What do they do to mitigate the roundtrip latency introduced by using a hosted database on a separate network? Is the idea that you wouldn't use this if you have pages that do many queries to render?

This is a very good question! One of the big reasons we partnered with Turso was for their edge network, so replicas are available close to your origin server. Being able to host anywhere is something that's important to Astro, so this felt like the right balance. Turso is known for its speed and one of the main thing that attracts people to use it.

Gotcha, so you guys use readonly edge servers and just eat the latency on writes

Doesn't this introduce a conflict of interest when it comes to ActivityPub? Not that I ever believed they'd really add support.

The video said that you'd be able to pull cables along, but I'm struggling to see how you'd do that given that the actual walls of the "robot" are static once deployed. Is the idea that you somehow turn the cables inside out too?

I can see it as a bootstrap method. You attach a string to the inside end of the vine that moves forward, and it spools out as the vine "grows":

    ____________________
                        \
    O=string============+   movement of growth ->
    ____________________/

        O = spool of thread

When you see the vine on the other end, you can open the end to get the string, attach the cable to the former spool end, and start pulling the string to pull the cable. You want a light enough string for the vine to pull through, while being strong enough to pull a cable through.

To affix your spool, or to access the string at all you need to go through the airtight robot and you need to know the exact distance that you need to run the wire since you can't access it unless the robot is fully extended. Kinda only makes sense if you have exact measurements and are willing to destroy the robot after it fully extends. Still probably a great way to run wire through long pipes.

Another possibility is just to leave the appendage of the robot in place. It's some form of inexpensive polymer, PET perhaps.

Nope, just attach the cable to the last bit of the robot to be grown. It’ll be pulled along with it until the cable runs through the length of the whole robot.

I've been wanting to get off of my s21 ultra because it's extremely bulky and has gotten slow over the years. The killer feature keeping me on it is the 100x zoom for identifying birds and I haven't seen another phone rival it yet. To be clear the 100x zoom is awful quality and not really useful for anything beyond getting basic shapes and colors but for identifying far birds it's awesome.

My S10 is still fast..

I'm fairly certain that you have to undergo a sleep study wherein they monitor your breathing and your blood oxygen levels. If your blood oxygen falls to levels that seem dangerous you will be prescribed. So my understanding is it's not as simple as going to the doctor and saying "I slept funny last night".

Breathing and blood oxygen at a minimum, for the take-home tests. If you go into a sleep lab, they also wire you up to an EEG as well to monitor your brain activity.

They're not just looking for blood oxygen to drop. They're also looking at how many times you wake up due to airway obstructions. Even if you wake yourself up before your blood oxygen drops, waking up 50+ times/night (this would be considered "mild", btw) is not good for your brain either.

But yes, you need a sleep study to be diagnosed with sleep apnea.

> waking up 50+ times/night (this would be considered "mild", btw)

50+ a night is mild? I thought I had it bad at 2 or 3. 50 times a night is every couple of minutes, do people who have it even worse than this actually manage to sleep at all?

When I did my study they said it as >90 times an hour when sleeping on one side and something like >60 on the other. I did a pulse ox test first and the results recommended a sleep study (which was not all that conducive to a good sleep they way it's performed)

Note, you don't actually regain consciousness that often (but if you wake up every 90-120 minutes to pee, and/or snore a lot, your body is telling you something) but it does keep you from getting a deeper sleep. I hadn't dreamed in years because I'd never drop into REM state.

I used to sleep 4-6 hours a night (but rarely felt tired) now I sleep 5.5 - 7 and am much better rested.

> do people who have it even worse than this actually manage to sleep at all?

It's sleep as in their eyes are closed and they are somehow unconscious but they are drifting on and off between falling asleep and being knocked out. There are no recollections of the events so it seems like sleep but very bad sleep (think: more tired when waking up than when going to bed).

From what I remember an apnea/hypopnia index of up to 5 is considered normal, 5-15 is mild, 15-30 is medium, and 30+ is severe. That's what I learned when I had severe sleep apnea, but it spontaneously resolved years ago so perhaps definitions have shifted. I'd be pretty shocked if 50 were "mild" now, though.

