TonesTones's recent activity
-
Comment on The Donald Trump US administration accidentally texted me its war plans (gifted link) in ~society
-
Comment on There are two kinds of credit cards in ~finance
TonesTones LinkThank you for posting this. I found the distinction between transactors and revolvers pretty useful and will keep it in mind. I think the article misses the truth behind revolver cards, though. In...Thank you for posting this. I found the distinction between transactors and revolvers pretty useful and will keep it in mind.
I think the article misses the truth behind revolver cards, though. In particular, the article seems to suggest that the poor are somehow subsidizing the rich here.
The people swiping their cards to pay for food and gas are also paying for wealthy cardholders’ upgrades to business class.
“When you talk to rich people who pay off their balance, they think that credit-card companies are losing money on them, and they’re the ones subsidizing the people who carry a balance,” Klein explained. “It’s the exact opposite.”
I do not know what "the exact opposite" means here, but I guarantee the credit card companies have internal calculations suggesting they are expected to profit off of each of their customers. It's just that the profits are made in different ways.
Everybody pays more for their goods; only fancy cardholders get juicy perks and cash returned to them.
Accepting a basic credit card costs perhaps 1.5 percent, and a fancy rewards card as much as 4 percent.
Fundamentally, the upper class (not the ultra-wealthy worth billions, but the high-value employees worth millions) drive the economy. Since they spend the most, exclusive high-value credit cards have much more negotiating power on fees. Also, high-end retailers are more likely to bear those high fees without complaint since they can pass them on to the consumer.
Transactors get much better perks because they spend a lot more. If you spend $200,000 with your credit card each year, credit card companies make between $2,500 and $7,500 (more if you have a "fancy" credit card). They are incentivized by competition to give some of that back to you so you don't switch to a different holder.
[Transactors] travel often, allowing credit-card companies to make lucrative deals with airlines and hotel chains.
Credit card companies can also incentivize spending with certain companies and become advertisers in a sense.
The industry argues that swipe fees, not interest and penalties on poorer customers, finance its cash-back and travel programs, Ted Rossman of Bankrate, the financial data and news site, told me. But he does not “totally buy that.” It’s all “part of the calculation.”
I'm remain convinced that if the "industry" found it was losing money on its highest-spending customers, it would raise its fees or lower its rewards; if customers left, those companies would make money!
Revolvers are subprime borrowers who use credit cards as a payment tool and as a short-term loan, to cover surprise expenses and groceries the week before payday. Such customers tend to take out no-frills cards, without lavish cash-back rewards and travel points.
The hard truth is that high-quality loans are only ever afforded to the wealthy with good credit, and this will always be the case so long as the loaners are profit-seeking. These consumers would not be able to afford their payments without credit cards, so they use them. Credit card companies try to extract as much value as possible before selling off those subprime loans to a debt collector if the balance is deliquent.
These loans and their interest rates are extremely predatory; these families' desparation affords them no negotiating power. However, if the predatory loans were made illegal, companies might just stop offering them, depending on their risk analysis. The worst case scenario is that families go to mobs and under-the-table loan sharks for money instead of banks. At least banks don't retaliate with violence.
It's also true that some of these families would be able to escape the cycle of poverty without such high interest rates. This is one reason why unconditional cash transfer is becoming a popular form of charity. I have not looked into it, but I wonder if a non-profit could try setting up credit cards that are less predatory as a form of good.
Naturally, they would have to run at a loss, and "we're a credit card company" is a pretty bad advertisement for a charity, but I wonder if it could work.
-
Comment on What programming/technical projects have you been working on? in ~comp
TonesTones LinkI have been learning Rust; I've done many of the introductory exercises I can find online and am nearly done with the Rust book. Now, I'm trying to build mimics of some command line tools so I can...I have been learning Rust; I've done many of the introductory exercises I can find online and am nearly done with the Rust book.
Now, I'm trying to build mimics of some command line tools so I can get a better grasp on the language. I'm starting with
lsand a subset ofglob. I might continue with hecto once I've finished those.- I'm quickly beginning to understand why so many programmers are Rust enthusiasts. The program structure and type system reminds me of functional programming in OCaml. I'm noticing that the borrow checker memory management system forces me to use strange programming paradigms sometimes, but usually they are equivalent or better than my first instinct.
