[rgmerk]

All copyright is way too long but authors go completely off-piste at anyone who suggests that it should be something more reasonable (I kinda like 21 years myself).

Part of the issue is that the loudest voices are that 0.01% of authors whose work still has some commercial value decades after its creation.

[nordsieck]

> All copyright is way too long but authors go completely off-piste at anyone who suggests that it should be something more reasonable (I kinda like 21 years myself).

> Part of the issue is that the loudest voices are that 0.01% of authors whose work still has some commercial value decades after its creation.

My favorite scheme is that all copyright last for 10 years by default. You can register it for $100 for another 10 years. And every 10 years after that, the cost goes up by 100x. That way commercial works that are very economically valuable can be protected for a pretty long time, but everything makes it into the public domain relatively quickly.

Maybe the numbers need to be tweaked a bit, but I think the idea is fundamentally sound.

[nostromo]

This scheme seems to protect the largest companies with the largest bank accounts the most. Why would we want a system like that?

I actually have the opposite view: I'm more worried about a small-time author that say, makes living on a low-volume text or training book, than I am protecting Mickey Mouse.

[mlyle]

It's nice to make works that no longer have commercial value enter the public domain by default.

Another justification: if the harms increase for giving an author exclusivity for a longer time, society should demand a larger payback. This may only be justified for the most prominent works

It feels a little out of calibration. How about free for 20 years; $500 for 10 more, and then 100x for every 10 additional years. Everyone gets 30 years to exploit a work for a reasonable price, and then the price ramps steeply so that very few works are registered beyond 40 years.

A tiny proportion of works will be worth the $50k step, let alone the $5M one. Even Disney will not pay $5M for most things.

And, of course, things that are forgotten or devoid of commercial value will lapse at 20.

[shiroiuma]

I think only 5 years should be free, then maybe $100 for another 5, then exponentially increasing after that.

A lot of stuff simply has no commercial value after 5 years. This is for software, BTW; for movies or books, different terms might make more sense.

[magicalhippo]

I think that if a copyright is held by the creator, it's fine that it's longer. If it's inherited to a person then the age should be inherited as well, so the exponential fee kicks in as if the creator had lived.

If it's sold or otherwise acquired by a non-person (company or similar) the exponential fee should kick in immediately.

Something along those lines.

[bluGill]

A lot of things take more than 5 years to create. If save all your garbage and find the notes from your rough draft that you started more than 5 years ago does that mean your work is not copyright, merrily a derivative work? Some authors write a novel in 6 weeks, but others take years to polish it.

[shiroiuma]

>Some authors write a novel in 6 weeks, but others take years to polish it.

I think 5 years might be too short for a novel, but regardless, the lesson here is: don't publish your work until it's ready. Copyright protection should start when it's published, not when the first word was typed.

>If save all your garbage and find the notes from your rough draft that you started more than 5 years ago

Maybe there should be a provision about rooting through someone's trash? This seems a rather rare edge case.

[nordsieck]

> It feels a little out of calibration. How about free for 20 years; $500 for 10 more ... Everyone gets 30 years to exploit a work for a reasonable price

IMO, it's important to have a short default period. The vast majority of content - posts, comments, videos, tweets, open source software, etc. will never be registered. It's a huge benefit to get that into the public domain as fast as possible while still providing a reasonable period of exclusivity to creators.

If you're really wedded to the 30 year thing, I think an initial 10 years, and 20 years for the first registered period is a better balance.

[mlyle]

On the other hand, some kinds of work may not really make it to market inside the 10 years. It's all tough compromises.

I'm sure we can agree that we can definitely do better than what we have now, though. Let's not make perfect the enemy of good.

[nordsieck]

> On the other hand, some kinds of work may not really make it to market inside the 10 years.

My understanding is that copyright starts when the work is first fixed to a medium. Which means that intermediate versions shouldn't start the clock[1]. But I'm not a lawyer, so I could be totally wrong.

