ABCLAW

History repeats itself here, with the difference that instead of paying for the data, the ai crawlers simply take it for free.

This discussion was broached originally when discussing whether or not search engines and aggregators had any compensation obligation in respect of news articles. This was a hot topic in the IP and policy circles for a few years.

When the Canadian government attempted to create a mechanism to compensate content creators for the scraped content, there was widespread outrage from tech circles, despite the same community agreeing, across extensive policy discussions, that action had to be taken to prevent this universal man-in-the-middle value capture by search engines.

I've had fairly extensive discussion with the individuals involved in the academic, policy and internal industry analysis of the issue. Watching industry agree to address the issue, then aggressively spend to shape public narratives in public was eye opening.

The recent shift into "AI is obviously going to hoover up all your data and there's nothing you can do now that the theft is laundered through an LLM" is just the latest example of the same trend of short sighted capital-over-everything decision making we've become used to in jurisdictions that have dysfunctional legislatures.

This isn't the danger with assuming good faith.

It's the danger with not updating your behavior when the balance of probabilities shows that the individual/group/organization you're dealing with is acting in bad faith.

Our modern era has so many megaphones blaring into the void that it's become significantly more difficult to determine which ones are overtly lying. Accordingly, it's far more difficult to establish a consensus about who's operating in bad faith.

Again, the failure point isn't with the assumption of good faith. It's with not recognizing bad faith when it's made manifest.

You can cultivate good taste by intentionally taking in a lot of information about what's in the field, and what you like and what you don't like about it. This could be commenting on elements of film, fashion, photography, but it can also be having a sense of what you like to see stylistically in a contract, in a framework, or in corporate culture.

I recall reading an interview about a legendary developer, and the majority of the interview was not focused on his coding decisions or the structures he built, but it was about a notebook that he kept with voluminous notes about what was good and what wasn't. That notebook is a materialized version of 'taste', and it's certainly something almost anyone could put together with enough effort and time.

No, most genetic drift that ends up being reflected in wildtype populations happens during those periods, because small errors taking up a sizable percentile of the allelic distribution is easier when there are less alleles.

When it comes to neutral mutations, we can literally see constant variance creation in plenty of non-coding areas of DNA over time.

Drift occurs at a fairly consistent rate that reflects the intrinsic error rate of the particular replication machinery that a given organism uses. You can measure the statistical error rate of different ribosomal complexes.

Different planets are going to have different selection pressures. They'll have different conflicts. Different crises. Different cultures. Different reproductive preferences. Imagining that these populations will converge on the same wildtype by sheer chance is lunacy.

Refusing to improve early, probabilistic diagnosis because today’s treatments are modest confuses sequence with outcome.

While you're right from the perspective of humanity taking the steps of gathering data then tackling the disease, most developed countries have single payer healthcare systems that require some level of cost-benefit analysis to approve covering new diagnostic systems.

Alzheimer's disease progression doesn't seem to have any notable preventative indications other than 'eat well, exercise and stay mentally active', all of which are standard recommendations.

Recall that this isn't an issue deciding between funding and non-funding. It's an issue deciding between funding Alzheimer's diagnostics, new GMP agonists, new screening options for highly preventable cancers, etc. Building out a dataset is nice but unless that's surplus money redirected from other programs it's going to come at a real flesh and blood cost.

I think the skepticism warrants more work than that. Darwin's finches are an entry-level concept to learn when learning about biochemistry and genetics. Separate planets would act to separate groups into distinct genetic populations which would then have different selection pressures put upon them. Even without selection pressure, genetic drift in both populations would result in differences compounding over time.

Humans aren't the endpoint of evolution. Something will come after us, and if we're spread out on a ton of planets, there would need to be explicit counteracting forces (genetic modification, tremendous volumes of interstellar human travel, etc.) to make sure whatever comes after us is uniform among our interstellar backyard.

America had a true famine; the dust bowl resulted in mass displacement, and the government took exceptional steps to create remediation programs to address the plight of those affected to maintain relations. The policies included measures that would be considered exceptional by today's standard, including the creation of a national organization to provide stock for relief organizations, buying out cattle herds above market value, other bailout measures for farmers, a massive work effort to create an erosion barrier and more. Most cultural histories indicate that these bailouts prevented widespread unrest in these communities.