- I have a sticky note with "Is it a move, copy, or borrow?" on my desk. It's really weird to think about memory this way instead of "stack or heap?". I think I like it, though.
- I love the way modules are structured, especially the way you can write inline modules. I don't find myself using inline modules very much, but I can imagine such a structure is very helpful when writing good, delineated libraries.
- The biggest shortcoming is the complexity. For example, good Python code can just read as a description of what it does. Sometimes in Rust I'm writing idiomatic code that isn't immediately obvious to me why it works or not (as a reader). Comparing it to Python is unfair, though; to me, C feels more readable right now, but once I get more comfortable with Rust, that might change.
Overall, this has been a good thing to occupy my free time. Definitely will be using Rust over comparable low-level languages in the future if I can.
-
Comment on How hard would it be to learn to code a Discord bot? in ~comp
TonesTones LinkOne of the CS educational teams I was on made a Discord bot the final project for one of our introductory programming classes. I think it’s a great first project. We had our students use...One of the CS educational teams I was on made a Discord bot the final project for one of our introductory programming classes. I think it’s a great first project.
We had our students use discord.py. If you want to continue learning Python, you can use that library to do most everything you want.
The other commenters suggested Javascript; I don’t know enough about JS to know if it is a good first language, but learning multiple languages early is pretty much always a good thing (so you don’t attach the syntax to the logic too much in your head).
If you end up using Python, feel free to DM with questions!
-
Comment on Generative AI tool marks a milestone in biology - Evo 2 can predict the form and function of proteins in the DNA of all domains of life in ~science
TonesTones LinkAt first glance, I wasn’t sure what to think: it’s a transfomer model that can predict nucleotide tokens instead of word tokens but still hallucinates sometimes; the purpose of that seems dubious....At first glance, I wasn’t sure what to think: it’s a transfomer model that can predict nucleotide tokens instead of word tokens but still hallucinates sometimes; the purpose of that seems dubious.
Then, I saw this quote.
The model is actually very good at distinguishing which mutations are just random, harmless variations and which cause disease.
To me, the kicker seems to be this differentiation. I’m not a biologist so maybe I’m wrong and DNA prediction is helpful, but I can absolutely see this being used to generate hypotheses to test with CRISPR experiments.
Cool stuff.
-
Comment on If you could go into hibernation and wake up in the future, would you? in ~talk
TonesTones LinkThis is a profoundly difficult question. My instinct is no, since I think trying to minimize regret is a pretty good heuristic for decisions, and I think with my current mindset, I could regret...This is a profoundly difficult question.
My instinct is no, since I think trying to minimize regret is a pretty good heuristic for decisions, and I think with my current mindset, I could regret going into hibernation and missing something. I do not think I would regret never hibernating.
The hard part is the possibility of hibernating until life extension technology is available. I think it’s highly likely humans eventually achieve indefinite lifespans unless we somehow cause ourselves to go extinct.
I would have to spend a long time thinking about this, but if I were to do it, it would be within the next year. I’m still relatively young and my brain is still somewhat malleable. I can probably adjust to a totally different culture right now, but this would be less true as I age.
I would also definitely ask to be woken up at regular intervals. Just in case I want to change my mind based on changes in the world.
My gut still says I wouldn’t do it. I don’t have a rational reason, and it’s not a fear response. I just don’t think I would.
-
Comment on Party City | Bankrupt in ~finance
TonesTones Link ParentAs far as I understand, private equity theoretically has no incentive to bankrupt a company they purchased, even with a leveraged buyout. They would much rather own a profit-generating business...As far as I understand, private equity theoretically has no incentive to bankrupt a company they purchased, even with a leveraged buyout. They would much rather own a profit-generating business than the temporary gains from the bankruptcy.
The biggest issue I see with a leveraged buyout is that a private equity firm also has no incentive to stop the company going bankrupt. If their company declares bankruptcy, the assets of that company just disappear. The equity firm loses nothing except the debt+asset bundle (which, upon bankruptcy, is a liability). All the risk has been shifted onto the lender.
This means there is a strong incentive for shareholders of a dying company to sell to private equity. The shareholders want to see maximum gains, and they can sell to private equity knowing that private equity will pay the full price even if they know bankruptcy is inevitable. That’s why this story is so common.
