Just a moment... https://www.garbageday.email/p/the-only-taboo-left-is-copyright-infringement

it's deeply weird to be on the internet rn. people who were saying 'information wants to be free' 15 years ago are now like 'the problem with AI is not the environmental impact but the copyright infringement'

Random link from the archives: "Lorde, Olivia Rodrigo, and the Next Phase of Music Plagiarism Battles - The Ringer" https://web.archive.org/web/20220109142059/https://www.theringer.com/music/2021/9/9/22663301/music-plagiarism-copyright-infringement-lorde-olivia-rodrigo-blurred-lines

AI industry horrified to face largest copyright class action ever certified - Ars Technica https://arstechnica.com/tech-policy/2025/08/ai-industry-horrified-to-face-largest-copyright-class-action-ever-certified/

Random link from the archives: "Public Domain Film History: A Cottage Industry of Copyright Mishaps" https://web.archive.org/web/20171025124247/https://tedium.co/2017/10/24/public-domain-film-history-copyright/ originally retrieved Wed Oct 25 12:42:47 EDT 2017

love that the internet archive gets a coordinated suit right before an administration that really needs reporting on it permanently archived at a granular level. just like anti-LLM rhetoric was spun into pro-copyright rhetoric right after the steamboat willy expiry.

Random link from the archives: "If Creators Suing AI Companies Over Copyright Win, It Will Further Entrench Big Tech | Techdirt" https://web.archive.org/web/20240219084023/https://www.techdirt.com/2023/12/04/if-creators-suing-ai-companies-over-copyright-win-it-will-further-entrench-big-tech/

Random link from the archives: "Why Is AI Art Copyright So Complicated? — Artnome" https://web.archive.org/web/20190327145731/https://www.artnome.com/news/2019/3/27/why-is-ai-art-copyright-so-complicated originally retrieved Wed Mar 27 14:57:31 EDT 2019

If Creators Suing AI Companies Over Copyright Win, It Will Further Entrench Big Tech | Techdirt https://www.techdirt.com/2023/12/04/if-creators-suing-ai-companies-over-copyright-win-it-will-further-entrench-big-tech/

A Bunch Of Authors Sue OpenAI Claiming Copyright Infringement, Because They Don’t Understand Copyright | Techdirt https://www.techdirt.com/2023/07/11/a-bunch-of-authors-sue-openai-claiming-copyright-infringement-because-they-dont-understand-copyright/

being critical of LLMs but not because you want to strengthen copyright is a difficult situation, if you want to interact with outsiders. actually this happens to me a lot: i criticize liberals but from the left, i think UFOs are a genuine phenomenon worthy of study but i think disclosure is bullshit & so is the ETH, i like low budget exploitation movies but totally unironically. coming to your opinions from the ground up instead of copying the opinions of your friends & slowly justifying or adapting them leaves you with a bunch of nuanced positions that don't fit well with the pattern-matching NTs usually do when trying to slot you into some category & figure out how to interact with you. and it's not like i don't understand why that's valuable: i totally spend a bunch of effort trying to satisfy the boolean 'is this person a fascist' when meeting someone new because i don't want to spend time and effort listening to a fascist. the problem is that a lot of the time, these frames are literally put in place as a form of power consolidation: whether you support strengthening copyright or you support widespread use of LLMs, you're supporting exactly the same large businesses for the most part, and only a nuanced position that doesn't frame it in terms of complete automation of existing commercial roles is capable of producing arguments that aren't functionally just shilling for pearson; the framing that UFOs are hostile extraterrestrials or hallucinations is basically just a story the US air force made up because it allows them to pull a paul bennewitz on people on the one hand but also justify military funding on the other, just by changing how the media spins an existing policy (the official air force position on UFOs has not changed since the 1950s); ironic enjoyment promotes bad craft while justifying dismissing good craft & interesting experiments as inherently unserious, & promotes the idea that a 'real movie' is defined by its budget and polish; i don't think i need to explain to masto why an anarchocommunist would dunk on liberals.

