And the vast majority of people read news for it's breaking content, not for its archived content from years before (and I say this as someone who has often recommended the latter, but has gotten very few people to do so). So giving people that free breaking content (either in its entirety like on Hacker News, or summaries like you see all over social media) is actually a direct competition to the news business in a way that training an LLM on an article from months/years back isn't.
As you noted it is not the norm to post pirate links here for IP other than news articles, but that doesn't mean that a lot of people think it is not OK to pirate those other forms of IP.
In nearly any big discussion that even remotely involves video streaming there will be numerous posts from people explaining why they pirate (usually with ridiculous justifications like "subscribing is not an option because even though this paid service does exactly what I want now at a price that is trivial for me they might someday later change").
The impression I've gotten is that piracy of nearly everything is widely felt to be OK here. Information wants to be free, yada yada.
About the only piracy that is consistently frowned upon here is piracy of open source software. When some company sells an embedded device that uses GPL code without releasing the corresponding source that's viewed as just a little short of a crime against humanity.
Briefly, something like:
1) Ycombinator could not tolerate HN becoming a site known for sharing IP-law-violating content. And the people who come here by and large are smart and socialized enough to implicitly understand why.
2) At the same time, a large number of folks here mostly wink and nod at that sort of consumer infringement. And there's a society-wide bias towards "things like news are less protected", so that gets to slide.
3) But people also have a need to tell consistent-seeming stories about how things work, thus the mental gymnastics.
It ends up being similar to trying to explain why people pretend to be prudish innocents about sex. It largely reduces to "a small subset of the population goes sufficiently ballistic about what I consider to be relatively trivial stuff as to make it not worth fighting over, even if I find that to be ridiculous."
There are a lot of different versions of this that become so normalized it can be hard to notice.
Like what you said...
> Information wants to be free
I’ve read and participated in many such threads and I’ve literally never seen this take. Often what I see is complaints about having to learn different UI for different services/apps, no offline, ads injected into paid services, having to figure out which service a show is on, and generally terrible UI you can’t change/fix.
I don’t think I’ve ever really seen someone use the argument “yes it’s great today but they might charge more later”. Not saying people haven’t said that but it’s far from the main thing people say in my experience.
[0] To be clear, I know of few who actually like copyright. Tolerate it? Use it as needed? Sure. The only people who actually defend the current broken-ass system are large media companies which are built to optimally exploit it.
For UFC, your complaint is you don't like their pricing. The whole point of copyright is to give someone the monopoly to control pricing so they can use that pricing power to incentivize them to create the product in the first place. Similarly to patents. Thus, complain about the format things are delivered in all you want (like the client) but pricing is inherent to copyright or patents for good reason. You are now just arguing that you as a consumer should be able to pirate if you don't agree with pricing. And that's ludicrous.
In that case, just read a news article about the event. Copyright doesn't cover facts, only creative expression. So a news article covering the facts of the UFC fight is able to be published without the consent of the copyright holder. Think of the digital video of the fight almost like buying a ticket to the fight. You're saying you should just be able to sneak into the fight and watch it for free without any justification for you're doing so.
Finally, you can also watch other people's videos of the fight that THEY recorded on social media as other sources of the fight information. But if you want the recording with all the right angles, coverage, etc, it clearly has value to you over written recaps or social media coverage. And you are just arguing over price, which they are the copyright holder have the right to set the price.
Then don't consume it and don't buy it. If you stop paying the abusive publisher, they'll be forced to change their policies.
The fact that you don't want to fund what is admittedly a rather abusive industry does not magically make it right to consume other peoples' work for free. That's theft-adjacent. You're not entitled to any piece of entertainment without paying for it.
Spotify hits this sweet spot where one subscription delivers almost all the music you'd want to listen to. Steam hits this for games where a couple clicks can play and launch almost any game with minimal hassle. Netflix mostly used to hit this, but most of the current streaming stuff feels overpriced if you want to get all content (unbundled cable bundle). News kind of feels similar to streaming where its unbundled, and there's a lot of interesting content out there, but there's no way I'm subscribing to 15 different newspapers, especially random local ones for cities I don't live in. If there was a news bundle subscription for a reasonable price I think I would pay for it.
People are understandably angsty about someone stealing credit. A NYT article is going to be a NYT article, not laundered around and presented as someone else's work.
Plus, there's the angle of enshitification and ads being injected into a paid service, and so on.