So who is lending? This private equity deal went through pre-COVID, so maybe the organizing banks were just naive about the future of brick-and-mortar.
However, the frequency of leveraged bankruptcies makes me think there’s some clever accounting going on from the lending side. I suspect there’s a tax loophole where the actual risk is being shifted onto the government from the lending banks. -
Comment on Anyone interested in trying out Kagi? (trial giveaway: round #2) in ~tech
TonesTones Link ParentSent.Sent.
-
Comment on Anyone interested in trying out Kagi? (trial giveaway: round #2) in ~tech
TonesTones Link ParentSent.Sent.
-
Comment on I didn't want to pay for an RSS newsletter email service so I built my own in ~comp
TonesTones LinkI enjoyed the blog post. I always like seeing small projects like this as a small act of rebellion against a host of subscription services. An interesting take that I definitely disagree with. I...I enjoyed the blog post. I always like seeing small projects like this as a small act of rebellion against a host of subscription services.
If an LLM wrote every line of your code but you've reviewed, tested and understood it all, that's not vibe coding in my book - that's using an LLM as a typing assistant.
An interesting take that I definitely disagree with. I could write a substantial bit about why I disagree with it, but I kind of want to see what people think if I just provide the following.
If an LLM wrote every line of your documentation and instructions but you've reviewed, tested and understood it all, that's using an LLM as a typing assistant.
If an LLM wrote every line of your blog post but you've reviewed and understood it all, that's using an LLM as an typing assistant.
If generative AI drew every part of your profile picture but you've reviewed it all, that's using generative AI as a pixel coloring assistant.
-
Comment on You are witnessing the death of American capitalism in ~finance
TonesTones Link ParentYou’re absolutely right: nobody can agree on what capitalism means. My (loose) definition is one where one can own and freely trade shares in corporations. (I think) Any economic structure with...You’re absolutely right: nobody can agree on what capitalism means. My (loose) definition is one where one can own and freely trade shares in corporations. (I think) Any economic structure with that capability approximately reduces to the current global economic structure, and any structure without it seems meaningfully different.
Hopefully, my perspective makes sense with that context. I certainly believe market economics and ownership are broader than just capitalism, has persisted throughout history, and will persist throughout history.
-
Comment on You are witnessing the death of American capitalism in ~finance
TonesTones Link ParentI think your comparison to feudalism is apt. Feudalism was clearly fairly stable as an economic system; it persisted over at least a dozen generations if not more. I think the difference between...I think your comparison to feudalism is apt. Feudalism was clearly fairly stable as an economic system; it persisted over at least a dozen generations if not more.
I think the difference between feudalism and capitalism is that feudalism was almost entirely a function of force. Lords controlled the land by paying armies with the rewards of that land (at least in my simplified understanding). States maintained power if the kings politically aligned themselves with enough lords.
Our modern global political system is not. Wealthy people do not pay private armies. Money is more efficient than force. States generally have a monopoly on violence, and so states have a disproportionate level of power. The courts resolve disputes because of that monopoly on violence (feudalist courts are more comparable to modern arbitrators).
Our modern political system is also not necessarily capitalist. The USSR had and the CCP has substantially more control over the allocation of resources than any private oligarchs. Legal ownership of companies (the definition of capitalism) is useful, but it’s just a legal concept and not enforced unless the state wants it to.
Why? Because we want it to?
I do not think capitalism is nearly as stable as something like feudalism because of the abstraction of force into the legal system. I also think historians would probably argue feudalism was not as stable as you describe, but I’m not educated enough to make that argument.
At the end of the day, power structures are just methods of human self-organization. Viewed through the lens of evolution, all we can really claim is that our current system is the most stable we have right now, but if a better system emerges, it will quickly outcompete other systems. We constantly live through this; the Cold War was the competition of the USSR and US, and now we’re likely seeing a transition into a new competition, with (at least) China and the US playing major roles.
-
Comment on US Department of Justice again files demand to break up Google’s search monopoly in ~tech
TonesTones LinkGoogle is a hard company to break up. An anti-monopoly ruling, while probably justified, would be truly unprecedented. This is because, as other commenters here point out, Google really doesn’t...Google is a hard company to break up. An anti-monopoly ruling, while probably justified, would be truly unprecedented.