You can take a look at the global hunger index; countries with less food security are certainly less stable than those that aren't, but by no means are countries like India and Pakistan undergoing constant revolution. By contrast, countries with comparatively solid food security like Egypt underwent revolution that toppled the government sparked by changes in the (comparatively affordable) price of food. Hunger itself doesn't tell the story. It's how society perceives it.

The zeitgeist matters more than whether or not everyone in society can eat, and you can change the zeitgeist with propaganda.

When you're truly hungry, nothing is beyond reproach.

When you're truly hungry, you can't plan a revolution. Anti-government efforts are generally spearheaded by groups that are fed, connected, and have the incentive to incite rebellion. It's more Navalny and less Oliver Twist. This means that both pro and anti-government groups will be engaged in a similar recruitment effort. The two groups will have competing accounts of why the hunger is occurring, complete with different evidence regarding the magnitude of the issue, the source of the issue, etc. Hunger doesn't short circuit that process, and propaganda doesn't lose it's force because it's a more persuasive and simpler motivator than, say, discontent over tax burden shifting or some other policy point.

It's worse than that; within a few generations our linguistic and biological systems will begin to diverge under conditions with little cross-pollination and different selective pressures. We will become aliens in the sci-fi sense very rapidly if we attempt to create a foundation-like diaspora of settlements.

I disagree. You can escape a disease, even during a global pandemic. And not every person that got COVID was on a ventilator or even felt that bad.

Propaganda works.

The knowledge worker class often believes their training will afford them some level of protection against it. Even then, with those warding effects, they're still susceptible. Consider further that most people in society are significantly less educated or trained in epistemological functions than they are - a large portion of society is defenseless against a liar with a megaphone.

Propaganda won't contest that starvation is occurring. It will claim that the reason for the starvation is a specific foe, internal or external e.g. It's China's fault we're starving or the immigrants have caused this food security crisis and once they're gone we'll have enough food for our own people, etc. They'll workshop and see which ones poll well, then run with the talking point that seems to perform best.

Since the government harnessing that discontent has no real desire to fix that problem, all they need to do is maintain the perception that they're the solution, while not addressing the problem itself.

I feel like this was more accurate a long time ago when the first rounds of YCombinator hopefuls were all piling in here and nerding out. The vibe, tone, and content has dramatically shifted towards the finance and ambition side of tech over the years.

Everyone here wants to look like the next Steve Jobs in front of YC, no one wants to look like Steve Wozniak.

I think it's interesting that this wasn't the vibe that was here around the time of the first few YC cohorts. Everyone posting here was chatting as if they wanted to embody the Richard Stallman/Wozniak prototypical hackerman. I think once YC grew and this became a place to network with successful industry insiders rather than tech savvy ultra-geeks doing it for the love of the game that the tone changed hard.

There were a lot of pre-internet 2.0 groups that were phenomenal in terms of competence density.

The first point I worry a bit less about but it does have moments when it's suboptimal - for certain specific discussions there's often a need for a more durable thread-space to continue discussion. Some of the heartbleed and cloudflare discussions, wherein there were ongoing developments day by day needed to be cut up into many threads and people discussing had to refer back to now dead-threads from earlier days.

As someone with a hard science background doing law, I agree with the second point. I agree and notice it fairly consistently where discussion moves into my areas of expertise. I feel like there's a lot of Bayesian overconfidence that bleeds into off-competence discussions on here. I think this fairly normal, where high-competence people are put into areas where they can't identify their own knowledge gaps.

I think Nobel disease is more of an apt moniker than the Dunning-Kruger effect to describe what happens here. People who are highly competent in some areas probably learn to have lower Bayesian uncertainty, so they speak in more confident terms and sanity check their own conclusions less.

If anything, they'll need to rely on more expertise now, so they can craft laws that aren't open to interpretation.

Every law is open to interpretation. If tech can barely secure the doors on machines that execute instructions near-flawlessly, you think we can construct flawless frameworks out of inherently ambiguous linguistic building blocks run and understood by deeply human executors? This just plain doesn't work when the rubber meets the road.

Someone's going to make a choice, and SCOTUS just decided unilaterally that it's going to be a body that hasn't been able to decide anything productively for a decade.

This isn't about creating better structures for the analysis of rules; it's about gutting the regulatory capacity of agencies.

Not really. When performing clinical trials, if you'd like to use the results of the studies in chinese/indian populations you'll need to prove bioequivalence in many cases, so you're going to need to collect a meaningful sample in the first place.