The problem is really that their business model sucks. They are working with fewer and fewer advertisers and much more competition and expecting business like they had before. And so we have a business that is attempting to fix itself with paywalls which don't work 100% of the time, but good enough to get the found newspaper analogy.
The internet simply exacerbated this as anyone could publish news on an equal platform to the big boys. Then we get paid-per-click, and that drives click-bait.
Stealing information absolutely is not responsible for that. People pay for junk, and that's the reason. We don't eat junk food because it's given away.
I'm not saying you've never seen anyone make an argument roughly like that, but I will certainly say that it is not at all representative of the argument that I see made. Complaints usually have to do with current behavior of the platform or the wider streaming ecosystem.
Gonna gamble and call bullshit on this.
My speculation: the most popular reason HN'ers give for pirating: they literally cannot get the content otherwise.
2nd most popular: it is such a pain to either to purchase the content or get it to run on bog standard software (like Firefox/Linux/etc.) that otherwise paying fans are driven to whatever the current equivalent is for bittorrent.
In fact, I don't believe I've ever seen a justification for using bittorrent or whatever due to what someone's favorite streaming service might do in the future. I'm assuming you saw at least one based on what you wrote-- care to give a link?
If this is true, it should be easy for you to link to an example. Could you do so?
If the story was linking directly to the "book, TV show, movie, video game, album, comic book, etc", and the link only worked for some people while others randomly got a login request or similar, you'd also see the top comment being a link to an archived version which avoids the login screen. That is: the main difference is that the archive link has the exact same content as the link submitted in the story, only bypassing the login screen that some people see. And the only reason the archive site has the content is that it didn't get the login screen; if everyone always got the login screen, what you would see on the archive site would be the same login screen.
> Are paywalls ok?
> It's ok to post stories from sites with paywalls that have workarounds.
> the archive link has the exact same content as the link submitted
No, articles are updated as new information comes in, retractions are made, etc. Especially breaking news (the type that would reach the top of HN). The archived versions are outdated.
> others randomly got a login request
It's not random, you get a number of free articles before the paywall appears ("soft" paywall).
The paywall is removed entirely for some topics/stories, especially matters of public health (common during the pandemic).
> the only reason the archive site has the content is that it didn't get the login screen
No, it's because they don't block archive crawlers, and prefer people bypassing the paywall and reading news at NYT. Hopefully users find the content valuable, and some of them subscribe as a result.
(opinions are my own)
There were some tweets the other day about how Midjourney could be prompted almost-exactly reproduce some frames of the film Dune. It wouldn't be shocking if these companies were using large databases of movies, with questionable legal status.
[1] https://en.m.wikipedia.org/wiki/International_News_Service_v...
[1]: https://guides.library.cornell.edu/evaluate_news/source_bias
[2]: https://www.newsmediaalliance.org/rise-of-opinion-section/ Interestingly there's a banner at the top of that link touting an agreement between Axel Springer and OpenAI.
EDIT: formatting
This is rather inaccurate. A fact is Hitler invades Poland. You're right, nobody can copyright this idea, as it is just a fact.
However, if I then write a 500-word article describing the scene of Hitler invading Poland, have short quotes from some civilians there, etc. that particular arrangement of ideas and words is copyright.
AP can't go and sue INS for just reporting the fact Hitler invades Poland, but if INS takes a whole article word for word and reproduces it that's still violation of copyright. The actual printed words of the news always had copyright.
The WSJ can't claim copyright on the markets going up yesterday. They can claim copyright on something like "After the bell rang in the NYSE, the tech industry ticked up 1.2% over last week. Meanwhile the whatever market took a hit of -0.5% ending the quarter slightly lower than our analysis expected. Blah blah blah..." If Investor's Business Daily wrote a different article that also talked about the markets ending up at the end of the day, that's not a violation of copyright. If they literally write "After the bell rang in the NYSE, the tech industry ticked up..." then they're violating WSJ's copyright. This was true before and after International News Service v Associated Press.
> INS members would rewrite the news and publish it as their own without attribution to AP.
So the case hinged on INS indeed reporting facts that differed in exposition.
I fully assume that if I was to post a magnet link to a torrent for whatever the link was about, I would be banned.
Morally speaking, I think it's perfectly reasonable to download a copy of something and either read the relevant info for my current task or to sample it to decide if I want to buy it. I see it no different to using the library or browsing at a book store.
Perhaps once news organisations can work out how to effectively wield the DMCA hammer against archive links we'll see the practice of posting them stop.
