Ted Chiang wrote about: that ChatGPT is a “blurry JPEG of all the text on the Web.” But the paper includes the math that proves the claim.

What this means is that text from before last year—text that is known human-generated—will become increasingly valuable. //

Tatütata • July 5, 2023 8:47 AM

What this means is that text from before last year—text that is known human-generated—will become increasingly valuable.

A bit like steel smelted before 8 August 1945… //

Tatütata • July 5, 2023 8:54 AM

The tails of the original content distribution disappear. Within a few generations, text becomes garbage, as Gaussian distributions converge and may even become delta functions. We call this effect model collapse.

Academia just discovered GIGO and the telephone game. Alleluia!

Just as we’ve strewn the oceans with plastic trash and filled the atmosphere with carbon dioxide,

and low-orbit space with débris.

so we’re about to fill the Internet with blah.

Isn’t it already? I just made my daily contribution. //

Winter • July 5, 2023 9:04 AM

I see a very lucrative market appearing for (high school) students working part-time as “real” human text producers. //

NC • July 5, 2023 9:27 AM

Hah, normal people don’t get paid! If a big tech company decides they want highschooler’s essays, they’ll just have Pierson or a pierson-alike company make essay-writing a part of the homework program they distribute with their textbooks, and thousands of teachers will require hundreds of thousands of students to submit millions of hours of work for free. For which Pierson might make a few bucks. //

Winter • July 5, 2023 9:49 AM

@NC

Hah, normal people don’t get paid!

Damn, my scheme is already torpedoed by those pesky capitalists.

But the matter is not really solved yet:
Who Owns Student Work?
https://designobserver.com/feature/who-owns-student-work/12667/

I know local Universities claim copyright to student’s works by way of some overarching educational contract (this is EU). I am not sure whether that has ever been tested in court. But I have never heard of schools being allowed to sell student work without getting the student involved. //

Valve boss says service, not price, the big issue for consumers and publishers.

Here's Valve's co-founder and CEO Gabe Newell talking about piracy in the games industry. The man behind online retail service Steam tackles those who claim that lower prices for games will make piracy go away, and he takes on companies that seek to fight piracy by inconveniencing consumers. //

"We think there is a fundamental misconception about piracy," Newell said. "Piracy is almost always a service problem and not a pricing problem. For example, if a pirate offers a product anywhere in the world, 24/7, purchasable from the convenience of your personal computer, and the legal provider says the product is region-locked, will come to your country three months after the U.S. release and can only be purchased at a brick and mortar store, then the pirate's service is more valuable.

"Most DRM solutions diminish the value of the product by either directly restricting a customer's use or by creating uncertainty."

He adds, "Our goal is to create greater service value than pirates, and this has been successful enough for us that piracy is basically a non-issue for our company. For example, prior to entering the Russian market, we were told that Russia was a waste of time because everyone would pirate our products. Russia is now about to become our largest market in Europe.

Kahle in a statement denounced the publishers' demands. "Here’s what’s at stake in this case: hundreds of libraries contributed millions of books to the Internet Archive for preservation in addition to those books we have purchased," he said.

"Thousands of donors provided the funds to digitize them.

"The publishers are now demanding that those millions of digitized books, not only be made inaccessible, but be destroyed. This is horrendous. Let me say it again – the publishers are demanding that millions of digitized books be destroyed. //

"And if they succeed in destroying our books or even making many of them inaccessible, there will be a chilling effect on the hundreds of other libraries that lend digitized books as we do." //

During the hearing, Judge Koeltl probed the arguments made by McNamara and IA's attorney, Joseph Gratz. He raised the defendant's point that there's no evidence of financial harm because there's no evidence the defendants would have paid to license electronic versions of their physical books.

McNamara responded that the harm is real. The ebook market is real and IA is just refusing to participate, she argued.

Gratz maintains that libraries have a right to lend a physical book they have purchased and that they have a right under the fair use exception to copyright law to facilitate digital lending so long as it's one copy per book.

