The tragedy is that they ever tried to pursue this. It was extremely foolhardy, and many people told IA leadership this from the beginning. Now that decision may destroy the entire project.
The tragedy is that they ever tried to pursue this. It was extremely foolhardy, and many people told IA leadership this from the beginning. Now that decision may destroy the entire project.
The greatest tragedy of our time is that we invented a device that makes copying so trivially cheap and easy that it happens by accident and as part of normal operation, and one of the first...
The greatest tragedy of our time is that we invented a device that makes copying so trivially cheap and easy that it happens by accident and as part of normal operation, and one of the first questions asked by the bosses was:
"How do we keep it from doing that?"
We sacrificed infinite abundance at the alter of profitability.
You could say the same thing about crawling the Internet to gather data for AI, and yet people are mostly against that, too. It's not just the bosses. Lots of people have sympathy for writers and...
You could say the same thing about crawling the Internet to gather data for AI, and yet people are mostly against that, too.
It's not just the bosses. Lots of people have sympathy for writers and artists. There are genuinely conflicting interests between creators and consumers, and they don't go away when technology changes.
My point is that we, collectively as a society, circa 1980something, had an opportunity to rethink how knowledge (and creative works, but that came somewhat later) should be funded and...
My point is that we, collectively as a society, circa 1980something, had an opportunity to rethink how knowledge (and creative works, but that came somewhat later) should be funded and distributed. We could have embraced post-scarcity, insuring that creators don't starve while creations get to be distributed freely. We hit this point again in the 90s with the advent of Napster going mainstream. It's always been there, lurking in the background, but addressing how much harm we're placing on the future by introducing scarcity where there doesn't need to be makes existing interests uncomfortable.
As such, we doubled-down on business as usual, and buried our heads in the sand.
I'd have 0 qualms with the total ingestion of content for AI training if we had already embraced the end of scarcity on computers. But since we already decided that artificial scarcity needs to be a thing, it is decidedly unfair that trillion-dollar corporations get to ingest for free while the creators get nothing and users get to pay for access.
Okay, I see your point. But maybe post-scarcity is a bit of an exaggeration? Just because we can copy data easily doesn't mean anything else is free, or about to become free. Often, it means...
Okay, I see your point.
But maybe post-scarcity is a bit of an exaggeration? Just because we can copy data easily doesn't mean anything else is free, or about to become free. Often, it means something else becomes the scarce resource.
And I think our attitude towards paywalls shows some contradictions. Computers don't make reporting free and it's not really post-scarcity. It's just that we're often unwilling to pay for it. I'm not sure it's actually a matter of principle. We're dickering over the price, who pays, how to pay.
It does mean that, often, government-funded services can be distributed a whole lot cheaper and often that's a good thing. If you want government-funded news, there is the BBC, Voice of America, and other news services like that.
I also don't see charging users for access to AI chat as a nefarious plot when really is expensive. In practice, free often means "not actually free, but cheap enough to be funded by advertising." And that's a whole different can of worms. Maybe LLM's will become free to use when researchers come up with enough performance improvements, though.
It's because the people who are in charge of this are good at poisioning the well in regards to these things. If a newspaper prints something that is wrong and it ends up hurting someone, they can...
It's because the people who are in charge of this are good at poisioning the well in regards to these things.
If a newspaper prints something that is wrong and it ends up hurting someone, they can be sued to oblivion. It's illegal to, for instance, print slander against people. There have been many news outlets that have closed because of these kinds of laws. It doesn't matter if it was one of their employees who wrote the offending text - they are responsible as the publisher. Fast forward to today, and there is slander all across the internet. But the publishers cannot be sued for it. Why? Because we made laws that expressly remove responsibility for what they publish. Why did we do that? Because most people did not understand the internet well enough to understand what they were doing and those who did had everything to gain, so they took advantage and "educated" the masses about what "needed to happen".
Large IP holders have done everything they could do for ages to get rid of consumer rights for essentially as long as copyright has existed. Every time a new digital physical format has come along, there have been built-in DRM schemes. Did you know that SD cards stand for Secure Digital, because the primary selling point to corporations was the encryption scheme? And now we have the DMCA, which makes circumventing DRM a federal crime. Bully.
Corporations have far too much power, and the more they have gotten away with the more powerful they have become.
Are they though? Or is it just a sentiment that gets over represented in certain social media circles? The closest I've seen to actual public opinion polling is a Pew survey that found that the...
people are mostly against that, too.
