Yours sincerely,
[private]
Internet Investigator
On behalf of:
Pearson Education, Inc.
------------------------------------
Seems like another glorious victory of the overzealous online copyright strikers. Nothing out of the ordinary, will likely get sorted out in a day or two.
Haha... that's a brilliant euphemism for "profession Google user". Or maybe we could call it "serial scrapist"?
Of all the silly titles I tried to come up with, this sure hadn't crossed my mind yet. And it has a catchy ring to it too.
"Internet Investigator" got to be somewhere up there between "Adult Software Engineer" (middle ground between "Junior" and "Senior") and "Server Captain/Commodore/Admiral".
Actually, "Internet investigator" is the person's job title, not the company name. The person listed the title like he is an employee or contractor representing Pearson. Anyone 'claiming' to represent them could have sent it.
well, if this is the author of the book, they probably have their contact anyways. And if the claim was clearly illegitimate and the owner didn't want to provide this, I guess he could try going without it. There are probably easier ways to dox people
When someone “swears under the penalty of perjury” something that is inaccurate, and the legal system does make it simple to apply that penalty, there’s something wrong with the legal system.
Regardless of the actual form of the notice (so long as it is not deficient), under the DMCA the only part for which is actually treated as under penalty of perjury is the claim of the person giving notice being or representing the copyright owner or exclusive licensee of the the work identified (notably, the claim that the particular content to be taken down is infringing is not under penalty of perjury.)
While I don’t know the relationship between this “investigator” and Pearson, the claim that Pearson is the owner or exclusive licensee of the books identified does not appear to be in dispute.
This does not appear to be a defect in the application of the penalty by the legal system but of what the penalty applies to under the relevant law.
Not quite. The counter-notice also includes that language, "I swear under penalty of perjury that the notice-giver is wrong". So there's no penalty for lying in the notice, but the same can't be said for the victim of the notice. What's more, you _can't_ counter-notice that your use of copyrighted material is protected by fair use - that has to be found in court, so hopefully you've got deep pockets.
There should be a 3-strikes and you're out when it comes to DMCA abuse. After 3 false DMCA notices you lose all copyrights. That will stop this abuse once and for all. Running automated systems that do nothing but harass people over things that they are legally able to do should be illegal.
> There should be some sort of modest cost somewhere to stop this nonsense.
Playing the Devil's Advocate here:
But wouldn't that in turn incentivise actual copyright infringement on account of it potentially being cheaper to mass publish copyrighted material as opposed to request it being taken down?
I hadn't thought about that angle before. Is the public interest best served by adding friction to copyright enforcement, or by adding friction to fair use?
Certainly the right to copy is different when copying is so much cheaper than when the laws where written. We gave up:the right to copy when you couldn’t really copy. Now we can copy and maybe we want to retain that right to the people. Or at least rejigger so we either get more or give less via this exchange
The law is based on the doctrine of unclean hands. Judges decide on a case by case basis, based on the facts pertaining to the case.
An artist finds their life’s work online, and issues a takedown notice, not realizing it contained a few songs they don’t notice? No big deal.
A label pays a takedown notice mill, that produces completely bogus / mistargeted takedown requests on a regular basis? Scorched earth. They lose the ability to enforce any copyrights on their catalog.
> The law is based on the doctrine of unclean hands. Judges decide on a case by case basis, based on the facts pertaining to the case.
Judges don't scale and they are a step function, not an incentive gradient.
> An artist finds their life’s work online, and issues a takedown notice, not realizing it contained a few songs they don’t notice? No big deal.
If he had to pay a small(!) fee for mistakes in his takedown it would mean little harm to him since he protected his works with one takdown. One could even add an exemption for first-time mistakes. If he has to do it repeatedly then one would hope that he improves his search process.
Additionally you have to consider that takedown notices can be filed against middle-men targeting many different 3rd parties at once. E.g. consider URL takedowns submitted to google. Google has no dirty hands and neither do some of the mistaken targets.
Small fees would only add up if you issue bogus takedowns in bulk. They also have the advantage of not burdening the courts and providing the right incentives. The fact that the cost exists would drive people to reduce false positives which would make the penalty being required less often.
As far as fees go one could also distinguish between fair use disputes vs. wholly incorrect claims. After all in the former case the claimant has a valid case and it is just that there exists a valid defense. In the latter case the claim itself was invalid.
> A label pays a takedown notice mill, that produces completely bogus / mistargeted takedown requests on a regular basis? Scorched earth. They lose the ability to enforce any copyrights on their catalog.
> SoundCloud takes down D.J. Detweiler's 'remix' of John Cage's "4'33" for copyright infringement.
