I think "prompts" are a much richer kind of intellectual property than they are given credit for. Will put in here a pointer to the Odd Lots recent podcast with Noetica AI- a give to get M&A/complex debt/deal terms benchmarker. Noetica CEO said they now have over 1 billion "deal terms" in their database, which is only half a dozen years old. Growing constantly. Over 1 billion different legal points on which a complex financial contract might be structured. Even more than that, the representation of terms in contracts they see can change pretty dramatically quarter to quarter. The industry learns and changes.
The same thing is going to happen with all of the human language artifacts present in the agentic coding universe. Role definitions, skills, agentic loop prompts....the specific language, choice of words, sequence, etc really matters and will continue to evolve really rapidly, and there will be benchmarkers, I am sure of it, because quite a lot of orgs will consider their prompt artifacts to be IP.
I have personally found that a very high precision prompt will mean a smaller model on personal hardware will outperform a lazy prompt given to a foundation model. These word calculators are very very (very) sensitive. There will be gradations of quality among those who drive them best.
The best law firms are the best because they hire the best with (legal) language and are able to retain the reputation and pricing of the best. That is the moat. Same will be the case here.
But the problem is the tight coupling of prompts to the models. The half-life of prompt value is short because the frequency of new models is high, how do you defend a moat that can half (or worse) any day a new model comes out?
You might get an 80% “good enough” prompt easily but then all the differentiation (moat) is in that 20% but that 20% is tied to the model idiosyncrasies, making the moat fragile and volatile.
I think the issue was they (the parent commenter) didn't properly convey and/or did not realize they were arguing for context. Data that is difficult to come by that can be used in a prompt is valuable. Being able to workaround something with clever wording (i.e. prompt) is not a moat.
Same. Never lived there- though almost moved in the 1990s- and now feel a pull to learn/feel the history. Did also just finish Grant's memoirs- and would strongly recommend Sherman's if you haven't read those, not only for the SF parts. Some of his letters are incredible and dare I say relevant today.
"We found strong correlations between greater consumption of fat and protein to lower anxiety and depression, while consuming higher percentages of carbohydrates was associated with increased stress, anxiety, and depression"
and
"We have shown that changing diet affects mood and happiness, that greater fat and carbohydrate intake is directly associated with anxiety and depression"
I am sure it is a fine paper and it is a preprint but a direct inconsistency in the abstract...
Your orthodontist is probably using an ass (term of art) specialized medical practice billing system. Not uncommon for smaller practices. Some of these are made by the dude down the block, kind of thing. Total shit. That will be the thing that is issuing the debit. It might be querying for the balance beforehand. The thing to do is find out what ass billing system they are using, might be associated with their ass practice management system, and then talk to those people. It might be stripe under the hood but stripe is not making this mistake.
That was another possibility I asked about but the bank said it's absolutely not possible to query the balance. Could have been misunderstanding what I meant, I guess, or just wrong - do you know for sure that it is possible?
Traditionally under ACH I don't believe balance inquiry is available. But if you did a Plaid like thing to "connect" your card, your balance is available to payees.
Yes. It allows criminals to move significant amounts of money without much effort, intervention or tracking. The current ransomware issue is a direct consequence of the attackers being able to utilize a relatively safe and cheap payment channel. Without it, the endeavor makes no economic sense.
It also helps people around the world to send and receive money that doesn't have the privilege to do so in a conventional way.
It is easy claiming it is useless because you don't need it, when you have no limitations or in the best case you don't need to swallow fees that start from 30%
It is by far the easiest way to pay people who are in another country. Trying to use a bank account for that always seems to get thwarted by a bunch of "fraud alert" false alarms.
Remote uses Stripe's Tempo chain [1] to offer USDC payments to remote workers. Revolut seems to use Stablecoins also in their Fintech products [2], though I'm not sure exactly where or how. Western Union launched [3] their own stablecoin.
One I am familiar with is Polymarket, a big prediction market, who uses USDC on Polygon [4] to denominate their bets and to post bids and asks.
These are all real uses, but also they are purely a regulatory arbitrage. Firms that are not allowed to deal with dollars instead deal Pokémon cards worth a dollar.
I'm not dissing it, fine, it's a real business, and maybe honest too. And indeed crypto is easier to send internationally, though overall presents more friction to the average user with access to high quality banking as an alternative.
