TonesTones's recent activity
-
Comment on A new era of intelligence with Gemini 3 in ~tech
-
Comment on Core Devices keeps stealing our (Rebble) work in ~tech
TonesTones Link ParentI think it is an open source thing because these kinds of disputes are precisely what non-permissive licenses are built to avoid. Each party is accusing the other of trying to hoard something...Maybe it's not fair to say that it's an open source thing, but it just seems to happen to stuff like this.
I think it is an open source thing because these kinds of disputes are precisely what non-permissive licenses are built to avoid.
Each party is accusing the other of trying to hoard something (data, an app store, code, etc.) for themselves. In any other business setting, the parties would get to do what they want with what they own. If there’s an ownership dispute, then it would be settled in court or maybe one party would buy the other out.
Instead, they feud publicly and say “no, I’m the one who can really be trusted by the community to maintain this software!” and various accusations get tossed around. It’s precisely because open source blurs the lines of ownership that this kind of dispute is even able to occur.
-
Comment on A Cloudflare outage is taking down large parts of the internet - X, ChatGPT and more affected in ~tech
TonesTones Link ParentI think being down at the same time as everything else is a positive from the business perspective. Imagine your service goes down. Would you rather your consumer think “Dang, this service might...When a service has downtime it sucks. When everything is down simultaneously it is worthy of some questions as to whether or not this is a good idea.
I think being down at the same time as everything else is a positive from the business perspective. Imagine your service goes down. Would you rather your consumer think “Dang, this service might be unreliable?” or “Oh, X and ChatGPT are down today; guess it’s just a bad day.”? Non-technical users aren’t going to see this as a shared point of failure, but rather business as usual.
It’s also not a bad thing from an employee POV. If a quarter of the Internet isn’t doing business, it’s a lot easier to take a few hours or a day off than if the one service you need has downtime and everything else is working.
-
Comment on That new hit song on Spotify? It was made by AI. in ~music
TonesTones Link ParentI’ve heard an interesting take on this. F.D. Signifier, in one of his videos analyzing rap (I apologize for not having the source), argues that record labels were actually helpful in how good...Now compare that to how they made music in the 70s, 80s, 90s. A lot of this can be blamed on the record labels I think.
I’ve heard an interesting take on this. F.D. Signifier, in one of his videos analyzing rap (I apologize for not having the source), argues that record labels were actually helpful in how good music was back then.
His argument goes as follows: because production and distribution was gatekept, artists needed to impress those with control over the production in order to have a shot at stardom. Now, artists and labels must compete with everyone in their bedroom making music. Production has been democratized. If one assumes that record executives have better taste than your average consumer, one can conclude that the democratization of production has actually decreased quality by virtue of competition. If they don’t make music that appeals to the lowest common denominator of the masses, someone else will, and they will win on the Spotify charts. F.D. Signifier argues that this is true of rap.
I just listened to Taylor Swift's latest album: overproduced, sterile, boring, unoriginal. Yet part of my friend group think it's great.
I’m not sure it’d be that hard to convince you that assumption is true.
-
Comment on Posts vs. comments. Where do you fall and why? in ~tech
TonesTones LinkAs a prelude, I think your choice of words is unfamiliar to me. I’ve understood posts to be all of the items I scroll through on the Tildes home page, but those aren’t responding to anything as...As a prelude, I think your choice of words is unfamiliar to me. I’ve understood posts to be all of the items I scroll through on the Tildes home page, but those aren’t responding to anything as far as I see. I think you are making a distinction between top-level comments/posts and replies to those comments, so that’s what I’ll assume, but please correct me if I’m wrong.
Overall, I think similar platforms with topic-comment-reply also have a substantial majority of top-level comments. I would not claim that Tildes is unique except insofar as there are fewer users in general, and so fewer replies in general.
Personally, I prefer making top-level comments because I often come off as critiquing or adversarial in my responses. I still need to develop my ability to think of and ask genuine questions about someone else’s opinions instead of immediately developing my own opinions as a default response. Therefore, if I have a thought, I often consciously choose to write it as a top-level so as to keep it independent and not feel like I’m trying to tell someone else they’re wrong even when we do disagree.
-
Comment on Denmark is facing one of the largest legal bills in English legal history, running into hundreds of millions of pounds, after the country lost a high-stakes tax fraud case in London in ~finance
TonesTones LinkGood thing I have today off work here in the States; this article sent me down a rabbit hole to understand how these traders were able to finagle this legal case. In the quotes, SKAT is the Danish...- Exemplary
Good thing I have today off work here in the States; this article sent me down a rabbit hole to understand how these traders were able to finagle this legal case. In the quotes, SKAT is the Danish tax agency and Sanjay Shah is the head of the arbitrage scheme.