The Atlantic https://www.theatlantic.com/technology/archive/2023/09/books3-database-generative-ai-training-copyright-infringement/675363/

i believe in abolishing all forms of IP law (and this is directly informed by my day job, writing the infrastructure that allows IP law to be executed at scale). if you are worried that abolishing IP law would screw over individual artists (though i would argue that small artists are already screwed over and IP law is mostly used against them, and in fact cannot typically be used to support them) then a reasonable half-measure is as follows: limit copyright terms to 7 years, and rule that copyright can only be held by a natural person, and cannot be transferred or inherited. this will prevent copyright from ever being used as a corporate cash-cow, and ensures that whatever benefits might go to the copyright holder always go to the artist instead. this doesn't address other forms of IP, but other forms of IP are really not available to regular people unless they're quite rich, whereas everybody you have ever met owns a billion copyrights in theory and doesn't know it.

an LLM is a plausible-sounding-bullshit machine because any purely-statistical generative system is a plausible-sounding-bullshit machine: it optimizes for least-surprise along the metrics it knows how to measure. it produces interesting output by breaking -- by failing to measure something important and therefore producing output that no human would produce, because no human could ignore whatever metric the model is ignoring. this is great for art: like random factors and constraints, an artist can use interaction with a statistical model to propel them out of a rut and onto a new path (as long as the model is sufficiently bad at predicting what a human would 'really' do -- i.e., so long as the training data in that area is sparse enough that the model's idea of the most-likely thing is absurd). it's horrible for anything where correctness is important: there is no negative feedback mechanism for factually incorrect or (in the case of code) non-functional output, only for output that is insufficiently boring from the perspective of event frequency correlations (where, in the case of an LLM or a markov chain chatbot, the event is a letter or word respectively). if you want reliably correct results, you need to be doing symbolic processing -- using a theorem prover, constraint solver, or expert system. this division between statistical and symbolic techniques in AI mirrors the system I / system II division in human psychology. with humans, it seems like system II is an extremely expensive epiphenomenon on top of system I -- that, through rigorous conditioning, humans are taught to sometimes simulate symbolic processing on statistical hardware. it's possible that computers could do this. but why would we want them to, when conventional computer hardware is much better at symbolic processing than statistical processing (and in fact, we are already doing it the other way around and implementing statistical modeling on symbolic hardware). (i say 'conventional computer hardware' because there are a couple dedicated neural net chips around, mostly built in the 90s; they are not heavily used, but if i didn't mention them somebody would probably complain.) IBM's 'Watson' was originally billed as something that could combine statistical and symbolic methods, but clearly, the statistical side has been given higher priority in ways that make it not just useless but actively dangerous for most of the tasks it is marketed for. the draw of statistical models for a corporation is clear: with an expert system, a human who is an expert in writing expert systems needs to interview domain experts (and all these people are typically paid very well for their time since there are few experienced expert system engineers and the domain experts are automating away some of their own future consulting contracts), whereas you just throw mountains of unstructured text at an LLM until it starts making coherent sentences (and there's a lot of unstructured text that's either in the public domain or where copyright claims are nearly unenforceable).

Librarians Are Finding Thousands Of Books No Longer Protected By Copyright Law https://www.vice.com/en/article/epzyde/librarians-are-finding-thousands-of-books-no-longer-protected-by-copyright-law

fair's fair: if AI- and monkey- generated art is not subject to copyright, then only natural persons should be allowed to own IP, and IP protections should dissolve entirely upon the death of the creator.

Copyright and Metropolis - Plagiarism Today https://www.plagiarismtoday.com/2016/10/19/copyright-and-metropolis/amp/

Random link from the archives: "A Mass of Copyrighted Works Will Soon Enter the Public Domain - The Atlantic" https://web.archive.org/web/20180430142206/https://www.theatlantic.com/technology/archive/2018/04/copywritten-so-dont-copy-me/557420/ originally retrieved Mon Apr 30 14:22:06 EDT 2018

The Legend of TUNIC: Obscurity, Nostalgia, and Copyright [CC] - YouTube https://www.youtube.com/watch?v=MTuD51hl1ec

Anyone can use this AI art generator " that’s the risk - The Verge https://www.theverge.com/2022/9/15/23340673/ai-image-generation-stable-diffusion-explained-ethics-copyright-data

U.S. Copyright Office Rules A.I. Art Can't Be Copyrighted | Smart News| https://www.smithsonianmag.com/smart-news/us-copyright-office-rules-ai-art-cant-be-copyrighted-180979808/

Random link from the archives: "You can’t copyright a cocktail, so what’s a creative bartender to do? | Ars Technica" https://web.archive.org/web/20190722091503/https://arstechnica.com/gaming/2019/07/rum-and-ginger-everywhere-how-intellectual-property-plays-out-at-the-bar/ originally retrieved Mon Jul 22 09:15:03 EDT 2019

because copyright terms are substantially longer than the average lifetime, the publication rights & possibly royalties for your writing will almost definitely outlive you. is it straightforward to, for instance, write your will such that your books fall into the public domain on your death? what about donating the rights to your books to a local public library?

