Yet DESPITE THAT they are running low on bandwidth on an 8xAGP. Furthermore, if DirectX magically solved the AGP bandwidth bottleneck then no games would be written in OpenGL - and nobody would be driving people into using PCI-Express.
So - NO! You will not be sending either OpenGL *or* DirectX commands over the Internet to play games.
Most of these names are purely for short-term convenience and are unlikely ever to be used beyond the handful of scientists who work on the data in years to come.
That some small rock has a name is irrelevent.
The name doesn't even have to be unique - so long as it's unique to a particular mission - which is just as well because if you took all of the words in all of the languages of the world, you couldn't name any significant fraction of the Martian landscape down to the level of detail that they are.
Big things like hills and major craters need permenant names - but that's unlikely to be a problem.
Slashdot Mirror
The sad thing about this is that people look to detect
'fakes' in order that they can be weeded out and derided
as "not good". But if an artist can paint as well as one
of the Masters, shouldn't we be excited to find a 'fake'
because it means that there is another great painter out
there who we know nothing about - and who paints so well
that even an art expert can't point out why that person
is a worse painter than Rembrandt?
We should be looking for other masterpieces by the same
guy and hanging those up next to the Rembrandts too.
Look at the problems FireFox had finding a name that wasn't already in use for something.
On a 110 baud teletype - touch-typing doesn't help.
That is true - some news does end up in Wikipedia.
However, that "problem" (well, perhaps not problem...but "issue") is being actively dealt with in the slowly growing WikiNews spin-off. WikiNews isn't yet strong enough to replace the kinds of information that Wikipedia has on natural disasters - but that's largely a matter of education - both of authors and readers.
It would (IMHO) be very wrong to exclude the newest 1% of members from posting to WikiNews. But when news turns into history - we can take a more leisurely approach to fixing problems.
Aside from the principles of the thing - that's three articles out of 800,000 that can't be edited by *everyone* - and with the number of members growing at a rate of 5 to 10% per month - anyone who has been a member for a week or so will out of the 'newest 1%' catagory. Sure, more articles will inevitably get added to that list over time - but it's never going to be more than a vanishingly small percentage of articles.
In terms of practical limitations, that's pretty minor - and if it keeps the site maintainable and useful - it gets my vote.
As a matter of principle - well, Wiki isn't about giving people the right to free speech - it's about getting facts into an encyclopedia.
It is believed that the encyclopedia will be better if everyone can edit any article at any time because 'Many eyes make all bugs shallow'. Even as an uninformed layperson looking up Aardvarks, I can spot a spelling mistake in an article and fix it right then and there...but in the case of the kinds of articles being restricted here, there are already PLENTY of eyes on them and adding more won't improve the encyclopedia.
From that perspective, how likely is it that someone who has authoritative knowledge about those few articles will know something that is verifiable that can't wait one week to be posted?
You might argue that (say) some insider in the pay of George Bush needs to be able to post especially incriminating evidence that he/she just discovered onto the Wiki page - and might need to do so either urgently - or anonymously. But that kind of information is unverifiable and falls under the 'no original research' criteria which would eliminate it from Wiki anyway. Wiki isn't a news site - information of that kind should be posted elsewhere first - and only end up in the encyclopedia when it's been verified, understood, etc.
People who visit the Wiki and search on 'George Bush' should not expect to find the latest, juicy tidbits about him there. It's an encyclopedia - they should expect to find historical information that's reasonably well established. It should contain information ABOUT any controversy without actually being controversial itself.
The VAST amount of work that goes on in the Wiki is far more mundane. The other day, I looked up Red Squirrels - found that a sentance about the number of young they bear was incomprehensible - so I looked the information up on half a dozen web pages about squirrels to find out the truth - and corrected the sentence right then and there.
Red Squirrels - not reigning US monarchs^H^H^H^H^H^H^H^Hpresidents.
Cool! Language Wars. Let loose the flaming trolls!!
If you are a Linux nut (as well you should be) then it's
gotta be Java since C# is a work of the devil.
