Plain old, madagascar vanilla artificial obsolescence.

I don't want to put information into a black box of mystery that can then be used for other monetization purposes. I am still waiting for a realistic local solution.

Have you tried qwen3.5 running locally? It’s quite “good enough”.

But on what time scale? Before a few connected entities make a profit or after?

The time scale is at least a thousand times faster than necessary for your retirement savings to be safe.

The problem of investment companies selling stupidly at 3am solves itself as they either learn or go bankrupt. And the counterparties making money off those dumb moves don't need to be 'connected'.

Not if I retire 300 times per minute

Connected is a euphemism for inside trading; someone with knowledge of future actions. With the pedotus giving access via his crypto grift who knows?

I know what the word means.

I'm saying that insider knowledge isn't necessary to make money in this kind of situation, and probably doesn't even help very much. If a bunch of people trade like idiots and cause wild swings when there's no liquidity, you can provide liquidity and make money.

Milliseconds?

Any overnight mispricing is going to become an arbitrage opportunity for market makers, hedge funds, and HFT firms...whom will then compete with each other to mine that arbitrage opportunity until profits go to zero, solving the market inefficiency and mispricing problem over time (and by over time, I mean like probably the first few nights and then it stops being an issue forever).

In other words, a liquidity-based mispricing that happens consistently every night is going to quickly stop being mispriced since its so predictable.

Extracting value from the market as they do it and leaving everyone operating at normal time/capital scales with less. Or isn’t that what you meant?

That's one way to think of it.

The other way to think of it is that these parties are essentially middle men who make a cut of the difference whenever someone buys/sells far from market price.

Basically if you mis price something they screw you rather than the counter party who would be interested in taking the other side at market.

I think this is stupid and does not add value, but I don't think it's harmful. It's like the stocks equivalent of a junk flipper.

Yes, when you correct a mispricing in markets that is a valuable service and you tend to get paid in proportion to the mispricing you correct (this is why markets work so well, they provide dynamic incentives in a decentralized way).

In case you aren't aware, a world outside the US exists on different time zones and also invests in US capital markets.

Having 24/7 trading a massive value-add for the entire world who also invests in US companies, which benefits US companies tremendously given they will continue sucking up the world's capital.

This is yet another reason why global companies will continue going public in US markets instead of their own. Meanwhile Europe will continue struggling to form a capital markets union over the next 50 years while they slowly translate legal documents back and forth to each other in 42 languages, growing the fine dining economy of Brussels more than their domestic economies.

If you're not trading overnight and there's a flash crash that corrects itself overnight... it's the people who are trading overnight taking money from each other.

It’s easy to forget that “overnight” trading is the middle of someone’s day, somewhere. Generally Asia.

There's nothing wrong with not retail trading. In fact, most people shouldn't be doing it.

That's the point.

You think some guy at a bank is trading for you.

In a hedge fund sure

Most people are in simple retirement year funds which have a set algorithm which decides by simply rebalancing to follow an index and appropriately mitigating risk by shifting some assets towards bonds.

There is no one trading on a whimsy. The mutual fund founding documents specify an exact time of day (or times) at which the fund gets rebalanced and it simply follows the algorithm..

The price shocks discussed here will not affect that

You missed the point of that comment. Nobody at your bank is trading for you or themselves at 3am.

But that doesn’t matter to your retirement. Don’t sell your retirement at 3am and the liquidity at 3am isn’t your problem.

Do you actively trade on a 401k or other savings instrument or do you leave it up to the bank/brokerage?

Makes no difference unless you are a daytrader

Did those headphones have a band over the top?

I troubleshoot thermal printers for work and the error is rarely in the hardware. I have no issue with this work request; TPs are deterministic and predictable.

Informed consent for a paying user is inconvenient?

Did you read the TOC?

Hiding something like this in the TOC rather than explicitly asking users to opt in is a dark pattern. You can't gain the moral highground by cackling that someone should have read the fine print.

This is technology, some politics, capitalism, and math that is trained on curiously gained data, where does non-selfish morality come in?

All TOS essentially boil down to "we owe you nothing and can change the product at anytime to anything we want at our sole discretion"

Obviously it would be unreasonable to accept such terms without further context. The further context in this case being that Anthropic will maintain Claude as an AI agent and seek to improve it's performance. What is at the heart of this issue is whether or not Anthropics recent A/B testing violated that context. Not whether or not they violated the TOS (they didn't, obviously)

Ultimately that just sounds like within their own TOC, they were just working on getting the best operational results.

If you wanted something more deterministic write it yourself or get it verified, all hosted llms as far as know does neither.

I read the article saying they were testing service changes on paying users without knowledge or explicit consent of the user, that the user had to test and determine why they perceived their sevice changed.

That is a dark pattern to inflict on users expecting consistent output.

Does anyone?

The profiles work fine in Apple Classroom mode. Just a question of SSD space for the active profile(s) and a cloud source that can be used to swap them around if the SSD saturates.

This 'red line' was quite apparent around 10.14 when macOS and iOS were set on the collision course we see today in Ta-hoe. So much wasted visual space in the last 5 releases, making room for touch.

I doubt we'll see a pseudo macOS mode on mobileOS, but the mirroring for iOS in the last 2 major releases of macOS is just a jump to the left of local emulation.

Yep. Information density and compact but efficient UIs were definitely an advantage of macOS. Now they are on track to emulate the big buttons of Windows so what's the point ?

Be good to your neighbor and try to mentor the motivated. Inculcate the same lessons they gave you in others.