The problem with the thinking in the root comment is that it implicitly assumes that people’s behavior is morally consistent, or that they even try particularly hard to behave in a morally consistent way. That’s not really how people work. If you ask them to discuss morality in the abstract, they’ll try to come up with a consistent system. But their actual behavior is mostly dictated by social norms. And if you try to pin them down on the morality of their concrete actions, they’re more likely to stretch their moral system to accommodate their actions than the other way around.
None of this is to say anything about my own opinions on news sharing or OpenAI’s situation. It’s just that someone decrying piracy but also posting/sharing/upvoting links to copies of news articles is neither surprising, nor indicative of some deeper nuance to how people view morality around IP.
It’s amazing the amount of books that copyright laws prevent us from finding
https://www.theatlantic.com/technology/archive/2012/03/the-m...
Definitely a grey area when that content is then used to train models though.
And everything is a grey area, determining the line is the existential purpose of these court cases.
We've been here before with hyperlinking, then indexing and then linking with previews and the Canadian Facebook stuff but I think this has more standing.
is it?
1) that you paid for news
2) that it included ads
both are just the price you want to pay. There are various state news outlets that you're probably already paying for - npr, pbs, bbc, cncb depending on your region
There are browser extensions that block ads. They are called ad blockers.
Are all the journalist layoffs a fever dream?
This is one of the more profitable ones, and only because they employ unscrupulous tactics:
https://www.macrotrends.net/stocks/charts/NWS/news/profit-ma...
This is NYT, the most successful news business:
https://www.macrotrends.net/stocks/charts/NYT/new-york-times...
As for movies/tv show/music makers, let’s just say most people in the software engineering business would look at their numbers and count their lucky stars that they are not in the movie/tv show/music business.
(It is also true that excessive copyright lengths have removed access to content that the public should have a right to).
You got my point backwards: AI companies will make it from the pirated content, that individual users don't make.
If only piracy would actually harm these businesses but alas as often demonstrated it has zero effect on their bottom line, if anything it increases their profits.
If you are advocating for a free for all libertarian dystopia, well, I have some bad news for you - they never work.
Not being able to un-see a movie and get your time and money back is one side of the coin. The other side is that information can be copied.
Both sides suck for one of the parties. There's no reason why one of them gets it their way, especially if it requires a contrived legal framework while the other way would require nothing at all.
And as long as you had the opportunity to experience the content, you’ve gotten what you paid for.
I don’t see “I don’t like it” as a valid reason for a refund.
Not sure about others, but I'm not.
It doesn't matter what you think you're paying for or should be paying for, the fact of the matter is that you're paying for the effort people put in bringing that to you. So you are, whether you want to be or not.
If I read an article in the NYT then I'm paying for what I took away from it, not for the amount of time that it allowed me to kill.
Incorrect. Many intellectual property has a certain merit that can be demonstrated before it is consumed. E.g. "This piece of software allows you to create 3d models". On the other hand, an article with headline "Will new batteries allow 10x more energy storage?" does not tell me anything.
Because those who own & produce such news articles asked to make them different. People listened and accepted their requests.
When you make a TV show or a video game, you don't get any protection from the Geneva Conventions and a long list of other international treaties for your rights on stuff other than the content you are producing. The same can't be said when you are producing news.
"Piracy is almost always a service problem and not a pricing problem."
edit: It didn't even occur to me to compare the time-cost of "just pay for the article", but: last I read, it's half an hour of work to cancel a New York Times subscription [0]. So, that option's not even on the table.
[0] https://news.ycombinator.com/item?id=26174269 ("Before buying a NYT subscription, here's what it'll take to cancel it", 812 comments)
I canceled mine two weeks ago. It was four clicks. One annoyed me because they tried to get me to stay with an offer, but I didn't drop them because of the price.
Probably because most print media is garbage and nobody in their right mind would actually pay to read them
NYTs revenue keeps growing though.
(It's the same reason for me. I have tried news site subs but eventually got so tired of the polemic that I cancelled. I won't sub again).
Norway has substantial public media funding across the political spectrum, but as you point out it always comes with conditions, even is less so than the funding for the state owned broadcaster.
Combining the two models and putting public funds into several perpetual trusts intended to provide funding from their profits at arms length from any sitting government similar to the (private) trust funding The Guardian might be an interesting alternative.
(EDIT: Norway also has its own variation over The Guardian model - the second largest media group was founded by unions but is now majority owned by the combination of two public benefit trusts)
Who pays?
