Often I get a lot more out of some YouTuber/streamers's slightly scuffed cover of a song than the original studio recording. "Perfect" music often makes me feel nothing. It's the tiny individual mistakes and flourishes that make it easier to relate to the performer and music as a human - and that's what puts a smile on my face listening to it.
> In short, shouldn't the amount of money you expect for a fine be at least vaguely proportional to the harms caused?
No. Fines are punitive and intended for deterrence. You're thinking of reimbursements (cost recovery and damages). The latter is intended to be proportional to real world harm/costs, the former need not be.
A traffic law violation could cause no harm, but you'll still get fined for it.
Some jurisdictions even impose fines based on how deep the defendants' wallet is, calculating the amount that might serve as appropriate deterrent/punishment based on their financial situation. Fining a billionare $50,000 for literring will still not serve as a deterrent, but doing the same to a minimum wage worker would unfairly financially ruin them.
OK fine, it's about deterrence. Explain why are trying 100x more vigorously to deter the category 'advertising-based infractions' (which the majority of the 'victims' still don't even know happened, let alone experience any pain and suffering from) than 'negligently giving people cancer and killing them'?
This is about emotion and resentment about social media. I get it. I hate social media too. But it's absurd to levy such a fine for petty data misuse - especially considering this is just going to pad state coffers. The state wasn't harmed here! If you take the position that New Mexicans will never be the same again, then that $11,737 per New Mexico resident should go to the residents. All of it. Write them checks.
I'm trying to puzzle together the belief system that makes your comment coherent.
You seem to assume that money being transferred from Meta's into New Mexico state coffers is worse than burning it. Also did the people of New Mexico not elect their government?
Additionally, what instance of "giving people cancer" is not being pursued because resources were allocated to this instead?
Finally, fines must be high enough to have the intended effect of deterrence. If only absurdly high fines work, which appears to be the case in Meta's instance, then absurdly high they must be. If they continue that pattern of behavior while ignoring every fine, then the only alternatives would be either shutting Meta down for being a criminal enterprise and putting people behind bars, or throwing the idea of "rule of law" out the window. A high fine is doing them a favor.
> money being transferred from Meta's into New Mexico state coffers is worse than burning it.
I do believe that. I'd say that about any state or federal government though. Think of all the good that money could do in the hands of people. At least if you burned it, it would offset inflation for a few days.
> shutting Meta down for being a criminal enterprise
Again, interesting to suggest that this is the other fair outcome here for advertising shenanigans, but not for the many corporations who outright kill people every day like health insurers.
If we're starting out from different axioms, there's really no point for us to argue about about policy and outcomes. We'll have to align our assumptions first.
> At least if you burned it, it would offset inflation for a few days.
Yes. And keeping it in Meta's coffers might ultimately redirect economic output towards maintenance of Zuck's yachts.
> Again, interesting to suggest that this is the other fair outcome here for advertising shenanigans, but not for the many corporations who outright kill people every day like health insurers.
You're conflating moral and legal arguments and read things into people's comments that were never said - only as a setup for some whataboutism. For your second post in a row.
There's an interesting discussion here, but this is not how to have it.
I see where he is coming from. I believe his point is not that tobacco companies never paid anything (they paid over $206 billion in the 1998 Master Settlement Agreement to cover health costs, according to my quick searches). What he is calling out is the disparity in how the state punishes these harms today. The reason Meta's penalty hit $25 billion is not because ad-data misuse is worse than giving people cancer anyways, it's just a quirk of software-era consumer protection laws (which I should be responding to him, but hopefully he reads this comment). In New Mexico every single automated digital interaction was counted as a separate statutory violation which multiplies a basic fine into an astronomical sum. His point is that state governments are aggressively leveraging anti-tech outrage and these statutory math tricks to extract billions for state coffers whereas deadly industries like tobacco were allowed to negotiate structured settlements over decades while continuing to sell. I think... Well, this is how I interpreted his comments.
TL;DR: I think he already put it this way, but if Meta gets hit with $25B for ad-data tricks then why are we fining social media far more aggressively per user than companies that literally gave (and continue to give) people cancer? The answer is partly in my first paragraph but I still do not find it particularly reassuring.
It's amazing how the anti-Facebook hate is so strong that it's not possible to even post a measured take like yours without getting downvoted to death.
And I acknowledge that a fine of a very manageable sum turns a violation of a law into a cost of doing business, like the proverbial rich man's parking ticket just looks to him like a really not-bad deal for a plum parking spot. So, yes, it's tricky to use fines to rein in businesses. But it's still very odd/suspect that Meta is fined 25 billion dollars for this, while Equifax, which lost the SSNs and financial data of every household in America, and they paid a settlement of 575 million dollars. Why? Because Meta's unpopular. That's the only reason.
