The original Copyright Act of 14 years, with an option to extend an additional 14 years, seems reasonable to me.

I've liked the suggestion of an increasing annual cost, perhaps double, every year past a certain baseline - so if the work is valuable enough, productive financially, then the owners can afford to retain/maintain full control over it.

Really, one of the major issues for old and abandoned work is that it can be impossible to figure out who the rights holders are for a particular work.

A step of active renewal by the rights holder would at least create a public registry of the current owner for a particular right.

This is a constant problem, publishers go out of business all the time. I have a lot of photocopied sheet music that cannot be purchased anywhere, at any price, but it will be decades before any of it can be legally shared. Just ridiculous artificial scarcity - I wonder how the composers would feel about it.

Yep, I really liked the copyright change to make copyright protection automatic, instead of requiring a filing.

But it should require active regular renewal, to create the record of continued ownership.

You should get a (reasonable) protected period of time automatically, to give you time to figure out which works are worth the effort to continue protecting. But it should require (minimal) active ongoing effort to maintain that protection.

For me, this is especially relevant when it comes to abandoned digital content (like video games and other applications). It'll be *so* much harder to archive and preserve content that's 140 years old, than it will be content that's 14-28 years old.

I realize it creates a legal quagmire to have abandoned works and I think registration should be required to solve that.

That being said, what in reality would happen if you release that sheet music? If it's abandoned, there wouldn't be anyone to come after you for it. You could just put it up online for free and then honor DMCA takedowns from people who can prove they are the rights holder.

> I realize it creates a legal quagmire to have abandoned works and I think registration should be required to solve that.

I really like the model of every work automatically receiving protection. I think creations should automatically receive a relatively short-term automatic copy-right protection (say 14 years). After that, I think extensions should require manual renewals by the current rights holder (including contact information for the rights holder)

This would just mean that rich companies/people will get to keep the copyright and others do not. We're in this whole situation where things don't enter the public domain for 100 years because of Disney's lobbying.

On an exponential curve, even companies would be thinking hard about renewal a few doublings into the process. Even Disney would start getting to the point where they'd be picking and choosing and not just forking over $(2^n × number of works) every year.

Also, I often advocate this, and to some extent I consider it a strategic retreat. Fine. You're a big company and you want to own your stuff forever. But behind the laws protecting those companies there's a ton of stuff that the owner doesn't care, nobody even knows who the owner is, etc. etc. (There's a lot of stuff that is de facto in the public domain because nobody owns it anymore in any practical sense, but there's no way to be sure what that stuff is, and the risk is too large to take.) The stuff the big companies are defending is just a small fraction of what exists. If we tuned the laws to give the big companies what they want (more or less) but stopped protecting everything else it'd be a win. And they can pay an increasingly steep fee for the benefit.

Or, to put it more prosaically, I don't really care how long Steamboat Willy stays under Disney's copyright, I'd like Steamboat Willy to stop shielding everything produced ever.

This would definitely be better than the current system of basically forever minus one day, but still leaves the problem that a few giant corporations own giant chunks of our hearts, minds, and childhoods.

The zeitgeist of our time should belong to us, at least after a decade or two.

This is just extending the rent seeking. If the work is so valuable the owner can continue to extend their copyright, the public should receive that value after the 14 years (by which point, the owner has been made whole through collecting rents for those 14 years) versus the owner ratcheting up the cost to pay to maintain the copyright monopoly on the work in question.

Eh, if the price rises enough year over year, and you dump that funding into the National Institute for the Arts, or other programs to provide grants and funding to low-income artists, I could see it being a potentially useful tool to spread the wealth around from very commercially successful works.

I'm not necessarily supporting it, but I could see some positive outcomes from the approaches.

I can see scaling it by change in life expectancy, to keep the bargain roughly the same.

Life expectancy of adults hasn’t changed much (ie your life expectancy when you’re in your 20s). The life expectancy number you’re most familiar with is largely driven by infancy because it’s the mean and high infant/child mortality skews that. We’re really good at saving the lives of children now but I don’t see why that matters for copyright.