I like crypto conceptually, but I'd love to finally see a use case that isn't just a regulatory workaround or recursive (something that shuffles other crypto around without an world-facing application).
Not to mention: banks are fiercely regulated FOR A VERY GOOD REASON. And all these stablecoin issuers are effectively unregulated banks. It's all fine if they are all honest, prudent, and don't take excess risks, and no one pulls their money. If either of those goes away, it's going to be a disaster.
> Not to mention: banks are fiercely regulated FOR A VERY GOOD REASON. And all these stablecoin issuers are effectively unregulated banks.
They're not. In the US stablecoins are regulated by the Genius Act and in the EU by MiCA. That's why interest in stablecoins is growing. The Genius Act mandates 1:1 backing of coins to treasuries or cash so the stablecoin custodians make money from treasuries and enjoy a risk free yield.
> I like crypto conceptually, but I'd love to finally see a use case that isn't just a regulatory workaround or recursive (something that shuffles other crypto around without an world-facing application).
I mean there's plenty but it's only interesting if you like finance. I've come to the conclusion that HN may have liked finance when it was populated by more startup folks but now that it's another tech forum, the average reader here doesn't really care for finance.
Prediction Markets are a big cryptocurrency usecase. There's also novel financial instruments like perpetual futures (patio11 is gonna write a Bits about Money post about them if you're curious.) From the sovereign currency perspective of Bitcoin, to the privacy of ZCash and Monero, to the subdivisibility of assets on programmable chains like ETH, the novelty is in the finance. If you're not into fintech then yeah you'll find the usecases all boring.
If you're curious about the tech there is a lot of interesting stuff around iterated consensus though.
> > I like crypto conceptually, but I'd love to finally see a use case that isn't just a regulatory workaround or recursive (something that shuffles other crypto around without an world-facing application).
> I mean there's plenty but it's only interesting if you like finance.
As it happens I do like finance. And all the use cases I know of are basically regulatory arbitrage.
There's nothing about prediction markets that makes them benefit from crypto. It's betting. Sports betting exists without crypto just fine. But the intersection of betting, real world events and US dollars is murky waters. So instead they work on crypto.
Monero is perhaps the one novel case, of untraceable payments. But is that the best crypto can do?
I agree the mechanisms are novel, it's just a shame to me that they didnt create a novel use case too. But I'd love to hear more tangible cases too.
Existing rails are slow and expensive due to anti money laundering laws and taxes. Enterprise customers want to move funds without those pesky laws and paying taxes.
Your linked article is solely about SpaceX customers paying with stable customers, as reported by Chamath Palihapitiya. Even if I disregard the fact that I strongly think Chamath must be lying because his lips are moving, everyone in that story has a clear biased incentive to pump crypto.
Point being, I don't think you can take this story and think it applies to a broader, general group of enterprise customers without more mundane examples.
While it’s easy to dismiss one story, stablecoin payments are gaining legitimate adoption.
> monthly B2B volume has more than doubled since February, rising 113% to about $6.4 billion. The expansion lifted the cumulative value of stablecoin payments since 2023 to over $136 billion, representing that on-chain money is no longer a niche settlement tool.[1]
And $9T in annual payment volume[2].
As it turns out stablecoin payments are actually a real thing, especially B2B.
I cannot for the life of me articulate why, because I actually don’t understand its value prop (despite lengthy conversations with ChatGPT and people on HN about it) or see any reason for its growth, but it is a thing whether we admit it or not.
I had a look at the second source. It goes to crypto investor A16Z's report, where they are looking at all crypto trading volumes. $46 trillion transaction volume but once they remove bots $9 trillion per year.
Note that forbes seriously misquotes. A16Z is counting transactions, not payments.
They compare it to visa which seems like a rubbish comparison. People use visa to buy things. Crypto trades are much more like forex, and A16Z seems to be measuring crypto for crypto purchases.
Global forex volume is about $9 trillion per day. But even that's a poor comparison as most people trade forex to DO stuff with the other currency.
Crypto is largely traded to trade crypto.
As for payments, supposing we take $136 billion cumulative lifetime payments as accurate. Mckinsey reports that global annual payments are $2 quadrillion dollars per year.
The crypto industry LOVES to throw big numbers around and tell tales of use cases just around the corner but the majority of actual use cases still seem to be skirting regulations.