I had an AHI of ~90 at diagnosis, and I couldn't figure out why I was so damn tired all the time. My sleep was very badly disrupted, but since an apnea doesn't typically bring you all the way awake it can be hard to notice.

Outside the edit window: I just realized that I missed the mention of the timescale. The apnea/hypopnea index is measured in terms of events per hour, and my doctors never discussed it in any other terms, so I automatically interpreted those numbers as per hour.

When I had my test, mine was over 50 per hour - and that was a decade ago, so I'd likely be worse now.

"Dad died of a heart attack last night, guess we'll never know why"

Health is something that is not just in the doctors purview these days. You can personally buy any number of electronic devices that monitor things like blood oxygen and how often you're waking up that will give you some idea of your sleep quality and likelihood of having apena.

Or, you can be a cynical old fart and continue accepting that dying at 60 of a heart attack is perfectly normal.

Not everything is a vast conspiracy.

I don't know about "vast conspiracy" and as a techie, I fully support testing in order to diagnose and repair.

However I just believe that the incentives are perverse and not improving, and it's begetting corruption. Doctors do not order tests out of personal concern for patient's well-being to start healing them; doctors are mandated top-down to tick boxes on a checklist and improve customer engagement.

As a techie, I also recognize how teched-up they are, and question why they need all that. I was shoved into a giant MRI and went through the whole theatrical adventure, and guess what - nobody touched that report. Nobody would read it, nobody would tell me what it meant, nobody would use it for treatment. It was hilarious. So glad it wasn't my pocketbook that time. And I had a real concern, not just a suspicion; my shoulder was bad, and fractures had already been ruled-out by a good old X-ray.

High-tech is able to supplant "safe, inexpensive and effective" by virtue of profit margins, patents, marketing, and customer perception. I've been honestly rather shocked once in a while, to see a nurse wearing a stethoscope, or checking BP by sphig.

I mean, do you realize how much of a fortune those guys are making by selling adhesive nasal strips, after we all developed a huge concern for apnea? What's that line about the American Space Pen and the Russian pencil???

>Doctors do not order tests out of personal concern for patient's well-being to start healing them

I don't order server rebuilds because out of personal concerns for a companies well-being. WTF kind of standard is this. Maybe if we had 10x as many doctors we could afford some personalism with our healthcare workers.

>High-tech is able to supplant "safe, inexpensive and effective" by virtue of profit margins, patents, marketing, and customer perception.

No, again, 40 years ago, you just would have died of your ailment, or had life long suffering. With increased insurance coverage along with increased treatment offerings more people want and are willing to get what affects them diagnosed. Meanwhile we've not increased the number of healthcare workers at the same rate, and increasing with individual workers is very difficult and expensive.

Honestly your post history is concerning. It appears everything is a vast conspiracy against you.

Except the doctors ordering and performing and evaluating these tests are not CEOs.

Exactly: https://en.wikipedia.org/wiki/Superior_orders

If Deno doesn't get adequate funding, the company developing Deno will not exist 5 years from now.

Skimming this post was hard when I had to read past "woke, Soros funded district attorney". How do you people take yourselves seriously?

There were three points I saw:

1. Fraud isn't a crime, no one was harmed so we should be allowed to commit fraud freely

2. Some of the inflated properties actually grew over a decade to be worth nearly half of the amount declared and so it's not actually as bad as it seems (Read: It's still bad)

3. Biden is being investigated and people don't have the ability to follow two news stories at once

These three points really seem to be reaching desperately for any way to what-about away this case. I personally haven't interacted with a single person who thinks that Biden or his son should be let off the hook because of who they are if any evidence is found but every single Trump supporter remaining (You really have to be a specific kind of person to support Trump at this point) seems to think that Trump is above the law.

How about we agree to all hold our politicians and community members to the letter of the law equally?

To be fair it is far more complex for a robot to grip a spatula and use that spatula on a griddle than to use dynamic motion to flip a pancake in a pan.