> I'm sure we can agree that we can definitely do better than what we have now, though.

100% concur.

---

1. As opposed to the situation with medication and patents, where the patents are ticking down while trials are in operation.

[Sakos]

Copyright was never meant to give perpetual rights, which is what your scheme would do (as it does now). So might as well not change the current system, because the conclusion is the same.

The final arbiter was never "does it have commercial value still?" until today. That's a purely modern perspective pushed by companies like Disney, because of course it is. That's all they care about. We don't have to care about that. In fact, I'd suggest it's completely immoral to accept that framing when we look at how important public domain has historically been to our culture.

[MacsHeadroom]

How many copyrights are worth half a billion dollars to register from 50-60 years? Presumably zero are worth registering for $50B to protect from 60-70 years.

This hardly seems perpetual in practice.

[mlyle]

> was never meant to give perpetual rights, which is what your scheme would do

Pretty quickly the cost to renew exceeds the amount of money in the world.

This scheme tosses the few powerful creators a small bone while shortening duration of all copyrights.

In addition to generating revenue from these large creators, and shortening their total term of protection... it also diminishes their power by creating a vibrant public domain with a whole lot of contemporaneous works in it.

[johnnyanmac]

Numbers can be tweaked but it adds up quickly.

Protect for 40 years for a million? Sure. Probably worth it for mega IPs

50 years for $100m? Still doable but it really starts to hurt.

60 for $10B? Maybe Pokemon and Mikey mouse can justify it, but very very few.

By 70 years it's simply not worth it.

>This scheme seems to protect the largest companies with the largest bank accounts the most. Why would we want a system like that?

Because the small time authors really won't care in 20 years anyway. Even if the cost to renew is a pittance. Are there authors trying to sell the copyright to a failed IP in 30 years?

[robocat]

My friend was getting royalty cheques 30 years after creating a song (slightly different from books) and the money mattered to him. The amounts were small and probably not worth any paperwork nor the risk of prepayment (paying for extension) which might not get back.

Hard to find the right balance, but what we have now is asinine.

[glimshe]

I'm sorry, but this isn't necessarily a good thing. Your friend takes this money and goes have a haircut, the person cutting their hair won't be making money for the haircut 30 years later. Okay, we can give some extra benefits for intellectual property, but 30 years is way too much.

[avereveard]

Heck even something absurdly low like 1$ per year would already be enough to salvage all those works from bankrupted companies that will lapse on payment. There are so many IP in limbo because their ownership is fragmented, unclear, or forgotten and having a payer renewal fees would clear up a lot of that real fast.

That said I like the exponential scheme.

[redwall_hp]

Exactly. Works that generate massive amounts of commercial value are therefore more culturally important and should fall out of copyright faster, to maximize the amount of creative derivatives that can be produced while it's culturally relevant.

Works with little to no commercial activity should be protected longer to provide more opportunity to do so and give more protection to unknowns.

[glimshe]

Large corporations won't bother to protect something that can't pay the 100x increase - Do you think Activision would have paid for Infocom rights if it cost 200K at this point? Maybe yes, but quite possibly not. But Harry Potter and Lord of Ring owners likely would, no matter whether they are small or large. This money could go to other services, such as education.

For small-time authors, why should they enjoy the intellectual property rights for the rest of their life? The vast majority of people need to keep working to keep generating money.

[mixmastamyk]

They’re not protected—they’re made to pay to keep the work out of public domain. The deepest pockets will likely pay the most.

[3seashells]

It's realistic to be adopted cause it acknowledges the takeover of democracy by buisness. What good is a policy, when it in implementation a maiden pure but has no chance in hell to become real? Purity signals are useless, when constructing the signal network that runs society.

[xlii]

I like the idea but would tie it to a yearly revenue + total for receiving entity and cut those in periods (10 years maybe less) in a form like:

Free, 1% of revenue, 10% of revenue, 100%, 1000% etc.