More hypothetically, let’s say human life spans got up to 500 years. Copyright is intended to promote the creation of art for the public good, not to fund the artist for their lifetime.

If you created a popular piece of art of music in your 20s, why is it useful to society to help fund the next 480 years of life in any way?

This is mostly an urban legend. Yes, some amount of the increase in life expectancy is due to reduced infant mortality. But it’s nowhere near the bulk of it.

In 1920 at birth a males life expectancy was ~54, at 40 a man could be expected to live to ~70. Current estimates are almost unchanged between birth and 40 at 76.23 vs 78.75. That’s a 25% increase in life expectancy at 40, but that a long way from the 41% increase from birth.

I chose 40 because it wasn’t medicine that made the biggest difference for 20 to 40 year olds but war. And actuarial tables can’t predict wars only account for past wars which make current estimates misleading. Aka we don’t have a lifetime of data for people born after 1920 only estimates.

Yes, your original post was indeed correct by hedging a bit, speaking of "adult life expectancy" vs. "infant/child mortality".

People read that and it registers as infants dying within days after birth. But life expectancy for 10-year-olds has doubled: https://ourworldindata.org/life-expectancy. I doubt that your comment makes people think of dead teenagers, even though your post doesn't contradict that interpretation.

Even for the 40-year-olds, life expectancy increasing from 68 to 82 years means a >50% increase in "time remaining" which isn't something to sneeze (cough?) at, although I have to reluctantly acknowledge that I remembered it being even larger than that.

I think you’re confusing me with someone else, this is my second post in this thread. Anyway, the data doesn’t support doubling life expectancy for 10 year olds unless you go really far back.

Look at the England a whales data in your link, in 1841 10 year olds could expect to see 57.6, or another 47.6 years. In 2013 that bumped to 82.2 or another 72.2 years. By 20, they could likely hit 60 or another 40 years worth of life, and by 40 they where up into ~67.4.

However again the difference is largely one of lack of war and famine rather than medicine. The chart is bouncing around because of events that killed large numbers of people, but it smooths out as you approach today.

PS: Their Spanish Flu suggestion is off, it was actually WWI and the flu that’s responsible for much of that drop as seen by 40 year olds being largely unaffected. https://en.wikipedia.org/wiki/Spanish_flu#/media/File:W_curv...

How are you defining urban legend? Here's from Wikipedia [1]

> Until the middle of the 20th century, infant mortality was approximately 40–60% of the total mortality

> Excluding child mortality, the average life expectancy during the 12th–19th centuries was approximately 55 years.

> ... even in preindustrial times as is demonstrated by the Roman Life Expectancy table, which estimates life expectancy to be 25 years at birth, but 53 years upon reaching age 25

Here are actuarial tables for people in the US today [2]. The life expectancy if you reach 25 is.... 52 or 57 depending on if you're a man or woman.

We're certainly better at reducing all cause mortality even in adults but it's not as stark of a difference as you might imagine, particularly its effect on the mean.

[1] https://en.wikipedia.org/wiki/Life_expectancy

[2] https://www.ssa.gov/oact/STATS/table4c6.html

You are not comparing these numbers correctly.

Your second link shows years of additional life expectancy. For example, at age 80 it shows 8.43 years of life expectancy, meaning that the average 80-year-old lives to age 88.43.

So the average 25-year-old in the US has 52–57 years of life remaining, and will live to age 77 or 82, according to your link.

This is a very big increase over the pre-twentieth-century adult life expectancy of ~55 years of age! You claim that "life expectancy of adults hasn’t changed much" but your own links show that it increased about 50% in very recent history.

Thanks for the correction. So we're at a 45% increase. Typically life expectancy is reported as the "from birth number". That's gone from ~20 to 80.

The claim was that the bulk of the life expectancy number being driven by the mortality rate prior to adult hood was an "urban myth". Life expectancy in adult hood is a ~1.4x increase. Life expectancy from birth is a 4x increase. I'd say the bulk is driven by saving the lives of children and is not an urban myth.