Ehhh.

Solving any one problem with robotic manipulation isn’t all that hard. It takes a lot of trial and error, but in general if the task is constrained you can solve it reliably. The trick is to solve *new* tasks without resorting to all that fine tuning every time. Which is what Russ is claiming here. He’s training an LLM with a corpus of one-off policies for solving specific manipulation tasks, and claiming to get robust ad hoc policies from it for previously unsolved tasks.

If this actually works, it’s pretty important. But that’s the core claim: that he can solve ad hoc tasks without training or hand tuning.

  > He’s training an LLM with a corpus of one-off policies for solving specific manipulation tasks, and claiming to get robust ad hoc policies from it for previously unsolved tasks.

It seems clear that many people do not understand that this is the key breakthrough: solving arbitrary tasks after learning previous, unrelated tasks.

In my opinion that really is a good definition of intelligence, and puts this technique at the forefront of machine intelligence.

Is the pancake and spatula problem actually that constrained though?

I know it isn’t as open ended as plenty of more important problems in robotics, but this doesn’t strike me as easy at all.

I’ve only dabbled in robotics as an entry level hobbiest, so I really don’t know the answer.

It’s constrained enough to be tractable.

Fair enough. When would you say it stops being tractable? What single, practical thing could we add to this problem to make intractable?

Flipping a pancake in a "random kitchen" would be much more difficult and have many of the same issues as the door problem.

It's hard to point to a single thing that would make "flipping pancakes" intractable, it's sort of the other way around, to usefully flip pancakes in the same way as a person takes a lot of skills chained together.

The "door problem" is a sort of compendium of many real-world skills, identifying the door, understanding its affordances and how to grip / manipulate them, whether to push or pull the door, predicting the trajectory of the door when opened, estimating the mass of the door and applying the right amount of force, understanding if there any springs or pulls on the door and how it must be held to traverse through it. Etc. There are also a ton of things I'm missing that are so fundamental one tends to take them for granted, like knowing your own size and that you can't fit through a tiny doorway.

I think you can ramp towards the "door problem" in difficulty by slowly relaxing constraints. A video linked above (not article) shows "can flip a pancake successfully with a particular pan (you are already holding) and pancake with a fixed camera and visual markers". Ok, now do it in varying lighting conditions. With no visual markers. With different camera views. Different pancakes. Real pancakes (which are not rigid, and sometimes stick to the pan). Different pans. Now you have to pick up the pan. Use a stove. Different stoves. Identify griddle vs pan and use the right flipping technique. Find everything and do it all in a messy kitchen... eventually you're getting to same ballpark as the "door problem".

physicist here (so very naive on these topics) - I’m wondering how to compare the steps you mention regarding the door problem (especially the predictive ones, e.g. about the trajectory of the door as it opens, etc) with how humans open doors? Surely people don’t stop in front of a door and begin planning things out, rather they seem to go for it and adjust on the fly, is this an approach that won’t work in robotics? Why not?

So classical robotics yeah, people used to write code for each step of opening a door. Practically speaking you would probably not do motion planning on the door, you would just code it up with a bunch of heuristics like, try to be over here in relation to the doorframe because that's a good opening spot and will probably work. Ok you're in the right place? Now, move gripper towards the door handle... etc. Bunch of hacks. Put enough hacks together and you can kinda sorta open (some) doors. Oh this is a SLIDING door? Damn we forgot to code for that...

The way things are going is sensors (cameras, force, etc) and neural networks. You let the robot try a bunch of ways of opening doors, sometimes it doors itself in the face, eventually it'll figure out good places to stand based on what the door looks like. The more doors you make it try to open hopefully the better it gets at generalising over the task of opening doors. The hacks/heuristics are really still there but the robot is supposed to learn them.

> Surely people don’t stop in front of a door and begin planning things out, rather they seem to go for it and adjust on the fly, is this an approach that won’t work in robotics? Why not?