A fine that amounts to 1/5 of the entire GDP of New Mexico, or 75% of its government's annual budget, seems farcical to me. Unless of course New Mexico will give a one-time tax rebate equalling to 75% of tax revenue to its people, in which case it's still farcical, but at least it's helping people.
> Where's the proof? How do you know it didn't search, find, and discard?
They found their page being cited in various AI generated slop around the internet, including an acedemic paper, investment research, an Indian publication, etc.
Models have proclivities, and many people use the same model.
Imagine all books on earth were written by just 5 people - and worse, they could not remember what they had written yesterday. You would quickly get tired of reading the same flourishes and patterns over and over.
Whatever style models currently in use have, that's what people are going to be tired of, and it's what they're going to recognize as "AI".
The most difficult thing in doing it yourself is knowing what to ask for. I can ask my model to talk like a caveman or like a Baltic pirate. But asking for something more nuanced, in the range of regular speech, requires the correct lingo
I'm sure it's trivial for somebody who has studied communication or contemporary linguistics, just as it's trivial for me to ask an LLM for tenant-isolation using row-level security in an app
I gave Claude several pieces of writing from myself, from notes to code comments to iMessage messages with my wife[1], and asked it to write skills that match my voice. I also incorporated the humanizer[2] skill into it.
Worked like a charm. This is the prompt I used[3] to write a version for the company I own with my wife, if it helps anyone. The personal version, which lead to the `jamenizer` skill, is very similar, but has some personal details in it.
Yes, we talked about it, I explained to her what I was going to do, and she said she didn't mind. She doesn't work in tech, but we talk quite a lot about it, and she understands the implications.
Why do you assume I would do anything against my wife's will?
And I never said anything about "optimizing myself out of existence", this comment was still written by me and no LLM touched it.
But when writing comments on code or editing articles, reviews, and whatever long-form writing I give an LLM to review or edit, it's nice to know it'll keep my writing style, voice, and tone.
I'm already obligated to use LLMs at work, otherwise I might lose my job, so why wouldn't I want to have it write in a way I can actually understand?
What I find amusing is that tech used to look outside itself for perspective. We did usability, ethnographic research was a big thing, I have taken paper prototypes (a long time ago) to the bar and chatted up strangers about them for a beer.
The fact that you're doing that with your wife, with her consent, is no different. The real value is in keeping you grounded by someone outside tech.
> Yes, we talked about it, I explained to her what I was going to do, and she said she didn't mind. She doesn't work in tech, but we talk quite a lot about it, and she understands the implications.
Respectfully, that is how one talks about their pet or their kid but not their partner.
But this is not my beer, but by god this platform.
You could try taking a body of your own writing and asking it to characterize it/come up with a prompt (or fine tune if you can) describing your style and quirks.
This is usually what I've had luck with. I've given it samples of my writing and told it to emulate my style. It did help, but it took some charistics of my speech and cranked them up to 10. I guess in hindsight I could have mentioned for it to take my writing quirks it finds and only apply them in moderation.
Reading what I did get out of it was just a bit uncanny valley strangely enough. Very much had some "hey there fellow kids" energy.
You should play around more, it's flawed but more than any other tech it rewards exploration.
But asking for something more nuanced, in the range of regular speech, requires the correct lingo
I like asking models to write things using only the N most common words in a language. It's probably never perfect, but it understands the assignment well enough. from there I notice things I prefer about that constraint, and things I don't.
> I'm sure it's trivial for somebody who has studied communication or contemporary linguistics…
But you know what you like, and you know what bothers you, and the good news is that you don't have to one-shot a solution. You can create a skill and refine it over time. As you test and/or use it, identify copy you don't like and tell the LLM what you prefer instead, and ask it to distill and generalize that feedback into the skill.
> I know what food I like, but that doesn't at all mean I know how to cook it.
Right, but you know what you like. The point is that you can tell the chef "too spicy" and send back the plate until you get what you want. He can then remember that, since you're his favorite regular.
I just went through this with a release notes skill. I brainstormed with the LLM about what I wanted, edited initial results when I saw something I didn't like or could be made better, than asked the LLM to capture that in the skill. Now I almost never need to touch it.
That’s why Claude had a customizable style setting that let you just paste in example text to create a new writing voice. Not sure if they still have it or not.
You don’t need a built in feature though, you can just paste examples into the prompt and tell it to write in the same style.
> I predict by the end of the year they'll have a style/voice selector for responses.
Why would they add it now? This has been a solved problem since GPT3.5, just prompt it to write in a different style and/or never use specific phrases.