You're still not doing apples-to-apples comparisons. The source that you cited says 12th–19th-century life expectancy at birth was 25–40 years when infant/child mortality is included, and 55 years when it is excluded. There has not been any “4x increase” in this time period. (Even if you start in the bronze age, life expectancy at birth has not increased four-fold.)

Since the time period you chose, life expectancy at birth has increased by about 40 to 55 years. Near-elimination of infant mortality accounts for 15–30 years of increase, while increases in adult life expectancy account for an additional 25 years.

North America has a life expectancy of ~80 years. I eyeballed it so closer to 3.2x the 25 year life expectancy in the 12-19th century. What am I missing?

Yes, if you start from a base of 25 years (which is the extreme low end of the range from your source), then we have 30 years of increase from reduced infant/child mortality and 25 years of increase from reduced adult mortality.

Can you really look those numbers and say that one of them is highly significant while the other "has not changed much"?

Translating this into percentages or "fold" increases is misleading because you are using a different denominator for the two increases. E.g. you are claiming that reduced infant mortality is a 2.2-fold increase (55/25) while reduced adult mortality is "only" a 1.45-fold increase (80/55), even though the number of years gained is almost the same.

And if you use the other end of the 25–40 year range, then reduced infant mortality is "only" a 1.38-fold increase (55/40).

I'm fine with that, if you use the life expectancy for an adult. Infant mortality has contributed the largest component of life expectancy gain since 1790, and we weren't incentivizing the creation of toddler art with the life expectancy gains.

So, if you scaled it from the life expectancy of an 18-year old in 1790, to the life expectancy of a 18-year old in 2021 then I'd agree with that.

50%+1 of a standard living persons actionable career. Do assuming 18 to 70. That makes 52/2 = 26 and +1 = 27. For sake of even numbers, you can fight over 25 or 30. I don't mind either way, both are far better than the current status.

I was going to go with 10 years. I mean, nobody cared about Madonna's "Ray Of Light" beyond 1999.

On the other hand, artists - like Dolly Parton - have made long careers out of songwriting that have carried them into their senior years, so it's a balance.

Clearly 99 years is absurd. Maybe 20 years is the balance to strike. Dolly would still be doing just fine with that term.

As the patron saint of my state, I'd also point out that she has used the money from her music to invest in other things. She is quite a shrewd business woman and I think even with 10 years she would have been fine.

Though I think 20 years is reasonable enough.

IMHO you have to start with the justification.

The previous justification was to incentivize creators by protecting their financial interests and the interests of their offspring. It’s very clear to see that those interests are no longer being met when you sell your blockbuster song to a record company for $500.

IMHO make copyright non-transferable and make corporate copyright 14 years with an application to double. That way it works for the little guy without big guys abusing it (e.g. Disney).

Life of the author+0 years.

Most people expect to inherit money and property from their parents but not to be paid for work their parents did years ago. I don't see why children of authors should have different expectations.

Creates the odd incentive for people to die.

And also to not die, since they earn money for their children. It's the reverse of the usual inheritance situation, where the state and the kids both want the person to die.

Maybe timeleft = (AVG_LIFE_EXPECTANCY - currAge) or till death if currAge > avglife

Is it the kids combining or the companies that bought the golden goose?

Maybe 10 years. We are moving at breakneck speed in just about every corner, but copyright protections. A decade is forever in today's world.

Is 10 years enough protection for the source code of a startup? Sometimes it takes longer than that to break even…

All law moves slowly. The law is supposed to move slowly, that’s (mostly) a feature, not a bug. You don’t want the existing laws to change every year, driving would be more of a nightmare than it already is.

Copyright was originally about protecting a creator’s ability to make some money from their works. Despite the fact it was extended in part to help Disney make money forever, we should still have some protection for artists and small businesses, right? We don’t want to shorten the copyright term just because NPM has a lot of churn, do we? That doesn’t seem like a good reason.