I see the gp post about pirating news as a very good point, while having no veleity to pay the New York Times, and being ok with not reading it in general.
But I also pay for my national (public) news outlet, and their articles are available to anyone anywhere in the world. I don't know how it should work, but I wish we could get to a system where the burden to keep news outlet alive is split thinly enough to have open but viable publications around the world.
Basically the same way weather stations collaborate all other the world and we pay for our local stations while getting acccess to all the forecast everywhere.
For example in Hungary there is an official news agency ran by the government, with (cumbersome) free access for everybody. Of course this does provide somewhat biased presentation of some facts, but on many topics it provides unbiased access to news for any citizen.
This is actually pretty common in Europe, often funded by mandatory fees (for some reason not branded as taxes) certain appliance owners need to pay (UK TV license, German Rundfunkbeitrag). For this fee people get access to news and cultural programmes for free via different media (radio, TV, internet).
The level of control governments exert on public broadcasting networks is widely different. Since Meloni, the RAI in Italy is facing similar issues, but Hungary is still the canonic example of government misinformation and propaganda.
People have free access to public roads all around the world, and the quality wildly differs in that as well. Also the quality of for-profit news services does differ wildly, you might have an opinion about that of fox news, for example, but that is also off topic in this discussion.
On the contrary, the quality of the news is very important to the discussion. There is no point in making trash freely available to the public, after all.
think about this: I will get mostly objective and useful reports of the flood approaching my home near the river regardless the narrative/interpretation they might have on some other topics, or the biased reporting on the merits of the government in handling the situation at the dams.
For me I'm not here to debate on the political policies of some governments, just gave a few examples of ways to fund public access to news. This discussion is over from my part.
Public donors ALA Patreon
People doing it in their free time because they care a lot about the subject (nowadays with things like Twitter its quite possible for an independent obsessive to write a good piece on, for instance, the Ukraine War by mostly referring to open sources and public announcements by governments and corporations)
Government sponsorship ala BBC
Instead of paying news outlets to provide ourselves with filtered feeds of content that match our own biases, we could instead pay news outlets to produce competing streams of explicit propaganda to be freely disseminated. The overall bias and quality of the news would be largely unchanged, even if the biases were more obvious; in fact, it may even improve.
Who should pay the journalists or the investigative reporters?
People post archive links even to fake NY Times.
How would you pay for news otherwise?
You could subsidise news via "public service" style stipends. Much like having a government owned "independent" news service (eg the BBC) this comes with a high risk of corruption. Don't bite the hand that feeds and all that.
You could implement a much lower friction non-recurring payment system. I'd be far more tempted to drop a little money on a fixed term (5 articles, 1 day, ???) setup than a subscription.
Realistically, I am not paying for more than 1 long running sub. And there are > that number of solid outlets.
This is somewhat what Apple News+ works like, but I doubt most news orgs want to be held captive by Apple.
I believe that’s often referred to as a newspaper, which should be available in all good newsagents on any given day.
(But yes, that is the model)
I understand that copyrights and patents are vehicles for ensuring a creator gets paid for their work, but they are flawed in not rewarding multiple parallel creations and that they last too long.
Much in the same way as when you read a book, your brain doesn't become a pirated copy of the text as you only store a hugely compressed version of it afterwards, a feeling for the plot, generated images and so on.
I don't want to read theguardian.com, or nytimes.com, or washingtonpost.com, or bloomberg.com, I want to read news.ycombinator.com. Paying an individual subscription to every possible underlying site that could be linked to from news.ycombinator.com is a non-starter.
I’m not going to switch to a new website where no community exists just so I can pay for news articles. To work it needs to be integrated into an existing, successful aggregation website.
Hypocrites are EVERYWHERE and are the majority.
Now a comment points out that HN News (and most of the internet) routinely does something much worse - allows people to bypass completely new articles in their entirety without paying - and almost all the comments are about how it's the New York Times fault for making it difficult to cancel subscription, the importance of news being available to everyone, the problems with copyright laws, etc.
Once the NYT pays reparations for the Iraq war, I'll be the first to stop pirating it.
I'd say the only real reason the Piratebay links thing you mentioned is not the norm is purely because those media sources have done a better job of striking fear into people doing that, so it's gone more underground. I.e. they're better terrorists.
There's no fundamental, moral reason why Piratebay links being posted and raised to the top would be wrong.
Can it do so more than a human can?