This is because, as other commenters here point out, Google really doesn’t have any true monopoly. There exist competitors in search, competitors in video hosting, competitors in web browsers, etc. I personally only use Google Maps and Youtube. This is just because Apple Maps has various UI idiosyncrancies that really bother me, and I use Youtube without cookies as effectively a video search engine.
Google just makes the most user-friendly product in every single category because of their integration.Their integration also gives them their real monopoly: ad services. If you want to pay for ads on the web, you are going through Google. This monopoly tax surely increases prices for consumers everywhere.
How do you break Google’s ad services up? You can separate ad services from the products, but then you literally tear all the product monetization from the products and the data collection from the ad services. It’d be a death blow.
You could split the ad services into several companies, but then the product side would just build out an ad service that can outcompete the rest.
In my mind, it’s hard to actually break up Google in a way that breaks up the monopoly and also doesn’t kill Google’s business model. (That being said, it’s probably ideal for the selling of personally target ads based on web activity to not be a business model.) That makes, in my mind, this anti-trust legislation unprecendented.
I doubt federal legislators have a nuanced enough picture of the tech landscape to be able to solve this problem effectively. I consider myself pretty tech-familiar and I really struggle to develop solutions.
The ones best equipped to design a transition to a healthy and competitive business plan are employees and executives at Google. And, yeah… -
Comment on I used to teach students. Now I catch ChatGPT cheats. in ~tech
TonesTones LinkI'm surprised by the number of responses arguing that this is a failure of education. Many commenters here seem to think that the professors need to adapt their curriculum to better fit the age of...I'm surprised by the number of responses arguing that this is a failure of education. Many commenters here seem to think that the professors need to adapt their curriculum to better fit the age of AI and to start testing things that AI cannot do.
The comparison to "you won't always carry around a calculator in your pocket" from @ThrowdoBaggins is poignant, but also important in where it falls short. We still need to teach basic arithmetic to children so they have some intellectual leverage with which to grasp the stronger concepts. We still need to teach basic reading and writing skills to students so they can learn higher-order skills. You cannot expect someone to become an architect of software systems without first learning how to program.
AI is fundamentally different than previous in that in can do most introductory-level work at a level that most students can't; in other words, LLMs are smarter than most students. If students learn the LLM-first workflow, they literally will not have the ability to adapt when LLMs fall short.
Part of my educational research was in designing programing assignments that LLMs struggled with that were still accessible to introductory programmers, and we came up with lots of introductory coding assignments that all 2023 LLMs consistently struggled with, even with good prompt engineering. Then GPT 4o came out and one-shot all of them.In my mind, there are two reasonable conclusions from the existence of systems with increasing intelligence that can complete student assignments. Either (A) humans cannot use LLMs to do higher-order tasks at work, in which case LLMs are actively harmful towards education, or (B) humans can use LLMs to do those higher-order tasks, in which case we are rapidly moving towards a time when most humans trained in white-collar language-based work are unemployable (since what's the human middleman needed for?).
My perspective is that (B) is what will hold given to what extraordinary extent AI researchers are using AI to make novel developments (and that we've only just begun using RL to actually improve LLMs). If that's the case, then we really will need to rethink education. -
Comment on Microsoft reported to be sharply reducing planned data center investment worldwide in ~tech
TonesTones Link ParentDeepseek did show more can be done with less with various optimizations, but as long as the scaling heuristics apply (those heuristics claim you get some predictable % return on capability with...Deepseek did show more can be done with less with various optimizations, but as long as the scaling heuristics apply (those heuristics claim you get some predictable % return on capability with extra compute), marginal economics state those optimizations actually incentivize investment instead of deincentivising investment.
For a basic example, if you can invest in some asset A for which $10M turns into 1000A, and your scaling heuristic says that for every 2x investment you get 50% more A, your payoffs look as follows.
- $10M = 1000A
- $20M = 1500A
- $40M = 2250A
- $80M = 3375A
- …
Now say a company shows you can get 1000A with $5M. As long as the scaling heuristics hold, you have new payoffs.
- $5M = 1000A
- $10M = 1500A
- $20M = 2250A
- $40M = 3375A
- $80M = 5062.5A
- …
If you decide that the $40M for ~1500A is worth it, but not worth it for ~1000A, in the first scenario you spend only about $40M, in the second scenario (after a company does more with less), you want to spend $80M.
