If you trust the other party, you can always write a derivative contract. If the other party was long, their position is closed. If they weren't, they're short now.
That's part of the problem though, this is kind of a no-trust transaction to start with since it's inherently fraudulent. As Matt Levine said this morning:
'Also, just, consider the situation here. I come to you and say “hey I have some Anthropic shares, I can’t sell them now but I’ll sell you a forward on them.” You say “isn't that not allowed?” I say “oh sure it’s not allowed but I am a rebel, I’m not bound by those pesky rules.” You’re like “okay I guess” and give me money, now, for shares in the future. How do you know I even have the shares in the first place? What if I’m just lying? You already know that I am willing to break some rules; what if I’m breaking the rules against fraud?'
I don't even think you could buy hedges or insurance against this counterparty risk, because Anthropic could probably get those voided in court too.
If you're afraid of buying derivatives from a trustworthy party that may have contractual obligations with Anthropic, but them from a trustworthy party that doesn't. If they expected a $100 post-IPO price one year from now, they'd short-sell you a derivative for $200 or whatever. Derivatives are legal, and as ethical as anything else in finance (more ethical than just about all crypto.)
Vacuum-tube amplifiers are not in the same class with techniques that are unlikely to have any perceptible influence on what you hear.
Among amplifiers that are not perfectly neutral, vacuum-tube amplifiers subjectively seem more pleasant.
Moreover, while an electronic audio amplifier made with modern components can be made perfectly neutral when terminated on a resistive load, i.e. it can reproduce any input signal without any changes except amplification at its output, once you connect loudspeakers at its output the amplifier-loudspeaker chain is no longer neutral, i.e. it no longer has a flat transfer function between the electrical input and the sound output and it is not at all clear which should be the output impedance of the amplifier as a function of frequency to ensure the least degradation of the sounds in comparison with the input signal.
So it may happen that a vacuum-tube amplifier - loudspeaker system has actually a better overall fidelity than a typical audio amplifier that was designed to demonstrate a much higher fidelity on a resistive load (because thus the transfer function is easy to measure and correct, unlike the complete transfer function to sounds).
In theory, one could make a modern amplifier reproduce any quirky behavior of vacuum tubes, e.g. a higher and frequency-variable impedance or certain kinds of distortions, but usually nobody bothers to do this, because it would be expensive and the normal amplifiers are good enough for the majority of people.
> Vacuum-tube amplifiers are not in the same class with techniques that are unlikely to have any perceptible influence on what you hear.
Yes, they are. The Carver Silver 7 was built to demonstrate this. [1]
It's a tube amplifier with 38 tubes per channel that costs $17,000.
It has all the important features - weighs 68Kg, vibration damping mounts,
takes four minutes to power up, and the wiring is silver.
Gets good reviews from the High End crowd.
Then Carver built the Silver 7 T, a transistorized amp with the same transfer function. As a demo, the Silver 7 and the Silver 7 T can have their outputs differenced, or wired up to cancel and drive a silent speaker. Same output. Gets terrible reviews.
Typical transistor-based audio amplifiers are different enough from traditional vacuum-tube amplifiers, but when the cost does not matter it is possible to design transistor-based amplifiers that are equivalent with vacuum-tube amplifiers.
The example given by you shows that there are indeed many people who do not truly perceive the differences or non-differences, so they judge based on prejudices. Of course, I agree that there are many such people and the gold-plated cables were intended for them. I agree that they must exist also among the customers buying vacuum-tube amplifiers.
> It is not at all clear which should be the output impedance of the amplifier as a function of frequency to ensure the least degradation of the sounds in comparison with the input signal.
> So it may happen that a vacuum-tube amplifier - loudspeaker system has actually a better overall fidelity than a typical audio amplifier that was designed to demonstrate a much higher fidelity on a resistive load (because thus the transfer function is easy to measure and correct, unlike the complete transfer function to sounds).
I don't know the electrical engineering at all, but I thought that this was a solved problem -- or that the degradation and mismatch were effectively negligible, well below the point of inaudibility.