Thus non commercial darlings could be kept relatively long without high cost but all commercial offerings would wither off relatively quick. Pipe dream as obviously companies in power are controllers here.

[thaumasiotes]

> That way commercial works that are very economically valuable can be protected for a pretty long time, but everything makes it into the public domain relatively quickly.

But wouldn't we want the valuable works to enter the public domain faster than the worthless works?

[tom_]

Once the free period expires, start at $1, then double each year.

[nordsieck]

> Once the free period expires, start at $1, then double each year.

That's going to create a lot of paperwork for not much value. Much better to have 10 year extensions. Your pricing scheme is functionally equivalent to starting at $2000 for 10 years, and increasing the cost by 1000x for each additional 10 years.

[audiodude]

Another annoying thing about the 1976 Copyright Act/Berne Convention is that copyright is automatic. You don't need to register with any organization, any "original work of authorship fixed in a tangible form" is copyright. This is the real problem with public github repos that don't have a license file, because they also likely lack a copyright imprint with a date.

This is not related, but another problem with current US copyright law is that there's no exception for companies that go out of business. The works are still copyright encumbered even if no one exists that can enforce that right.

Based on the former two points, you end up with works where you can't find out who holds the copyright, and it doesn't matter if they're dead or whatever because it's still copyrighted. This leads to the "assume everything is copyrighted" posture that stifles so much creativity.

[kmeisthax]

That's not the most annoying thing about the 1976 Copyright Act.

The most annoying thing about the 1976 Copyright Act is that copyright is only nominally automatic. To be clear, it casts a huge shadow over all creativity that would have otherwise been uncopyrighted. But on the other side, there's still a registration system. You need a registration in order to sue infringers, and you don't get statutory damages on infringements that happened before registration[0]. This trips up loads of creatives, and especially photographers, because it's a lot of boring bureaucracy that never gets explained to them up until they've already talked to a lawyer who says "no you can't get $$$ out of this big company that used your Facebook uploads without permission, because copyright lawsuits are never worth pursuing without statutory damages on the table".

The cruel irony of US copyright law is that, while we only half-implemented Berne, we still use Berne as a thought-stopping cliche for why we can never claw back copyright protection from the half-dozen publishers and creative artists that actually benefit from owning your childhood. Because the base assumption of copyright is that only the creative upper class is worth protection. Protecting artists as a class requires syndicalism and mass unionization, not atomizing everything into individually held psuedo-property rights that are financially ruinous to assert against anyone who won't fold immediately and settle.

The orphan works problem you're talking about is deliberate. Publishers like the idea that when they knock out a less-scrupulous competitor, their creative works spill out onto the ground like Diablo loot, and they can collect all that up and just idly hoard it forever. At the very least, a work that nobody knows how to license or can't afford to license is a work that has been taken off the market and can't compete with them.

Funnily enough, GitHub repos without a LICENSE file are covered by a fallback license in GitHub's ToS that basically says you're allowed to fork and PR. They've probably also explicitly added a "and we can train GPT on your code too" EULA ruffie in there too.

[0] There is a short grace period for this, of course. I think it's 90 days.

[jpc0]

> But on the other side, there's still a registration system.

This a thing in the US? It isn't in my country, you just need to be able to prove you own the copyright, a stamped letter or certified document is good enough.

You cannot make blank statements about the US on a platform that has an international audience.

Copyright law is cross border and the US does respect other countries Copyright acts just like other countries respect the US

[kmeisthax]

Yes. As far as I'm aware the US is one of the few countries that still has some vestigial copyright formalities.

>You cannot make blank statements about the US on a platform that has an international audience.

I'm not sure what you're saying - I was very clear that this is US specific law.

That being said, while I don't exactly know how it works for works created by Berne signatory country citizens, I suspect foreigners still have to register in the US before suing in the US. "International copyright law" is generally just a pinky promise to let foreigners use the copyright system as it exists in other countries, so the rules you have to follow depend on what forum you sue in. For example, if you can get standing to file a copyright lawsuit in Japanese court, then your defendant can't mount a fair use defense, because there is no fair use there[0].