Are they though? Or is it just a sentiment that gets over represented in certain social media circles? The closest I've seen to actual public opinion polling is a Pew survey that found that the overwhelming majority of respondents believe AI applications should credit the sources used in their models. Otherwise, I have not seen anything objective indicating widespread disapproval over the practice.
Letting people borrow books is what libraries do. Giving copies of books away for free is different. Somehow the Internet Archive thought they could make it legal, but that was never going to happen.
Letting people borrow books is what libraries do. Giving copies of books away for free is different. Somehow the Internet Archive thought they could make it legal, but that was never going to happen.
I don't understand. I've been borrowing books from the Internet Archive for years, and they are indeed borrowed - I don't get to keep them. And the time limits are much more strict than with...
I don't understand. I've been borrowing books from the Internet Archive for years, and they are indeed borrowed - I don't get to keep them. And the time limits are much more strict than with physical libraries.
The difference is that the libraries have contracts and agreements with publishers. Whereas the IA just scanned some books and are now trying to “lend” them unilaterally.
The difference is that the libraries have contracts and agreements with publishers. Whereas the IA just scanned some books and are now trying to “lend” them unilaterally.
Why should they need the publishers permission? If I buy a book, I'm allowed to lend it to someone. After they give it back I can lend it to someone else.
Why should they need the publishers permission? If I buy a book, I'm allowed to lend it to someone. After they give it back I can lend it to someone else.
I think the issue here isn't 1:1 lending of the ones they have in their own possession. The problem is that they teamed up with a number of physical libraries and added their collections to the...
I think the issue here isn't 1:1 lending of the ones they have in their own possession. The problem is that they teamed up with a number of physical libraries and added their collections to the number they'd lend. That's the main part the court ruled against - saying that they couldn't count the third party ones.
I'm not totally sure, but to me this sounds like one-to-one lending is illegal: There is also this in the footnotes of the conclusion:
I'm not totally sure, but to me this sounds like one-to-one lending is illegal:
This appeal presents the following question: Is it “fair use” for a
nonprofit organization to scan copyright-protected print books in their
entirety, and distribute those digital copies online, in full, for free, subject to
a one-to-one owned-to-loaned ratio between its print copies and the digital
copies it makes available at any given time, all without authorization from
the copyright-holding publishers or authors? Applying the relevant
provisions of the Copyright Act as well as binding Supreme Court and
Second Circuit precedent, we conclude the answer is no. We therefore
AFFIRM.
There is also this in the footnotes of the conclusion:
IA makes a final argument that, even if its Open Libraries project did not qualify as a fair use,
we should restrict the injunction to the Open Libraries project and allow IA to continue CDL for
books that IA itself owns. Appellant’s Br. at 62. In support of that argument, IA argues that the
fourth factor analysis would be more favorable if CDL were limited to IA’s own books. Id. In
our view, the fair use analysis would not be substantially different if limited to IA’s CDL of the
books it owns, and the fourth factor still would count against fair use. So we decline IA’s
invitation to narrow the scope of our holding or of the district court’s injunction.
That is true of physical books, but it's not true of digital copies, at least not in general. If you own a digital copy of a file, you generally own some sort of license or right to use that file,...
That is true of physical books, but it's not true of digital copies, at least not in general. If you own a digital copy of a file, you generally own some sort of license or right to use that file, but it's not a physical good that you can own in the same way as a book.
This is why I buy digital books DRM free directly from authors where I can. Or from digital book stores, then grab a DRM free copy from elsewhere. This bullshit about you don't own what you buy...
This is why I buy digital books DRM free directly from authors where I can. Or from digital book stores, then grab a DRM free copy from elsewhere.
This bullshit about you don't own what you buy because they try and gaslight you that what you "purchased" was some crappy revocable licence needs putting in the bin.
That's not exactly true, as those prohibitions are bounded by fair use. Libraries have always been permitted to scan books for archival, it's seen as fair use. As well as scanning for the purposes...
That's not exactly true, as those prohibitions are bounded by fair use. Libraries have always been permitted to scan books for archival, it's seen as fair use. As well as scanning for the purposes of indexing and searching. The scanning itself is not illegal, its more of the distribution where the fair use gets a bit murky.
Libraries were digitizing books since the 1980s, pioneers of the ebook, and the publishing industry has been trying to stop it for decades, but that pesky "fair use for public benefit" really puts a cramp in their style.