The conclusion is not that SKAT should have identified that is was being misled — it is not a defence to a fraud claim to argue that the alleged victim should have spotted the fraud. The conclusion is that SKAT was not being misled by any of the misrepresentations it claimed were being made to it, which for the most part the judgment finds were not being made to it anyway.
The crux seems to be SKAT claims specific areas where it was being misled, and the court found that none of those claims held up. This is despite Sanjay and co. knowing that their refund claims were invalid, and submitting them anyway.
The judgment finds that none of the 4,000+ tax refund claims examined at trial was a valid claim under Danish tax law, which means that SKAT would have been entitled not to pay any of them.
The court does not accept claims made by Sanjay Shah and some others that they had a reasonable belief that their cum-ex trading models generated valid tax refund claims under Danish tax law[.]
This raises the question of what the actual arbitrage scheme was, since the court agrees that Sanjay knew the claims were invalid, yet submitting them anyway supposedly did not constitute misrepresentation to SKAT. I’ve spent about an hour and a half digging through the judgment. Naturally, I originally thought it was much more complicated but the scheme is simple.
“Explanation of dividend comepensation payments
Short sellers who sell shares cum-div and settle ex-div are liable to pay a dividend compensation payment from their custodians. …
Long buyers who buy shares cum-div and settle ex-div must receive a dividend compensation payment from their custodians. Such payments are documented as dividend credit advice notices..”
Here, cum-div means before a dividend payment and ex-div means after a dividend payment.
The judgment takes nine and a half pages (21-30) to explain the scheme, but while the technical details are difficult, I think the basis is simple.
Through some financialization or asset construction, a vendor V and a buyer B are able to gain short and long exposure, respectively, to a Danish company D. This is equivalent to the buyer B loaning shares to the vendor V except that neither of them actually own a share of D.
I believe that it is possible this financialization is done with multiple options, but the judgment doesn’t make any reference to them, so I won’t say for sure.
Simultaneously, a lender L borrows this asset representing a share of D from V and then sells it to B, and suddenly the credits/debits of all three companies are net-zero, though no debts have been settled.
Now, depending on the timing of these transactions, there may be a dividend payment owed from one of these lenders to another. This typically manifested during what the judgment calls the “unwind phase”. Essentially, they have to settle like the Three Stooges.
the stock loan recall and reverse equity trade would be internally settled at C4 by another balanced, share-less settlement loop, with a complex wrinkle on the ‘cash’ side of things.
Notably, whoever is owed this money receives it in cash, records it as a dividend payment of D, and submits it to SKAT. They then received the refund despite no shares of D having ever been owned by B, V, or L, so no dividend was ever paid by D (and therefore no tax was collected).
I have no clue why this is not considered fraud by the court. It’s manufactured in a bunch of legalese, but nobody seems to deny that the defendants purported to hold shares of D when they did not.
-
Comment on Who can name the bigger number? in ~science
TonesTones LinkThis Numberphile video about subcubic graphic numbers gets to big numbers using (in my opinion) a much more interesting method than just continuing to recursively repeat operations. I like the...This Numberphile video about subcubic graphic numbers gets to big numbers using (in my opinion) a much more interesting method than just continuing to recursively repeat operations.
I like the notion of ideas that we can define yet also defy the scale of numbers we can write down.
-
Comment on Playing with words in ~creative
TonesTones LinkPerhaps you’ll like Raddle from Tom Scott’s weekly newsletter. It’s kind of similar though more focused on word association than splitting up words.Perhaps you’ll like Raddle from Tom Scott’s weekly newsletter. It’s kind of similar though more focused on word association than splitting up words.
-
Comment on OpenAI moves to complete potentially the largest theft in human history in ~tech
TonesTones LinkI think the symbolism here matters a lot more than the financial or legal details. The articles speculates on various outcomes and explores what actually constitutes a public theft, and I...I think the symbolism here matters a lot more than the financial or legal details. The articles speculates on various outcomes and explores what actually constitutes a public theft, and I appreciate that extensive analysis. Still, this story isn’t really about what OpenAI’s theoretical profit will be or what the legal method was. The parties with controlling interest seem to want this to happen, and power will do what power will do. I’m not sure there’s a real dispute here.