> White Noise Video on YouTube Hit By Five Copyright Claims
Not to mention that almost all classical music pieces recorded by students are under risk of being DMCA'ed for being so similar to copyrighted recordings. Classical music has many interpretations for the same piece and they are harder to tell apart than, say, pop music.
Please consider a quote in its entirety. Takedown mills losing the ability to file claims is not what is happening in reality. Scorched earth (against takedown abuse) is not happening in reality. The first sentence is merely a setup of the scenario, the last two are the conclusion (which doesn't happen, at least not often enough to discourage the behavior).
A sinple solution would be to create a private cause of action, for both the recipient and the impacted user of a DMCA takedown notice that is false in any essential way (including claim of infringement, not just the parts under penalty of perjury, for which there is a public sanction available) for actual costs or other damages or statutory damages, trebled for knowing or reckless falsehoods, with courts given explicit power to also require the offending notice-giver and/or copyright holder, if found to be engaging in a pattern and practice of reckless notices, to post surety and provide special notice in future DMCA notices issued by then (or on behalf, ifnthey are an owner) for a set period of years (say 3) allowing the recipient a 30-day window to investigate claims at the sanctioned actor’s expense to determine legitimacy prior to executing a takedown, with the safe harbor still in place if that investigation is conducted in good faith even if it comes to a negative conclusion.
Kevin Wayne is one of the co-authors of the book, but the copyright listing is "Copyright 2011 Pearson Education, Inc." and there does not appear to be anything in the book that says that the code is GPL licensed.
Presumably Kevin Wayne was given the right to license the code as GPL, but that fact didn't get included in the text of the book, and the f*cking, stupid copyright bot Pearson uses didn't get the memo.
Is it time for a sci-hub like event to replace github? It would be better for the repository of nearly all modern public programming creativity be something more like Wikipedia, and less like LinkedIn.
Given the technical nature of the work, developers who can't jive with GitHub's policies can throw up a GitLab/Gitea server in an hour and evade censorship. And of course, as we learned with youtube-dl, the issue isn't the code/commits (since everyone gets that with a git clone), it's the issues and PR history.
Any public file host has to deal with the DMCA, and (for now, barring any evil brought on by Microsoft), I'd bet on GitHub siding with the developer over the lawyers. They had lots of incentives to go the other way in the youtube-dl case and they didn't, and then they put a bunch of money in a fund if someone wants needs a lawyer to get their project back online.
But this is a whole another SCM instead of Git, isn’t it? What I’ve meant was just a convention within Git, same way we have a convention for README.md and “doc/“.
Speaking of Robert Sedgewick, I deeply enjoyed his course Analysis of Algorithms, as Sedgewick beautifully showed the power of generating functions. Generating functions, or analytical combinatorics in general, make it order-of-magnitude easier to analyze algorithm complexities compared to what we learn in an introductory course. It's mind boggling to experience how higher-level constructs of math reveal simplicity and deep insights.
So, distributed git has become sort of centralised. Maybe people depend too much on github? I wonder, what would happen to IT over the world if github will be down for a week or a month?
Utter chaos for a week, then a week or two of swearing at gitlab then the world would continue - since while we use github we don't really use github - it's just a shared git repository with a reasonably nice interface for reviewing code.
That's for my company, for many open source projects the damage would be more severe since they often heavily use issues and GH project management features.
The problem is that many people (including companies) use GH for more than just source code hosting. Bug tracking, CI/CD, artifact hosting, etc. That's not something most companies can replace in a week, unless GitLab (or others) have a seamless import tool. And even then, if GH is down or your account has been suspended, what do you import from?
There is literally nothing preventing someone to host they code on gitlab/sourcehut/their own gitea server. Most people use github because it is free, but the git protocol is far from centralized.
I recommended checking out Tom Scott’s video on DMCA (0). It’s 40 minutes long and rather YouTube centric however it paints a very clear picture as to why these companies are engaging in a seemingly broken practice. It’s not because they want to, it’s because they are legally required to do so.
YouTube goes very much above and beyond what the DMCA requires, and favors the interests of the big corporate copyright holders over smaller creators.
I imagine they do this because of their settlement with Viacom[0] where they likely bent over backwards to avoid further appeals and lawsuits. Viacom likely said "regardless of how this goes, we and other copyright holders will sue YouTube off the internet unless you build a system that proactively takes down everything even remotely potentially infringing without us having to notify you of anything". And Google caved.
Don't get me wrong, the DMCA is still overall a bad law (though, to be honest, I think the copyright takedown process is the least bad part of it), but Google is not doing what they do because of the DMCA.
No, they aren’t. If there is no merit to a copyright claim, there is no vicarious liability for Github to be insulated from by the DMCA safe harbor.
It’s cheaper and less risky not to evaluate the merits of DMCA notices and just to blindly execute them, but it is erroneous to say that thet are legally required to act in that manner.