That being said, I do have to wonder if you could get SCOTUS to agree to "foreigners don't need to register but Americans do", because that's the sort of blatantly stupid ruling that only a law-huffer would love.

[0] To be clear, fair use is generally a concept borne of the Anglosphere and it's weird obsession with common law and precedent. Other countries don't have it. But those countries will still codify exceptions to copyright that do similar things. Japan is just unique in that they didn't even bother doing that, so it's illegal to, say, review anime if any Japanese fans might be watching your reviews.

[jpc0]

> I suspect foreigners still have to register in the US before suing in the US

Generally that wouldn't be needed since it's no US copyright law that is being violated but the country of the copyright holder. Being in violation you would not be tried in the US but in the copyright holders country.

There is this whole thing where in general Copyright is unenforceable but when it matters it is enforced, since nobody really cares if you as an individual infringed on my Copyright since to be fair there just isn't enough to be claimed in damages for it not to be a frivolous suit.

But when the amounts become significant enough it is quite serious.

https://www.wipo.int/wipo_magazine/en/2006/02/article_0006.h... for reference to an influential case on copyright law, there are other more modern cases as well

[TheRealDunkirk]

It is a fundamental truism of American "capitalism" that, if there is a law for something, it was written to protect established "capital" from competition. There are no free markets in the US.

[kmeisthax]

Capitalism is a necessary transition from feudalism to feudalism.

[bruce511]

I doubt any would describe my voice as "loud", but I suspect I fall into the 0.01% you mention.

I sell code (programming libraries, and commercial business systems) which have been under continual development sint 1996. So 27 years.

Copyright in this context would be complicated in a "20 year" model. Last month saw a significant upgrade to a product first released in 2000. Should the 2K version be public domain now?

I get the main complaint, especially in the context of old games like infocom (which are abandoned). It feels like those should move gracefully to public domain. But on the other hand there are those of us who do still make a living from old, but active, code.

I do agree that current copyright is absurdly long, but I also understand why Disney et al see ongoing value in their creations and are prepared to lobby for that value.

I don't think there's a simple fix here, one which covers such a wide set of circumstances. All the fixes I've seen proposed are "great for x, terrible for y".

[dagw]

under continual development

Copyright timer would be reset for every new release. So the version of the library you released in 2000 would be out of copyright, but the version you released last month wouldn't be. I guess there is a risk that someone will take your version from 2000, reverse engineer it, and release a competing product, but it seems like a minor risk.

[mkesper]

Would you be restricted/harmed if the version from 2000 was public domain by now? How?

[bruce511]

For yhe most part, Code-wise I would not be harmed by that. (We supply source code anyway so there are no secrets there.)

Market-wise I could see a fair bit of confusion as an old (now public domain) version floats around with the same name (but different version number) to the commercial version.

Of course this is likely immaterial anyway. Unless someone bothered to archive the version at the time, that version no longer exists. We can date it (from the release notes) but I no longer have that build in the archive.

So maybe my wariness is unfounded.

[elondaits]

The conversation is hard because we’re thinking about being able to copy and distribute an old game… and some authors will be thinking about not losing rights to their original characters, music, story, etc.

… so maybe it would be useful to have a law that protects the preservation of old media, allowing the public to legally break DRM, copy distribute and shift it to new platforms after a number of years… without necessarily granting rights to economic exploitation of the characters, story, etc. or making new derivative works.

[Sakos]

> some authors will be thinking about not losing rights to their original characters, music, story, etc.

But this didn't used to be an issue. It was accepted that things would fall into the public domain and you couldn't benefit from something for eternity. It's unacceptable that we've finally reached a point where only the creator or owner can ever do something with a creation, regardless of how long ago it was created. Just because you created a thing doesn't give you an inalienable right to own it forever. We're just making it so because "commercial value" and other bullshit.