To my understanding, OpenAI arose from the 2010s community who believed artificial intelligence was our greatest threat, and that it needed to be developed for public safety. The original structure was never designed to be robust—no social construct is robust against human action—but it did show a commitment to do things a bit differently than the social media and Internet companies dominating the landscape at the time. OpenAI was built not to do profit-maximizing value extraction, but OpenAI has continued to stray further and further from that mission in recent years. This move is really just the final public resignation that they really aren’t different, and that they plan to use their dominating market share and new technology to generate value for the shareholders, regardless of the cost.
That makes me scared, but more than anything, it makes me sad. The world would be a nicer place if power did not corrupt.
-
Comment on OpenAI says hundreds of thousands of ChatGPT users may show signs of manic or psychotic crisis every week in ~health.mental
TonesTones LinkMy feeling are mixed on this. I’ve been concerned about AI psychosis for a few months, and I think the risk that chatbots pose to human relationships is much greater than the risk of...My feeling are mixed on this. I’ve been concerned about AI psychosis for a few months, and I think the risk that chatbots pose to human relationships is much greater than the risk of superintelligence or mass unemployment, at least right now. I’m worried what it means for someone to choose to talk to ChatGPT over anyone else in their life.
However, it’s also worth noting that these statistics include people who are voicing these thoughts to ChatGPT without being prompted. I’m only just acknowledging there’s people who are unwilling to talk to anyone about their issues except for ChatGPT. I can see why an immediate response available at any time in a venue that seems private is very appealing to these people.
I despise that the technology is in the hands of a company that needs to deliver substantial returns to all of its investors. Yet it’s also true that their reach opens up an opportunity to help people or get earlier intervention in a lot of cases that would normally fall through the cracks.
-
Comment on What the hell are we doing with hierarchical tags? in ~tildes
TonesTones Link ParentI just want to say I’m grateful for all the tagging that you and other members do. I think I speak for many when I say that I do not care that much about the details as long as they are internally...I just want to say I’m grateful for all the tagging that you and other members do. I think I speak for many when I say that I do not care that much about the details as long as they are internally consistent, and you all have been doing a great job with that.
-
Comment on Changes to Advent of Code starting this December in ~comp.advent_of_code
TonesTones LinkI usually only make it 10-ish days anyway before getting burnt out or needing to focus on other things for the holidays. I hope that harder problems come earlier now! As an aside, I’m sure the...I usually only make it 10-ish days anyway before getting burnt out or needing to focus on other things for the holidays. I hope that harder problems come earlier now!
As an aside, I’m sure the advent of AI programming means that the global leaderboard was a lot less meaningful anyway.
-
Comment on 'This is definitely my last TwitchCon': High-profile streamer Emiru was assaulted at the event, even as streamers have been sounding the alarm about stalkers and harassment in ~games
TonesTones Link ParentI see. I didn’t understand that it was a different incident. My best understanding is if the stalker sued, the bodyguard would have a legal defense if Emiru could show that she was legally...This is not the situation in the video, which is clearly assault and battery, but a previous a case. A known stalker was loitering around her.
I see. I didn’t understand that it was a different incident.
My best understanding is if the stalker sued, the bodyguard would have a legal defense if Emiru could show that she was legally assaulted: in CA, this would require a visible attempt to touch or harm her. (In other states, I believe you only need to reasonably fear being harmed for it to be assault.)
I don’t think a stalker just “standing” would constitute an attempt to touch or harm, so I’d agree that the bodyguard would have a hard time making a legal defense.
-
Comment on 'This is definitely my last TwitchCon': High-profile streamer Emiru was assaulted at the event, even as streamers have been sounding the alarm about stalkers and harassment in ~games
TonesTones Link ParentObligatory I am not a lawyer, but I’m pretty sure that assualt and battery law is clear here in the U.S. (although each state has its own nuances). In general, one can use reasonable force to...Obligatory I am not a lawyer, but I’m pretty sure that assualt and battery law is clear here in the U.S. (although each state has its own nuances). In general, one can use reasonable force to protect themselves or someone else from imminent harm. Here, reasonable meaning equal or lesser force that the attacker is using, or force to the extent necessary to protect (I think, and I forget which one takes precedence).
I’m less sure about what the precendents and laws are for restraining someone else for a long period of time, but I am fairly confident those precedents have been clearly established for quite some time.
-
Comment on Silksong should’ve came with a guide from developers in ~games
TonesTones Link ParentI 100% agree. I think these design criticisms are reasonable, but it’s clear Team Cherry is full of big fans of metroidvanias (the original Metroid games suffered from this “problem” too), and...I 100% agree. I think these design criticisms are reasonable, but it’s clear Team Cherry is full of big fans of metroidvanias (the original Metroid games suffered from this “problem” too), and this is definitely an intentional artistic choice and not an error.
