I'm pretty sure most portable ACs sold in Europe these days are R290 (propane) which has a GWP of about 1/14th of CO2, so for those that's no longer much of a concern.
I do the same as the poster above, fastmail supports it directly and makes it very easy to manage. All you have to do is bring your own domain (they'll even manage your DKIM/SPF records etc as necessary if you want).
> Excess deaths actually seem to line up better with the vaccine rollout than with the initial viral outbreak.
This seems to be a bizarre statement. I can't speak to other countries, but at least in England the widespread rollout of vaccines (just beginning in January 2021) only started at the end of the final major spike in excess deaths.
It's simply not possible for excess deaths to line up better with vaccine rollout than with the initial outbreak, because both major spikes in excess deaths were before mass vaccination rollout.
NIST does not refuse hybridization, they will be publishing guidance on hybrid schemes in the draft of SP 800-227 at the same time as the final standards. They don't impose it though, because at a large scale it's more efficient to run just (fast) ML-KEM instead of (fast) ML-KEM + (slower) ECDH, which more than doubles your computation time for what they see as no benefit.
Generally recycling aluminium is incredibly cheap, especially if it's just a solid block without the regular potential contaminants, although I don't know what specific alloy they're using. If it's something like 6061 then it's especially easy to recycle, afaik, because it's very low in silicon and iron.
Do you know of any large tech companies that have more environmental ethics? I'm not sure Apple is good about it, but sadly I don't know of any better.
There's a throwaway comment in here about the earth's magnetic field affecting CRTs and turning the TV upside down, is that actually true? I would have thought the earth's magnetic field would be incredibly weak in comparison to the deflecting coils.
Interesting. uname -a reports x86_64, and lscpu also reports x86_64, although perhaps that's just the kernel being patched to lie about the architecture.
I regularly use Orbstack to develop for x64 Linux (including kernel development). It works transparently as an x64 linux command line that uses Rosetta under the hood, so performance is reasonably good.
It can also run docker containers, apparently faster than the normal docker client, although I haven't used that feature much so I'm not sure.
RCS has no end-to-end encryption in the standard, though. That's a non-standard google messages extension. I think a better example would be the cross-device messengers like signal/whatsapp/etc.
Traffic restrictions are not part of the "15 minute city" initiatives. I think they are often confused because of Oxford's traffic filter scheme. As the article you link says, tying walkable city initiatives to one particular city's traffic scheme is indeed a conspiracy theory.
If you have valid criticisms of the traffic filter trial in Oxford then that's fair enough, although I imagine Oxonians won't be particularly interested in hearing it unless you're a local resident.
Notably it does bound the energy of gamma rays over long distances (as the higher the energy the more likely it will annihilate with other photons along the way.)
The wikipedia article on the Breit-Wheeler process has some history of the work on experimental observations, although I don't know how accurate or up to date it is https://en.wikipedia.org/wiki/Breit–Wheeler_process
The more recent 4 bit quantizations are almost along these lines. Q4_1 in ggml for example takes a block of 32 weights and gives each block a scaling factor 'd' and takes the minimum of the weights 'm' to be the quantized '0', so the final weights from a quantized weight 'q' is q * d + m, and taking a relatively small block size makes it more likely that those are all within a reasonable quantization range. Notably, d and m can be stored with more accuracy without sacrificing too much space, since the overhead is divided by 32. Q4_k goes a bit further, and takes 'superblocks' of 8 blocks, and applies another scaling factor 'd_s' and minimum 'm_s' to that, so the final weight is (q * d + m) * d_s + m_s, and the additional factors are stored as 6 bits instead of 4.
In practice this seems to get very good results, while being cheap to implement and relatively space efficient, Q4_K for example works out to 4.5 bits per weight instead of 4. The PR adding it has more details: https://github.com/ggerganov/llama.cpp/pull/1684
QEC implies phase error correction (if you don't care about phase errors you could just use a classical error correction code). A 'logical qubit' refers to an abstraction layer that implements some sort of idealized qubit behaviour (including the phase) that is used as a qubit in computation without considering the underlying implementation. The underlying implementation of a single logical qubit may be something like 5 'physical qubits' using https://en.wikipedia.org/wiki/Five-qubit_error_correcting_co... or a entire structure of many physical qubits using something like a https://en.wikipedia.org/wiki/Toric_code
You can run it on either now (for example, MochiDiffusion allows you to pick https://github.com/godly-devotion/MochiDiffusion#compute-uni...). Anecdotally, the GPU seems to be faster for an M1 Max or up GPU, the ANE is a touch faster on anything smaller, and more power efficient in general.