>In theory, one could make a modern amplifier reproduce any quirky behavior of vacuum tubes, e.g. a higher and frequency-variable impedance or certain kinds of distortions, but usually nobody bothers to do this
> Vacuum-tube amplifiers are not in the same class with techniques that are unlikely to have any perceptible influence on what you hear.
I don't disagree. Not in the same class, but the audience overlaps.
> one could make a modern amplifier reproduce any quirky behavior of vacuum tubes, e.g. a higher and frequency-variable impedance or certain kinds of distortions, but usually nobody bothers to do this, because it would be expensive
In other words, the willingness to pay for OG tube amplifiers exceeds the willingness to pay for the sound thereof. I'm not sure you disagree with me either.
As a child, I have used some very good vacuum-tube audio amplifiers, which had been built by my father, at a time when they were still the cheapest solution, instead of being a luxury product.
They were excellent, so I feel nostalgia remembering them, and I would like to experience again listening through such an amplifier. Nevertheless, if I had so many thousands of $ to spare, I would rather buy some memory modules ... :-(
When I was young I made a few unusual transistor power audio amplifiers, e.g. with the output transistors biased in class A and designed for high output impedance instead of low output impedance. I was very satisfied with their sound and some of them resembled more some vacuum-tube amplifiers than typical transistor-based audio amplifiers.
However, despite their high audio quality they would have been completely impractical as commercial products, because they needed very big power supplies and they produced an enormous amount of heat. Semiconductor devices are much more difficult to cool than vacuum tubes, for the same amount of heat (because the temperature limit for the former is much lower than for the latter).
Nowadays, switching amplifiers can cover all the audio bandwidth with excellent energy efficiency. With an appropriate combination of linear and non-linear feedbacks, one could reproduce both the distortions and the output impedance of vacuum triodes, to make an amplifier hard to distinguish from true vacuum-tube amplifiers.
Even after we've all retired (pretty soon for those who can afford it) or transitioned out of software engineering (for those who can't), we'll still get to amuse each other with home-brew projects like this. Warm fuzzy feeling - I'll take it!
I harbor some hope that the (sad) fall of human SWEs will at least be accompanied by language defragmentation. We don't need 38 systems languages once human taste is mostly out of the picture.
Since the LLM craze started I have always assumed it would end up in a place where programming languages are dead and LLMs generate something more low level.
Programming languages were always designed as an abstraction to allow humans to more easily instruct a computer than by writing binary or assembly. If humans write natural language and don't check the generated code, there's no reason to take the hit of generating C, JS, etc that still has to be compiled and/or interpreted.
If anything LLMs should use something higher level because it compresses the context and makes programming closer to natural language they are trained on.
Forcing LLMs to do a shitty job of what a compiler can do deterministically is not a good approach IMO.
Low level was the wrong term for me to pick there. I was meaning more along the lines of "purpose built". I.e. I could see a language, potentially still an abstraction requiring a compiler, that isn't meant to be particularly meaningful or inspectable by humans. For LLMs your right, conciseness would be important and that would likely mean it would be compiled.
you're both under and over estimating this technology I think, we are further away from SWE becoming a niche job than you think, but also LLMs really are less impacted by programming language than you think, this is a very pre-2010s approach to NLP, it really matters quite little to a language model what language the code is in, you could have it write it in Python using exclusively Chinese logographs for names, and exclusively French transliterated into Mayan script for comments and it might barely perform worse than if you asked it to do something normal, but that also goes in the other direction, trying to design a language that's "friendly" to LLMs will likely not do much good either. The reality is there's a massive amount of data on existing languages and they are good enough for LLM usage.
I think if changes will happen in this direction it will be around formal verification, it's more difficult to trust LLM generated code if you aren't completely brainroted on press releases, and traditionally formal verification was seen as more effort than it's worth, but a tool that's great that generating insane amounts of text is quite well suited to formally verifying code, and assuming you can figure out how to make the specification readable to people, you just don't (in theory) need to ever look at the code or the proof, even if it might be helpful to anyways in practice. I'm quite excited the future of software might be fully unhackable software being the standard, with hacking becoming something that's talked about as a brief criminal fad in the beginning of the 21st century, like how you might hear about coin clipping from before fiat currency.
reply