It took 7 months for the Sonny Bono Act (of Sonny and Cher and at the time congressman) to extends the copyright from 50 to 70 years. I would call it pretty fast when you think it took several years for congress to vote the last "infrastructure" law (remember 2017's infrastructure week?).

By the way, that law was sponsored by Mary Bono, Sonny Bono's widow (Sonny Bono died during that year) and also congresswoman.

All this is a neat little circle of people voting laws for their own benefits. This has nothing to do with artists or small businesses.

I think it is fair to say that once the original author of the work is dead, the copyright should go with it. And if a corporation buy the copyright, then they should be able to recoup their investment in a timely manner and 10 years is not that huge with all the means of electronic distribution we have today.

> they they should be able to recoup their investment in a timely manner and 10 years is not that huge with all the means of electronic distribution we have today.

It’s neither fair nor accurate to point at the speed distribution as an indicator of the speed of building a viable business. It’s a fact that it’s common for a business to need a decade to build it’s business pipeline before the business is viable and fully supporting its employees. Don’t forget this system needs to support small businesses and not just megacorps.

From the perspective of law, and of copyright specifically, it’s also a mistake to assume that all business can move at the pace of electronically distributed software. Physical hardware goods frequently take more than a year to design, manufacture, and stock prior to distribution. Today we can add in an extra year of COVID related delays for the ordering and distributing pipeline … how many things are on insane back order delay as we speak? Copyright needs to be long enough that things like today’s situation don’t prevent a business from even starting to recoup, right?

> I would call it pretty fast when you think it took several years for congress to vote the last “infrastructure law”

The point I was making is that law moves slower than tech. You would agree that infrastructure funding happens on a glacial pace compared to what’s hot on GitHub or what new ideas are being incorporated into neural network training, right? It’s fine to think of a few years being “fast”, but that’s a subjective opinion that isn’t relative to something specific. I was trying to make a comparison between law and the things presumably parent was calling “breakneck speed”, which is clearly referring to things that move faster than law.

I also mentioned “existing” laws for a specific reason - new laws can and sometimes do come up quickly. But laws we need to know in order to comply with need to be mostly stable. And by and large, most of our law doesn’t change every year, we have small nips and tucks mostly in places that don’t affect the majority of the population.

> From the perspective and of copyright specifically, it’s also a mistake to assume that all business can move at the pace of electronically distributed software

I think you are confusing copyrights with patents or trademarks. Copyright do not protect physical hardware or goods.

As far as I know, almost all copyrightable works can be transferred electronically. After all, we are talking about the right to copy (and distribute) a work (make more of them to sell it). Goods that cannot be copied are not copyrightable.

> Copyright do not protect physical hardware or goods. […] almost all copyrightable works can be transferred electronically

Don’t conflate “can be” with “are”. Copyrights absolutely protect books and posters and phonograph records; physical goods is how the law originated, and physical copies is a big part of what the law’s language still protects to this day. There are big problems with copyright law because it hasn’t yet been adequately adapted to the invention of the internet.

There is plenty of overlap when it comes to the business of physical goods. Copyrights do protect parts of the designs of many physical goods, and the sources and driver software of computer hardware, for example. Copyrights also protect marketing and advertising material and many other parts of a fledgling business.

What are we discussing at this point? I agree with you that copyright should probably end with the author (though I would say there should be a minimum term, to account for the many reasonable possible scenarios like being part of a business, accidental death, old age, etc., etc.). I don’t think 10 years is enough, and I don’t think that tech moving quickly is a valid reason to shorten copyright. There are valid reasons to shorten copyright, so we should discuss those instead.

But think of all the businesses that might be built on the work of the original business that never took off. What we are trying to maximize is overall benefit to society, so we grant a monopoly to a creator for a limited term to incentivize new work.

Ten years may or may not be too short. How often does ten year old software make the difference for a company taking off this year?

> we grant a monopoly to a creator for a limited term to incentivize new work

That’s a nice goal, but not entirely accurate. Copyright is intended to protect individual works of authorship, and allow the author to leverage their work, even when they only produce a single work.