In the end, once you know one OOP language, you know 95%
of what you need to work in any OOP language - so if you
learned Java - but needed to pick up C# or C++ or something
in the future, it wouldn't be that hard.
I guess you could do the course in C# and teach yourself
Java in parallel on your Linux box...but that's more work.
The issue is that of the tradeoff between the cost of launching a base camp and the cost of maintaining it.
The cost of putting a simple air-filled cylinder into orbit - with enough consumables to keep a handful of people alive for a week or two is MUCH smaller than the cost of getting the ISS up there. But the trouble with simple approach that is that every time you need a base station, you have to launch a new one. So over N missions that need such a base station, you pay N x LaunchCost.
However, with the ISS, having launched it, you have to maintain it. The original plan for ISS was that it would pay for it's maintenance with science results - but because the crew have to spend 100% of their time maintaining the place - zero science ever comes out of it. So we are paying solely for maintenance. That wouldn't be so bad if we had a heck of a lot of missions using it as a stepping stone to wherever - but we aren't. We have not had a single mission actually use the ISS - and we're unlikely to have one for another 10 years.
Even after 10 years, we'll probably only fly new missions that rely on a base camp every 5 years or so (and that's pretty darned optimistic).
So we're paying decades worth of maintenance for just one or two uses (or maybe none at all).
The cost of repeatedly launching cheap, disposable stations (such as SkyLab) is much less.
NASA had all the right technologies in their hands with the Saturn V. It could launch moon missions, short-lived space stations, it was capable of reaching the Lagrange points - the works. The shuttle on the other hand can't even reach synchronous orbit where the big money is at for commercial satellite recovery and repairs! It's orbital payload is pathetic in comparison to the Saturn V and it's cost per mission is more - despite it's much vaunted reusablility!
If the ISS and Shuttle debacles have taught us anything, it's that disposable spacecraft and disposable space stations are the way to go. IMHO, we knew that just a few years into shuttle operations. The ISS taught us nothing.
De-orbit the sucker ASAP.
Whilst it's true that most reputable media wouldn't go with something as extreme as this - they do get successfully sued for libel all the time.
The ability of the victim (or anyone else for that matter) to correct the problem at source the very moment it's detected is entirely new - and offers a more effective and more amicable way to sort out small mistakes, larger errors and malicious bullshit.
In most cases when someone says or prints something nasty and incorrect about you, you have a major struggle on your hand to get it retracted.
This is the one - possibly the *only* - place where you can simply get in there and fix it yourself. Yes, someone can then go back and trash you again - but there are Wiki mechanisms to get that fixed.
If someone had said this stuff about the guy on Slashdot - or in the New York Times - or on radio or TV - he'd have had an enormous fight on his hands to get his good name cleared - and in all likelyhood, never have gotten clear retractions. A retraction in a newspaper doesn't retract all of the copies already in print - an erratum or even a full apology is going to go unread by the vast percentage of readers and would possibly occur weeks or months after the damage was done.
In this case, a dozen keystrokes would have fixed the problem within minutes of the problem being discovered - and REPLACED the offending material burying the original maligning text where most people will never look - and those who do will understand clearly what happened from the document history. Furthermore, the fact that nobody noticed the problem means that almost certainly nobody read the darned article in the first place.
This should never have happened to Wikipedia - it's the one place where this kind of thing isn't a real problem.
Since anonymous users can still edit articles, it is perfectly possible to log in using a legitimate account, create a new page with minimal content then log out and put whatever crap you want into it using an anonymous account.
Perhaps a better approach is to somehow disallow access to disconnected pages. When the last link to the page goes away, the article is put into hibernation until someone again links to it.
In the Mythbusters book, there is a list of "The 12 myths that you'll probably never see broken on Mythbusters". Some of them are obviously untestable ("Does your soul weigh 21 grams?").
Number six on the list says: "Flat VW: Is it possible that a flattened VW bug was found between a head-on collision of two massive semi trucks."