On Tuesday, the US Copyright Office declared that images created using the AI-powered Midjourney image generator for the comic book Zarya of the Dawn should not have been granted copyright protection, and the images' copyright protection will be revoked.

In a letter addressed to the attorney of author Kris Kashtanova obtained by Ars Technica, the office cites "incomplete information" in the original copyright registration as the reason it plans to cancel the original registration and issue a new one excluding protection for the AI-generated images. Instead, the new registration will cover only the text of the work and the arrangement of images and text. Originally, Kashtanova did not disclose that the images were created by an AI model. //

Based on the record before it, the Office concludes that the images generated by Midjourney contained within the Work are not original works of authorship protected by copyright. See COMPENDIUM (THIRD ) § 313.2 (explaining that “the Office will not register works produced by a machine or mere mechanical process that operates randomly or automatically without any creative input or intervention from a human author”). Though she claims to have “guided” the structure and content of each image, the process described in the Kashtanova Letter makes clear that it was Midjourney—not Kashtanova—that originated the “traditional elements of authorship” in the images. //

Despite precedents for earlier algorithmically generated artwork receiving copyright protection, this ruling means that AI-generated imagery, without human-authored elements, cannot currently be copyrighted in the United States. The Copyright Office's ruling on the matter will likely hold unless it's challenged in court, revised by law, or re-examined in the future.

https://twitter.com/Dorialexander/status/1566489664961347589?s=20&t=C-1gw5GAR6GWccj9h9z_Fg //

IncorrigibleTroll Ars Praefectus
6y
3,626
Subscriptor
Somebody should let ChatGPT know. I was asking it about copyright recently, and it was quite insistent that it is a mere tool and copyright of its output would be automatically assigned to the operator. I asked it if this was its conjectured interpretation or settled case law, and it got offended that I might doubt it.

Identical fork is restored after DMCA counterclaim. //

adespoton Ars Tribunus Angusticlavius
9y

In general, reverse-engineering source code from a compiled binary is less straightforwardly illegal than simply cracking a game's DRM for piracy purposes, for instance. As the EFF explains, US case law includes certain fair use exceptions that can allow for this kind of decompilation work for research or interoperability purposes.

In the case of Grand Theft Auto, though, the game's End User License Agreement specifically asks players to agree not to "reverse engineer, decompile, disassemble, prepare derivative works based on or otherwise modify the Software, in whole or in part." Back in 2005, the 8th Circuit Court of Appeals upheld a similar anti-reverse-engineering EULA to take down BnetD, a reverse-engineered version of Blizzard's Battle.net that allowed the service to be emulated on private servers.

That's not quite how US Copyright works. If something's fair use, it doesn't matter what the EULA says, as Fair Use means copyright doesn't hold sway, which means any agreements based on copyright claims are meaningless.

The big difference here is that bnetd was used to create competition for the official battle.net server, so it didn't pass the Fair Use test. If it were released on Github today, there'd be no issue, as battle.net is no longer hosted by its creator, so there IS a fair use claim in play today.

And before people argue that copyright doesn't care about intent and competition... copyright of course does not; but Fair Use, which precludes copyright, does.

So while Take Two could argue that this game engine release that's binary compatible with their assets competes with current offerings of other games, Fair Use argues that Take Two no longer supports their game on modern hardware, so this third party code steps into that gap to enable assets to be used in the way intended at time of original purchase.

So at the end of the day, this isn't really any different than Right to Repair regarding John Deere and Apple: it's repairing an existing (and purchased) product so it can still be used. Take Two can't (successfully) argue that since they make a new tractor they want people to buy, they can stop people from upgrading the last tractor they sold them.

Mods tell SCOTUS that Reddit's special formula depends on Section 230 immunity.