forgive me if this is obvious but it just now struck me as salient for the first time: collage-based pop-art became a thing in the 50s and 60s because of mass-distribution high-resolution photo-mags like Life providing a wealth of material to draw from, and likewise, the 'postmodern' mixing of styles from different decades and genres in music videos and later films in the late 80s and through the 90s were probably enabled by VHS releases of older films & cable channels like Turner's re-running old movies (i.e., even though footage wasn't necessarily used & editing wasn't necessarily done on VHS, VHS re-releases created an awareness of a variety of styles and helped situate them, so that one could draw from a style that wasn't part of a current 'scene' and expect some of the audience to also have a context for it). in other words: technical advances in communication technologies democratized access to visual ideas from groups distant in space or time, preparing people's imaginations for both creating and interpreting new variations. both photo-mags and video tapes are time-binding technologies, and i think this might be specific to time-binding: long before VHS existed, films got re-released for new theatrical runs, and although that occasionally led to artistic inspiration when a particular film ended up being very popular in re-run (ex., Disney's Alice in Wonderland, originally not terribly popular, got re-released in the late 60s and marketed to heads, & 50s re-releases & TV re-runs of 30s Universal horror flicks created the 'monster kid' generation), these hyper-specific genre riffs mostly only happened when it was possible for large numbers of people to re-experience media in detail repeatedly at their own leisure -- especially to rewind, use slow motion, go frame by frame, and eventually to zoom. likewise, you don't get hip-hop without records and tapes (not just the equipment but the actual media -- if records decayed & became unusable in 10 years then hip hop would have looked very different because crate digging would not involve digging so far back in time) and you don't get jazz without cheat books (i.e., jazz has its origins as a popular form in a certain set of process-technologies -- notational technologies in addition to techniques of performance and concepts about improvization -- and those notational technologies have a material basis and economic-social constraint-incentive structures that influence them). videocollage existed before the 90s, because at least some people had film projectors, film splicing equipment, and access to large quantities of developed film, but until the 80s everybody in that category was a Film Person (and probably a Film Industry Person); videocollage only became A Thing when regular people had cheap access to a huge variety of film (and even then, the editing technology was only accessible to Film People), and videocollage will only become democratized when recorded video and the film editing tech is not just cheap but the technical skills widespread. video (and to some extent audio) has actually gone backwards in this respect for reasons of copyright protectionism: streaming is so common that younger people often don't have copies of any film or music (either on media like DVDs or pirated files), and as understanding for how to get access to video in a form that can be edited becomes an additional hurdle people have to cross before editing (on top of getting and learning to use editing software, which is generally quite hard to learn to use) the number of people from a new generation who bother doing any of it drops. IP holders see DRM and (to some degree) streaming as a win because they see piracy as lost sales, but this is actually only a short term win for them (if at all) because the most skilled people in any domain are folks who started young and got a lot of practice -- in other words, people who didn't get 100% of their training on-the-job but instead are former amateurs -- and those people are also the ones most likely to bring new and innovative ideas into a stagnating field. In other words, when streaming and DRM is promoted to the exclusion of media that can be easily edited, the next generation is starved of new talent. the same thing has happened to computing: in the 80s, lots of people learned BASIC without intending to ever become professional programmers because it was one of the things machines could do out of the box; in the 90s, lots of people learned HTML and javascript without ever intending to become professional web developers because it was one of the things a web browser could do out of the box; today nearly all development tools & development tool documentation is aimed at professionals or would-be professionals and put in places where only people who are interested in learning to code will ever come across it, with few opportunities (excel being a rare exception) where non-programmers can 'naturally' slide into doing a little programming in order to make their lives easier. creative industries that don't get fresh ideas stagnate and eventually crash as they hit the limit of what their intellectual monoculture can imagine. because we have misused computer technology for professional protectionism, we are headed for a creative-stagnation crash in software, film, and music within the next 15 years, unless we take some drastic measures (like putting programming & multi-channel editing into the curriculum of mandatory schooling)

Random link from the archives: "" https://web.archive.org/web/20181221105221/https://www.smithsonianmag.com/arts-culture/first-time-20-years-copyrighted-works-enter-public-domain-180971016/ originally retrieved Fri Dec 21 10:52:21 EST 2018

Lorde, Olivia Rodrigo, and the Next Phase of Music Plagiarism Battles - The Ringer https://www.theringer.com/music/2021/9/9/22663301/music-plagiarism-copyright-infringement-lorde-olivia-rodrigo-blurred-lines

Random link from the archives: "GitHub - NYPL/cce-renewals: Tab-delimited versions of Catalog of Copyright Entries renewals" https://web.archive.org/web/20190515084653/https://github.com/NYPL/cce-renewals/ originally retrieved Wed May 15 08:46:53 EDT 2019