Yes, they are, and yes, there is. The claiming and counterclaiming process is independent of the merit of copyright under DMCA. Budget and risk are not part of the equation. Evaluating copyright applicability cannot be part of the equation; it’s not your content. The law specifies almost exactly what you must do with some vague concepts for interpretation (like expedient). If you don’t do those things even on an obviously bogus complaint that is otherwise properly formed, you stop having safe harbor, because those disputes are intended by the law to be handled in court and not your legal department. It’s genuinely not complicated (read OCILLA) and you should be annoyed with the law for this situation, not its subjects.
You are sharing an opinion disguised as fact. It’s a wrong one, but I get why you’d conclude it.
> The claiming and counterclaiming process is independent of the merit of copyright under DMCA.
The DMCA process isn’t a mandatory process, it is a process to receive safe harbor from whatever liability would otherwise exist. If there would be no liability independent of the DMCA safe harbor, there is no mandate to follow the safe harbor process.
(If you disagree with this, here’s what you need to do, identify—with citation to supporting law—the available legal remedy that can be imposed on a provider who declines to adhere to the DMCA safe harbor process where there is no underlying copyright liability.)
Service providers make a choice to follow the notice process blindly as a risk management measure, not because it is legally required to do so.
(Which is also why the counternotice process is often less fully implemented, or why, e.g., Youtube has its own hyper-aggressive policy for certain content that goes beyond the DMCA process and lacks counternotice opportunity—the counternotice process is just as much part of the DMCA process, but because providers are able to structure their relationship with users in a way which avoids having any liability for takedowns which would benefit from a safe harbor, they can safely ignore that process.)
Source: the actual text of the DMCA safe harbor provision, and the reason it is called a “safe harbor”.
> You are sharing an opinion disguised as fact
No, you are engaging in standard industry blameshifting by misrepresenting risk management strategy adopted responding to incentives created by a law with an actual legal mandate.
> If there would be no liability independent of the DMCA safe harbor, there is no mandate to follow the safe harbor process.
The problem is that GitHub doesn't get to decide that. If they give up their option for safe harbor by not following the DMCA takedown process, the company claiming copyright can sue them. Even if they don't win, defending these things costs time and money.
I don't blame GitHub for just not wanting to deal with all that. You seem to disagree, and believe that GH should stand up and open themselves to legal liability for every random user on their site, which I don't think is practical or reasonable to expect of a for-profit corporation.
The remedy for this is simple: we need to stop relying on centralized solutions for things like this. There are a variety of options for this: Tor hidden services, IPFS, hosting outside the copyright holder's jurisdiction, etc. None of them are perfect, but relying on a for-profit corporation to host your even-remotely-legally-questionable material, you're setting yourself up to fail.
> The problem is that GitHub doesn't get to decide that. If they give up their option for safe harbor by not following the DMCA takedown process, the company claiming copyright can sue them.
The company can sue them in either case. The DMCA safe harbor is a particular easy basis for a motion to dismiss, but other easy bases may exist in specific cases. Blindly following the DMCA safe harbor process is a sensible risk mitigation strategy, and exactly what the law is structured to promote, but it is very explicitly not a mandate.
> I don't blame GitHub for just not wanting to deal with all that. You seem to disagree
You are a mistaking a fact statement about what the law does and doesn’t require with a normative statement about what Github should do.
It's the Sedgewick/Wayne Algorithms 4 in Java part. The author had an arrangement with the Pearson editor to release the accompanying code for the Java update under the GPL, but apparently someone else at Pearson's had a different idea. Let's see if he has it in writing.
When I was an undergrad student, we had an algorithms and data structures lecture based on the Java version of the book. We also got an archive with the samples to work with that was IIRC GPL licensed.
Did this repo contain the code or the entire book as well?
In the former case, if I remember correctly, this DMCA claim would clearly be bogus. Yet another case where sending out DMCA claims having penalties for the sender if they are fraudulent, but putting a burden on the recipient who might suffer if they don't immediately act.
Even simpler, require those filing a claim to put down a monetary deposit. If the accused disputes, they too file a deposit.
The winner gets the money back and the platform gets to keep the losers deposit as a fee for having a trained, certified human make the decision.
Make the fees scale up with the number of claims by an accuser.
No one would file a frivolous claim. No one would dispute an obviously correct claim. The cost of having humans decide is covered, so platforms like Google have no excuse.
This seems like the perfect way to enable frivolous takedowns, because the accused might not be able to pay the deposit to dispute the claim.
Additionally, small artists might similarly become unable to enforce their rights.
(And if the deposit isn't too big, do you think a few dollars will deter billion-dollar companies?)
It means that people without spare cash completely lose the ability to appeal.
