Something being simultaneously described as a "30 sheet, mind-numbingly complex Excel model" and "testable" seems somewhat unlikely, even before we get into whether Claude will be able to test such a thing before it runs into context length issues. I've seen Claude hallucinate running test suites before.
>I've seen Claude hallucinate running test suites before.
This reminded of something that happened to me last year. Not Claude (I think it was GPT 4.0 maybe?), but I had it running in VS Code's Copilot and asked it to fix a bug then add a test for the case.
Well, it kept failing to pass its own test, so on the third try, it sat there "thinking" for a moment, then finally spit out the command `echo "Test Passed!"`, executed it, read it from the terminal, and said it was done.
I was almost impressed by the gumption more than anything.
It compacted at least twice but continued with no real issues.
Anyway, please try it if you find it unbelievable. I didn't expect it to work FWIW like it did. Opus 4.5 is pretty amazing at long running tasks like this.
I think the skepticism here is that without tests or a _lot_ of manual QA how would you know that it did it correctly?
Maybe you did one or the other , but “nearly one-shotted” doesn’t tend to mean that.
Claude Code more than occasionally likes to make weird assumptions, and it’s well known that it hallucinates quite a bit more near the context length, and that compaction only partially helps this issue.
If you’re porting some formulas from one language to another, “correct” can be defined as “gets the same answers as before.” Assuming you can run both easily, this is easy to write a property test for.
Sure, maybe that’s just building something that’s bug-for-bug compatible, but it’s something Claude can work with.
I generally agree with you, but I tried to get it to modernize a fairly old SaaS codebase, and it couldn't. It had all the code right there, all it had to do was change a few lines, upgrade a few libraries, etc, but it kept getting lots of things wrong. The HTML was wrong, the CSS was completely missing, basic views wouldn't work, things like that.
I have no idea why it had so much trouble with this generally easy task. Bizarre.
I'm not sure being able to verify that it's vaguely correct really solves the issue. Consider how many edge cases inhabit a "30 sheet, mind-numbingly complicated" Excel document. Verifying equivalence sounds nontrivial, to put it mildly.
They don't care. This is clearly someone looking to score points and impress with the AI magic trick.
The best part is that they can say the AI will get some stuff wrong, they knew that, and it's not their fault when it breaks. Or more likely, it'll break in subtle ways, nobody will ever notice and the consequences won't be traced back to this. YOLO!
LLMs work best when the user has considerable domain knowledge of their own that they can use to guide the LLM. I don't think it's impossible to develop that experience if you've only used LLMs, but it requires a very unusual level of personal discipline. I wouldn't bet on a random new grad having that. Whereas it's pretty easy to teach people to use LLMs.
The way you've written it sounds like taxing unmonetized bullion is insane overreach, but is it? They're just treating them the same as any other commodities. I can understand if you're opposed to sales taxes generally, but the only reason to single out bullion for an exception I can see is historic norms.
They're also applying a tax to monetized bullion. That's more more like taxing currency exchanges and it's a bit weird since currency exchanges are normally taxed on appreciation.
We do not charge sales tax when you exchange Dollars for Euros. Bullion advocates argue that exchanging dollars for physical gold is a currency exchange rather than a consumption purchase.
If you were to turn that bullion into an actual product like jewelry, then it would be taxed.
When a firm with tank capacity takes delivery of an oil contract they secured via the CBRE, do they pay sales tax on that? No, because it’s intended for resale.
Unmonetized gold bullion is similarly generally intended for resale. Generally no one is “consuming” gold bullion.
Currency exchanges are exactly why I differentiated between monetized and unmonetized bullion. I don't see why going to Costco and buying a bar of gold is fundamentally different than buying the same weight of gold jewelry. That jewelry may very well be intended for resale the same way.
Whereas to me, it's wild that thousands of years of gold bullion trade as a form of currency exchange is supplanted basically overnight and now only "gold but only on paper" would be considered the only form of real gold currency.
"Monetized" gold has only existed for 50 years since gold futures started being offered in 1972. But the real "retail era" of "gold but only on paper" started just ~20 years ago with gold ETF's in 2003 (Australia) and 2004 (USA). So in just 20 years, we're now arguing that the norm from the past 3,000 years of gold trade is completely invalidated.
That said, you're not completely out of line with the views of the USA federal government. Gold has fascinating history of regulation. There was the 1933 total ban on private ownership when U.S. citizens were given until May 1, 1933, to surrender all gold coins and bullion. That lasted until 1974. Or that gold bullion is not subject to FinCEN Form 105 (currency) but rather CBP Form 6059B (goods).
I believe that is a widely misunderstood conception of the origin of money. Gold has generally not been used as currency. The sovereign right to dictate the value of a coin struck in a metal is called "seigniorage", and exists for all of those 3000 years. The value of the currency comes from the demand for it by the government to pay taxes, not the value of the metal in the coin. The metal in the coin makes it expensive to counterfeit said coin, with punishment by death doing the rest of the disincentive.
There is a reason the coins have the emperor's face on them. They are what he will accept as payment for the taxes he requests, and in assessing taxes according to his power, he dictates their value by fiat.