There's a reason publishers are honing in on Internet Archive's worldwide-emergency violation, and not the university libraries that do the exact same thing with the limit to the physical copy restriction.
If you bought yourself a book scanner, as a private person, it would not be illegal for you to scan every book you own and upload it to your ereader and leave your books on the shelf. Nor would it be illegal to loan your ereader to your friend. If you got rid of all the books, and/or never got that ereader back from your friend and made an additional copy, that would start treading away from fair use into violation territory. And the very existence of the lending library itself leans on that fair use construct.
This fight will almost certainly make it to the supreme court, and even with the heavy conservative bent, I would be surprised if they fully overturn the ideas behind digital surrogacy.
I agree, except that it seems like more of a loss for Google compared to what they wanted to do. But it seems pretty clear now that search is allowed, letting users read the whole book is not....
I agree, except that it seems like more of a loss for Google compared to what they wanted to do. But it seems pretty clear now that search is allowed, letting users read the whole book is not. Unless the publisher permits it.
Eh. I mean, if they decided tomorrow that all your physical books were also just licensed physical files and that you weren't allowed to do what you pleased with them it be equally stupid.
Eh. I mean, if they decided tomorrow that all your physical books were also just licensed physical files and that you weren't allowed to do what you pleased with them it be equally stupid.
This has not always been true. Scanning physical copies and lending the scan was originally protected, and IA isn't the only one doing it. It'd be different if they were taking the publisher's...
This has not always been true. Scanning physical copies and lending the scan was originally protected, and IA isn't the only one doing it.
It'd be different if they were taking the publisher's ebook, cracking the DRM, then handing it out. But that's not what was happening.
I haven't used it so perhaps I misunderstood something. I've read that during the pandemic, they let people borrow an unlimited number of copies, more than they actually had? Normally a library...
I haven't used it so perhaps I misunderstood something. I've read that during the pandemic, they let people borrow an unlimited number of copies, more than they actually had?
Normally a library has to buy each copy of a book that it lends out, and books sold mostly to libraries are often more expensive.
During the pandemic they allowed any given digital book to be borrowed an unlimited amount. Individual users could still only borrow a limited number of books, but each book no longer had a cap on...
During the pandemic they allowed any given digital book to be borrowed an unlimited amount. Individual users could still only borrow a limited number of books, but each book no longer had a cap on how many users could simultaneously borrow it.
Otherwise, their controlled digital lending program remained unchanged:
Each user could only borrow so many books at once
Each digital book file was encrypted and could not be redistributed
Each digital book file expired after two weeks
Literally the only thing the Internet Archive did differently with the National Emergency Library program was lend out more digital copies of books than they had physically acquired by buying or donations. That's it.
The Internet Archive made it clear they were only removing that one particular restriction, and only for a limited time, until libraries, universities, and bookstores could re-open.
But four of the largest publishing houses in the world smelled blood in the water and saw an opportunity to fundamentally destroy the ability of libraries to be libraries in the twenty-first century, so they filed a lawsuit.
Publishers never liked that libraries were leading the charge on ebooks, and very much dislike that doing what IA does, at a 1-1 lending/owning ratio, completely undermines their ability to charge...
Publishers never liked that libraries were leading the charge on ebooks, and very much dislike that doing what IA does, at a 1-1 lending/owning ratio, completely undermines their ability to charge absurd sums to libraries.
Yes. My hope is indie archivists also build and maintain their own archival collections of various groups of these works (and others) so that more heads are added to the hydra. They may take down...
Yes. My hope is indie archivists also build and maintain their own archival collections of various groups of these works (and others) so that more heads are added to the hydra. They may take down the central repository but it's only one head.
If a regular archival digital library cannot exist within the law (tbf yes they made missteps that put themselves in this position) then shadow libraries can pick up the slack.
soulseek ? does it have books too ? and isn't that really old. I agree with you about the rest of them and they do have torrents as well so everything shouldn't be lost in the worst case scenario
soulseek ? does it have books too ? and isn't that really old. I agree with you about the rest of them and they do have torrents as well so everything shouldn't be lost in the worst case scenario
Can’t say I’m surprised, but it’s still a tragedy.
The tragedy is that they ever tried to pursue this. It was extremely foolhardy, and many people told IA leadership this from the beginning. Now that decision may destroy the entire project.
The greatest tragedy of our time is that we invented a device that makes copying so trivially cheap and easy that it happens by accident and as part of normal operation, and one of the first questions asked by the bosses was:
"How do we keep it from doing that?"