-
Comment on AI slop is killing our channel in ~tech
TonesTones LinkI’m not very convinced that the problem of “fake information” is that new to the age of AI. There are lots of great stories in the study if history about “facts” appearing from thin air. Searching...I’m not very convinced that the problem of “fake information” is that new to the age of AI. There are lots of great stories in the study if history about “facts” appearing from thin air. Searching for primary sources and then debating about the validity of those primary sources was my favorite part about the few history classes I did take.
I guess random stuff just being fabricated about today is new. Realistically, I think AI will just reinforce the practice of “trust trusted sources” that you see before the age of pictures and videos.
-
Comment on Microsoft is plugging more holes that let you use Windows 11 without an online account in ~tech
TonesTones Link ParentAren’t online installs generally preferable for businesses? I’m not in IT, but afaik online installs make it way easier to do business device management at scale. In my experience, businesses use...Aren’t online installs generally preferable for businesses? I’m not in IT, but afaik online installs make it way easier to do business device management at scale. In my experience, businesses use Windows because of these policies (and price), not in spite of them.
I agree that this is consumer-hostile, and I’ll never have a Windows machine as my personal computer. I’m just hesitant to say that this is corporations “putting up with shit” rather than Microsoft giving them what they want.
-
Comment on The hydrostatic paradox in ~science
TonesTones Link ParentI think the video was well-paced. I likely have a stronger background in physics than the average Youtube viewer, but I haven’t taken a fluid mechanics class so the second demonstration in the...I think the video was well-paced. I likely have a stronger background in physics than the average Youtube viewer, but I haven’t taken a fluid mechanics class so the second demonstration in the video was new to me. The video didn’t belabor one point but showed different applications of the same principle, which is a great educational technique.
I also think your perspective on education is a bit miscalibrated. Most high schoolers only ever make it to Physics 1, where force diagrams are first introduced. Many students in those classes likely couldn’t explain what’s going on here, especially in the second demonstration because of the implicit pressure gradient.
The video certainly is trivial to someone who has taken a introductory fluid mechanics class and is used to those forces, but I think that’s fewer people than you’d think.
-
Comment on Introducing Kagi News in ~tech
TonesTones Link ParentIt’s worth pointing out that they only take what is publicly available on RSS. They are not scraping past paywalls or on websites. So, I wouldn’t consider it stealing in the same way that lots of...since Kite/Kagi News is based on other outlets’ output, with articles authored by typically human and thus typically opinionated writers, I’m not sure how sustainable the project can be long-term. Who pays for the pre-LLM work?
It’s worth pointing out that they only take what is publicly available on RSS. They are not scraping past paywalls or on websites. So, I wouldn’t consider it stealing in the same way that lots of LLM scrapers operate (except insofar as one believes all LLMS are trained on stolen data).
I agree the funding pathway is not obvious, but there’s certainly an incentive to fund publications if you want to push a certain perspective available. A place like the Guardian believes in free access and reader donations, and they might be willing to make content available via RSS to push their own reporting over someone else’s.
It’s also worth noting that providing content (like a summary) via RSS might make your website more likely to show up in the cited sources of Kagi’s feed, potentially driving clickrate.
I believe that news fundraising is already a problem in the age of social media and LLMs without something like Kagi News, and I don’t think this makes it worse.
-
Comment on Spotify removed 75m spam tracks over the past year as artificial intelligence tools increase the ability of fraudsters to create fake music in ~music
TonesTones Link ParentAh, this explains some weird tracks I’ve noticed under artists I enjoy but clearly aren’t their music, especially on platforms that are not Spotify. I assumed it was just a case of shared artist...Ah, this explains some weird tracks I’ve noticed under artists I enjoy but clearly aren’t their music, especially on platforms that are not Spotify.
I assumed it was just a case of shared artist name and immature database practices, but track metadata fraud makes way more sense.
Verification does not seem that hard of a problem to solve, luckily.
I see the post-mortem impact, but doesn’t the bubble, while active, stimulate high job growth and high stock growth? People both lost and made retirements depending on if they (or their financial managers) made winning or losing bets. That’s just playing the stock market. People lost jobs they may never have gotten without the bubble. In the modern U.S., AI is pretty much the only short-term bulwark against a total recession considering the recent economic impacts of tariffs, deportations, gov. shutdown, etc. We should be doing worse, but the bubble just delays the consequences for a bit longer. I understand those consequences are real, but I think the economic picture is more nuanced.
AI is driving a lot of our economy in the short-term. I personally believe the long-term impacts of the technology on human beings (education, socialization, creativity, etc.) will be worse, but I wouldn’t consider those economic impacts (not directly).