[ethbr1]

There's an unexplored (to my knowledge) middleground where copyright expires more rapidly with regards to derivative works, but not the original.

E.g. At x+10 anyone could make a new sequel to Zork, but the rights to the published Zork are retained for x+25

That feels fair.

[mlyle]

And there's a sibling to your comment proposing the exact opposite https://news.ycombinator.com/item?id=37895648

I can think of arguments why derivatives should get more favorable and less favorable treatment than overall copying. To me it feels like a wash and they should probably be treated similarly-- though trademark can provide some limited protection beyond copyright terms if the work's use in commerce has been continuous.

[ethbr1]

My reasoning is:

The original thing has already been created, and proceeds benefit whoever created it (or who they sold rights to).

Whether they choose to continue to actively market it or not, while under copyright, can't be influenced by copyright. Abandonware under copyright is still abandonware.

But! There's a class of derivative works, built on top of the original, that currently can't get made.

If they were, we'd all be better off.

Arguably, I can even see an aggressive sunsetting for derivative works being net positive for the original work's commercial value ("See the original").

[dagw]

At x+10 anyone could make a new sequel to Zork

Would the copyright holders of Zork get royalties in this scenario, like with cover versions of songs?

The other problem is that in this scenario people will just be slapping the names of things that were popular x+10 years ago on literally anything. People will just make the lowest effort loot box laden pay-to-win mobile games you can get onto the App Store, and advertise them as sequel or tie-in to that old popular game/book/movie.

[ethbr1]

> royalties

Debatable. Some sort of FRAND-level payment, for a limited period, seems fair? Not enough to torpedo the economics of anyone using a property. But enough so an originator has a revenue stream for wildly-popular IP.

> slapping popular names

Would this be that bad? If there were Harry Potter crap... how would that be different? Expect there'd be more stuff out there.

[dagw]

Would this be that bad?

It's essentially a case of brand dilution. Today 'Harry Potter' is a brand that has certain values. If a new Harry Potter book shows up on the shelves tomorrow I can be very sure it's a kid friendly, easy to read, book about wizards. People can feel safe buying as a present for their niece or grand child that likes Harry Potter. In this alternate future a 'Harry Potter' book could be literally anything. If that is 'good' or 'bad' is left as an exercise to the reader, but something would definitely be lost.

[jccooper]

An explicit renewal regime with some reasonable fee would allow most works to pass into the public domain, and avoid objections from owners of still-valuable properties. (I like the concept of renewal fees escalating each period, a kind of dutch auction against the public domain, but I'd accept even a fairly trivial flat fee as the price of avoiding ever-longer universal copyright.)

[toast0]

Explicit renewal is also nice, because there's more likely to be updated contact information, which makes licensing more likely.

[TulliusCicero]

20 years, same as patents.

People aren't writing books because of the profits they might make 25 years from now.

[zuminator]

Sure they are. Plenty of film and streaming projects are based on older work. Sapkowski's Witcher series comes to mind. If his rights to the IP had expired after 20 years then not only would Netflix been able to rip his work off for free, but he wouldn't have benefitted at all from the resurgence in popularity that his novels garnered. It would've just been gravy for some publishing company -- how would that be fair?

I do think that the current copyright regime is too long. I guess I'd like to see something like ~25 years from publication for human authors, but easily/cheaply renewable to lifetime copyright; and 25 years from creation for copyright held by companies, or for works for which the copyright has been assigned to someone else other than the original author, with no renewal.

[thaumasiotes]

> If his rights to the IP had expired after 20 years then not only would Netflix been able to rip his work off for free, but he wouldn't have benefitted at all from the resurgence in popularity that his novels garnered. It would've just been gravy for some publishing company -- how would that be fair?

Presumably in the same way that if you sell the mineral rights on a plot of land to someone else, and then they find oil or rubies there, it's thought of as fair that they get the money from that and you don't.

Or how if you sell someone a stock, and it goes up, they get the money from the appreciation and you don't.