> How often does ten year old software make the difference for a company taking off this year?

It happens all the time in software companies! The early software foundation of a company is often integral to where it is when it starts making money. It’s common for the core ideas and algorithms to sit around for many years while the company is building marketing and payment and integrations and customer support and waiting for business to catch on. Sometimes companies wait on their patents to be granted before even starting to try to monetize some software, and it can take a couple of years just to get the patent.

From a copyright perspective, it’s important to understand and recognize that business, like law, often moves very slowly. While I’m firmly in agreement that 99 years is too long, I also think 10 years is too short. And to reiterate: the term length shouldn’t be set because tech moves fast, it should be set by considering what’s long enough to recoup an investment and when it’s fair for the copiers to come take your work and use it for their own financial gains.

> That’s a nice goal, but not entirely accurate.

I based my statement on https://copyrightalliance.org/education/copyright-law-explai...

> it should be set by considering what’s long enough to recoup an investment and when it’s fair for the copiers to come take your work and use it for their own financial gains

I don't think we disagree really. I said the goal was to maximize overall benefit to society and I think you did too. Nobody will know if 10 years is too short until the issue is studied.

> Nobody will know if 10 years is too short until the issue is studied.

I disagree. Plenty of people already know 10 years is too short, and it doesn’t take a study to show it. Lots of businesses take longer than that to get established, even the very fastest-moving fastest-distributing businesses like software startups.

It might be bad for you if code you wrote ten years ago is no longer protected by copyright, but what about society in general? Maybe five new start ups can take that old software and build something better than you were able to. Would that disincentivize creators? Maybe. Would the copyright expiration make a thousand flowers bloom? That's what would need to be studied. What the net benefit or harm is, isn't obvious.

True, this is a valid point. I’m under the impression that it has been studied, though I don’t have a source off the top of my head. It’s definitely been debated vigorously and often.

Anyway, keep in mind that this net benefit needs to incentivize people to get over the static friction of starting in the first place. We don’t want to make a point of handing content and ideas over to incremental producers because that will disincentivize people from trying in the first place.

Also important is that part of the net benefit to society is a trickle-down argument that we need to reward the authors first in order for society to reap the net benefit. The point is not to encourage business at the fastest possible speed be letting people borrow the ideas and inventions and content of others, the point is to protect the people who do the actual original work for a reasonable period of time, and to allow for the fact that they might be less good at business than someone who’s only reselling content.

Might be worth stating that study the economics of social net benefits is incredibly difficult, and can take a very, very long time, longer than you and I have left. The environment is a great example of how we’ve made disastrous business choices that were rationalized by “social benefit”, but turned out to be wrong.

> Is 10 years enough protection for the source code of a startup?

How many startups today make their money by sharing actual source code and not cloud services? Unless its AGPL then you need not know the source for most services anyways.

But there's plenty of things made 10 years ago that are still very relevant and still generating money for the creators. I think that's the concern.

Yeah. New Order are still making good money from Blue Monday from 1983, and clearly there’s a lot of interest in The Beatles still.

Maybe a shorter period of time with an option to extend it like a snooze feature on an alarm clock.

The option to extend it being priced by the royalties collected and requiring substantial accounting of such.

Patents, for example, have maintenance fees that are quite high. So even that 20 year length is not really automatic and guaranteed, while already seen as problematically long to many.

But does having a work enter public domain automatically mean that eg New Order will stop getting paid for streams via legitimate services like Spotify?

Why would Spotify pay anybody for playing a public domain song?

Presumably if they want to have access to the non-PD stuff they’d have to figure out something.

Good point! :D

That’s something that’s probably not been figured out yet. Probably won’t be figured out until popular stuff from the 50s and 60s starts hitting PD.

Honestly I think we should just work at doing away with the concept. The idea of it was to encourage the creation of works by providing a limited term of artificial scarcity so said works could participate in the market. But in the information age, where copying information is as close to free as makes no odds, I think that makes a lot less sense than it used to. That's without even considering that giant corporations have subverted this system and exploited many producers of works while profiting well past the deaths of the creators.