I think that's the key here. If an AI is no more precise than a human telling you about the news article they read today then ChatGPT learning process probably can't be morally called copying.
Feeding someone else data into your system is usually a violation of copyright. Even if you have a very "smart" system, trying to transform and obfuscate the original data.
In some circumstances, yes, but often it's not, especially if you're not continuing to store and use it (which OpenAI isn't).
So I'm a living breathing copyright violator. As a person I should be banned.
Fortunately, copyright is a bullshit fictitious right with no basis in natural law. So I don't lose much sleep over it.
The court could ask to show the training dataset.
And if OpenAI were selling the reproductions, that would be infringement. But that's not what's happening here. It's selling access to a system that can do countless things.
When you tell people about some news article you read earlier you repeat it exactly verbatim? You also give this out to potentially millions or hundreds of millions of people for commercial purposes?
Furthermore, there's Google research on extracting training set data from models. More specifically, Google found out that if you ask GPT to repeat the same word over and over again, forever, it eventually starts printing fully memorized training set data[0]. So it is memorizing stuff, even if it's not regurgitating it.
[0] When told of this, OpenAI's response was to block conversations with large amounts of repeated words in them.
I do not pay for any news websites because I read very little of what they produce, and it tends to pop up more on aggregator sites like HN than me actually going to them.
I actually did have a subscription to The Telegraph for a few months at one point because initially I wanted to read a full article (without cheating). But eventually I cancelled because so much of it is polemic trash.
That's my justification: I pay for things that have value to me.
It's funny because I use PayPal for any unknown-to-me site where I don't want to give out my card, but the only site where I've needed their help to cancel something was the New York Times.
Assuming you just said five minutes figuratively... Do you live in California or some other legal jurisdiction that forces them to play nice? Did you subscribe through some other company, like Apple?
Horror stories about unsubscribing from the NYTimes are easy to find in the archive if you search for it. They make you call and chat to a retention specialist on the phone. This should help you have an idea of what he's talking about: https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...
On ads it's acceptable to distribute them freely and it is advantageous to the company. Can we also see good journalism as an ad for the quality of a broader product?
If the news/magazine doesn't want this they can simple serve a cut down or zero length article to all non-paying viewers! But they want that SEO, and they want that marketing.
https://garymarcus.substack.com/p/an-artist-fights-back-and-...
With text articles behind paywalls the relevant information is hidden and only hinted at as a teaser.
https://news.ycombinator.com/from?site=amazon.com
None of these have the Pirate Bay or Library Genesis or Anna's Archive or the equivalent as the top comment.
Compare that to...
https://news.ycombinator.com/from?site=nytimes.com
And almost all of these have an archived version as the top comment.
If I could just buy one article for a coffee without entering a bunch of PII or go through a time-wasting process I would agree on the moral equivalence between the examples.
This makes a news story copyright murky in the eyes of wider society unlike a clearly 100% creative work like a TV Show or Movie.
Further the news themselves self cannibalize, how many stories are just rewrites of stories from other outlets? why it is OK for the Washington Post to copy the NY times, but not ok for OpenAI or Archive.org?
Copyright is a complex subject, and not as vast as many believe, at the same time ironically it is more vast than i believe it should be. copyright should be much more limiting than it is. Which is at odds with people that believe copyright should be maximized.
Keeping in mind commercial success of a work, author or company is not why copyright exists. For the US, the only reason copyright can exist in our framework of law (i.e the constitution) is for the promotion of the useful sciences. No other purpose for copyright would be constitutional under the US Constitution
That is a General Article about Copyright world wide, I Specifically stated US Copyright, which is Authorized by Article I, Section 8, Clause 8 of the United States Constitution[1], implicitly for the promotion of the useful sciences. That is where congress derives its power to pass copyright laws, and to enforce copyright on the people of the United States. No other purpose is authorized by the US Constitution
[1] https://www.law.cornell.edu/wex/intellectual_property_clause
It is not just for sciences.
> Keeping in mind commercial success of a work, author or company is not why copyright exists.
Lets take a look at the clause again:
> To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries.
Lets go ahead and skip over the fact you're consistently ignoring "useful arts" part as well and keep going.
What exclusive rights do you think they're talking about here? Do you really think they didn't mean the economic rights related to their writings and discoveries? How do you imagine this would "promote" the sciences if not by allowing the creators to share their works and ideas while still retaining economic benefits of their labor?