Early attempts at implementing transcopyright (essentially, a system by which quote attribution is… https://scribe.rip/@enkiv2/early-attempts-at-implementing-transcopyright-essentially-a-system-by-which-quote-attribution-is-4b76d9edf081

KLF assert justified and ancient copyright claim to block documentary | The KLF | The Guardian https://www.theguardian.com/music/2021/oct/19/klf-assert-justified-and-ancient-copyright-claim-to-block-documentary

Horace Goes Copyright Striking | Boing Boing https://boingboing.net/2019/11/18/horace-goes-copyright-striking.html

Cops are playing music during filmed encounters to game YouTube's copyright striking https://mashable.com/article/police-playing-music-copyright-youtube-recording

Random link from the archives: "YouTubers and record labels are fighting, and record labels keep winning - The Verge" https://web.archive.org/web/20190528095412/https://www.theverge.com/2019/5/24/18635904/copyright-youtube-creators-dmca-takedown-fair-use-music-cover originally retrieved Tue May 28 09:54:12 EDT 2019

Random link from the archives: "" https://web.archive.org/web/20181221105221/https://www.smithsonianmag.com/arts-culture/first-time-20-years-copyrighted-works-enter-public-domain-180971016/ originally retrieved Fri Dec 21 10:52:21 EST 2018

Random link from the archives: "GitHub - NYPL/catalog_of_copyright_entries_project: NYPL Project to transcribe and parse pages from the US Catalog of Copyright Entries" https://web.archive.org/web/20190515084628/https://github.com/NYPL/catalog_of_copyright_entries_project originally retrieved Wed May 15 08:46:28 EDT 2019

Random link from the archives: "A Mass of Copyrighted Works Will Soon Enter the Public Domain - The Atlantic" https://web.archive.org/web/20180409101127/https://www.theatlantic.com/technology/archive/2018/04/copywritten-so-dont-copy-me/557420/ originally retrieved Mon Apr 9 10:11:27 EDT 2018

The Truth Cannot Be Copyrighted

A Twitch streamer was banned for not owning the rights to her own body - The Verge https://www.theverge.com/2020/11/24/21591488/projekt-melody-twitch-ban-copyright-strike-digitrevx-vtuber

replace copyright with UBI

Random link from the archives: "Part 1: Copyrightability of #RPG Stat Blocks #DnD #copyright #iplaw – Frylock's Gaming & Geekery" https://web.archive.org/web/20200120100553/https://gsllc.wordpress.com/2019/08/12/part1statblocks/ originally posted Mon Jan 20 10:05:53 EST 2020

GitHub - cdanis/sandia-public-license: This is not a license of honor. No highly esteemed copyright statement is written here. https://github.com/cdanis/sandia-public-license

Random link from the archives: "" https://web.archive.org/web/20181221105221/https://www.smithsonianmag.com/arts-culture/first-time-20-years-copyrighted-works-enter-public-domain-180971016/ originally posted Fri Dec 21 10:52:21 EST 2018

Random link from the archives: https://web.archive.org/web/20171025124247/https://tedium.co/2017/10/24/public-domain-film-history-copyright/ originally posted Wed Oct 25 12:42:47 EDT 2017/

Random link from the archives: https://web.archive.org/web/20190826101930/https://medium.com/@enkiv2/early-attempts-at-implementing-transcopyright-essentially-a-system-by-which-quote-attribution-is-4b76d9edf081 originally posted Mon Aug 26 10:19:30 EDT 2019/

Random link from the archives: https://web.archive.org/web/20181221105221/https://www.smithsonianmag.com/arts-culture/first-time-20-years-copyrighted-works-enter-public-domain-180971016/ originally posted Fri Dec 21 10:52:21 EST 2018/

It occurs to me that, because (xanadu-space era) transcopyright uses one time pads, it's possible to rubberhose it with different OTPs in order to create alternative plaintexts so long as they are of the same length & all subsections still make sense in their link/transclusion contexts...