We sacrificed infinite abundance at the alter of profitability.
You could say the same thing about crawling the Internet to gather data for AI, and yet people are mostly against that, too.
It's not just the bosses. Lots of people have sympathy for writers and artists. There are genuinely conflicting interests between creators and consumers, and they don't go away when technology changes.
My point is that we, collectively as a society, circa 1980something, had an opportunity to rethink how knowledge (and creative works, but that came somewhat later) should be funded and distributed. We could have embraced post-scarcity, insuring that creators don't starve while creations get to be distributed freely. We hit this point again in the 90s with the advent of Napster going mainstream. It's always been there, lurking in the background, but addressing how much harm we're placing on the future by introducing scarcity where there doesn't need to be makes existing interests uncomfortable.
As such, we doubled-down on business as usual, and buried our heads in the sand.
I'd have 0 qualms with the total ingestion of content for AI training if we had already embraced the end of scarcity on computers. But since we already decided that artificial scarcity needs to be a thing, it is decidedly unfair that trillion-dollar corporations get to ingest for free while the creators get nothing and users get to pay for access.
Okay, I see your point.
But maybe post-scarcity is a bit of an exaggeration? Just because we can copy data easily doesn't mean anything else is free, or about to become free. Often, it means something else becomes the scarce resource.
And I think our attitude towards paywalls shows some contradictions. Computers don't make reporting free and it's not really post-scarcity. It's just that we're often unwilling to pay for it. I'm not sure it's actually a matter of principle. We're dickering over the price, who pays, how to pay.
It does mean that, often, government-funded services can be distributed a whole lot cheaper and often that's a good thing. If you want government-funded news, there is the BBC, Voice of America, and other news services like that.
I also don't see charging users for access to AI chat as a nefarious plot when really is expensive. In practice, free often means "not actually free, but cheap enough to be funded by advertising." And that's a whole different can of worms. Maybe LLM's will become free to use when researchers come up with enough performance improvements, though.
It's because the people who are in charge of this are good at poisioning the well in regards to these things.
If a newspaper prints something that is wrong and it ends up hurting someone, they can be sued to oblivion. It's illegal to, for instance, print slander against people. There have been many news outlets that have closed because of these kinds of laws. It doesn't matter if it was one of their employees who wrote the offending text - they are responsible as the publisher. Fast forward to today, and there is slander all across the internet. But the publishers cannot be sued for it. Why? Because we made laws that expressly remove responsibility for what they publish. Why did we do that? Because most people did not understand the internet well enough to understand what they were doing and those who did had everything to gain, so they took advantage and "educated" the masses about what "needed to happen".
Large IP holders have done everything they could do for ages to get rid of consumer rights for essentially as long as copyright has existed. Every time a new digital physical format has come along, there have been built-in DRM schemes. Did you know that SD cards stand for Secure Digital, because the primary selling point to corporations was the encryption scheme? And now we have the DMCA, which makes circumventing DRM a federal crime. Bully.
Corporations have far too much power, and the more they have gotten away with the more powerful they have become.
Are they though? Or is it just a sentiment that gets over represented in certain social media circles? The closest I've seen to actual public opinion polling is a Pew survey that found that the overwhelming majority of respondents believe AI applications should credit the sources used in their models. Otherwise, I have not seen anything objective indicating widespread disapproval over the practice.
Letting people borrow books is what libraries do. Giving copies of books away for free is different. Somehow the Internet Archive thought they could make it legal, but that was never going to happen.
I don't understand. I've been borrowing books from the Internet Archive for years, and they are indeed borrowed - I don't get to keep them. And the time limits are much more strict than with physical libraries.
The difference is that the libraries have contracts and agreements with publishers. Whereas the IA just scanned some books and are now trying to “lend” them unilaterally.
Why should they need the publishers permission? If I buy a book, I'm allowed to lend it to someone. After they give it back I can lend it to someone else.
My understanding is they were lending out more digital copies at a time than they had purchased.
That was the case for a few months during the pandemic, but the court has now ruled against even one-to-one lending.
I think the issue here isn't 1:1 lending of the ones they have in their own possession. The problem is that they teamed up with a number of physical libraries and added their collections to the number they'd lend. That's the main part the court ruled against - saying that they couldn't count the third party ones.
I'm not totally sure, but to me this sounds like one-to-one lending is illegal:
There is also this in the footnotes of the conclusion:
Ah, seems I missed that part of it. I had tried to read the ruling, but am programmer, not lawyer ;)
Thank you for the correction.