On November 4th, a class action lawsuit — Doe 1 v. GitHub Inc., N.D. Cal., No. 3:22-cv-06823, 11/3/22 — was filed in the US District Court in the Northern District in California, alleging against Microsoft and GitHub (a Microsoft subsidiary), inter alia: violation of the DMCA; breach of contract; tortious interference in a contractual relationship; unjust enrichment; unfair competition; violation of California Consumer Privacy Act; and negligence. Also sued were a confusing mishmash of for profit and non-profit related entities all using a variation of the name OpenAI (OpenAI, Inc., OpenAI, LLC, OpenAI Startup Fund GP I, L.L.C.; you get the picture). OpenAI received one billion dollars in funding from Microsoft although they seem “officially unrelated.” //

Plaintiffs allege that OpenAI and GitHub assembled and distributed a commercial product called Copilot to create generative code using publicly accessible code originally made available under various “open source”-style licenses, many of which include an attribution requirement. As GitHub states, “…[t]rained on billions of lines of code, GitHub Copilot turns natural language prompts into coding suggestions across dozens of languages.” The resulting product allegedly omitted any credit to the original creators. //

As a final note, the complaint alleges a violation under the Digital Millennium Copyright Act for removal of copyright notices, attribution, and license terms, but conspicuously does not allege copyright infringement. A material breach of a copyright license can give rise to an infringement claim, so this is an interesting move. While the plaintiffs’ attorney indicated that an infringement claim might be added later, I suspect that this was done to avoid a messy fair use dispute. The complaint includes a statement by GitHub asserting an expansive, almost global fair use assertion which is at odds with explicit relevant law in many countries and frankly at odds even with US law. Nonetheless, fair use as a defense is expensive and complicated to litigate, so perhaps they chose to focus on something that is beyond factual dispute, and still provides the same damages.

Stoltz told Ars that EFF considers copyright trolling a major issue, especially since "the opening of the 'Copyright Claims Board,' a quasi-court for copyright infringement claims run out of the US Copyright Office in Washington DC." Stoltz told Ars that this board "created a new, more streamlined forum for copyright trolling. So the demise of Malibu Media doesn’t mean the end of the troll problem."

Copyleft Through Copyright

The primary goal of every GPL enforcement action is to gain compliance, which means getting to users complete and corresponding source code so they can copy, share, modify and install improved versions. The GPL itself is a copyright license that does a weird hack on copyright: it uses the copyright rules to turn them around, and require people to share software freely (as in freedom) in exchange for permission to copy, modify and distribute the software. A GPL violation occurs when someone fails to meet the license requirements and thereby infringes copyright. The copyright rules themselves then are the only remedy to enforce the license — requiring that the violator come into compliance with the license if they want permission to continue distribution.

Up until now, almost all the enforcement I've done has been purely under GPL version 2 (GPLv2). GPLv2§4 says that upon violation, the violator loses permission to engage in those activities governed by copyright: including copying, modifying and distributing the software. The only way to get those permissions back is for the copyright holder to grant them back.
Speaking For the Users

Copyleft's unique way of using copyright means the parties who may enforce are copyright holders (and their designated agents). However, the victims of the violation are typically thousands of users who have bought a product that included the GPL'd program. The goal, therefore, is to get source code that these users can actually use to compile and install the software. In GPLv2-speak, the goal is to get the all the "complete source code", which includes "the scripts used to control compilation and installation of the executable".

Cory Doctorow's book, Radicalized, is up for a CBC award. To celebrate, here's an excerpt.
by Cory Doctorow - Jan 22, 2020 9:05am EST //

entropy_winsArs Scholae Palatinaeet Subscriptorreply3 years agoreportignore user
RickRoyLeonPrisZhoraRachael wrote:
I'm not Canadian and I knew what CBC was! :)

Somehow, the movie, Brazil, lends more hindsight in having a character that DeNiro played, showing up to perform an unauthorized repair on the furnace. Eluding the IoT Enforcement Squad...

+1 for the Gilliam movie reference.

We live in a world that has become the embodiment of the imagined absurdity of the past... //

MisterManoArs Praetorianreply3 years agoreportignore user
To think there are many, many powers pushing so we reach exactly that kind of dystopic IoT crapshow because biznizz and money. A great read that leaves me even more worried about the future. //

CuriouslySaneArs Praefectuset Subscriptorreply3 years agoReader Favreportignore user
AusPeter wrote:
I saw the headline and thought "Hmm .. yep jailbreaking IoT devices is a thing now". Then I realized that it was a fictional story.