The reality is that for 99% of people and use cases, they just do not fucking care. They're either using chatgpt as a 1:1 chat interface where the AI style is fine, or they're sending off homework/generating reports to send to their boss/etc, which they know will never be read and just summarized by an AI anyway, so what's the point.
>I predict by the end of the year they'll have a style/voice selector for responses.
Style permeates all aspects of writing: word choice, sentence structure, literary devices, pacing, plotting, characterization, dialog, theme, etc. Its a culmination of these things that is reinforced over again every sentence, paragraph, and chapter for potentially hundreds/thousands of pages.
I don't see how a prompt could reliably "tune" all of those different variables over such a large context window without it feeling like a surface level facsimile at best.
Human language isn't verifiable. If I ask for something "dark and gritty", there are endless outputs that are indeed "dark and gritty" yet completely miss the style I had in mind.
Trouble is, it’s still the same model, just with a request to sound different. It can only go so far. It’s much like asking an author to imitate the style of another author. They might do a decent job of it but they’ll never match perfectly. And that’s for humans who can actually learn. LLMs can’t.
> My objection to the Trolley problem is that it assumes that people are interchangeable like ants and that the only thing that matters is "lifes saved".
That's very hard to do in a work environment where you have someone above you who doesn't understand or care about this. Instead of keeping things sane so they can run smoothly, you will be seen as slowing down progress.
If I work for someone else it's not my problem. I should point the problem out to them, but since it will hurt them first and foremost, it's no major concern of mine.
It's quite hard to rise above this stuff. The only time I was really able to do it was while contracting because I knew I would be gone in a short while. Even then I was usually fairly blunt about it.
That's a pretty tall assumption when every other customer is trying to copy their own data at the same time through the same tiny pipes and off the same spinning rust.
The typical customer also doesn't do their own backups, but rather relies on the infrastructure provider (that's a major selling point after all).
If we take the damaged Amazon data center in SA or the OVH fire as data points, the timeline appears to be measured in weeks and months, rather than hours.
In any case, I doubt that in 2 days most larger companies could even figure out where their data is and develop a strategy to copy it.
If you make 90k in Germany, you pay around 36k to mandatory (varying degrees) insurance/pension plus taxes if you're single. About 19.5k of that are actual taxes. If you're married/have children that total can drop to ~30k and will include health insurance for your entire family unit.
I just gave you the actual numbers. The same numbers you will quickly get by using any of the many available net income calculators for Germany[0].
If you have a different idea, feel free to break the 90k down yourself. Don't lazily push the work on other people, asking them to puzzle together where some neglected Wikipedia articles' table of magic numbers actually comes from. Hint: It's a number made up by an interest group called the "German Taxpayers Federation", representing mostly entrepreneurs and wealthy taxpayers. The number being opaque is intentional, because its purpose is to mislead and support their advocacy for tax cuts.
But by all means, don't let me stand in the way of your right to blindly believe whatever someone else wants you to. God forbid we input a five digit number into a calculator.
What will it take? A notarially certified copy of tax returns mailed to your address?
You have to add in all the other taxes as well. For example VAT, gasoline, and property taxes are significant. That's generally what tax freedom day is meant to mean - when an average citizen is done paying all taxes.
Edit: I was curious what the answer might be and couldn't find much. One source calculates the tax burden to be about 53% of income.
The link you provided is the web page of the interest group I mentioned. They will literally try any trick in the book to get that number up, including adding a bunch of things that aren't even taxes, a boatload of things they call "quasi-taxes" with numbers presumably pulled from their ass, and adding expenses your employer incurs only to subtract them again immediately. Anything to get that percentage number up.
But even they don't even call the resulting number "tax burden" - that'd be transparently stupid. They more generally call it a "burden ratio" (Belastungsquote).
Just to give you an idea of how ridiculous their calculations are: If my wages are 90k in Germany, they do their calculations as if I earned ~105k just so they can immediately subtract the entire 15k again and call it a burden (for expenses my employer pays on my behalf). Does this still sound like something you can easily compare one country to the next and draw useful conclusions from? Of course not. In the next country it might come from an interest group with a different goal, a government calculation, or a Fox News writer.
> You have to add in all the other taxes as well. For example VAT, gasoline, and property taxes are significant. That's generally what tax freedom day is meant to mean - when an average citizen is done paying all taxes.
What you're trying to get at - hopefully because the alternative is a nonsense number with no purpose other than being a fun fact supporting a discussion about degrees of individual autonomy in modern societies - is called discretionary income, and it's only meaningful when adjusted for cost of living and compared across similar living situations.
> If my wages are 90k in Germany, they do their calculations as if I earned ~105k just so they can immediately subtract the entire 15k again and call it a burden (for expenses my employer pays on my behalf).