That is true of physical books, but it's not true of digital copies, at least not in general. If you own a digital copy of a file, you generally own some sort of license or right to use that file, but it's not a physical good that you can own in the same way as a book.
This is why I buy digital books DRM free directly from authors where I can. Or from digital book stores, then grab a DRM free copy from elsewhere.
This bullshit about you don't own what you buy because they try and gaslight you that what you "purchased" was some crappy revocable licence needs putting in the bin.
In this case, this is lending scans of books, which are different from publisher-created ebooks.
Books have long prohibited reproduction by scanning, I don't see how that would change things.
That's not exactly true, as those prohibitions are bounded by fair use. Libraries have always been permitted to scan books for archival, it's seen as fair use. As well as scanning for the purposes of indexing and searching. The scanning itself is not illegal, its more of the distribution where the fair use gets a bit murky.
Google got a pretty hard win for the legality of book scanning and distribution in 2015, and that fight dragged for years, the service first launched in 2004....and back then you could access a whole lot more of said book. Incidentally, Google Books was one of those projects that I cited as "see Google isn't evil like the rest" back then.
Libraries were digitizing books since the 1980s, pioneers of the ebook, and the publishing industry has been trying to stop it for decades, but that pesky "fair use for public benefit" really puts a cramp in their style.
There's a reason publishers are honing in on Internet Archive's worldwide-emergency violation, and not the university libraries that do the exact same thing with the limit to the physical copy restriction.
If you bought yourself a book scanner, as a private person, it would not be illegal for you to scan every book you own and upload it to your ereader and leave your books on the shelf. Nor would it be illegal to loan your ereader to your friend. If you got rid of all the books, and/or never got that ereader back from your friend and made an additional copy, that would start treading away from fair use into violation territory. And the very existence of the lending library itself leans on that fair use construct.
This fight will almost certainly make it to the supreme court, and even with the heavy conservative bent, I would be surprised if they fully overturn the ideas behind digital surrogacy.
I agree, except that it seems like more of a loss for Google compared to what they wanted to do. But it seems pretty clear now that search is allowed, letting users read the whole book is not. Unless the publisher permits it.
Eh. I mean, if they decided tomorrow that all your physical books were also just licensed physical files and that you weren't allowed to do what you pleased with them it be equally stupid.
This has not always been true. Scanning physical copies and lending the scan was originally protected, and IA isn't the only one doing it.
It'd be different if they were taking the publisher's ebook, cracking the DRM, then handing it out. But that's not what was happening.
I haven't used it so perhaps I misunderstood something. I've read that during the pandemic, they let people borrow an unlimited number of copies, more than they actually had?
Normally a library has to buy each copy of a book that it lends out, and books sold mostly to libraries are often more expensive.
During the pandemic they allowed any given digital book to be borrowed an unlimited amount. Individual users could still only borrow a limited number of books, but each book no longer had a cap on how many users could simultaneously borrow it.
Otherwise, their controlled digital lending program remained unchanged:
Literally the only thing the Internet Archive did differently with the National Emergency Library program was lend out more digital copies of books than they had physically acquired by buying or donations. That's it.
The Internet Archive made it clear they were only removing that one particular restriction, and only for a limited time, until libraries, universities, and bookstores could re-open.
But four of the largest publishing houses in the world smelled blood in the water and saw an opportunity to fundamentally destroy the ability of libraries to be libraries in the twenty-first century, so they filed a lawsuit.
Publishers never liked that libraries were leading the charge on ebooks, and very much dislike that doing what IA does, at a 1-1 lending/owning ratio, completely undermines their ability to charge absurd sums to libraries.
There's Anna's archive, libgen, z-library, soulseek 🤷🏾♂️
Yes. My hope is indie archivists also build and maintain their own archival collections of various groups of these works (and others) so that more heads are added to the hydra. They may take down the central repository but it's only one head.
If a regular archival digital library cannot exist within the law (tbf yes they made missteps that put themselves in this position) then shadow libraries can pick up the slack.
soulseek ? does it have books too ? and isn't that really old. I agree with you about the rest of them and they do have torrents as well so everything shouldn't be lost in the worst case scenario
Books aren't really the main focus, but it is a sharing platform so you can put anything you want there. I can find books easily.
Ugh. Damn gubberment. I wonder if IA will consider moving to a different country with more lax (no?) copyright laws.
Preferably a nuclear-armed nation, so they can’t be strong armed by gunboat diplomacy.