Cory's writing has a distressing habit of not staying fictional. //

I can't think of anything else with the acronym "CBC".

complete blood count

(Does not apply to Canadians as their 'blood' is actually a mixture of maple syrup and double-double)

SJW Legal Scholars, Funded by Google, Seek to Eviscerate Copyright Protections Through 'Restatement' of Law
By Jennifer Van Laar | Jan 17, 2022 9:30 PM ET

(Stefan Rousseau/PA via AP)
We know that the progressive left engages in Astroturfing. What many are just realizing or admitting is that what you see on the surface when an issue suddenly becomes hot is just that, surface. There are years of effort behind what you see, and it’s not just in organization. Astroturfing even consists of creating the scientific and legal “research” pointing to the desired conclusion. We know that the oil and tobacco industries did this decades ago, but the practice wasn’t limited to that time and those industries. More recently, we’ve seen this happen with regard to gun control (Everytown and Bloomberg-funded “research), criminal justice (Soros-affiliated groups and think tanks fund research that is then used by District Attorney candidates to claim the “science” backs their “reimagining” efforts), and, of course, Critical Race Theory.

Now a massive astroturfing effort is happening in the field of copyright law, and anyone who produces creative content should be paying close attention. As I wrote last week, there are major problems with the American Law Institute’s Restatement project, starting with how the project was initiated (at the request of a Google-funded, anti-copyright law activist professor) and the Lead Reporter’s unacknowledged conflicts of interest. Major flaws in the draft Restatement have been pointed out by other prominent copyright law scholars, a bipartisan group from Congress, the Register of Copyrights (the director of the Copyright Office), and even Advisors to the project. All of these groups essentially claim that the Restatement incorrectly interprets the Copyright Act and the policy preferences of the Reporters are improperly included. A letter to ALI from 10 of the project’s advisors states:

The Draft has a variety of problems including inaccurately summarizing the text of the Copyright Act, confusing explanations of important principles and misleading illustrations. Moreover, as further detailed below, these problems uniformly reflect an unduly restrictive view of copyright law that is not consistent with the robust statutory rights granted by Congress to authors.

Is the problem that the Restatement draft is flawed, or is it that “progressive” forces and their capitalist enablers, such as Google and Spotify, are using the Restatement to create a new legal “gold standard” that their attorneys can use in court to get favorable rulings and set new precedent in the arena – effectively changing the law through judicial activism and not through the legislative branch?

A “work made for hire” is—

(1) a work prepared by an employee within the scope of his or her employment; or

(2) a work specially ordered or commissioned for use as a contribution to a collective work, as a part of a motion picture or other audiovisual work, as a translation, as a supplementary work, as a compilation, as an instructional text, as a test, as answer material for a test, or as an atlas, if the parties expressly agree in a written instrument signed by them that the work shall be considered a work made for hire.

TL;DR: Technically, the SO and YouTube licenses are (probably) incompatible, so if you want to use copyrightable elements from SO in a YouTube video, you'll have to get permission from the original author. Certainly it never hurts to ask, even if you might not strictly need it.

People have always explored and modified the technologies in their lives, whether crystal radios, automobiles, or computer software. Reverse engineering is one expression of this tinkering impulse. Unfortunately, legal regulation of reverse engineering can impact the Freedom to Tinker in a variety of ways. This FAQ gives some information that may help coders reduce their legal risk.

in order to use the App Store, you have to agree to the iTunes Store Terms of Service and/or the App Store Terms of Service. You can confirm this yourself just by reading the documents: they say as much in their all-caps preambles. The two documents are pretty similar; this post will give section numbers from the App Store Terms of Service, but the same language appears in the iTunes Store Terms of Service and so our analysis applies identically to it. You can read both those documents on Apple's site, and we have a copy of that page as it exists today to provide this commentary.