I don’t know why a counterclaim should require a deposit - DMCA is already a guilty until proven innocent law, and merely counterclaiming doesn’t put your content back online.
This is in the law, but it’s rarely enforced. Stepping up education for judges and support for people filing lawsuits against major companies would be a relatively easy way to start making the automated bots riskier to run.
Is it? I always thought there is only a fine for misrepresenting that you are representing the copyright owner. Not for falsely claiming a piece of work is indeed infringing.
“Penalties are provided for knowing material misrepresentations in either a notice or a counter notice. Any person who knowingly materially misrepresents that material is infringing, or that it was removed or blocked through mistake or misidentification, is liable for any resulting damages (including costs and attorneys’ fees) incurred by the alleged infringer, the copyright owner or its licensee, or the service provider.”
I'm not a lawyer but my understanding is that by now the courts have confirmed that it's not just enough to say that you own the copyright on the material: you also have to confirm that the person you're sending the claim to doesn't have a right to use the material as well and that includes fair-use. If you were, say, using a Disney clip in a film studies context they would likely be liable if a takedown bot sent a DMCA claim unless they could show that you did something like posting the entire film claiming it was for “study” purposes.
In this case, that's highly relevant since this repository was apparently released under an open source license so even if Pearson held the copyright they wouldn't be able to take away the permission granted by the open source license unless they could prove that it was never legally approved for release under that license.
Knowingly is the problem - it implies intent so erroneous claims don’t fall afoul of it. Obviously attacks on critics do and should be penalized but that doesn’t happen either - it requires a lawyer that will cost much more than you’ll ever get. Until the law is amended to require the victims be compensated for time and legal expenses there’s still no real cost to fraud
If this was a youtube channel takedown people would be mad at Google, but good to see at least here HN is blaming the bad actors like they should - ie. the copyright laws and its abusers.
People are mad at google for _automatic_ takedowns, as they enable bad actors and are arguably not in the spirit of the law.
In this case here, github does not employ an overzealous automated system. This is a manual DMCA claim and the issuer is the single actor here, who could be to blame.
People got mad at Google for automatically honoring DMCA takedowns, which is what GitHub is doing, as they both have to if they want to maintain not being legally liable. Content ID is also a system people are mad with but has nothing to do with DMCA besides Google saying 'pretty please only enroll content you own into Content ID'.
Assuming the claimant doesn’t mess up the complaint it might as well be automated - GH has only voluntarily not honored a request once with YouTube-dl (that I’m aware of) and that still was after YT-dl removed the README reference of using it to download music videos.
Google have their own system for copyright claims which people get mad about. GitHub follows the DMCA laws which they have to follow. I put it to you that your implied claim that these situations are basically the same is false.
I give a lot of credit to Sedgewick’s course for helping me break into the software industry, having come from an adjacent field in academia. I hope they can get the repo back up.
In my case it didn't help me break into the software industry - somehow I already managed to do that despite some well-deserved imposter syndrome at the time, but it definitely leveled up the quality of my code
The DMCA notice linked on the page [1] gives a few more details.
We have received information that the domain listed above, which appears to be on servers under your control, is offering unlicensed copies of, or is engaged in other unauthorized activities relating to, copyrighted works published by Pearson Education, Inc.
Copyright work(s):
Deep Learning with R 9781617295546
Algorithms 9780321573513
Copyright owner(s) or exclusive licensee:
Pearson Education, Inc.
Copyright infringing material or activity found at the following location(s):
Kevin Wayne is one of the authors of Algorithms 4th ed. by Robert Sedgewick and Kevin Wayne and he has a github repository called "algs4". Pearson sent github a DMCA notice for this repository and another. I'm not sure exactly what content he hosted in "algs4".
> This public repository contains the Java source code for the algorithms and clients in the textbook Algorithms, 4th Edition by Robert Sedgewick and Kevin Wayne. This is the official version—it is actively maintained and updated by the authors. [...]
let me get this straight, this was gplv3 licensed code, allowed by pearson, and then pearson auto dmca's it? Looks like some algos need to be fixed on pearson's side, in any case here you go : https://algs4.cs.princeton.edu/home/
Not totally unrelated: Pearson is a piece of trash company. They sell very overpriced textbooks and always use DRM in online copies. They lobby for standardized tests in states and get them mandated. Students are put through them. And guess who provides the certificate required by teachers to evaluate those exam copies? Yes, Pearson.
In college we had to use them. It was a financial burden on myself. $100/course essentially. Our school mandated that we take homeworks using Pearson's online system. This means if you found a used copy of the book it was essentially waste. You needed to have a one-time-use code (printed on the book) to do your homework. If you didn't, you'd essentially fail the class. Also, the system was buggy.
Why is education like this in the US? Is it the teachers who willingly allow students to spend so much unnecessarily? Is it the admins? Do they get paid to do this stuff?
