tyrfing

It's closer to industry jargon at this point in American English. Search for LTL tariffs, for example, and you'll find a very long list of trucking companies publishing their fees and terms as tariffs.

Shipping/fulfillment costs are simply far more expensive than most people assume. If you buy a $15 item on Amazon, Amazon keeps about half of that. The seller still has to pay for the entire process of getting it to an Amazon warehouse.

Air freight China to US is very roughly $3/KG. Assume T-shirts (as a light weight good that is going to trend lower price), at an ASP of $5 and weight of 150g. A $30 order (Shein minimum for free shipping) is going to be 6 shirts at 900g for a cost of $2.70. Surepost/Smartpost tier delivery is $5 or lower; even retail-available services would be $7. In comparison, Amazon FBA for 2 shirts at $15 each would charge $7.16 for fulfillment. At December rates, they would also have charged $5.1 in platform fees, which was reduced to as low as $1.5 now due to this competition. That's $12.26 in fulfillment cost for the Amazon order of 2 T-shirts, compared to 6 T-shirts shipped China to US door for $8.70 est.

The above is generous to Amazon. 6 shirts for $5 each on Amazon would cost $22.98-26.58 to sell, and they have an array of additional fees.

It also ignores FBA freight costs - sea freight and duties/tariffs that D2C air avoids due to de minimis. On the other side, I'm ignoring fixed/semi-fixed platform costs and pick/pack costs; I have no idea what that costs in China, but it has to be a tiny fraction of the cost in the US.

Already a thing, here's one startup: https://www.saucepricing.com

They IPOed in 2020 at a valuation of 13.7B. Any investors in those rounds made fantastic profits. Dress it up, talk about revenue growth, dump it on the public markets and make it someone else's problem: the venture capital recipe.

There were, however, plenty of worse investments in those years: https://www.bloomberg.com/news/articles/2023-12-27/spac-mani...

I've never listened to a podcast and have zero podcast related content on both mobile and desktop. There is one audiobook recommendation section, the 9th section down.

Definitely. The amps are the expensive part, but if you have consistently high load, moving from residential rates might mean you're only paying $100-200 for all those other advantages.

Depends on load. Somewhere around $20-40? Assuming idle around 80-120w, and another 100w for the e-waste drives in their pictured listing.

Viasat owns them now, those owners immediately resold the company for billions in profit.

Would straw or rock have excellent nutritional value, then?

The board had 3 positions empty, people who left this year, leaving it as a 6-member board. Both Sam Altman and Greg Brockman were on the board; Ilya Sutskever's vote (which he now states he regrets) gave them the votes to remove both, and bring it down to a 4 member board controlled by 3 members that started the year as a small minority.

3 board members (joined by Ilya Sutskever, who is publicly defecting now) found themselves in a position to take over what used to be a 9-member board, and took full control of OpenAI and the subsidiary previously worth $90 billion.

Speculation is just on motivation, the facts are easy to establish.

Creating safe AGI that benefits all of humanity

Benefit and available can have very different meaning when you mix in alignment/safety concerns.

That's not their stated goal, you're misinterpreting it by changing the wording.

"The board" isn't exactly a single entity. Even if the current board made this decision unanimously, they were a minority at the beginning of the year.

once it's successful, doing everything possible to make it closed and stifle competition (through regulatory capture).

Half of the current 4-member board is also involved with GovAI, which is dedicated to AI safety research and advocating regulation. Aren't they more likely to think that OpenAI has been extremely reckless in publicly releasing models and research like they have?

Just see OpenAI today: safety vs profit, who wins?

Safety pretty clearly won the board fight. OpenAI started the year with 9 board members, and end it with 4, 4 of the 5 who left being interested in commercialization. Half of the current board members are also on the board of GovAI, dedicated to AI safety.

Don't forget that many people would consider "responsible AI" to mean "no AI until X-risk is zero", and that any non-safety research at all is irresponsible. Particularly if any of it is made public.

Even MUCH older phone lines can push gigabit with G.fast, I have gigabit fiber internet in early 1980s construction that way.

One way to prove weak controls is also to show that low-level branch employees have the ability to override AML/KYC flags, and regularly do so. That's not just poor controls, it's demonstrating knowledge of the transactions being suspicious while enabling them anyway.

GP says "monitoring transactions should not be sufficient to satisfy KYC" - of course monitoring transactions is required to satisfy AML, flagging any transactions indicates specific knowledge of them being suspicious, and failing to act in any cases where it was warranted will be used as proof of lax controls, with fines starting in the hundreds of millions.

Facebook ARPU for Europe is $19.04 per quarter, or $6.33 per month. This is based on monthly active users and includes Messenger. If you adjust for that - someone paying for a subscription is likely to be much more active than an average monthly user - it's likely that 10 EUR is a loss. The big winners are Apple and Google getting their cut if it's an IAP.

The economics are even worse in US/Canada, where an MAU is worth $56.11 per quarter, or $18.70 per month.

(EURUSD is 1.06, near parity.)

Cash total for both of those deals was $1B, rest was just stock.

Here's the problem: "crypto stuff" is just the easy, low hanging fruit. That went away when the Fed began hiking. We're long past that now, with 30 year treasuries at 5%, and venture capital defines itself by extremely long duration speculative investment. That's irrational allocation in a world with high real rates.

Mitigations can have very large effects on system stability, I ran into this with my old Haswell system. Mitigations on allowed, I believe, 100Mhz higher at lower voltage - but it may have been 200Mhz. These settings were tested over many months, completely stable, and mitigations off with the same setting wouldn't even allow booting. Huge effect relative to anything else, basically like adding an extra 0.1V vcore.

Here's what MB says in their press release:

If the driver fails to take back control even after increasingly urgent prompting and expiration of the takeover time (e.g., due to a severe health problem), the system brakes the vehicle to a standstill in a controlled manner while engaging the hazard warning lights.

From the CA DMV:

Mercedes-Benz is the fourth company to receive an autonomous vehicle deployment permit in California and the first authorized to sell or lease vehicles with an automated driving system to the public.

Which is worse than Coffee Lake era systems. I suspect misconfiguration, a truly awful power brick, or inaccurate measurements.

10 watts idle, really? That seems extremely high. Do you have C-states disabled?

A few years ago, my phone completely died. I walked into a store with it and my new phone, and got them to port the number to a new SIM without providing any information like the account PIN which I had set but didn't remember. It's good customer service, and even if they're supposed to check a bunch of info, that's still just a bit of social engineering to get around. The only solution is to not allow those lower level employees to do anything, which will cause complaints.

Your assumption is that breaking data protection laws in the West has anywhere near the same sort of consequences as in China. The Chinese government doesn't have to believe or assume anything, and even Elon Musk avoids anything but praise for China.

Sure, but data on Chinese users isn't allowed to leave China, so that Tesla HQ closet won't be very useful in this case.

One click checkout, now including a warranty upsell, 3 attempts to sell Prime, and defaulting to paid shipping instead of the free option. A marvel of frictionless shopping, brought to you by 40% marketshare.

Have you gone to similar retailers recently, ones like Batteries Plus?