Along the same lines, we'll be talking about GPLv2 specifically in this blog post, since that's the license at issue, but this analysis would apply to all versions of the GNU GPL and AGPL. Section 6 of GPLv2 says:

Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein.

(Emphasis added.) This last sentence is a crucial part of the strong copyleft in the GPL and AGPL: it prevents distributors from using separate legal agreements, like Terms of Service or NDAs, to take away the freedoms that the license is supposed to grant. This is the license condition that Apple is violating when it distributes GPL-covered software through the App Store. //

That's the problem in a nutshell: Apple's Terms of Service impose restrictive limits on use and distribution for any software distributed through the App Store, and the GPL doesn't allow that. This specific case involves other issues, but this is the one that's most unique and deserves explanation.

We would've liked to see Apple do the right thing and remove these limits, but it looks like that's not going to happen. Apple has removed GNU Go from the App Store, continuing their longstanding habit of preventing users from doing anything that Apple doesn't want them to do. As we said in our initial announcement, this is disappointing but unsurprising; Apple made this choice a long time ago. We just need to make sure everybody else gets the message: if you value your independence and creativity, you should be aware that Apple doesn't. Take your computing elsewhere.

Nonfree (proprietary) software is very often malware (designed to mistreat the user). Nonfree software is controlled by its developers, which puts them in a position of power over the users; that is the basic injustice. The developers and manufacturers often exercise that power to the detriment of the users they ought to serve.

This typically takes the form of malicious functionalities.

The “jails” are malicious operating systems that are designed to impose censorship of which applications the user can install. The image of the iPrison illustrates this issue.

These systems are platforms for censorship imposed by the company that owns the system. Selling products designed as platforms for a company to impose censorship ought to be forbidden by law, but it isn't.

This page lists a few jails, along with some of the methods they use to censor apps, and includes specific examples of apps that were blocked using this censorship power.

Guiding Principles in Community-Oriented GPL Enforcement

• Our primary goal in GPL enforcement is to bring about GPL compliance.
• Legal action is a last resort. Compliance actions are primarily education and assistance processes to aid those who are not following the license.
• Confidentiality can increase receptiveness and responsiveness.
• Community-oriented enforcement must never prioritize financial gain.
• Community-oriented compliance work does not request nor accept payment to overlook problems.
• Community-oriented compliance work starts with carefully verifying violations and finishes only after a comprehensive analysis.
• Community-oriented compliance processes should extend the benefit of GPLv3-like termination, even for GPLv2-only works.

Copyleft itself is not a moral philosophy; rather, copyleft is a strategy that software freedom activists constructed to advance a particular set of policy goals. Specifically, software copyleft was designed to ensure that all users received complete, corresponding source for all binaries, and that any modifications or improvements made anywhere in the chain of custody of the software were available in source form to downstream users. As orginially postulated, copyleft was a simple strategy to disarm proprietarization as an anti-software-freedom tactic.
The Corruption of Copyleft

Copyleft is a tool to achieve software freedom. Any tool can be fashioned into a weapon when wielded the wrong way. That's precisely what occurred with copyleft — and it happened early in copyleft's history, too. Before even the release of GPLv2, Aladdin Ghostscript used a copyleft via a proprietary relicensing model (which is sometimes confusingly called the “dual licensing” model). This business model initially presented as benign to software freedom activists; leaders declared the business model “barely legitimate”, when it rose to popularity through MySQL AB (later Sun, and later Oracle)'s proprietary relicensing of the MySQL codebase.

In theory, proprietary relicensors would only offer the proprietary license by popular demand to those who had some specific reason for wanting to proprietarize the codebase — a process that has been called “selling exceptions”. In practice, however, every company I'm aware of that sought to engage in “selling exceptions” eventually found a more aggressive and lucrative tack. //

Most proprietary relicensing businesses work as follows: a single codebase is produced by a for-profit company, which retains 100% control over all copyright in the software (either via an ©AA or a CLA). That codebase is offered as a gratis product to the marketplace, and the company invests substantial resources in marketing the software to users looking for FOSS solutions. The marketing department then engages in captious and unprincipled copyleft enforcement actions in an effort to “convert” those FOSS users into paying customers for proprietary licensing for the same codebase. (Occasionally, the company also offers additional proprietary add-ons, improvements, or security updates that are not available under the FOSS license — when used this way, the model is often specifically called “Open Core”.)