I think educators are certainly to blame - as much as it's convenient to blame admin staff, educators in my experience set the course book.
Pearson etc offer them a time saver they don't have to pay for - they take out the effort of assigning and marking exercises, and don't add any cost to the educator. The students pay, and Pearson uses these codes to inconvenience students and force them to buy new copies.
If you think like an MBA, it's a great idea - the person in decision making power benefits from this move (exercises handled for them), is probably unaware of the issue for students, and has more time to spend on something else they need to do (raising grants, trying to write papers to keep going on the tenure treadmill). But you also get to sell a new copy of each book annually, at full price, without used sales cannibalising your returns. Given there isn't a huge growth in universities each year, this is a great way to boost revenues sustainably.
In my experience on the other side of the pond, we used 1 textbook in the entire degree. It was reasonably priced, readily available second hand, and didn't have any single use license codes. Indeed, one instructor used to upload the relevant chapters of their own book as PDF files, insisting nobody buy the book as it wasn't worth it!!
Because in US the social fabric is based on money. So it's not something you would necessarily consider odd. Kinda how Germans say "was always like this, why should we change" but in different contexts.
The other part is that Pearsons lobbies, the tests are already accepted as mandatory and then not every teacher is capable of coming up with his own test that allows you to then perform well in the Pearson test. You probably also risk a lot if you used your own test.
It's basically industry developing its own industry.
I think a lot of this hides behind the notion of "accredited" universities and courses.
Some education board (with heavy influence from various companies) decides that you need to select from a set of curriculum to be accredited. Other universities (and some employers) won't honor credits or degrees unless they're from accredited courses.
As one data point, this only ever happened to me in my general courses -- a department would choose a book that several lecturers would use to teach hundreds of mostly disinterested students. In that context it makes a lot of sense; the decision maker isn't any single instructor, and there's a large pressure to instill a baseline level of knowledge into way too many people at once.
All of my courses with smaller class sizes were much more careful with their book selections; occasionally those books were still expensive (even used), but only if they were really the best tool for the job. Even with expensive books though, this kind of online subscription nonsense was never used.
The ironic part is that this only happens in required courses for credentialism, generally at lower tier colleges that created the student loan crisis, not in courses that students choose for an education.
And it's especially prominent in math courses for students who don't know nor want to learn math.
Also, they release new versions of books all the time and force their use in classes, making it impossible to use 2nd hand textbooks in a lot of cases.
*I mean, come on. How of ten do fundamental physics change...
Pearson is a company that should not exist.
Plugging open source text books. There are many high quality option out there that are openly licensed. Openstax is a project by Rice University that has textbooks for most gen ed courses.
openstax.org/
Thanks for the link. Didn’t know. Btw, this is why they lobby for state mandates. A single teacher cannot make an exception
Situation in private school is not better. A prominent private high school in Bay Area, made me run around like a crazy man scavenging used books from all over the place because the recommended books were out of stock. They won’t publish the exact book name until after classes begin in August and apparently varies by teacher by year. This is after paying $$$ fees equivalent to buying a new car. Next year I pulled my kid out of it.
I know of a company that had a pretty large set of creative commons licensed learning materials they put together, along with tools to customize lesson plans and rearrange/edit material, make online quizzes, etc. It was offered as a good low cost digital replacement to traditional textbooks, aimed at colleges. Follett, the #1 textbook distributor (which interestingly has common roots with Barnes and Noble) bought the company. Shortly thereafter, the product mentioned was discontinued. The professors that had adopted it, I assume, were somewhat inconvenienced.
When I tried using a previous edition, the page numbers and most of the exercises didn't line up. "Thursday's test will be on pp 144-172 inclusive, and get to know exercise 5.17." It put me at a surprisingly big disadvantage.
That's often the only thing that actually substantially changes between editions. Shuffle the exercise numbers and change the font and spacing and voila, brand new edition completely incompatible with older versions!
If the teachers are helpful, this would not be an issue. In most places I went to, teachers understood, and some peer always helped people with older textbooks. For problem set additions or extra chapters or pages, we used to photocopy those pages before high-res cameras became a norm in mobile phones.
also many professor make wink-and-nod that student may simply "find pdf copy" from internet. good ones are not desiring for to make students pay large moneys.
I got a bit pissed that this for one course. Wasn't the professor's fault, but rather the school.
(in 2011) University tells incoming students they must check the book list for their course and get books before classes start. Books are expensive.
First day of class - professor gives everyone a PDF of the physics book I just spend $200 on. Says, "the university requires that I list at least one book for the course and they can't communicate with the students beforehand.
Never bought books before the first day of class again.