Well, that seems *AWFULLY* similar to the show I saw this evening. I'd like to hear the process by which this changed status from impossible to actually happening? Why did you change the myth to require that the trucks would be fused together - and most important: Why didn't you use a VW bug? How can we expect any scientific validity without a squished bug?
I *really* want to see you do "Birds in a Truck" though - "Can a truckload of 1000 pigeons cause a truck to levitate if it hits a bump?"....well, how about "Does the truck weigh less if you bang on the sides to make them all fly around inside?" (Yeah - I know the answer).
Great show though - the world needs more entertaining science on TV.
Any serious or even semi-serious Lego builder will tell you that MegaBloks are *AWFUL* compared to the genuine article.
They are made of a much harder plastic and after being snapped together and pulled apart just a couple of times, they wear out to the point where they hardly stick together at all.
Lego parts are of a slightly more 'rubbery' plastic - they feel almost oily to the touch. I have Lego bricks from 40 years ago that still work just as well as they day they were first used.
When my son was given a bunch of MegaBloks as a present, they 'polluted' our vast Lego collection. Every time I find one, I toss it straight into the trash.
About the only use for MegaBloks is in making large sculptural pieces that you want to glue together to make permenant. The hard polystyrene in MegaBloks can be glued together with polystyrene cement - and the issue of wear becomes irrelevent!
Yet other Lego clones exist - but they tend to have poorer tolerances than either Lego or MegaBloks and can actually damage your real Lego if you mix them.
Pick up a good book of quotations.
You'll find a bunch of quotations - many from copyrighted works.
They are there in part so you can find something witty to say at parties (plagiarism in it's worst form) - or you can use them to find the book from which some famous quotation comes.
This seems to be no different from what Google is doing - except that they are able to produce even the most obscure quotations as well as the famous ones (that's just a much *better* dictionary of quotations - it's not logically different)...and Google has copied the books in their entierity in order to do that.
Copying those books is only illegal if:
1) The book is still in copyright...and...
2a) They don't own a copy of the book...or...
2b) They distribute copies of the book beyond what is allowed under fair use.
But Google aren't distributing copies - so as far as I can tell, so long as they buy copies of any copyrighted books they might index, they are on solid ground.
Dunno whether they do that or not.
There is definitely an argument that Google should purchase one copy of every still-in-copyright book they index (for all I know, they may actually be doing that - does anyone know for sure?)
Whilst that would represent a significant cash outlay for them, it's not going to assuage the fears of all the authors and publishers out there who stand to sell exactly one additional copy of their precious works.
However, I don't see how Google could be accused of breaking any law by doing that - any more than the Oxford Dictionary of Quotations is breaking the law by printing quotations from the works of famous authors.
> That said, with a decent OCR program, it would be pretty trivial to
> write a script that just dumps a book from the snippets provided by
> Google.
As a practical matter, I'm not sure it's even possible.
If you searched on the title of the book and (say) the last sentance of the last piece of text you got from the Google Index quotation, you *might* get the next couple of sentances - but you might also get a quotation from a review of the book. Your search mechanism would also tend to get stuck in a loop if the book used the exact same sentance in two or more places.
As a practical method for stealing content, it's really not gonna fly.
However, one must be concerned with theoretical issues - so I guess your point is taken.
> The big difference between a "book data" cache and a web content cache is
> that web data is already freely available on the web.
The big difference between a "web data" cache and a book content cache is
that book data is already freely available in public libraries.
Same difference.
You are allowed to read library books freely - but you aren't allowed to
make permenant copies of them. You are allowed to read web sites freely -
but you aren't allowed to make permenant copies of them.
There is complete symmetry here - both legally speaking - and in practice.
There isn't a different copyright law for the Internet.
> Indexing a printed work in no leads to the user actually doing anything that will
> make money for the person(s) responsible for that work.
How not?
How is this different from the list of citations at the end of a scientific paper?
How else do you lead someone to something that'll make money for the author?
1) You quote the title of the work, the name of the author and the name of the publisher. I'm pretty sure Google will do that.