It is my opinion that we need an alternative mechanism for funding the creation of information-works that doesn't rely on artificial scarcity.

You have said it was originally to help out the market, and you have said copying information is now nearly free, but you forgot to say how or why the first part is incongruous with the second. Why does it make less sense than it used to? Which of our new capabilities could be leveraged by a new system?

You used to have to recoup the costs of publishing as well as fund the author. Binding books was and is not free, but copying bits effectively is.

You still didn't answer the question. Copyright was introduced because it was amazingly cheap to copy as opposed to create. If it's gotten even cheaper, that means it's more relevant, not less. So, again, how does the first part connect to the second? Explain fully why the current system isn't relevant, because there's clearly something obvious to you that isn't obvious to me.

What I'm saying is, the motivation to print pirated copies of books was to sell them to people for profit because as cheap as it was to copy books compared to writing them, it was still quite a bit above 0.

However, copying cost today is effectively 0. No one would pay money for a pirated copy of an ebook. One could (somewhat pedantically) say that this means the creator's 'exclusive right to profit' from their work is not violated because no one profits in this way from piracy[0], but that's not the point. the point is that the concept barely even made sense the first time around, but allowed a creator who didn't own a giant expensive distribution mechanism the leverage to negotiate compensation. Now the distribution is about as free as it is ever likely to get.

[0] or at least they wouldn't if it weren't for laws creating barriers that allow people to profit by providing mechanisms around said barriers.

But how would that begin to work in practice? Someone makes an album or a book and everyone an just copy it without any fee?

Yes. And why not? That I can do that so easily today even with copyright just goes to show how post-scarcity information is.

The problem, of course, is that the creators of the work still need to participate in a scarcity economy to live. There are examples of efforts being made to fund creators outside the traditional scarcity-based mechanism, such as patreon funding, github sponsorship, and the like, although I admit they appear to only currently be useful for those who would be unable to leverage the traditional system effectively anyway.

While I may not have the answer to funding creators, I firmly believe that we'd be going backwards to insist on forcing scarcity into a post-scarcity space.

How do you ensure the copy you buy isn't adulterated and modified? One benefit of copyright is that the author can ensure the "real" one is shared... which you may care about when viewing content.

https://www.nytimes.com/2019/08/19/technology/amazon-orwell-...

Trivially solved with hashing or signing. Copyright as a legal mechanism is unnecessary to ensure 'legitimacy'.

Copyright is useful for preventing others from using characters you created, but frankly I don't see the value in that. Moreover, it isn't even effective as fan fiction is incredibly common anyway. We can see how this would work with the proliferation of open source and the commonness of forks. It really isn't a big deal.

Copyrights are not only unnecessary for ensuring that the work you receive matches the original as copyright holders are free to make modifications. Worse, with the current copyright system the copyright holders can prevent the original from being distributed at all.

Then we would be well advised to listen to the creators of such work, who largely tend to describe the type of system you want to do away with.

I don't think that follows. If we listen to car manufacturers we should replace our cars every year. Creators who make a lot of money with the current system are incentivized to perpetuate it, and some of those who don't make much imagine that they someday might.

Honestly, I don't mind the idea that you the creator, personally, holds the copyright for life. If I write a song in my 20s that sets the world on fire and I can still live a good life at 80 off the back of it, great.

The minute I sell that copyright to someone else (or it's inherited), a time limit begins to apply. Perhaps here you move into the '20 years' or something category.

For popular works, tracking the life of the creator is doable. For less popular works, figuring out who the creator is can be rather difficult. And then figuring out if they're alive isn't always easy either.

The screen actors guild has somewhat of a highlander policy, but if I release something Copyright My Name, there's hundreds or thousands of people it could be, including a Pulitzer Prize winning author who will likely predecease me.

Thats what registration is for. We know the "true" authors/owners of most patents, so why not copyrights.