Part 1: Copyrightability of #RPG Stat Blocks #DnD #copyright #iplaw – Frylock's Gaming & Geekery https://gsllc.wordpress.com/2019/08/12/part1statblocks/

Random link from the archives: https://web.archive.org/web/20171102145144/https://motherboard.vice.com/en_us/article/59ydmx/copyright-law-artificial-intelligence-bias originally posted Thu Nov 2 14:51:44 EDT 2017/

https://medium.com/@enkiv2/early-attempts-at-implementing-transcopyright-essentially-a-system-by-which-quote-attribution-is-4b76d9edf081

Random link from the archives: https://web.archive.org/web/20190327145731/https://www.artnome.com/news/2019/3/27/why-is-ai-art-copyright-so-complicated originally posted Wed Mar 27 14:57:31 EDT 2019/

You can’t copyright a cocktail, so what’s a creative bartender to do? | Ars Technica https://arstechnica.com/gaming/2019/07/rum-and-ginger-everywhere-how-intellectual-property-plays-out-at-the-bar/

YouTubers and record labels are fighting, and record labels keep winning - The Verge https://www.theverge.com/2019/5/24/18635904/copyright-youtube-creators-dmca-takedown-fair-use-music-cover

GitHub - NYPL/cce-renewals: Tab-delimited versions of Catalog of Copyright Entries renewals https://github.com/NYPL/cce-renewals/

GitHub - NYPL/catalog_of_copyright_entries_project: NYPL Project to transcribe and parse pages from the US Catalog of Copyright Entries https://github.com/NYPL/catalog_of_copyright_entries_project

Random link from the archives: https://tedium.co/2017/10/24/public-domain-film-history-copyright/ originally archived Thu Oct 26 08:16:35 EDT 2017

Why Is AI Art Copyright So Complicated? — Artnome https://www.artnome.com/news/2019/3/27/why-is-ai-art-copyright-so-complicated

'Kingdom Hearts III' is a Big, Beautiful Mess - Waypoint https://waypoint.vice.com/en_us/article/gyaqwm/kingdom-hearts-iii-review-lore-copyright

Tedium Trends 2019: Copyright Law is Shifting https://tedium.co/2018/12/27/tedium-trends-2019/

The Quietus | Opinion | The Quietus Essay | Rude Citizenship: Jamaican Musical Challenges to Copyright(ed) Culture http://thequietus.com/articles/25494-music-copyright-jamaica-dub

Translit is a little like bittorrent, if bittorrent used literally the same bytes for AMVs and trailers as for the corresponding scenes in the show/movie. Transcopyright is like assembling AMVs from your own DVD boxed set based on a list of instructions & leaving a gap if you're missing a disk.

Come to think of it, gluing some unrelated poem to some unrelated tune was all over the place before the 20th century, wasn't it? Ode to Joy, Jerusalem... Maybe we just stopped because of tin pan alley copyright expansion?

Actually, if I did a series of articles on (non-secret parts of) Xanadu internals, would anybody care? Not just transcopyright -- I've got detailed knowledge on how ZZOGL / FloatingWord did layout, for instance. I think I can explain these things in a clearer way than Ted does, in some cases.

I should probably implement an example version of transcopyright for my repo of 'important data structures, algorithms, & ideas that have no easily-understood isolated example implementations' (https://github.com/enkiv2/misc/tree/master/ds-lib) since it's already mostly xanadu shit

A Mass of Copyrighted Works Will Soon Enter the Public Domain - The Atlantic https://www.theatlantic.com/technology/archive/2018/04/copywritten-so-dont-copy-me/557420/

Is English Wikipedia’s ‘rise and decline’ typical? – copyrighteous https://mako.cc/copyrighteous/revisiting-the-rise-and-decline

A Mass of Copyrighted Works Will Soon Enter the Public Domain - The Atlantic https://www.theatlantic.com/technology/archive/2018/04/copywritten-so-dont-copy-me/557420/

Bad idea of the day: require a payment of a dollar (held in escrow by the copyright office) to send a DMCA takedown notice. The dollar goes to whoever is determined to be the legitimate owner, in case of a dispute. If there is no dispute, it goes into the federal education budget.

Copyright Law Makes Artificial Intelligence Bias Worse - Motherboard https://motherboard.vice.com/en_us/article/59ydmx/copyright-law-artificial-intelligence-bias

Public Domain Film History: A Cottage Industry of Copyright Mishaps https://tedium.co/2017/10/24/public-domain-film-history-copyright/

Public Domain Film History: A Cottage Industry of Copyright Mishaps https://tedium.co/2017/10/24/public-domain-film-history-copyright/

How Apple’s Early Clone Wars Reshaped Copyright Law - Motherboard https://motherboard.vice.com/en_us/article/59za4x/how-apples-early-clone-wars-reshaped-copyright-law

The Wikipedia Adventure – copyrighteous https://mako.cc/copyrighteous/the-wikipedia-adventure

Do Androids Dream of Electric Copyright? Comparative Analysis of Originality in Artificial Intelligence Generated Works by Andrés Guadamuz :: SSRN https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2981304