Aren't you talking about social contributions that your employer is forced by law to pay on your behalf? You realize that if this was not in place, those money would go directly to you since your salary is whole sum your employer is paying towards your employment? Doesn't seem you grok it fully from your words.
I am not claiming if its good or bad (dramatic global financial changes will test viability of all these systems rather well in upcoming years, I'd expect many people will be unhappy when their naive views hit reality of numbers). I am outside of such system, in mostly private social system in Switzerland (aka I save for myself, mostly) and wouldn't want it any other way.
If that's the real wage, it should be used when comparing wages internationally. It's however not used as such, because trying to figure out what should and shouldn't be included in that number for each country will result in an even bigger headache. Do we add the average cost of employer-provided health insurance to US wages to keep the number comparable? What about those who self-insure?
In any case, the resulting individual "burden" is meaningless if you don't also quantify individual benefit. As an extreme example, imagine a country in which your income taxes pay for things such as transportation credits usable on gas, public transport and roads, while the neighboring country has mostly private toll roads and low taxes. Another example is health insurance. In Germany it is pretty nearly mandatory with most having some form of insurance, and total spending per citizen is around $9,000. In the US more than 90% are insured and spending is higher, but within the same ballpark.
What does the raw tax/burden tell you here in an international comparison? Fuck all. It's at best usable as a prompt for a philosophical discussion about autonomy, but often merely used for masturbation or making dishonest arguments designed to go over well in establishments specializing in alcoholic beverages - like those of a certain interest group.
> Do we add the average cost of employer-provided health insurance to US wages to keep the number comparable? What about those who self-insure?
Obviously yes, and obviously no. Just as we don't add the price of groceries or rent to wages unless the employer subsidizes those in which case obviously yes we must.
> What does the raw tax/burden tell you here in an international comparison?
It tells you how expensive it is to live there, relatively speaking. Obviously (as you note) that's only half of the discussion just as an expensive appliance might or might not be worth the premium to any particular individual. But you're wrong to dismiss it out of hand as you do. Just as it's disingenuous to dismiss the benefits of (for example) universal healthcare it is also disingenuous to attempt to derail discussion that focuses directly on total cost or effective tax burden or what have you.
What you are dismissing as a "nonsense number" of an "interest group" I would consider one of the most accurate and easily understood quantifications of total cost.
Imho, parent's point about attempting to quantify costs without balancing them against benefits is valid.
It's meaningless to say "Germans pay X, Americans pay Y" if the two receive completely different benefits in exchange for those rates.
Furthermore, the point that the German system would be paying to individuals if not to the government is somewhat disingenuous. There are a few options.
1. The government doesn't do it, so private options do at some cost
2. There's some government-private hybrid
3. It's not done at all
An example would be parental leave. In the US, it largely doesn't exist to the European extent. Employers don't offer it. You can't buy insurance for it. It just doesn't exist. That's something Germans are getting for existing in the German system.
Yes, as I acknowledged it is only half of the story. Certainly you need more information to make an informed decision about anything. But that doesn't make it meaningless. And generally where people level such dismissals there is conversational context being conveniently ignored (such as happened earlier in this thread).
I'm not sure what point you were trying to make with parental leave as an example. I completely agree with what you wrote but don't see it being at odds with anything said previously.
I think we all know that Germans enjoy substantially more free at point of service public services than those in the US let alone those living in a developing country. Obviously that costs something. Yet when someone brings up the price tag you get a bunch of partisan screeching about how prices are meaningless.
Prices _aren't_ meaningless. The US military (for example) is absurdly expensive. There are lots of things received as a result of that expenditure. Regardless, the relative cost is high and that's not meaningless to point out given that everyone involved in the discussion has at least some conception that there are important differences in the downstream impact.
I can spend that in whichever tax jurisdiction I want and the VAT rate might be whatever depending on what one buys. If you want to go that route you'd then also have to adjust for cost/quality of living to stay intellectually honest, which means you'd then have to come up with some statistical consumer basket. It's a stupid idea to even start down that route in what fits into HN comments.
It's called PPP and yes, it makes Europe look sightly better but it's still behind the US and that's for median household income not tech income.
I've managed staff all over the US and in Europe and have had access to large corporate pay scales across zones. The difference is so significant that CoL is a total rounding error for software developers.
You always have to cost/quality of living if you go that route. Because that's literally what taxes are for. Anyway I don't think it's fair to not include VAT. For every euro you want to utilize(modulo some exceptions) you have to pay it.
3€ fee on every line item for imports says no you cant'.
Yes the exact cost will differ depending on what you buy but that doesn't remove the cost, it just obscures it - which is the point of having so many different taxes.
reply