Reading between the lines, the whole point of IP is to help protect the potential commercial success of sharing your ideas. It doesn't guarantee the idea will actually be a commercial success, but it does give them the exclusive right to the commercial success for a limited time.
> [the United States Congress shall have power] To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.
...
> Some terms in the clause are used in archaic meanings, potentially confusing modern readers. For example, "useful Arts" does not refer to artistic endeavors, but rather to the work of artisans, people skilled in a manufacturing craft; "Sciences" refers not only to fields of modern scientific inquiry but rather to all knowledge.
"Science" refers to knowledge, and conveying that knowledge entails creative expression. Copyright covers expression of knowledge; facts and ideas themselves are not copyrightable. "useful Arts" refers to inventions. Patents cover useful inventions and novel implementations of practical ideas, not creative expression and not unimplemented ideas. (Which is one reason most software patents shouldn't have been granted.) Congress's authority to make copyright law and patent law is conditional on promoting the spread and advancement of knowledge, creativity, and inventions in the long term. The means of achieving that end is short-term restrictions on how people can use others' creative works and useful inventions.
But copyright does not prohibit mere usage of someone else's creative works [2]:
> To win a claim of copyright infringement in civil or criminal court, a plaintiff must show he or she owns a valid copyright, the defendant actually copied the work, and the level of copying amounts to misappropriation.
If the output of an AI model is not similar to any creative work in the training set, then the output cannot infringe on copyright. And even where the training set contains illegally obtained materials, the act of illegally obtaining those materials has nothing to do with including legally obtained materials in the training set.
[1] https://en.wikipedia.org/wiki/Copyright_Clause
[2] https://en.wikipedia.org/wiki/Substantial_similarity#Substan...
If the Washington Post printed an article from the NY Times nearly verbatim and without attribution, it would not be OK and surely they would take legal action.
This reasoning would presumably apply to any neural network, including one made of neurons, dendrites, and axons. So any human reader of the NYT who is capable of accurately summarizing what they read is an evil copyright violator, and must be "deleted".
Effectively, the NYT legal department is setting the stage for mass murder.
Here’s a service for the UK providing paid access to copyrighted materials to schools: https://www.nlamediaaccess.com/newspapers-for-schools/
This implicit permission for the archive links to exist, gives some of us the implicit permission to pirate the content.
Disclaimer: I am a happy subscriber to the NYT (and other digital newspapers).
My uncle used to distribute daily newspapers and his saying was "News ages like a fish".
OpenAI is allegedly using NYTimes articles to train a computer and sell its services. I see different use scenarios.
I guess another way to look at it is that human just reads the pirated material. A computer makes a verbatim copy and analyzes it to the point to mimicry and sells fuzzy versions.
Isn't this what the Associated Press is intended for, a stream of news trying to report just the facts and happenings of the day? That's quite a bit different than a NYT article intending to inform but also convince someone of a position of some sort.
Feeding an AI opinionated news compared to "just the facts, ma'am" seems risky from a bias perspective.
The articles themselves are indisputably not a part of the model, because it doesn't store text at all. OpenAI's position is correct; people just underestimated how well the AI learns from reading, especially when it reads the same text in a bunch of different places because it's being quoted/excerpted.
but somehow storing the words and their links is not storing the actual text? What is text but words and their links?
If I had a database of a billion words, and I had a list of pointers to words in a particular order, and following that list of pointers reproduces a copyright text exactly, isn't the list of pointers + the database of words just an obfuscated recreation of that copyrighted work?
Of course, if you read NYT's argument, they're also mad when it's incorrect about the text, or when it hallucinates articles that don't exist. Essentially they're mad that this technology exists at all.
I mean this is still a link, no?
Like, sure, it is a probability. But if each of those probabilities is like 99.9999% likely to get you to a chain of outputs that verbatim reproduces the copyrighted text given the right prompt, isn't that still the same thing?
And yeah, it hallucinating that the NYT published an article stating something it didn't say is concerning as well. If the model started telling everyone Matticus_Rex is a criminal and committed all these crimes and started listing off hallucinated court cases and news articles proving such things that would be quite damaging to your reputation, wouldn't it? The model hallucinating the NYT publishing an article talking about how the moon landing was fake or something would be damaging to its reputation right?
It is stored in a somewhat hard to understand way, encoded in weights in a network but it must be stored otherwise it would not be possible to reproduce it.
