dumbfoundded

I created an ai cross-stitch pattern generating tool: https://stitchwink.com/admin

If you try to use image generation directly, you don't get real patterns. StitchWink makes real patterns that you can also edit manually or through text/images with AI.

I just tried a pure text: "Can you make a cross stitch of a very round and big tuxedo cat? He should be laying down and look very confident"

I got this one: https://stitchwink.com/share/7a9c5c69-b45a-4d4c-9fb0-2792189...

It definitely takes a few tries sometimes to get it right. I've been having trouble with a "brain in a jar" for halloween right now.

I haven't tried it in firefox. Thank you for finding that.

Sometimes it does take a couple tries to get a good result but I just tried my tuxedo cat (a close up of him sitting on my car) and it seemed to work pretty well: https://stitchwink.com/share/5bf3d6f0-f064-4476-83e6-f75c3a3...

It's absolutely wrong. We also have: Alexander Fleming, Howard Florey & Ernst Chain who contributed significantly to antibiotics.

You could even add politicians. How much wealth did FDR create?

It sounds like they gave you a feature for free you didn't want, and now are trying to charge for it. Very much a dark pattern.

Have you tried creating skills for your agents to follow your patterns?

Will organizations want to control their own proxy or use OpenRouter?

I'm working on Ito.ai : https://www.ito.ai/

It's Agentic QA + auto-provisioning sandboxes. Makes it plug and play to do code reviews that actually run your code instead of looking at it really hard. B/c the agents control all of the environment (ie running all of the services), it's able to collect runtime evidence about pretty much everything.

A couple open source examples: (Excalidraw) https://app.ito.ai/share/d1cb1475-fbe5-4c71-901b-409ba2aa6d6... & (n8n) https://app.ito.ai/share/bb7d73aa-fd08-482d-9938-87938e2a232...

If the question is storage, bitcoin itself provides a perfectly good mechanism. idk the exact costs but it'd be in the range of ~$0.45 to store a commitment. That's cheap enough to enable good users with small numbers of keys but also expensive enough to prevent spam. It's kind of the whole point of blockchains.

As for verification being expensive, it sounds like you don't know the actual costs. It's basically a hash. Finding the pre-image of a hash is very expensive to the point of being impossible. Verifying a pre-image + hash function = a hash is extremely cheap. That's the whole point of 1-way functions. Bitcoin itself is at ~1000 EH/s (exahashes per second)

Again, this isn't a technical problem. It's a coordination problem.

I don't think you're understanding how cryptography works. A commitment is basically a hash that is both binding and hiding. In this example it's probably easiest to think of it as a hash. So you hash your post-quantum public key (something like falcon-512) and then sign that hash with your actual bitcoin private key (ecdsa, discrete-log, not quantum safe) and then publish that message to the bitcoin network. Then quantum happens at some point and bitcoin needs to migrate but where do funds go? Well you reveal the post-quantum public key and then you can prove that funds from the ecdsa key should go there. From a technical perspective, this is a complete and fool proof system. DoSing isn't really a concern if you publish to the actual bitcoin network and it's impossible for someone to use up the key space (2^108 combinations at least).

The reason this is a dumb idea is because coordination and timing. When does the cutover happen? Who decides which transactions no longer count as they were "broken" b/c of quantum computing? The idea is broken but not from technical fundamentals.

It's an electron app but latency sensitive bits are written in rust: https://github.com/heyito/ito/tree/dev/native/audio-recorder

Corporations will require everything going through an LLM to meet company standards.

It's definitely a middle ground, but PR reviews, are not perfect. So it's easy to miss a lot of things and to have a lot of extra baggage. From reviewing code it's not always easy to tell exactly what's necessary or duplicate. So I agree, this is a middle ground of using LLMs to be more productive. Removing one bad line of code is worth adding a hundred good lines of code.

The difference between a government and a corporation is the ability to use violence. A government is just a corporation with a monopoly on violence (police, military, jails...). The structure of how people are organized is more significant. Are we talking about a dictatorship or a functioning democracy? Are we discussing a non-profit or a publicly listed company?

I'm not sure. Somewhere around 10^12 kg of initial mass would be evaporating today (1). So perhaps there is no meaningful minimum, only a minimum initial mass. If it's just about to evaporate, it could perhaps be arbitrarily small. Earth is ~10^25 for reference.

(1) https://en.wikipedia.org/wiki/Micro_black_hole#Expected_obse...

It's not a practical possibility. The black hole wouldn't last long and would be too small to actually absorb anything. It's the equivalent of asking if a nuke would set the atmosphere on fire.

Even a "large"ish primordial black hole would probably just pass straight through the Earth without anyone noticing.

Strange matter on the other hand...

Best analysis I've seen: "Balwani might have been perceived as someone “a little bit older and wiser” who should’ve known better."

https://qz.com/theranos-sunny-balwani-jail-time-elizabeth-ho...

Programmers just move up the stack with some staying behind to manage the edge cases and optimizations in the underlying tech. Same thing happened with the move to cloud.

It's unclear which jobs with be enabled with more productivity vs replaced completely.