It’s my understanding that these scaling heuristics (not with my numbers, naturally) are commonly accepted in LLM research, with your asset being some “intelligence” and money being some combination of compute and data. You might see them called scaling laws, but there’s no reason (yet) why these laws should hold forever, so I call them heuristics.
My point is that this is likely not a response to Deepseek, since Microsoft is theoretically incentivized to further maximize AI spending with Deepseek if scaling heuristics hold.
It’s possible Microsoft has evidence these scaling heuristics fail? That would cause ripples across the AI community if true.
My personal speculation (assuming the author is credible) is that companies see the turbulence of the new administration and their economic decisions, and are choosing to reduce their financial exposure. (Go lean before going bust.) -
Comment on Inheriting is becoming nearly as important as working in ~finance
TonesTones Link ParentContext: US I’m wondering what your alternative to ownership and inheritance is. The intuitive is to have the state take property and sell it off to fund social programs. Sounds great on the...Context: US
I’m wondering what your alternative to ownership and inheritance is. The intuitive is to have the state take property and sell it off to fund social programs. Sounds great on the surface, but who are they selling the house to?
Perhaps the highest bidder, like a corporate real estate investment fund.
Doing away with it would level the playing field, increase social mobility, put a damper of the concept of "property as investment"…
No, it wouldn’t. It would actually reinforce those properties. Now, instead of having to just, you know, convince their parents to pass down property, your average middle class American has to outbid a Blackrock property investor to keep their family home.
Property as investment isn’t what most families are doing. Most families want a place to live. Property as investment is a corporate phenomenon.
I really can't think of anything but benefits for the average person.
I can’t think of anything but harm for them. Social mobility from the lower to middle or upper class (get a good education and job) is much easier than mobility from the middle and upper classes to the capitalist class (start a company that sells for 9 figures).
Getting rid of inheritance would entrench legal corporate ownership, which (in the US) are legal persons who do not die. Inheritance allows people to not sell family assets if they don’t want to, or to choose who to sell them to.
-
Comment on Algorithmic Complacency: Algorithms are breaking how we think in ~tech
TonesTones LinkI think the creator makes some salient points. I agree that trying to search for what you want to see is good “exercise” in trying to filter bad information from good information. I also believe...I think the creator makes some salient points. I agree that trying to search for what you want to see is good “exercise” in trying to filter bad information from good information. I also believe in the idea of “algorithmic complacency”; I think it’s common to just kind of accept whatever the algorithm feeds you because it is easy.
However, I’m going to push back a bit on “this is breaking our brains”. I do not think the author makes a compelling argument that this is at all a new phenomenon. In fact, they briefly touch on the snarky “I googled it” response on forums, which I think is evidence for the opposite.
Someone with enough technical knowledge to join a forum on the early Internet still didn’t think to google something. If you assume the “complacency” around searching for information is new, this should be really surprising, since you’ve already conditioned on being able to access the Internet, use a web browser, find a forum, sign up for that forum, and post. Googling should be really easy.
I think humans, in general, tend to just gravitate towards what information is easy to access. In fact, I bet the algorithmic feeds led to increases in the userbase, because before those feeds, people who did not want to search would just avoid the platform entirely. Pre-algorithms, Youtube had substantially fewer users.
So, I don’t buy that algorithmic feeds are breaking our brains in the way the creator describes. There’s probably strong arguments that social media is more harmful than the pre-Internet ways of wasting time (like watching TV on a channel or reading a magazine in your house or listening to the radio or just being bored), but that is pretty explicitly not the argument the author made: the comparisons being made are between pre- and post-algorithmic feeds online.
For those of us who do want to practice searching for information to satiate our curiosity, this video makes good points. I had already turned off cookies on Youtube (so every time I boot it up, it’s just a search bar), left every other platform with algorithmic feeds, and taken other steps to mitigate this issue before watching the video. I hope the video inspires more people to take their steps. I just think the segment of the population who actually want to make that change are, and have always been, in the minority. The change the creator noticed, in my mind, is the remainder of people joining the Internet and the products responding accordingly.