Or how if your neighbor finds buried treasure on his land, and you find none on yours, he gets a big finder's fee and you don't get anything. (Indeed, in this case, there are commonly accusations of unfairness, but those tend to focus on the treatment of the neighbor being unfair. You deserved the nothing that you got.)

What's a scenario where this kind of event wouldn't be called fair?

[TulliusCicero]

You seem to have missed what I said. I didn't say nobody gets money 25 years down the line, I said people don't write because of the profits they might get 25 years down the line.

Yes, Netflix might've been able to make The Witcher without paying him, and also random Youtubers, complete nobodies, would've been able to make derivative works as well. That's the whole point of expiring copyright.

Does that favor big corporations? Well, it also means anybody and everybody can make Mickey Mouse content, or use the popular Marvel and DC characters, etc. Do you think Disney is a fan of that idea?

So while corporations would take advantage of it, so would random people, including indie artists.

[zuminator]

I still have to disagree on both counts. First of all, plenty of writers write with at least the hope that their work will continue to generate residual income for the rest of their lives.

Second, no, even with the expiring copyright on Disney's earliest works, people can't just go out and make their own Mickey Mouse content, because there is trademark protection on Mickey Mouse and other major Disney IP, and that can be renewed indefinitely.[0] So even a quickly expiring copyright regime would still favor the deep-pocketed large corporations.

And you might say, well, the little author can trademark their work as well. But most won't. So all Disney and Warner would have to do is sit on the sidelines, trawl through the 25 year old fantasy section for interesting stories and characters that have not been trademarked, and just start use NG them without compensation. And what's worse, by introducing minimal changes to the characters (“Frobo Daggins“) they can then copyright and trademark their derivative works such that even the original author could not turn around and try to benefit from the resurgent popularity of their creation.

Not expecting to change your mind or anything, but I hope I have communicated my misgivings.

[0] https://blogs.luc.edu/ipbytes/2023/08/13/is-disney-losing-mi...

[rgmerk]

Yes, but people who wrote commercially successful books 25 years ago will scream blue murder if you take what they see as their retirement nest egg away from them.

[vlz]

I have to disagree somewhat. People write books for a lot of reasons, but one alluring thought in many an author's head certainly is: "Sure, it might not be commercially viable directly now, but over time…"

Writing books can be a pretty risky proposition from a financial point of view. Copyright should at least span an authors lifetime in my opinion.

Thinking that an author could profit not at all from a work if it gains popularity 20 years after it was written also seems unfair, seldom though as that case may be.

[TulliusCicero]

> Writing books can be a pretty risky proposition from a financial point of view. Copyright should at least span an authors lifetime in my opinion.

Why is it accepted that patents can last only 20 years then? Like, do you think creating new inventions is easy? Why for one and not the other?

But the bigger thing is that books making significant amounts of money after twenty years is an extreme exception. That's not the motivating thing. It's a "maybe that'd be nice" on top of the desire for revenue in the immediate future, in the first few years of the book coming out.

[rgmerk]

The proportion of books that make money 20 years after publication, after languishing in obscurity for that time, is a rounding error.

Like, how many such books can you actually name?

I think it’s a little different for music, where people sometimes sample fairly obscure old songs and turn them into hits. But then, a lot of those old songs shamelessly ripped off even older works…

[andyferris]

Shorter copyright would be great.

I could imagine an argument for harmonising the duration with patents for even more simplicity.

Another interesting consequence is a lot of GPL code would become public domain. I haven’t really thought about that before.

[dagw]

authors whose work still has some commercial value decades after its creation.

Sure, but who should capture that value for those works. Is it more fair that all the value goes to the publishers and TV/movie studios making adaptations, rather than some of it being shared with the author.

[atmavatar]

Make a copyright cost 1$ to register. You have to re-register every year. Each year, the registration cost doubles.

[sneak]

Getting the congress to pass laws that say copyright is whatever you want it to be is cheaper for Disney than 2^100 dollars to keep the mouse under their control.