The reality is that most clinical trials aren't successes. If you can get a huge cohort of people for relatively cheap elsewhere, you can screen a lot of promising but doomed tests at a cheaper price point, then only re-create similar testing on the most promising candidates in your lucrative markets.

100% what I was going to say.

This is a just a misstatement. There's plenty of valid things to critique sama on, but this isn't one of them.

Kinda? Lawyers are myopic, vain, and we don't really do much to innovate.

We wanted to make sure there would be no cross pollination between legal advisory services and other professional services in most jurisdictions, but the only thing that division did was significantly restrict our ability to widen our service offerings to provide more value.

The end result is that we protected our little nest egg while our share of the professional services pie has been getting eaten by consulting and multi-service accounting firms for the past 20 years.

We should be seeing many people holding mortgages at HSBC not able to pay.

Not really. Lying about the source of cashflow doesn't mean the cashflow isn't real.

The end objective for a lot of these frauds isn't to sink the bank with fake loans. It's to launder money.

Vaccination generally uses an adjuvant to increase immune response to the target antigen in order to provoke a response strong enough to produce lasting immunogenic memory. Antigens alone in small numbers aren't enough.

Taking random adjuvants consistently after minimal exposure to environmental antigens is more likely to give you deleterious allergies or issues associated with chronic inflammation.

I think most people recognize that you don't get every job you apply for, but there are people who contact hundreds to thousands of organizations, apply with custom crafted letters and tailored resumes, put in thousands of hours of job searching time over months of time and get jobs well under the median of their cohort.

The assumption that other people's experiences are like yours is a key cognitive bias. A 90% fail rate means 10 offers after a hundred applications - that's an incredibly solid success rate. Now imagine some people have orders of magnitude more difficulty because their name is strange, or because their social class shows through in their dress or mannerisms.

If I can offer a comparison, most people on this forum have it made. It's like asking a group of models how difficult online dating is and having someone say "I can't fathom anyone having trouble keeping their weekends full of new people to date... It's just so easy!"

Also, I note that you seem to assume I'm talking about my own experiences; I'm not. I am a very well established professional in a lucrative field.

but does that make it surprising that people hate doing it?

When an activity validates you, it's tough to imagine other people having the opposite experience with it. Imagine every time you went to a round of interviews you were offered riches and status. It would be tough to understand why others just didn't muster up the gumption to go downtown and get what they're worth.

This is only true to a point. Evaluating incremental cost benefits on the basis of the delta of energy loss along specific lines ignores the state change that occurs when main trunk elements of the grid become lossless and energy generation and storage solutions can be deployed in a near-location agnostic manner.

As with all toy models being applied to the real world, there are important factors to model in that aren't immediately obvious.

Retail banking is low margin high volume, but a bank's CMG or IB divisions are NOT low margin operations. Trading desk margins vary depending on the strength of the group and the clientele. If you're a primary desk for a trillion dollar AUM entity, you make a very pretty penny off your Bloomberg terminal subscription.

The number of banks is not a good indicator of how much competition exists in a market. The elasticity of prices in that market is a better measure, but also has flaws. In general you need to look at how much margin a given industry is pulling in to determine how much leverage it has.

On the bank metric: There are plenty of quasi non functional FDIC insured banks that are used as vehicles for reverse takeovers to allow a market entrant to avoid the hassle of obtaining their licenses.

Additionally given the substantially increased variability within different European markets we'd expect significantly different competitive dimensions between regional and international tier banks.

Interac is a great example of the opposite. The government of Canada had to strong arm the financial sector to create a non profit third party to allow zero cost transfers and a robust debit payment system. The banks have been lobbying to dramatically increase interchange and transaction fees on Interac payments for years to allow for an expansion in lucrative 'premium' credit card processing fees.

Competition did nothing to stop industry alignment against consumer interests.

This is misleading. Banks have capital sufficiency requirements following the implementation of the Basel 3 capital sufficiency framework. This has created an international growing market for bulk capital availability and reinsurance services.

In short, due to international capital sufficiency regulations, there's now a market for excess regulatory capital. The attempt to prune low performing assets is due to a market as an alternative to low return on capital investments.

It's not because of charges and overhead.

Banks aren't worried about bad assets more than usual, they just have another arena to make money in now.