Yeah, figuring out how to do this is basically "the problem". Most people don't have a sense or feeling of "planning things out" as they open a door because we reached "unconscious competence" at that task. We definitely have predictions of what is going to happen as we start opening the door based on prior experience and our observations so far. If reality diverges from our expectations we will experience surprise, make some new predictions, take actions to resolve the surprise, etc.

Not sure that anyone has ever studied how people open doors in detail, it'd be interesting. I bet there are a ton of subtle micro behaviours. One that I know is, if you hear kids running in the house it is a good idea to keep a foot planted in front of you as you approach the door, because those guys will absolutely fling or crash doors open right into your face.

Thank you, great answer. As soon as I had asked my question I realized that we must have a lot of unconscious behaviours. Very interesting points about surprise/expectations. And top marks for the advice about kids

What makes you think a kitchen would have to be random? We regularly design physical spaces to accommodate robots.

I was responding to address why the "door problem" is more difficult than "pancake flipping under controlled conditions".

(I also ignored that door opening is generally done by mobile robots of a certain weight class which tend to be more expensive than a stationary arm with enough strength to pick up a spatula or hold a pan).

There is a steep difficulty gradient from "works in the lab" to "works under semi-controlled real world conditions" to "works in uncontrolled real-world situations".

> However, the human should receive a decent standard of living, regardless of their employment, which may be politically impossible

They'll get pie in the sky when they die

Yes there is a lot of gatekeeping happening across the board. From my wife's experience living just outside of Portland Oregon:

- Prescription has been cancelled twice for no real reason.

- Pharmacists have refused to give more than one month at a time more than once.

- Even if everyone is on board and not hostile she's only able to get a few months at a time so she has to do this multiple times a year.

I can't imagine how it'd be if we lived in a more conservative state or in a more rural area.

I've had some prescription issue for other meds as well. Usually an issue of the insurance or pharmacy, not a political motive. They don't like to give long term supplies given the expiration is 1 year from manufacture.

At least once she's been lectured about starting a family by pharmacists when trying to pick up her prescription which caused us to switch to a different pharmacy.

It's possible it's all imagined, but I don't really understand why a pharmacist would only give one month when the prescription is for multiple months, or why the prescription can't be for longer as to require less doctors visits / bills. We certainly haven't had the same experience picking up other prescriptions.

"why the prescription can't be for longer as to require less doctors visits / bills."

They aren't just renewing your prescription? Most places see you once, or once per year, and then just keep renewing it.

Most insurances/pharmacies even have partner mail order services where you can turn on auto-renewal and auto-refill. That might be a good option to look into.

Your original question was whether there were barriers. Now you're clarifying you mean political barriers determined by the intent of the provider, which is of course opaque to us. Is it possible under these constraints to describe a barrier you'll accept?

Generic issues are not barriers, especially if they are not universal. There are online services that can cheaply provide these services conveniently. Just because someone uses a provider that provides poor service is not a barrier when alternatives exist.

The one month at a time thing is coming from insurance companies not the pharmacy.

Not necessarily. Pharmacies have arbitrary power.

My doctor gives me a three month prescription for a scheduled substance and I've had a couple pharmacies reject it because it's "sketchy". Of course, they never offer to call my doctor to confirm anything. They just smirk and cross their arms.

Even OTC amphetamines have purchased limits, some imposed by law depending on the state.

I guess you're suggesting that they rejected my prescription because of supply issues, they openly said it was because it was suspicious (3-mo in cash), and also Vyvanse hasn't had shortages like generic amphetamine in 2022-2023.

Also, I think you mean OTC amphetamine alternatives since amphetamine is sched I.

Amphetamine (Dexedrine) is schedule II, not schedule I. Doctors cannot generally prescribe schedule I substances like LSD. Pseudoephedrine is a substituted amphetamine so referring to it as "an OTC amphetamine" is technically reasonable although potentially misleading.

These pharmacists have very real personal liability if they end up getting fingered as being a supplier of components used to make illegal drugs. That's why they look askance at this prescription.

Not supply issues, meth manufacturing suspicion. Yeah, precursors to their manufacture.