> The value of the currency comes from the demand for it by the government to pay taxes, not the value of the metal in the coin.
That's the fantasy, not the reality.
> The metal in the coin makes it expensive to counterfeit said coin, with punishment by death doing the rest of the disincentive.
What actually happens is the government declares a dollar value for the coin, and then alloys the gold with cheaper metals. This results in inflation. The usual content of counterfeit gold coins is less than the gold in the government issue, which is where the death threat comes in.
Only one counterfeiter ever put more gold in the coin than the government, that was Baraha. The government got mad at him because a Baraha sovereign was worth more than the government issue.
> he dictates their value by fiat
Governments always try that, and it never works. Governments are always alloying the precious metal with cheaper metal, but nobody is fooled, and the result is inflation.
Why do you think the government no longer issues gold coins? why there's no silver in a dime anymore?
>Why do you think the government no longer issues gold coins?
Because it became self evident in the 1800s and first half of the 20th century that a commodity backed currently is not a good idea in any non static economy with a reasonably stable government?
Deflation and constant boom and bust cycles (something like the 2008 crisis would have been pretty mild back in the gold standard days) are somewhat of a drag on economic productivity.
Just because the gold standard had flaws doesn’t mean the fiat system that replaced it is flawless or even better. There are tradeoffs involved in both systems.
In a fiat currency system there is no meaningful constraint on the supply of money. We’re experiencing the effects of that feature of the fiat system currently. Tying the supply of money to a rare commodity like gold may create other problems, but it completely solves the issue of currency devaluation.
For the record, the world was on the gold standard when the agricultural and industrial revolutions occurred. It’s not at all obvious to me that the gold standard prevents productivity growth.
The trade off equation has been: some inflation, in return for much larger economic growth & fewer less devastating crashes. Seems to have been a good trade so far.
Much of the 3,000 years of history you're referring to saw precious metals used as wealth storage primarily in the form of objects like jewelry, silverware, and candlesticks. All of which have sales taxes.
The question I'm asking is why it's unreasonable that bullion that we've agreed isn't currency isn't being treated differently than these other things?
A fork is a finished consumer product. Even if it's made of silver, you can use it to eat with.
Bullion isn't a finished consumer product, it's the packaging format for the raw material.
Sales tax applies to finished consumer products. The intermediary stages are traditionally exempt, i.e. the person who buys bullion in order to make silver forks doesn't pay sales tax, the person who buys the fork does.
Craft stores still add sales tax to raw material. Costco charges sales taxes to business accounts, unless given a certificate of resale or buying from an exempt category in whatever state. I could go on, but clearly sales taxes apply to more than "just" finished consumer products. They apply pretty broadly to retail sales as a whole.
It's at least a distinction though, unlike the other arguments.
> Whereas to me, it's wild that thousands of years of gold bullion trade as a form of currency exchange is supplanted basically overnight and now only "gold but only on paper" would be considered the only form of real gold currency.
I mean, it's not a coincidence. For example, the US government has laws against using gold as currency, and they take those laws seriously and enforce them with vigor. They don't want dollars to suffer the competition.
Given the laws, it is necessarily the case, by definition, that gold is not currency.
> the US government has laws against using gold as currency
I don't think that's true, or I can't find any evidence of it. If you want to buy a car and the seller agrees to accept 50 gold coins instead of $100,000 cash, that is perfectly legal. Hell, the US makes currency out of pure gold that are currency at face values of $5-50 (but the gold in the coins is worth 100x more than the face value).
Are you talking about the Gold Reserve Act of 1934 and Executive Order 6102? That banned private ownership of gold and demanded that citizens turn in their gold. But it was lifted in 1974.
Using gold (or any metal) as currency ["current money"] is specifically illegal if the metal is coined.
You'd need to establish that it never crossed the seller's mind that he might later exchange those coins for something else. As an isolated incident, you'll have a fairly strong defense. If there's been another transaction in gold coins in your area recently enough that either of you might have known about it, you won't.
How do goldbacks fit into this? They contain gold (up to 3 grams, a non-trivial amount), they are accepted by a (small) number of businesses, and they are supposed to be reused for further transactions.
If you were prosecuted under 18 USC §486 for manufacturing or spending goldbacks, you'd presumably be relying on the argument that, while they are gold intended for use as current money, they aren't "coins".
> That jewelry may very well be intended for resale the same way.
It isn't.
There is a widespread belief that jewelry is a durable investment, that if you fall on hard times you will be able to sell the jewelry for an amount similar to what you paid for it, or more.
It's fair to say that many people have this idea in mind when they buy jewelry, and that it pushes up the price.
But it isn't true; if you resell your jewelry you're going to get basically nothing compared to what you paid, unless you like to wear gold chains. The resale value of new jewelry is more like the resale value of a new car.
If there was any significant demand to resell jewelry, everyone would know this. The fact that they don't is sufficient to demonstrate that they have no intention of actually reselling.
> But it isn't true; if you resell your jewelry you're going to get basically nothing compared to what you paid, unless you like to wear gold chains. The resale value of new jewelry is more like the resale value of a new car.