Your rebuttal doesn't really address the substance of my point, but to put a fine point on it, "It's a crime, therefore it isn't speech" isn't a good argument - if anything it demolishes your own position that speech should never be restricted by government.

The very substance of your original argument is that the power to restrict these things is inherently dangerous, yet you don't seem to apply that standard in current cases where the acts of communication are criminalized.

Accordingly, your original argument falls flat based upon your own argument; the government can criminalize some speech and you don't seem to have an issue with it where it's obviously pro-social to do so.

So, again, work a little harder on refining the original position - the absolutist approach of complete governmental restriction doesn't exist in real life.

If you want to try redefining speech to only include vocal discussion between two people (which very much is not what freedom of speech entails), you'd still fall flat on almost all of the fraud related restrictions mentioned above in my previous post.

Anyways, that's all from me on the topic, the substance of the reply ignoring literally every area of nuance and actual productive conflict is just nauseating. It's like watching some tech guy tell a banker how bonds should work.

This is a Motte and Bailey.

No it isn't. The motte and bailey tactic is when you advocate for a very broad position, then retreat to something more defensible. I've stated that the broad statement (that all censorship is bad) is false, so as a result we need to have a more nuanced discussion about when it's appropriate to exercise this power and how we can de-fang potential abuses of it. Your position regarding all censorship being bad is your bailey, which quite frankly, I've razed to the ground. Your motte is the position that censorship can be used by a government to manipulate their citizenry, which is true, but significantly more limited in application than the original concept that all censorship is antisocial.

I’ll engage with your second point only if you yield that the _only_ form of information that is acceptable for a government to censor is speech that is directly and explicitly calling for genocide.

I mean, I wouldn't concede that because I can think of a ton of situations in which banning speech is appropriate. We can start with the classic 'fire in a crowded theater', we can discuss automated scam-driven robocalls, we can look at email spam filter, content moderation on websites, restrictions on the spread of child pornography, etc. What about speech designed to disenfranchise people of their right to vote by lying about where they're supposed to vote? What about financial fraud or any form of censurable misrepresentation?

From a tech perspective, should you be able to DDOS domains without authorization? Does your

therwise we are talking about the “known bad information” from your first paragraph.

Does “bad” mean false? Or just malicious? Who decides it’s bad? How does it become “known.”

As a citizen, are you allowed to challenge the classification of “bad” or does challenging the classification of “bad” itself count as “bad information?”

Great questions. This is where real debates about how legislation and law intersect with censorship occur. Different jurisdictions have different approaches, each with their own strengths and weaknesses. None are perfect, but if you're actually interested in this topic, you can look up review articles looking at comparative law on the subject and I think you'll find a lot of substantive meat to chew.

Even the United States, which values freedom of speech incredibly highly in it's hierarchy of rights have a very large panoply of situations in which speech is restricted.

I’m suggesting that’s exactly what people are advocating for.

Yes, that's a strawman, which is why people who actually work on the issue don't take the argument seriously. It's the equivalent of someone outside tech saying 'you can't host a website because you can get DDOSed' then being mystified that the internet is still chugging along. The threat is real, but the value of taking action is important, and finding out how to mitigate the threat in real world is where the actual complexity is; identifying a threat everyone knows exists isn't.

Anyways, this is boring. It's the same 101 level discussion every time on this topic and it really gets tiresome.

When you trust your government with the power to control the flow of information, you trust them with the power to hide their own actions from you

I think this is as false dilemma - no one's advocating that any institution or system that can regulate the flow of known bad information can operate without any constraints, oversight, etc.

If we assume any exercise of power will ALWAYS be misused under any circumstance, then taken to the extreme we literally shouldn't let people exercise under the notion that they can and will use physical force to coerce people. So we can't run on that assumption; sometimes governments can use power over information to do good things, like prevent the spread of pro-genocide propaganda.

We need to have a far more nuanced discussion about whether or not THIS instance of censorship is more positive or more negative.

Not the poster you're replying to, but despite the fact that I think you're right, the mistake you've highlighted on is pretty reasonable given that there's people from many countries here and plenty of people default to C vs. F. No need to go on the attack to make a helpful correction. :)

I'll be over here cooking an egg on the sidewalk if you need me.

But is your worry for the authors, for other platforms, or for customers?

All of the above.

Don't take this as a slam, but given your responses in this thread, it's probably a good idea to get a primer about how anticompetitive practices work.