>But even that's a poor comparison as most people trade forex to DO stuff with the other currency.
No, most of this volume (>99%) is institutional speculation. If people actually do something with this money you would see payments of over 9T a day, which is not happening.
Then you're doing it wrong? Solana has low fees. Lighting on Bitcoin has low fees. L2s on Ethereum have low fees. Curious when you did these transactions on what chain?
Goldman did a gold <---> painting swap via ETH like 12 years ago. The ship has sailed. It's okay, you don't have to like the calculator code as a DLL to use the calculator
So does literally everything other than Bitcoin. Bitcoin was a first draft that decided to freeze development in beta. There are much better technologies out there.
What alternatives actually scale to the level that they could seriously replace fiat currency?
I feel like every time I hear about scaling crypto the answer is to build a system that’s actually decoupled from crypto and occasionally writes back a little bit of info.
Edit: It seems like Solana was actually built for decent scale
and claims ~3k tps. This is surprisingly high, though, given, that Visa does about 7k tps on average. What are all these transactions on Solana?
I don't know what you mean by replace fiat currency, exactly, but that wasn't the post I was replying to. parent post eluded to BTC's famously extremely low throughput rate. All I was saying is that this low throughput is not a feature of crypto, it's a feature of the very bad design of Bitcoin that was never improved. This is to be expected- BTC was a basically brand new thing and nobody expected that it would ever be in a position that it needed to provide visa-level throughput.
It's problem is that the Bitcoin development community is allergic to improvement in a way that, for example, Ethereum is not. Ethereum never would have existed if the BTC community was open to developing the protocol.
RE: Other crypto with proper consensus mechanisms that have better throughput: ETH's L2 ecosystem is reasonably fast - something like 19000 TPS sustained was achieved recently, with an average around 10000 TPS and an instantaneous peak of around 24000. This is just an organic number - it certainly could go much higher, there just isn't demand for those transactions right now. It's hard to say how fast it could go.
RE: your last point, Solana is junk - it abandons proper consensus mechanism in favor of throughput. It's subject to all the same problems that delegated POS systems have because that's what it is.
How does this work? It feels like moving all of this off the blockchain loses a bunch of the supposed value.
If I move to a “side chain” do I get to keep the guarantees around settlement? I would assume no, not until this gets worn back to the block chain.
I think I understand (theoretically) how the side chain would increase throughput and reduce transaction fees, but these benefits as I understand them come specifically from making the side chain high volume. If I need to do one transaction with you, the side chain is redundant, right?
> RE: your last point, Solana is junk
Good to know. I only know what I learned in maybe 20 minutes of research.
Not all L2 are about throughput exclusively, some are, for example, more focused on private ZKSnark based transactions. There is a huge quantity of development out there.
Generally the security guarantees on L2 range from identical to the main chain to 'good enough' to 'I pinkie promise not to take your money'. It's a permissionless protocol so that's kinda what you get.
Absolutely agree, but most know nothing other than the almighty BTC.
I want 3 things from crypto:
* throughput that _could_ scale to something like visa. 50k/second christmas.
* privacy by default for everyone - "public ledger" for a currency replacement is INSANE
* reasonable transaction completion time ( < 1 minute)
A good currency is stable. Bitcoin has gone up over 5x in the last 5 years. This is not a good currency, though it’s certainly been a good speculative bet.
What is BTC inflation? How much BTC was printed by government in recent years? And of course - how much it grew in value?
In other words:
* deflationary nature
* independence of any government
* speculation
Governments will keep printing money. Stock market is only good in US (check Japan and China). Bitcoin is good alternative investment in places where options primarily limited to property.
You can equally ask what was inflation of natural gold and diamonds. It was zero since there is fixed amount of gold and diamonds on the planet. Same with bitcoin.
What?? There has been more Bitcoin today than there was yesterday since the first BTC was mined. It's the definition of inflationary. You have no idea what you are talking about.
Crypto values are arbitrary decree of crypto holders
We're just creating techno religious sects preaching from their pulpit
The only thing unique about anyone is their biology. Same old gibberish coming out their mouth, keyboards; not their values! My values! My line go up! Not their line!
Ooh yeah "fiat money" you mean. An arbitrary decree to constrain the definition. Infinitely endless arbitrary rhetoric. Incompleteness theory of primate language.
reply