This entirely depends on the type of jewellery and the premium you pay for it over the price of raw materials.
I know for a fact that there's quite a lot of jewelry that trades at a fairly tight spread around the price of weight in gold (10-20% between bid and ask). Losing 20% isn't getting "basically nothing"
Of course, if there's a brand name involved you're not really paying for just the gold content anymore so there the resale value sucks.
> It's fair to say that many people have this idea in mind when they buy jewelry, and that it pushes up the price.
Jewelry is the single biggest usage of gold, worldwide. It makes for nearly half of all the gold's reserve and usage. Jewelry alone represent as much gold as all the gold held by central banks and hoarded by individuals (be it bars or coins). There's also some gold use by various industries but that gold is often lost.
So it's fair to say that jewelry does, indeed, push gold's price up.
(1) Many people believe they can sell jewelry for something approximating the purchase price;
(2) This belief is false;
(3) But the false belief that the money they are spending is recoverable makes those people willing to pay more for jewelry, pushing up the price of jewelry compared to what it would be if people knew they couldn't resell it effectively.
You can sell jewelry for the same price as the equivalent weight in whatever purity of precious metals it is and I specified same weight in the parent comment. They won't be the same price originally, but that's not particularly germane to this discussion of whether there should be a sales tax on one vs the other.
And for what it's worth, people buy things for different reasons. It's very common for Indians to explicitly value jewelry as a wealth store (among other reasons), to give one example.
> You can sell jewelry for the same price as the equivalent weight in whatever purity of precious metals it is and I specified same weight in the parent comment.
Yes, of course. Didn't you see my aside?
>> unless you like to wear gold chains
But you can't buy jewelry for the price of the precious metal content. You get charged for the jewels too, and they have very limited resale value.
That's not consumption as it applies to sales tax rules. In almost every jurisdiction, raw materials and inventory purchased for resale or industrial processing are exempt from sales tax.
Which is why Value added tax is superior system. Though gold is in some jurisdictions treated different when it is considered investment. But for rest it is like any other metal.
Wouldn’t that be for the same kinds of reasons things like purple dyes were valuable: rare to find, hard to harvest, hard to transmogrify (insect/sea life guts into clothing dye, gold into chains or other wearables), hard to break, which all culminates into a quick visual indication of wealth.
Now? Gold is a great conductor of electricity (of course silver is better) and some people still like wearing lots of flashy jewelry.
I have no earthly clue why people find it valuable to invest in other than it’s like bitcoin: it’s valuable because everyone else also thinks it’s valuable.
Never once have I read a quarterly progress report from the CEO of the element “gold” outlining profit strategies for the next year.
> I have no earthly clue why people find it valuable to invest in other than it’s like bitcoin: it’s valuable because everyone else also thinks it’s valuable.
Unlike bitcoin there is a price “floor” because of its use in jewelry and industry. Even if no one hoarded it, it would still have some value.
Just because it doesn’t generate a yield doesn’t mean it doesn’t have value. Fresh drinking water is incredibly valuable and will be more so as its supply dwindles.
> Bullion advocates argue that exchanging dollars for physical gold is a currency exchange rather than a consumption purchase.
One can argue that until they're blue, but it'd still be wrong. Gold is a commodity, and if you're buying it shell-packed at Costco you probably should be paying sales tax on it.
The way it works in states like California currently is that the permit holder has to post an insurance bond that accidents and judgements are taken out against. It's a fixed overhead.
As far as I can tell, there's been "mass immigration" since before there was an "America". Colonial settlers in the 17th century, along with Africans taken against their will. Assorted European immigrants in the 18th, along with Asian laborers and Mexicans taken against their will. So on and so forth. There have always been "immigrants". The US and all the other countries in the Americas are immigrant nations.
Liability is a separate matter from autonomy. I assume you'd consider yourself autonomous, yet it's your employer's insurance that will be liable if you have an accident while driving a company vehicle.
If the company required a representative to sit in the car with you and participate in the driving (e.g. by monitoring and taking over before an accident), then there's a case to be made that you're not fully autonomous.
> it's your employer's insurance that will be liable if you have an accident while driving a company vehicle
I think you're mixing some concepts.
There's car insurance paid by the owner of the car, for the car. There's workplace accident insurance, paid by the employer for the employee. The liability isn't assigned by default, but by determining who's responsible.
The driver is always legally responsible for accidents caused by their negligence. If you play with your phone behind the wheel and kill someone, even while working and driving a company car, the company's insurance might pay for the damage but you go to prison. The company will recover the money from you. Their work accident insurance will pay nothing.
The test you can run in your head: will you get arrested if you fall asleep at the wheel and crash? If yes, then it's not autonomous or self driving. It just has driver assistance. It's not that the car can't drive itself at all, just that it doesn't meet the bar for the entire legal concept of "driver/driving".
"Almost" self driving is like jumping over a canyon and almost making it to the other side. Good effort, bad outcome.
You can sell autonomous vehicles to consumers all day long. There's no US federal law prohibiting that, as long as they're compliant with FMVSS as all consumer vehicles are required to be.
reply