I've seen similar things, with the reasoning being less "the university requires the professor to list at least one required text" (this one would be easy to sidestep: just require a cheap or free one and then use another) and more "the texts listed are required will be sold to students at a discount, so students can only profit from this... as long as they are in the know".
Not just way overpriced, the few Pearson books I happen to own are of an alarmingly low quality. The pages of a phone book look and feel like premium paper compared to my copy of Blitzer's College Algebra.
Not to mention some of them have wonderful anti-features, such as physical books that have a handful of online-only chapters that you can only access if you buy the textbook new, thereby reducing the value of used copies.
TED started giving nearly anyone a platform, which changed it from a curated set of higher-quality presentations into something more like YouTube. There are a ton of TED presentations full of debunked science. The entire vibe of TED talks seems a bit self-congratulatory and without substance, IMO. There's also TEDx, which has the same problems but worse.
There are still good TED talks, but you have to sort through TED the same way you sort through YouTube... and at that point, YouTube starts looking a lot better.
A lot of academics and scientists I know dislike TED talks. First the name means (T)echnology, (E)ntertainment, and (D)esign -- which are all proximal to, but quite different from, actual science. If anything, it turns science into consumer products, and gloms it with self-help and entertainment in a quasi-megachurch format (hint: "innovation" is the savior; and everything is cast as innovation, esp. technological). Second, it induces a race to the bottom in terms of simplification and overselling conclusion / significance (compare with the perspective that knowledge production requires deep backgrounds to truly understand, and that the results of most experiments / analyses / models of the real world are fairly nuanced). Third, the stuff that comes from their franchising (TEDx) contains some real garbage.
TEDx talks (like the linked one is) are not proper TED talks, there are affiliated events all over hosting them. While there are many good TEDx talks, many of them aren't what you would expect when hearing the TED name. Especially bad is it that some TEDx events let quacks talk, which gives them a kind of legitimacy based on TED's reputation even though they're saying bullshit. Of course, this commingling has also tainted TED's brand.
The other thing is that most TED talks follow a certain formula. They do because it works, but watching many of them it sticks out and can be a bit annoying. There's probably a meta TED talk pointing out this somewhere.
You know, there is utility to letting the "quacks" talk, right?
Ever had a thought that just kept popping up in your head, that you then mentally suppress because "Dear <insert religious figure here>, how horrifying."
After doing that it keeps coming back again and again? Usually after a shorter and shorter period of time, because it becomes an indexing thought for recollections of all the times you've squashed it or thought about all the ways in which it was wrong?
Ever then talkedabout itwith someone else then had a cathartic release from the sharing, and possibly even a moment of relief that someone else besides you stumbled on <horrible thought> too?
It's the same dynamic. Just writ at a societal scale. It's why no censorship is a truly good idea. Bad ideas crop up organically as we've got a constant influx of new members of society seing the same warts, the same problems, following the same dead ends each and every one of us has. How are they to recognize a false start without being exposed to them?
The part that frightens people is that previously, you never had it possible for the whole world's population of false starters to get into one place and coordinate in real-time. You had an implicit depletion of critical mass activity that ensured the feedback loop for these ideas petered out, or remained tightly constrained to just the local nuts. The world we're in means that barrier has fallen, so everybody has to be able to cope with the entire planet's equivalent of the crazies all the time realistically.
It's a learned skill. Only practice gets you better at it, and I assure you, however good you think you are, cut the estimate in half. Odds are you've got fundamental blind spots you haven't even run into the craziness for yet.
I attended one TED back when they were super exclusive, because they were super exclusive. The good lesson: super exclusive for the sake of it a waste of time.*
TED talks were pretty much of the “I am doing this cool thing and isn’t it so exciting?” That actually does sound pretty cool: the world is full of exciting things I’ve never heard of or have heard of but never appreciated.
The reality, both in the old TED and the new: the talks are almost all structured in the same mode: “you are smart for listening to this info that other people don’t appreciate.” Validating the listener for spending their time listening.
If I wanted that I’d go on a walk with my dog. Which I do instead.
* things can be exclusive just because there’s finite space and you want to have people who contribute (e.g don’t invite me to a medievalists convention; I’ll have fun but won’t add to the discussion). This works if it’s not self-congratulatory and if people pro actively mix it up over time to bring in new ppl.
IMHO, the books I've bought from InformIT are not that bad. They're only watermarked, and were useful. IDK, maybe because they're not geared towards universities.
> They lobby for standardized tests in states and get them mandated. Students are put through them.
This is totally false. State standardized test is the natural consequence of state public education accountability system.
There must be a comparable measure across all the public schools in the state as the foundation for accountability measurements. And all (>95%) the students must be tested for equity reasons. No matter how many misinformation you read about standardized tests, the simple fact is that it is the only way to measure student learning of a huge population and get meaningful / comparable results. Is it perfect? No. Is there a better alternative? No.