Why We Must End The Proprietary Relicensing Exploitation of Copyleft

This business model has a toxic effect on copyleft at every level. Users don't enjoy their software freedom under an assurance that a large community of contributors and users have all been bound to each other under the same, strong, and freedom-ensuring license. Instead, they dread the vendor finding a minor copyleft violation and blowing it out of proportion. The vendor offers no remedy (such as repairing the violation and promise of ongoing compliance) other than purchase of a proprietary license. Industry-wide. I have observed to my chagrin that the copyleft license that I helped create and once loved, the Affero GPL, was seen for a decade as inherently toxic because its most common use was by companies who engaged in these seedy practices. You've probably seen me and other software freedom activists speak out on this issue, in our ongoing efforts to clarify that the intent of the Affero GPL was not to create these sorts of corporate code silos that vendors constructed as copyleft-fueled traps for the unwary. Meanwhile, proprietary relicensing discourages contributions from a broad community, since any contributor must sign a CLA giving special powers to the vendor to continue the business model. Neither users nor co-developers benefit from copyleft protection. //

Given the near ubiquity of proprietary relicensing and the promulgation of stricter copylefts by companies who seek to engage (or help their clients engage) in such business models, I've come to a stark policy conclusion: the community should reject any new copyleft license without a clause that deflates the power of proprietary relicensing. Not only can we incorporate such a clause into new licenses (such as copyleft-next), but Conservancy's Executive Director, Karen Sandler, came up with a basic approach to incorporating similar copyleft equality clauses into written exceptions for existing copyleft licenses, such as the Affero GPL. I have received authorization to spend some of my Conservancy time and the time of our lawyers on this endeavor, and we hope to publish more about it in the coming months.

We've finished the experiment. After thirty years of proprietary relicensing, beginning with Aladdin and culminating with MongoDB and their SS Public License, we now know that proprietary relicensing does not serve or extend software freedom, and in most cases has the opposite effect. We must now categorically reject it, and outright reject any new licenses that can be used for it.

How do we know Google knows they stole Android from Oracle-Java?

The E-Mail That Google Really Doesn’t Want A Jury To See:

“Lawyers defending Google against a patent and copyright lawsuit brought by Oracle are trying desperately to keep a particular engineer’s e-mail out of the public eye-but it looks like they’re unlikely to succeed.

“The e-mail, from Google engineer Tim Lindholm to the head of Google’s Android division, Andy Rubin, recommends that Google negotiate for a license to Java rather than pick an alternative system….

“The second paragraph of the email reads:

“‘What we’ve actually been asked to do by Larry [Page] and Sergey [Brin] (Google’s founders) is to investigate what technical alternatives exist to Java for Android and Chrome.

“We’ve been over a bunch of these and think they all suck. We conclude that we need to negotiate a license for Java under the terms we need.’”

Except Google never did negotiate for Android “a license for Java under the terms we need.”

But they released Android anyway.

That’s not legal.

Ron Coleman
@RonColeman
Today the New York Supreme Court dismissed the "toddlers meme" lawsuit for "misappropriation" against our client Carpe Donktum (Logan Cook) https://drive.google.com/file/d/1l1kM8lGbFJh-eUWKJLD0eQSmyd83GA4V/view?usp=drivesdk

Harmeet K. Dhillon
@pnjaban
·
Jul 9, 2021
Congratulations to my @dhillonlaw partner @RonColeman for defeating the lawsuit brought against client Logan Cook aka Carpe Donktum and 45 over the infamous toddler video, commentary upon which is a matter of public interest! Congrats to President Trump as well! #FirstAmendment