You can ask "please provide me with the first paragraph of the carl zimmer article on the oldest DNA" and it produces it, verbatim. This is not possible unless the model contains, encoded within it, the NYT's copyrighted text.
First, Open AI is the one doing the pirating here. Hacker News is the host, they aren't doing any pirating or posting any archival links to the copyrighted information themselves.
Second, Open AI charges subscription fees and profits off of the copyrighted material they have pirated, whereas Hackers News does not, nor do the people who post the links.
NYT, or someone's blog? Meh, fair use, and if you say no, you're in the way of progress.
But if you wanted to scrape ChatGPT answers to tweak your network, uh oh, violation of T&C!
For LLMs you're essentially teaching them language by showing them lots of examples of written language - newspapers are of course a great example of written language.
The goal of OpenAI is not to reproduce newspaper articles verbatim when asked questions (even if the answer could be a newspaper article) and the fact that it can happen is a side effect of how LLMs work.
When a HN participant shares a (pay walled) link to a NYT article, I do want to read the exact article linked verbatim because while the facts of the article may be reproduced elsewhere in a form that's free, specific word choices or whatever might be a focal point of the discussion on HN, and therefore I can't realistically participate in a discussion without having read the article being discussed.
And as an aside, I have no problem with paying to read news, or whatever media, however it's impractical for me to subscribe to every news source HN participants link to, and therefore I gravitate to archiving services instead. I do wish there was a better solution - for example Blendle with more sources.
This is an excellent point. A properly functioning LLM should not return the original content it was trained on. When they return original content, I believe the prompt is tightly constrained and designed to extract or re-create original content. Another reason that occurred to me recently is that maybe the training set is too small, and more general prompts will re-create source material.
Another question would be, are LLMs regurgitating what they were trained on, or are they synthesizing something very close to the original content? (Infinite Monkeys, Shakespeare). Court cases like this increase the need for understanding the "thinking processes" in an LLM.
Seems like a nice split-the-baby resolution would be to send the NYT Corp a single article read amount anytime GPT plagiarizes more than what’s allowed at an academic institution.
Or is it "you can't talk to someone about an article they read".
This is really saying you can't call up your buddy and have them tell you a summary of what they just read. Maybe my buddy has a good memory and some of the text is actually nearly duplicate. But I wouldn't know because I didn't read the original, I just asked for a summary from someone else that read it.
A lot of of that is going to stem from the fact that respect for "journalism" is pretty low. More than 99% of news articles are copies of the <1% of original work that happens in that field. In news, everyone is already lifting content from everyone else.
(Books, to me, are separate still, in that I like to have a physical copy (and generally see the authors as humans who deserve compensation, rather than mega-orgs that deserve eternal torment), so I'll frequently use the digital copy as a kind of preview, then purchase it once I see it's a good book I want to read.)
I've only been reflecting on this difference for a few minutes, but, to me, I think the major difference boils down to:
1. Netflix series (movies, albums, etc) are non-essential, fictional works that take a long time to produce - think: fancy chocolates and caviar.
2. News, generally, contains timely, important information - more meat and potatoes.
3. While much of the super-critical news is not paywalled (e.g., product recalls, election dates, COVID stats, etc), a lot of information that is advantageous to know (discussions on interest rates, details on legislation, etc) is paywalled, compounding information asymmetries.
You're not depriving anyone of anything. Unauthorized copying is not theft. There's no equivalency. You can't copy and paste a cake. If you take a cake from a bakery, you're depriving the bakery of a thing. If you take a picture of the trademarked bakery's sign, copy its the copyrighted text from its website, and print them out, you haven't stolen anything. Nobody has lost anything. Nothing was damaged. No person, place, or thing was harmed.
Current copyright law is offensively absurd. Patenting of software, effectively eternal content copyrights, ridiculously broken DMCA, music publishers taking 99 cents of every artist's dollar, and so on and so forth.
If you support the dissolution of archaic institutions and broken laws favoring those with entrenched wealth over individual rights, you support piracy.
There is a legitimate case for laws respecting and protecting intellectual property rights. Such laws do not currently exist. These laws do not deserve to be followed or respected, and should be broken as a matter of course. Civil disobedience is called for. Refuse to participate in an exploitative market immovably entrenched in governments all over the world. Pay artists directly and commensurately if you feel they've brought value to your life. Copy whatever you want. Share those copies with whomever you want. Nobody gets hurt. Only conglomerates of already wealthy individuals and corporations are "deprived" of the potential transaction with you that they feel they are entitled to, as a matter of course.
