> Which leads us to the question of why it should indeed ever expire?

It's the 21st century, nothing stops us from simply copying works with absolute impunity. The question is why even recognize copyright as legitimate to begin with?

Society is doing creators a favor by pretending their works are artificially scarce. We allow creators to profit from their works. Public domain is actually how things are naturally: after data is created, it can be copied infinitely at negligible costs. We enable the artificial scarcity that allows creators to profit from their works.

There has always been an understanding that eventually the works will return to the public domain. Creators are supposed to make money for a while but eventually the party is supposed to end. We're not here to enable their eternal rent seeking. If they want more money, they gotta make new stuff.

Every time copyright duration is extended, it's a slap in the fact of the public. We have been blatantly robbed of our public domain rights. No one's ever been bold enough to suggest straight up eliminating our rights though. At some point copyright infringement becomes civil disobedience and a moral imperative.

Limited time because it is in the Constitution. But also it allowed others to build on your creation. Sherlock Holmes is a wonderful example. Most of Holmes is in public domain. And recently the has been a rich at of culture created. Movies, television series, additional books. Personally I think the main thing that is broken is the the expiration date. The Constitution says "finite" and while there is a finite limit any limit that is longer than the average lifespan is essentially unlimited. Basically nothing created in your lifetime will be in public domain. And to me that goes against the spirit of "finite"

I agree, the time limit seems arbitrary. And sure there are a handful of works that are famous enough to allow for the creation of new, obviously derivative works.

Not that we are exactly short of original movie and TV content - but now we can have a few more derivative sherlock homes, so yay...

So if finite means a time limit, then isn't any limit arbitrary? Should an author live to see their work exploited in their own lifetime? If we make it say 40 years, how will RMS feel when oracle starts deriving Gnu Hurd code to make a new OS?

The first release of GNU Hurd was in 1990. So Oracle could use that release to build from in 2030. Do you think Stallman's copyright on the 1990 release of GNU Hurd is still giving him the fuel he needs to keep creating?

From https://copyrightalliance.org/education/copyright-law-explai...:

> The primary purpose of copyright is to induce and reward authors, through the provision of property rights, to create new works and to make those works available to the public to enjoy.

I think knowing copyright ends in 40 years would do more to incentivize new work than the current set of rules. If a copyright is held by a corporation, the duration should probably be half of that. Having copyright last beyond a creator's life makes no sense to me at all because nothing new is created by that person.

> So back to copyright - you made it, you control it, the only question seems to be - for how long? Which leads us to the question of why it should indeed ever expire? Patents have a public good when they expire - but what public good happens when copyrights expire seems less clear to me...

A good question indeed! And, fortunately, one history is well-equipped to answer. You might find the history of the Statute of Anne interesting - https://en.wikipedia.org/wiki/Statute_of_Anne In short, perpetual copyright means that publishing is essentially governed by private laws rather than public ones.

Some, particularly Americans, may also find it relevant that non-expiring copyright is quite clearly forbidden by the constitutional clause that enables copyright to exist.

I'm not sure the linked article explains the public good in copyrights expiring - perhaps that the public had access to topical works improving their lives? Would that be true now? Even allowing for the fact that many copyrighted works are currently distributed freely?

Maybe I could phrase the question this way - after how much time should, um, Microsoft be allowed to take a GPL licensed work, and use it in their own closed code? 10 years? 20? 50? Never?

Some would contend that preventing perpetual private censorship improves the lives of the public.

Personally, I would be very impressed with any code still useful for incorporation into current things after 20 years. Certainly worth unshackling the original work from copyright.

Prepare to be impressed!

All the following are still actively using code that I'd more than 20 years old;

Gnu user tools, as used in Gnu Linux. Linux kernel. Pretty much all databases. (sqlite for example is 21 years old). Much of Windows, and Ms office.

Not to mention mountains of my own code, still in production. 20 years is nothing in coding terms.

Most developrs I know are small or single, and have one or two products they've been working on for over 20 years.

Linux 2.4 was pretty good, but even most of my IoT junk runs 2.6 which isn't quite 20 years old.

MySQL 4.0 was a lot better than 3.23 IIRC, and isn't quite 20 years old either.

IMHO, it's not so bad for the old version to become unrestricted; more recent changes can still be restricted. And 20 years of updates are often compelling enough to submit to restrictions.

What number seems right to you though? 95 years for work for hire or death of author + 70 (which is gonna be fun to track down for open source works with many author) seems rather long to me.

the parent posted,

>> I would be very impressed with any code

And I would argue that _lots_ of the code in today's shipping products (which I listed, and others) use lots of code laid down more than 20 years ago.

Sure there's lots more in Linux now, but lots of the code is also basically unchanged from 20 years ago. It's not like GREP gets a complete rewrite every 5 years...

I'm not saying that I'd be up for using a Linux install from 20 years ago, but hey "any code" is a lot more than "complete install".

> public good happens when copyrights expire seems less clear to me

People can build upon the original work to make it better or use parts of it in their own work, instead of making things from scratch (things like music remixes, free translation, restoration of obsolete software etc.)

I hear this argument, but eh. You want to write a song, go write on, you don't need to mix someone else's. And if you do, go pay them something instead of just freeloading.

Translations from one language to another I get, but again, if you want to make money translating books, go talk to the author and give them a share. Or write your own book in chosen language.

Unless you have the source code to obsolete abandoned software you can't build on it anyway. And if you do, and it is abandoned, then who's to stop you building on it now?

And again, what _public good_ comes from any of this? If you want to create, then create, don't remix..

What do you define as a public good? All of these can be done with lower friction when copyright expires. Is that not a public good? Maybe it's a minor public good in your eyes that is outweighed by the benefits to the individual, but it certainly qualifies as something advantageous for the public.

> and it is abandoned, then who's to stop you building on it now?

A copyright troll who just sits on copyrights without doing anything about them. Again maybe you argue that it's their right to do so, which is fine and is argument that perhaps the hit to an individual good is not worth the public good provided, but it's clearly also a public good, however small, to remove that copyright troll.

(Also in the particular case of music there is a very finite cap on the number of copywritable musical phrases, since there are comparatively very few musical phrases that are compatible with Western musical theory and aren't just noise, we're already seeing lawsuits over this. With an infinite copyright system it's very possible that effectively all new music would have to pay royalties within a few decades)

EDIT: To put it another way, freeloading is basically just another way of saying "public good" (a library is allowing patrons to freeload off publishers, a public park is freeloading off property, etc.). It's just one that implicitly prioritizes individual good as well (which is totally fine, the two are often in tension and must be balanced against each other).

First time I've heard of copyright trolls, so it's worth unpacking that..

If you knowingly reuse a work, then I'd argue they are not a troll, they are a copyright holder, so go figure out a license fee.

What about unknowingly using something they claim? That's more subtle, and I buy the argument there. Copyright expiry would help to avoid nonsense claims. But maybe sane courts can adjudicate that - then again Sane Courts (laughing).

But aiming overall copyright expiry to limit nonsense claims, would also prejudice real works that need real protection. So do we lose more than we gain?

So assuming expiring is good - lifetime of author plus a bit?

> You want to write a song, go write on, you don't need to mix someone else's.

Why I don't generally like bringing up fiction as proof of my words, Melancholy Elephants story has a very clear and explicit answer why this is a bad idea: http://www.spiderrobinson.com/melancholyelephants.html

On top of the examples already in the story: Disney's empire is built directly on the back of directly using public domain works or remixing existing works. And yet, you want to prevent anyone from doing just that.