2) You might provide a link to a website where you can purchase the book. How long do you think it'll be before Google are making money by pointing you to Amazon and taking a click-though fee for that?
I'm 100% certain that indexing printed works will be a win for the authors and publishers.
What's much less certain is how much of the work Google could quote without destroying sales - their usual two lines of contextual quotation is *occasionally* enough to prevent me from clicking through to the underlying web site. If you were looking for the atomic weight of cadmium and enough of the target were quoted to tell you the answer - why would you click through? (Well, you might...but you wouldn't have to). In that case, the original web page author might be denied some advertising revenue or other valuable statistical data that he was seeking when putting up that information for free access.
I think a couple of sentances of content is enough to verify that the web site is talking about the chemistry of Cadmium - and is not an artist supplier selling Cadmium Yellow paint. But it's a thin line.
But what about books? I think the threshold for a book is much larger than for a web site. I would be buy the book just in order to read the rest of that paragraph? I think not. I think quoting *more* of the book is useful to both the author and to the person doing the searching.
The cache is a different matter...there are many good reasons (listed ably by others earlier in this thread) why I might want my copy of the page to be the only one people access. I let them read it - but I might prefer that copying were not allowed.
There is also a distinction here between what IS legal and what OUGHT TO BE legal. Both are worthy of discussion in the context of "Should the copyright law be changed in order to *clearly* allow indexing?" - which is obviously a huge public good.
I don't see any difference between what Google are doing here and what they do to index web sites.
The roam the web - they take local copies of every web page - they index those pages - then they display a 'snippet' of the page in response to a search query.
Same deal with the books. Scan them into a private archive, index the archive - display the title and a sentence or two of content to provide context. I see no problem with that.
What is problematic (both with the Web indexing and Book indexing) is the Google 'cache' - where you can get the content of the web page from Googles cache if the original web page is missing or slow. That is (in my opinion) a breach of the Web page owner's copyright - and would be a breech of the book's copyright too.
However, the indexing service that Google (and others) provide for the Web is the only thing that makes the Internet useful. Doing that for books would be of HUGE benefit to mankind and absolutely must be allowed - even if copyright law has to be changed to make it happen.
Let's think carefully about the 'Google cache' thing though - that's dubious because it allows people access to content without going through the content provider's access mechanisms. That's the thing that deprives the author of value. Indexing actually increases the value of a work because it allows people to find it - and therefore increases the pool of potential purchasers by an enormous factor.
Google indexing should be the savior of printed media and authors should support it.
Google caching is morally dubious.
The USA can only exist by the strength of it's economy.
It's economy can only exist if it's industry can continue to be profitable.
But every industry the USA develops is eventually understood well enough by other societies with lower wage costs - so to preserve it's high cost of living, outsourcing is inevitable.
The only way to survive waves of outsourcing is to develop new industries that are not yet accessible to low-wage countries. This is a never ending cycle - invent, exploit, outsource, abandon.
New Industries are driven by new technology.
New Technology is driven by new science.
And new science is driven by high standards of education coupled to the kinds of blue-sky research that pretty much only comes from government and university programmes.
Pull away the rug at the bottom - and the whole edifice comes tumbling down within maybe one or two human generations.
> Or we could remove any CSS mandate from the DVD spec.
> Release an open DVD, get distribution protection.,
> Release encrypted DVD, you're protecting yourself.
This goes beyond the DVD standard...which carries no legal force anyway. This problem exists in all digital media.
> Obviously the better choice is copyright protection since
> no encryption works anyway.
That's only true as long as future computers are much faster than present day machines.
It typically takes a vastly more powerful computer to break a code than the computer that does the encoding - or the one that decodes using the official key. However, when Moores' law runs out of steam, computers will no longer be getting faster and the most sophisticated encryption scheme that can be devised at that point in time will be well-neigh impossible to crack unless there is some horrible loophole in the encryption math. With all the people studying encryption these days, that prospect vanishes too.