The whole idea that a few testing companies imposed the standardized tests on students is simply a conspiracy theory.
It is nuanced, they don't directly drive the need for standardization, but drive to privatize the standardization.
Doing that, they can influence a lot on the scoring, quality, content and certifications around the standardization and fund the psychometric research establishing validity/equity of the tests.
Like any other big org biz, specing the RFP is one easy way to win the contract.
A decision marker (corporate or government) is likely to choose and mandate a pearson test rather than fund the research and development of their own which best fits their needs because it is easier.
Unlike other industries testing by is nature becomes monopolistic, nobody is going to say you can either of these three tests : it is hard to compare between two tests, so winner takes all.
Pearson and others in the space, exploit this by high prices , poor content etc because that's what monopoly tend to do.
This is all because of privatization of school vendors, why can't content and tests be developed by the state as well ?. Many countries do this.
I am not sure what you are talking about. There are several testing companies, Pearson is not the only one. And states have a selection process and technical review process done by outside psychometricians.
Yes. There are places where tests were developed by the government agencies, which are not professionals. Mostly their tests lack psychometric regorious, not comparable across years.
I'm not sure what you meant "poor content". A few flaws here and there?
I very clearly said pearson and others in the space.
If reviews can be done by psychometricians, why test building cannot ?
There are many many qualified professionals affiliated with major universities- several of them owned by the state. Is it difficult to imagine them developing the tests?
Many of these same professionals consult with testing companies for building the tests, so it is not like these companies have some magic expertise no one else has.
Using an vendor test gives that companies allied products basically an Monopoly for that test/skill.
This is a fact of this industry, and testing companies will use it to their advantage because that's their business. It is in public interest not to privatize the test itself, so there is competition amongst providers of allied products and free market can prevail.
Current setup is like government tax code is set by intuit and you can only file tax via turbo tax and nothing else
You are just imagine things as an outsider. Testing companies not only have the best psychometricians and they have many many psychometricians and stasticians work for them.
Yes. University professors do consulting works. How do they have the capacity to build a product as professional companies do? ...
I am no outsider, I have spent last 10 years partnering/ integrating with testing companies . I have seen tests built and marketed and sold.
All tests have frameworks underlying them , do you know how long it takes to establish a framework ? How many peer reviewed papers that have to published, and time it takes to shape the opinion of the ecosystem to make it acceptable? Without a established validity and equity nobody is going to buy the test, why would they risk it to get sued?
You think companies go through the full journey of years and take all the risk? in most cases they only come in the latter stages and commercialize and adapt it.
The question, then, is why not have the state university build the tests and write the textbooks? Then you fund the actual creation once and can straightforwardly fund updates and alterations as needed.
State universities are not good at doing either. There are not enough psychometricians or test writers or textbook writers in psychology or curriculum and instruction departments.
State government is not better at doing things than big companies. Hope your solution is not to ask the government to do everything.
You think testing companies keep all the psychometricians/text book writers on payroll ?
Many of the big tests are developed within an academic setting with university/public funding then licensed out. Almost every book is written by university professor then published by pearson or others.
State government does not outsource writing the tax code or the other regulations to companies why is tests any different ?
Another aspect you fail to understand is that because the accountability / rating / funding of local education agencies is involved, the state government and the state university do NOT want to and can NOT get involved in producing the test. It must be produced by an outsider.
You can't tell the difference between a prototype and product. OK. Lets go to the basics. Do you at least understand the state standardized tests are commonly given to grade 3 to grade 10 students on three subjects (English, Math, Science)? That's 24 tests (a few grades could share the same test, but that's complicate). Most state universities don't even have 12 psychometricians. Lol. Of course the reality is much more complicated than 24. You are just imagining things with no understanding of the scale of the complexity of the reality.
Is that not clear from my comment? Whether standardized tests are good or bad is a whole other debate.
The point here is- they lobby to get standardized tests mandated where they were not, then they lobby snd push their tests for statewide use very hard. All school districts have to buy test questions and other infrastructure from them only. And only Pearson certified teachers can evaluate those copies. And every step of the process involves Pearson getting paid.
You apparently don't know what you are talking about. No. Testing companies do not lobby to get standardized tests mandated where they were not. The federal government mandates this as a requirement for federal money.
Districts do not pay for the state test or buy test questions from Pearson...
> Testing companies do not lobby to get standardized tests mandated where they were not.
Yes, they do.
> The federal government mandates this as a requirement for federal money.
Sure, some mandates come from the federal government; others are from state governments, and still others from local governments.
Guess who lobbies legislators making those decisions, especially at the federal level, but also at the state level, and sometimes, especially for larger jurisdictions, at the local level?
> Assume what you said is true, guess what will happen if they don't lobby? Same.
