ulnarkressty

Did you get custom molded ones or are there more comfortable generic earplugs?

An experiment from an old physics textbook: trace a convex shape on a piece of paper, wet its contour with a brush and drip some water on it until it beads up. You then place the paper on a still water surface and touch the surface of the bead with a needle. The surface tension will move the shape under the needle until it points at the geometric center of the shape.

Germany as well, this year there are huge swarms of mosquitos everywhere on the hiking trails. In the past you could find them deep inside forests, now they are even around the fields in broad daylight. It's unprecedented.

Not sure what is going on, maybe they all hatched before their predators came around. Haven't seen one dragonfly yet.

There are no exercises that restore joints which are degenerating. Most likely you would have gotten 5-10 more pain-free years without the running, but it would have come anyway at some point. The increased cardio health and bone density does help sweeten the terrible deal though.

+1 for fanless. I got one of these optiplexes as they were recommended everywhere. They are indeed more powerful but very closed down. For example one cannot even control the fan speed on them and I hear it whine at idle all day long. If I try to unplug the fan from the motherboard, the whole computer won't boot.

Also, what I could gather from a youtube presentation is that they are electricity hogs (compared to a 5W Pi) unless you get a >9th gen box which has some better low power states (which may or may not be supported by your distro).

You wouldn't be the first to think of this

https://www.theatlantic.com/international/archive/2014/07/fl...

Storing heat in bedrock sounds like a good idea, but there are risks, e.g. https://www.thelocal.de/20170818/this-historic-german-town-i...

As a backer on the original Oculus kickstarter, I have such a sinking feeling in my stomach every time this comes up. My money went to enable Luckey to achieve this and I hate myself for it.

Microsoft's takeover of Mistral makes a lot more sense if this lawsuit has a chance to succeed.

Take this with a grain of salt until someone reproduces it. Improvements such as these require extraordinary evidence. Not to mention extreme quantization has been tried before.

I do hope that they have a documentary team embedded in this company, like DeepMind had. They're making historical advancements on multiple fronts.

To put it into perspective, the Will Smith eating spaghetti video came out not even a year ago --

https://www.youtube.com/watch?v=XQr4Xklqzw8

Oh man, I had this idea a long time ago when I was still commuting to work during sweltering hot summers in trams with no AC. Glad to see someone implemented it.

If you want to take it to the next level, use GIS information to account for building shadows and waiting times in stations and at traffic lights.

I just started using the metro instead :)

On non-accelerated tasks the performance is very close[0] and at half the idle power consumption. For the specific case of a low-power always-on platform, it's still the best.

[0] - https://www.youtube.com/watch?v=hekzpSH25lk

As exciting as it is to design a low power system, it's kind of pointless in the case of a NAS that uses spinning rust as storage media - as the author later writes, the HDD power consumption dwarfs the other system components.

If one uses SSD or M.2 drives, there are some solutions on the market that provide high speed hardware RAID in a separate external enclosure. Coupled with a laptop board they could make for a decent low power system. Not sure how reliable USB or Thunderbolt is compared to internal SATA or PCIe connections though... would be interesting to find out.

Why not try it for a month or two, if it's not any better you can go always go back. I read comments here that some take it during the week to get stuff done and go without on the weekends when they want to be more creative or whatever.

Any ideas on how to help my engineer colleagues write good documentation that actually helps? Most of the problems that I see come from a lack of empathy - they assume the reader has previous knowledge about things that are only inside their head, so after reading the docs there's always the need for further clarification.

I suggested some technical writing courses, but they got weird ideas from them, like writing documentation in a conversational style, which makes it somehow even worse...

Someone needs to invent something similar for sleeping. I would pay good money to have a completely silent sleep environment. Right now ear plugs are the best way, and not at all comfortable to sleep with.

Traffic, neighbors, construction - there's always something going on in a big city. Even having a detached house in the suburbs is not guarantee of a quiet night. Sleep interruption due to noise can have disproportionate impact later in life [0].

I had some ideas about a 'room within a room' spanning just the bed, and I couldn't find a good source of materials. Best bet would be to fill the whole thing with concrete and hope it's acoustically isolated from the building itself.

[0] - https://sleepreviewmag.com/sleep-health/sleep-whole-body/bra...

Successful CEO does successful CEO things. From another thread: https://news.ycombinator.com/item?id=38360906

Eh, at this point are there any successful people that don't have these 'qualities'? We raise our children with stories about great men and fair leaders, when reality is nothing of the sort.

How would this actually work in practice? Do I ask the user to utter specific words then train on that? How is it different from the traditional speech recognition that I need to 'train' to work better on my voice?

The Holy Grail would be to train the model while using it, without any friction. I don't think these methods support that though.

I have improved my deep sleep by taking lavender oil extract capsules. They help relax and give me a 'knock-out' feeling that rapidly induces sleep. I also don't wake up in the middle of the night anymore when I take them. Unfortunately the quality of the extract varies wildly, I would get the same effect with 80mg as I would with 400mg from another supplier. Doubly unfortunate is that one rapidly builds tolerance to it. I use it every 3rd day, so I get a deep, restful sleep twice a week.

Quite a few sleep supplements have started adding it to their formulas recently, using melatonin to initiate sleep and lavender oil to sustain it. It remains to be seen what side effects it has taken long term - there aren't any studies about this available as far as I could tell.

> It's not that onerous, except that I assume going under propofol + midazolam comes with cognitive risks each time, and this might be cumulative.

Do you have any sources for this? I'm also wary of going under, even if it's once every 3 years.

I'm curious how the rewards in the "best places" to work look like. There was a famous study that found that developers are happy when they get enough money not to care, but in real life this doesn't seem to be the case. Whenever I speak with my management buddies, they always lament about engineers who complain they're not being appreciated enough. When asked what proper appreciation would look like, they always give various answers which always boil down to money (i.e. "you could praise me in front of everyone for the great job I did, but actually want a raise / more money than the guy I perceive to be a slacker").

If we're being honest, one is employed by the company to make them money, so it will always more or less boil down to this. Would anyone be satisfied that they made their boss a million more, while getting a pat on the back and the same salary?

Is there any proof that doing this regularly helps with anything? Not necessarily making one 'smarter', but at least keeping dementia at bay for older folks? All I could find is that playing board games seems to help with cognitive decline, but this could also be the social part of it.

If it's 'composite' then it most likely contains some binding material apart from wood, which will still get into your food.

More targeted ad campaigns, as per the article. The interests will always be there.

Most of the improvements apparently come from training larger models with more data. Which is part of the problem mentioned in the article - the probability that the model just memorizes the answers to the tests is greatly increased.

AI is getting subjectively better, and we need better tests to figure out if this improvement is objectively significant or not.

They look interesting but are they durable? I can see the bolts ripping through the plastic if riding in rough terrain. All panniers that I've used have the mounting points reinforced so that they are not loaded in only a couple of places.

What I find even more amazing is that he's been inhaling oil and plastic smoke for 50 years and yet he seems to be in good health.