Remember, nobody 'cracked' the CSS scheme - the original author was given the key-code by somewhat nefarious means. If the CSS key had remained secret, DeCSS would not exist. Notice for example that the X-box encryption scheme has not been broken - despite an awful lot of effort. In that case, the key is only required by one manufacturer - so it's remained a close-kept secret. The CSS key was required by dozens of manufacturers of DVD players - so it was eventually leaked.
> As a result this will never happen because our government likes
> its corporate sugar daddies.
Indeed.
The copyright office will probably do due diligence, collect public comments, pass them on to the law makers - who will ignore them and extend the life of copyrights again and again until they are effectively infinite.
Try to imagine a day when Mickey Mouse becomes public domain. Now try to imagine a world in which he never does. Which seems more plausible?
That strikes to the heart of what we're going to call 'The Work'. If the encrypted digital version is identical to the analog version then I guess the fact that the analog version is freely available means that the public has access to the work in the future so copyright protection is a fair trade. Of course the temptation would be to make and sell just one copy of the VHS in order to meet my hypothetical copyright law - then ship the remainder as encrypted digital copies - so I guess that wouldn't work. Even if you required an unencrypted version to be lodged at the library of congress or some such repository, you could make that copy be so blurry and noisey as to be essentially unwatchable.
However, if the digital version is significantly different (in the way that you can improve on an existing patent and patent the result for example) - then it must be treated as a different work.
A third option would be to say that the work is conferred copyright privilages only so long as it's not shipped in an encrypted form...so having released the VHS version under copyright law would effectively prevent you from ever releasing an encrypted digital version without losing your rights to the analog version. I don't like that option at all.
I think my preference would be to say that the digital version is treated as a separate work. So the analog version is protected by copyright law - but if you, the owner of the copyright, subsequently decided that you'd prefer to rely on encryption to getting copyright law protection - then you could release an encrypted version which (as you say) could be legally copied if the encryption was broken.
It is difficult - no doubt.
Copyrights and patents exist in order that the holder may gain some period of commercial value - protected by the state, in return for which, the work returns to the public domain once that period has expired.
It's a reasonably fair trade.
The DMCA (and encryption in general) overturns that trade. It allows the owner the protection of the law - but it doesn't give the public their rights to eventual open use.
That's not fair.
IMHO, people who encrypt their products should be denied copyright protection - in just the way that people with technical innovation can choose to keep that innovation as a Trade Secret instead of getting Patent coverage.
In the absence of this, historians of the future are going to have a very hard time finding out about our society. If most or even all media of the next century ends up locked away behind almost unbreakable encryption - with DRM hardware locking that encryption to a particular computer - it may become impossible for someone a few hundred years from now to read our newspapers, books or watch our movies. That would be a terrible thing.
It's bad enough that we've been primarily responsible for screwing over the planet for our descendents - but now we're working hard towards denying them access to our greatest musicians, authors and other media?
Right now, it seems like a computer from a few hundred years into the future would be able to break any of our codes quite easily - but when Moores' Law runs out of steam, the most powerful encryption systems we have on that day will probably be *FOREVER* uncrackable.
Yep.
The nasty thing about latency is that the laws of physics prevent much advancement. The speed of light becomes noticable in Ping times.
OpenGL has an analog of 'Retained mode' - it's called a 'Display List' and the games are already using that. Yet DESPITE THAT they are running low on bandwidth on an 8xAGP. Furthermore, if DirectX magically solved the AGP bandwidth bottleneck then no games would be written in OpenGL - and nobody would be driving people into using PCI-Express.
So - NO! You will not be sending either OpenGL *or* DirectX commands over the Internet to play games.
Most of these names are purely for short-term convenience and are unlikely ever to be used beyond the handful of scientists who work on the data in years to come.
That some small rock has a name is irrelevent.
The name doesn't even have to be unique - so long as it's unique to a particular mission - which is just as well because if you took all of the words in all of the languages of the world, you couldn't name any significant fraction of the Martian landscape down to the level of detail that they are.
Big things like hills and major craters need permenant names - but that's unlikely to be a problem.