boulos

> Protobuf is another library to handle cross-language serialization, but wouldn’t offer us the flexibility to add all the other features we now have — plus, it was a binary serialization when I last used it.

Protobuf has a well-defined text format, which is the top search result for "protobuf text" [1]. Proto3 also has a clearly defined set of rules for supporting JSON.

[1] https://protobuf.dev/reference/protobuf/textformat-spec/

Yes, at 1% margin on those sales, that's more like $125M in revenue. It's important to remember that things like Prime Day are basically marketing that results in revenue outside the event.

Try Mama Ji's, Mister Jiu's (upscale) or Mamahuhu (casual). The mapo tofu at Mamahuhu is great, but be prepared for the spice.

Edit: looks like the author included Mister Jiu's :).

Yep. This was going to be the next phase: there isn't anything special about the SI system, but it is consistent. The poster didn't seem prepared for "or any other velocity units you want".

For the original poster, you could equally use c in miles per hour, mass in "pounds", and get whatever that produces for energy (I'm sure there's a name for it).

This is also why Imperial units produce such amusing things as Foot-Pounds (for torque). The math all works out in the end, but you get some amusing numbers along the way.

Do you have a more specific example? Xiaomi's site, Mi.com, for instance doesn't seem to cover every EU or European country (e.g., Bulgaria, Croatia, Norway, Switzerland, etc.). Are you referring to OPPO or OnePlus as a "cheap no name Chinese phone"? They account for about 10% of the global sales (as does Xiaomi).

Assuming this is in good faith, the units are:

- c: speed of light in meters per second

- m: mass in kilograms

- energy: joules

The majority of the EU population can buy it directly... https://www.androidauthority.com/google-pixel-7-availability... says:

> The full list of countries includes the US, UK, Canada, France, Germany, Italy, Spain, Ireland, Netherlands, Sweden, Denmark, Norway, Japan, Australia, Singapore, Taiwan, and India.

I get that plenty of countries don't have official Google Store support, but it's a bit like saying there isn't a Samsung store? I assume you're free to buy the phone elsewhere in the EU, or even in countries that aren't directly supported by Google, and activate with your own carrier.

Related: what experiences have people had with middlemen services that handle the EU claims process for you?

Right, the closest word is the turkic "pide" which isn't Arabic but is from the Aramaic for bread (https://en.m.wiktionary.org/wiki/pide)

There are several chutneys that include vinegar. They aren't in the same category as you listed, but the parent included an ellipsis, which I'll include chutney under.

Yes. Somewhat famously, Sutherland named it that to troll Evans, to see if he'd flinch. He didn't and thought it was funny.

Disclosure: I used to work on GCE.

"Renewable" is a broad set of terms. Importantly from your first link on GCP it says:

> Google is carbon neutral for our operations today, but aiming higher: our goal is to run on carbon-free energy, 24/7, at all of our data centers by 2030.

Carbon-free, 24x7, is a stronger commitment than carbon-neutral which would include using renewable power offsets. Amazon's methodology is at https://sustainability.aboutamazon.com/renewable-energy-meth... which makes this clear as well.

Just a bit, but the $1500 price point didn't stick around for long. So really it's a $999 headset now vs $3500 "next year".

> I can’t wrap my head around it

Pun intended?

The direct announcement (posted at https://news.ycombinator.com/item?id=36113129) is probably more useful.

Rate limit and return a 429?

Ahh, I hadn't found that one! That's a great example, thanks for highlighting it.

The stuff at https://github.com/besm6 is great! Being much later, the programs are substantial, but it gives a better sense than the simplified examples like https://github.com/xldenis/besm/blob/master/examples/chapter... .

Yeah, that's why all the recent sprinter van conversions are using them for the "house" batteries.

If someone just says "AVX" they usually mean AVX(1). And that's what the article is discussing.

AVX-512 forking into tons of different beasts is a separate, but related, problem. But it's more like how SSE2, 3, 4, 4.1 and so on existed.

Sometimes people said "SSE" and they might have meant one the later versions, but I don't hear the same statement with AVX, since people very explicitly seem to say AVX2 and AVX512.

Some of the results on https://rhasspy.github.io/piper-samples/ are super impressive (via https://www.home-assistant.io/blog/2023/04/27/year-of-the-vo...).

Absolutely. Most development work for these things is just fine on a laptop.

Each processor family has its own supported SIMD instructions though. For a long time, SSE / AVX / AVX2 were the only game in town on x86. AVX-512 introduced in Skylake was primarily available only on server parts, and was a mixed bag, and didn't have support on AMD machines until just recently (with Genoa / Zen4).

A modern Mac laptop with an M-series chip will have ARM NEON vector instructions, just like on an iPhone or similar. There is a new-ish ARM vector instruction set called Scalable Vector Extensions (SVE) that Apple doesn't support, but Graviton 3 does. But generally improvements on your laptop often translate onto the server side, and you can definitely test correctness.

There are a few least-common denominator SIMD libraries out there, but realistically for many problems, you'll get the most mileage from writing some intrinsics yourself for your target platform.

Today, they come back to the depot to be recharged. People plug them in and unplug them when done.

Yep! Our research team publishes a lot of work (waymo.com/research).

If you're interested in Transformers and attention in particular, the excellently named Wayformer [1] paper is pretty easy to follow and SWFormer is impressive on a lot of levels [2].

The main challenge for autonomous vehicles has been (large) multi-modal inputs, while delivering real-time inference performance on the road. Many of the large-language model things you see produce cool results, but would be much too slow or expensive for a real-time, safety-critical application.

[1] https://arxiv.org/abs/2207.05844

[2] https://arxiv.org/abs/2210.07372

And yet... https://www.fox10phoenix.com/news/caught-on-camera-teen-jump...

Part of this announcement is the completion of retiring our Pacifica fleet. So while we used to have the vans on the road in Chandler, you'll now see the same vehicles that were in Downtown Phoenix and San Francisco.

Yep, three in the back and one up front, driver seat empty. (The door won't open for you anyway)

> The question is whether it'll ever get out of cities? While storage is constantly getting cheaper, but is it feasible to map and update the maps sufficiently for a cross-country trip?

That's certainly the intent with Waymo Via for trucking, and we drive multiple routes today including say Dallas <=> Houston for beer [1]. The video at [2] includes some footage of the trucks on a highway.

[1] https://www.ttnews.com/articles/waymo-ch-robinson-driverless...

[2] https://www.youtube.com/watch?v=Lon1vRv2wQ8

Good summary! (I was away for a bit)

I would only caution that people shouldn't overindex on mapping costs. When you're talking about entering a major metro area, you need lots of vehicles and depots and so on anyway just to provide the eventual service. So the upfront initial map build out is certainly a component of starting in a new geography, but I think people assume it is somehow dominant. It's not.

The current approach then for ride hailing, which requires depots, means that:

> just turn on the tap, buy a million cars and cover the whole country

is not happening anytime soon.

We currently only operate the public service in Phoenix and San Francisco. There isn't a wait list in Phoenix, anyone can ride, but there is in San Francisco. So... technically they're currently independent.

But our goal in any new area would be to focus on the people that live there, because they would use the service regularly. Of course, the broader goal is no waitlist anywhere :).