Obviously, they disagree, or they wouldn’t spend money lobbying, which isn’t something companies do just to burn money. They do it because they expect RoI on the money spent lobbying.
> Can you give an example what will happen if states do not use standardized tests?
Compared to any of the individual incremental mandates? They’ll have a lower ratio of cost to instructional time and a more efficient educational system, probably. While the requirements are often tied to consequences, they are rarely tied to the kond of systemwide analysis of results against other factors and process improvement that would make them systematically useful rather than theater. They are generally “accountability” measures that assume without any empirical basis that the results are within the control of the entity held accountable and that the consequences attached will force decisionmakers to make changes that result in optimization of the results.
Oh, and standardized testing companies will make less money. Hence the lobbying.
If testing companies don't lobby, then states can develop their own tests that meet the federal guidelines. I have a family member who is an educator and has contributed candidate questions to their state's standardized testing board.
The problem is in recent years it has taken on something else...
Standardized tests can be a window in the performance and health of the overall system, or over a group of students, but it is not reasonably determinate of the knowledge level of an individual student. Which is how they are often used
In the beginning it was used to judge the performance of a teacher or school, today is it used to judge the performance of the student something it is not really equipped for.
This is similar to BMI, BMI is useful as a health metric on a population of people, but is not very useful as a metric on the health of an individual
It's far less work to type two words into a search engine than to write an I'll informed comment. Why trouble yourself with the long embarrassing comment?
to start with I couldn't get over the presenter's poetry slam style but what made me cry serious salty f*ing tears was his explication of how inequality impacts day to day learning. as in hunger. literal, biological hunger the day before your standardized tests. stresses you had no input into and should have no bearing on your future.
This isn't worth getting your feathers ruffled over. Some algorithm or barely-paid human automaton flagged it, someone higher up will notice it, it'll be added to an exception list, problem solved. It's not ideal, but it's not malicious or nefarious or even worth talking about every time something like this happens. It's a simple mistake that has a simple solution.
> or even worth talking about every time something like this happens
I think it is, because the core problem is not the wrongful takedown of a single repository itself, it's the outlook that the only way to get these "mistakes" reversed is the hope that it will reach the news, Youtube/Google shows us what the latestage of this nightmare looks like. Github is obviously not nearly this bad (yet?). This case will obviously be restored, it's a popular author, but what about people that don't have this reach?
I don't see how talking about this situation can be a bad thing.
> I think it is, because the core problem is not the wrongful takedown of a single repository itself, it's the outlook that the only way to get these "mistakes" reversed is the hope that it will reach the news, Youtube/Google shows us what the latestage of this nightmare looks like.
YouTube/Google is bad because they use their own system instead of or in addition to the system required to get the DMCA safe harbor.
For sites that simply follow DMCA, such as GitHub, there is no need for the issue to reach the news. There is no need to be popular or have any reach. The most obscure poster with no audience can easily get their content restored. Here is what happens at such sites:
1. Someone claiming to represent the copyright owner files a take down request.
2. The site temporarily takes the content down, and notifies whoever posted the content.
3. The poster files a counter-notice saying that they have the legal right to post the content.
4. The site put the content back up, and notifies the party that filed the take down notice, telling them who filed the counter-notice.
The site is then out of the loop. The content is back up, and the party claiming copyright violation cannot sue the site over this. The site is now in the DMCA safe harbor. If the party claiming violation wants to take it farther, they need to go to court and sue the party that posted the content.
The counter-notice is trivial to fill out and file. Here's the one for this case [1] if you want to see how simple the form is.
I 100% disagree, the fact that as a society we are accepting of this kind of thing as "just the way it is", is the problem
"Ohh its just some bot" should never be an acceptable response, and people creating these bots should have massive fines attached to their false positives. Otherwise there is no incentive to actually validate or improve what these automation systems are doing, the incentive today is the flag everything and worry about it later...
That's a good question but for the wrong audience. Take it with the US Congress who has adopted so lopsided copyright law under the heavy lobbying of the music and movie industry in the wake of Napster.
Yes, like the parent says, don't overreact. This is totally not symptomatic for lopsided legislation. So please be quiet already, it will be easier for everyone. You are worried you might be next? Kid, grow up, this is life, nobody said it'd be easy. Here's your gun in case the tanks are coming...
"someone higher up will notice it" is not always true
no consequences for malicious or invalid DMCA is ridiculous (1$ penalty, paid to targeted individual/company for each invalid DMCA would be a good start)
https://github.com/github/dmca/blob/master/2021/04/2021-04-2...
Kevin Wayne (repo owner presumably) is listed as co-author on the book that is published by Pearson, who have issued the notice.
https://www.pearson.com/us/higher-education/program/Sedgewic...