Many other places have solved this problem. San Francisco has relatively low population density compared even when only looking at US cities. NYC is very livable.

It's in CA best long term interest to support as many people as possible. Remote work isn't going to solve the problem that someone who works at Walgreens cannot afford to live within 20 mi of the city.

I think cryptographic AI will become a reality. The use-case I was thinking is more of immortality/digitizing human consciousness. If you could be uploaded (like the show Upload), what would that actually look like?

Well, plain text representation would just be too dangerous. Companies could mine your consciousness, duplicate it at will or whatever else they wanted. It's a scary thought. FHE provides the solution.

This is already doable with most wallets today. Most wallets enable you to create 2^64 addresses from the same seed phrase. These are hardened and can't be linked together by just creating them.

So if Alice wants to send Bob an NFT, Bob creates a new address (recoverable with the same seed phrase) and Alice sends it there. Bob can then fund the wallet with tornado cash to use the NFT.

It's a stupidly complex way to achieve privacy and Tornado Cash is illegal. That's why we need private by default chains like Aztec & Aleo

College education is ~30x more expensive (1). Home prices (2) & Health care (3) are ~22x more expensive. Farm land is up 20x (4)

> The main[1] answer to your riddle is that the economy grew about ~6x faster than we have been mining gold. The dollar is closer to being worth 7x less than 45x less.

How are you measuring it? It's a circular argument if you measure it in dollars. If it measure it in anything that can't be made more efficient due to automation & offshoring, it's no where near a 7x decrease.

> [1] The secondary answer to your riddle is that late-night-infomercial-manufactured demand from goldbugs and other morons can easily raise the price of gold significantly above where it 'ought' to be. Beanie babies, baseball cards, etc.

Gold is simply a good that's impossible to mass produce with technology. Use land, housing, healthcare, education or whatever you feel is most representative. Using toothpaste and tv's for CPI is a bad measurement in the last 50 years, our technology for mass producing them has lowered the true cost.

(1) https://educationdata.org/average-cost-of-college-by-year

(2) https://www.in2013dollars.com/Medical-care/price-inflation

(3) https://fred.stlouisfed.org/series/ASPUS

(4) https://www.statista.com/statistics/196400/average-value-of-...

I agree that the Fed was right in letting these crypto scams fail. The problem inherent to the system is inflation. As the Fed expands its powers, it can avoid significant recessions/depressions but one day it won't work and the dollar will fail like every other fiat currency in the world has eventually failed. This system "working" isn't eliminating risk, it's polarizing it.

I wrote a longer comment a year ago but here's a piece: "The price of gold was $45/oz in 1970. 52 years later it's $1,800/ounce. That's roughly 7.6% a year or 45x increase. If you use the inflation provided by the government, CPI, (1), they say inflation is only 3.6%/year or roughly 7x since 1970. Obviously we have a discrepancy. Is the dollar worth 45x less than 1970 or 7x times?

When we look at prices of things like education, housing, and healthcare, the 45x number makes a lot more sense. Education has 30x in price over the same time period (2). If you're comparing prices in dollars, it feels like education got really expensive compared to the basket of goods the BEA tracks but in reality, education requires less gold than it did in 1970. Our incredible supply chains and manufacturing automation have lowered most consumer prices such that we don't really notice inflation but when you look at things that can't get much cheaper like housing, healthcare, education, asset prices of all sorts, you can't miss the fact that they correlate more closely with gold than the USD."

(1) https://news.ycombinator.com/threads?id=dumbfoundded&next=29...

Monero & Zcash are the best examples right now but like bitcoin, they are volatile and not practical for actual commerce outside of use-cases (like crime) where it's worth the cost.

Stablecoins are a large use case in crypto now but there isn't a private stablecoin yet (I'm ignoring all SGX based technology). A private stablecoin, fully backed by audited bank reserves would enable something like a private Venmo. This private Venmo would know how much you brought in and took out but would have no idea what your transaction history looks like. I think in the next few years we'll see a private Venmo.

I responded to: "Name a financial firm that had a significant negative impact on consumers?" I didn't respond to which financial firms stole money from customers. What SBF did was more like Bernie Madoff.

The only reason we didn't have bank runs and frozen accounts is because the Fed stepped in to provide liquidity to the whole market. The banking system would've collapsed similar to the Great Depression without such action. Lots of people lost their savings in the Great Depression.

It's still useful for now, but it will get less and less so. The tornado cash dev is in jail for writing code that made such efforts easier. Since coins can be marked by "good" actors as having been involved in illegal activity, they can't ever be used again.

Let's say a US company gets ransomware attacked by a Russian national and demands payment in bitcoin. US company pays in bitcoin and later reports crime. All accounts get flagged. Anyone helping that address offramp the money will be in trouble. So there has to be a Russian exchange dealing with Russian banks that's creating a track record of provably aiding in crime. It's either going to be such a small amount no one cares or the Russian companies involved will be sanctioned.

Because of the public permanent record, the utility of ETH and BTC are decreasingly useful. Things like instant international settlement, fast payments, and the alike can always be better done with tradfi. Privacy and avoiding regulation was the only real utility ever.