One of the rights that copyright protects is the right to make derivative works--such as sequels and film adaptations. If you consider that someone like Terry Pratchett can productively write a literary universe for decades, it feels to me like the best answer is around 30-50 years.

Isn't that trademark?

IOW, when Steamboat Willie's copyright expires, it doesn't give you the right to make a new Mickey Mouse film. Disney still has a trademark on Mickey.

No, trademark is entirely different IP. Copyright very explicitly protects the right to make derivative works.

(See 17 USC §106 (https://www.law.cornell.edu/uscode/text/17/106) for the full list of what copyright actually protects.)

But if your copyright has expired, you can still prevent derivative works via trademark. Disney has trademarked Mickey Mouse.

Copyright should be taxed every year based on the cumulative earnings for the first 10 years.

This allows the rights holders to milk the bulk of the profit and yet quickly pass the orphaned works to the public domain - both benefitting the development of arts and sciences as stipulated by the US constitution.

Heck, even taxing works for $1 a piece per year would free up a huge number of orphans.

Not OP, but I think it makes very little difference whatever number you pick between “none” and “forever”.

Earlier this year a few authors whose work I enjoy tweeted furiously about this topic. Similar arguments may apply to audio and visual copyright.

If, for example, copyright lasted 14 years, then The Atrocity Archives (Stross) would already be public domain, which reduces incentive to create more in the series (not sure why, I would have expected new novels have new copyright, but I trust an author to know what their own motivations are, and I refuse to second-guess).

Now, these various authors on Twitter argued that being an author is poorly paid (it is), and that they would like to offer someone an inheritance comparable to a widow/er pension.

I can sympathise with that.

But, sales in most things decline over time. Sequels more, apparently (I wouldn’t know, one of the authors tweeted that). If income falls off exponentially with time, half the income for some specific work will be in the first x-months, 75% in 2x months, 99.9% in 10x months — I’m saying months rather than years because of how fast best-seller lists change, I don’t have real data.

Conversely, the reason for such a decline is that almost all the people who care will have bought the work soon after it is published. Making copyright last 100(average of all x) fails to harm society by preventing the work from becoming public domain, but only by the same[2] as it fails to help creative people commercialise their creativity, and for exactly the same reason: almost nobody cares any more. This remains true even if the income curve isn’t exponential, as the benefit to society is simply not having to pay for it[3].

Certainly I’m not rushing out to download some copy of inter-war music, radio broadcasts of whatever category, or newspapers; and I don’t even know if that’s for lack of money because I don’t care. I suspect I’m not alone in not caring.

Indeed, I only oppose zero/infinite duration because of the edge cases.

The point of copyright is to give creators an incentive at all besides the default of creative people being patronised or commissioned. This is already a thing — furry art by commission; stories[0] and YouTubers supported on Patreon. As this is not the status quo, if we switched to “no copyright” I expect this will break stuff and bankrupt people who have come to rely on it (this also means I oppose sudden changes in general).

If copyright was infinite, all works world eventually become lost in a legal quagmire of e.g. mergers and acquisitions where the actual owner ceases to be known.

[0] https://www.patreon.com/HamboneHFY

[1] except where the income per unit is too small to process ($0.001 per person times US population is still a lot of money).

[2] or losing track of who owns it, or [1]

> Making copyright last 100(average of all x) fails to harm society by preventing the work from becoming public domain

That does not follow. Even if literally everyone who was interested in the work already has a copy, copyright prevents others from building on top of it. That is the harm.

I don’t see this being a huge cost/limit.

Yes, sure, it would be a copyright violation to write today a novel set on a giant space artefact containing a flattened full-scale map of Earth, where a motley crew (including a bald human on anti-aging drugs) crash-lands and investigates, finding the machinery of the world breaking down, where those things are “the Ringworld”, Luis Wu, superconducting wire destroyed by a synthetic microbe, and Boosterspice; but that didn’t stop Pratchett writing Strata with “the flat world”, Kin Arad, general entropy, and Days.

25 years