The NYT is just as complicit as any other legacy media institution in the enshittification of journalism and laying waste to the potential value of their content. The "Gray Lady" is not a person, or a valuable institution. It's a soulless corporate construct not deserving of our empathy or high regard simply because of the reputation of human individuals who previously produced quality content. Stop pretending these institutions serve some higher purpose than to fatten the wallets of shareholders.
The good journalists have left. The ones left behind are naive, or are desperately clinging to an illusion of legacy and institutional legitimacy that no longer exists.
All that is left for these media dinosaurs is to leech off the success of others, to use their reserves of wealth and influence to arbitrarily insert themselves into the market, with no regard to the fact that they no longer have value or prestige or purpose in the context of modern technology and communication.
Anyway. Copying isn't theft. Don't give them the linguistic territory. Call a spade a spade, and media companies the desperate corporate leeches that they are.
Who thinks this? I don't. I think copyright is wrong across the board. I would love if the same pattern of posting archive'd articles held for books, movies, et cetera.
I would love to change my mind on this, as it is a very unpopular opinion to have. But I have _never_ seen a morally or scientifically sound argument in favor of copyright law, and I've spent decades looking.
I think it subsidizes the creation of junk food content (superhero movies and clickbait news for example) while not contributing anything to the progress of science (paywalled scientific journals and textbooks). I shudder how much time I have wasted in my life consuming crap attention grabbing media and advertisements. I like to think if we lived in a world where everyone could be a publisher if they wanted to, the quality filters would be better, and information reaching us all would be more likely to be in our best interests.
You speak of "the author". But the current system does not benefit "the author". 1% of authors profit off copyright. 99% lose money on copyright (they pay more for copyrighted media than they earn from it).
Your question should be "How does that benefit monopolist authors"?
I agree, my idea would not benefit monopolist authors. They would lose the bulk of their revenue stream.
But it would benefit the average author whose cost of living would fall and information would start serving them more than serving business.
I am not downplaying the talent and hard work of successful monopolist authors. But I do not think the works they create are worth everyone giving up their rights to reshare and remix information. I believe the world would look very different post-IP. You'd probably have a new profession--small independent librarians (similar to data hoarders today)--who would help their local communities maximize the value they got from humanity's best information.
Maybe I'm wrong! Maybe the information ecosystem is better controlled and the genetic differences of monopolist authors are so stark that without the subsidies to this gifted class we'd all be worse off. But that's an argument based on outcomes and not principles.
> without their permission
The oxygen I'm breathing right now mostly was created by trees on land owned by others. But I don't ask for their permission to breath. Some things are just not natural.
I am not saying plagiarize. It is always the right thing to do to link back and/or credit the source. But needing to ask permission to republish something seems to go against natural laws.
However, this doesn't apply to organizations that freely share copyrighted information while making money in the process, or to organizations that share copyrighted information in a way that specifically disadvantages or does harm to the original creator of that information.
In 1990 it would have been considered normal and appropriate to clip an article out of a newspaper and post it on a communal corkboard. What are the key differences between that form of IP and others, and that analogy and the present situation of HN allowing archive links?
As for ease of distribution, that might address OP's original question: It's easy to make and click an archive link, but it's a lot more effort to make or find a Pirate Bay link to another form of media, and for someone else to download and view it.
https://news.ycombinator.com/item?id=23735026
Even talking about it will get you scolded for talking about something "off topic"
On the other hand having an archive link to a times article in order to discus it is not really a substitute for a times subscription as a news paper has to walk a line of letting some of it's articles be read while requiring payment for others (the times actually allows you to create a "gift link" to do exactly what the archive links do).
The current arms race got us scrapers, and then paywalls, and then ad-blocking archivers ...
But in reality, I might drop a penny to read a NYT article. Maybe a nickel. There's no reasonable way of performing microtransactions right now. Everything is still in hefty increments, so nobody can work out what the market would bear.
And this seems to be tolerated as the norm. And yet, whenever there is a submission about a book, a TV show, a movie, a video game, an album, a comic book, or any other form of IP, it is in fact very much _not_ the norm for the top-rated comment to be a Pirate Bay link.
I think that's something worth reflecting on, about why we feel it's OK to pirate news articles, but not other IP.
And the reason I bring this up, is that it seems like Open AI has the same attitude: scraping news articles is OK, or at worst a gray area, but what if they were also scraping, for example, Netflix content to use as part of their training set?