-
Comment on The terrorist propaganda to Reddit pipeline in ~tech
TonesTones LinkEven with the article’s slant, this looks like young Western activists coordinating; mostly because they are primarily communicating via Discord, something a Middle East state-sponsored initiative...Even with the article’s slant, this looks like young Western activists coordinating; mostly because they are primarily communicating via Discord, something a Middle East state-sponsored initiative would never do.
This article dances around a pretty critical conclusion, though. If this is something online activists can do effectively, I guarantee you there are state-sponsored groups trying to do the same thing. You do not need to censor speech if you can algorithmically silence it, and control over ideas is an authoritarian golden goose.
I stay off of Reddit, Twitter, Facebook, and the like. Articles like this one convince me that the most reasonable future for online discourse will be a combination of well-known, deanonymized public figures with microphones and invite-only, anonymous communities like Tildes.
You cannot let anonymous actors coordinate to algorithmically manipulate information without leaving the free exchange of ideas ripe for manipulation. At least news organizations have known figureheads who risk reputational damage with their words. Random anonymous accounts have nothing to lose.
On Tildes, even though it’s anonymous, the small size means I get a sense of who the regular commenters are and what their perspectives are. People still risk reputational damage, albeit to a lesser extent. There are still avenues for exploitation, but this is why you need a benevolent dictator (and the site’s reputation depends on their behavior).
I’m not exactly sure how to construct a healthy system of discourse without some trust factor.
-
Comment on Murdoch family US legal fight over trust could change the future direction of Fox News (gifted link) in ~finance
TonesTones LinkMy high school physics teacher had a saying that’s stuck with me. As I assume most people are, I’m occasionally envious of these wealthy familes with their exorbitant money and control. Think of...My high school physics teacher had a saying that’s stuck with me.
The smartest people are the ones who find a way to be happy.
As I assume most people are, I’m occasionally envious of these wealthy familes with their exorbitant money and control. Think of all the power they have! The good they could do!
I’m reminded by articles like this that so many of the people who win capitalism just find misery at the end of the road. I cannot imagine being so broken by the end of my life that I decide to go into legal battles with my own children, all for the sake of preserving a legacy. A human legacy, which in the grand scheme of humanity, is still pretty pitiful. I also can’t imagine fighting over control of a vast corporation with my siblings, all for the sake of political ends.
I guess their castles are made out of sand. Every desperate effort to make something solid that they can hold onto just crumbles. Not only has someone like Murdoch not found what is important; I’m not sure he even knows what to look for. A small part of me feels sorry.
-
Comment on Breakfast for eight billion in ~enviro
TonesTones Link ParentAbsolutely. I should make sure to say that I do believe in advocating for change. In my view, even the most left-wing politicians are not really “burn it all down” advocates until you get to...Absolutely. I should make sure to say that I do believe in advocating for change.
In my view, even the most left-wing politicians are not really “burn it all down” advocates until you get to anarchy or communism. Taxing the rich is a fairly safe intervention; it’s not like people won’t still want to be rich.
On the other hand, the policies implemented by the current admin, like firing huge numbers of federal workers and cutting off funding, reek of the mentality: “the entire current system is awful and corrupt, and we need to get rid of it, consequences be damned.” And I think many supporters of the admin would actually endorse most of that perspective. I’m sure there is corruption, irresponsible grants, fraud, etc. However, getting rid of the cancer should not also destroy the vital organs.
I wonder if this was a bad whistleblowing attempt by Michael Waltz.
I do not know how Signal works, but if this journalist hadn’t specifically outed that Waltz invited him, was there any way for the administration to figure out who did it? I think Signal wipes most logs of communication if you want. If so, the editor maybe shouldn’t have disclosed who invited him.
The fact that this thread exists at all is insane. But adding an experienced journalist to it seems very convenient; so convenient that it seems intentional. (Is there anyone in Trump’s administration with a similar name?). Waltz is Trump’s national security advisor; perhaps he understood how bad it was that these plans were being communicated over Signal but couldn’t push back against dear leader.
Edit: Seems like Waltz is an experienced combat veteran. He also worked for the Bush admin. It’s possible he’s not a diehard Trump loyalist, and he understood the danger these chats were putting soldiers in.
I hope we didn’t just lose somebody who would have stood up to Trump if he tried to use the military to perform a coup.