pyentropy

I don't want to read the whole thing because it looks like it's very cynical - it judges TCS scientists who believe in Grover speedup as naive because they are unaware of real life noise, without actually realizing that's the point of TCS.

We don't know how noise will scale IRL so the job of theoretical scientists is to design the basic units of quantum computation regardless of how it may or may not work IRL. It's like judging XOR and NAND in 1920s because transistors maybe won't be able to simulate them.

> the authors haven't used a typewriter

The authors are the famous Kernighan and Brailsford (!). They spent their life on stuff like this, and know him personally very well. They are talking about spacing much finer than half-character.

No you can't. Read the post.

Wait... zsh which is quite a powerful shell that's now by default on macOS too, would say "(" is equal to ")" until 2016?

Was it just single [ (test) or [[ as well?

I respect that. Can you elaborate a bit on the routing protocol thing? I assume you used WAN gossip?

I love the simplicity of fly.io & wish you all the best improving Fly's reliability!

Almost half of the issues are caused by their use of HashiCorp products.

As someone that has started tons of Consul clusters, analyzed tons of Terraform states, developed providers and wrote a HCL parser, I must say this:

HashiCorp built a brand of consistent design & docs, security, strict configuration, distributed-algos-made-approachable... but at its core, it's a very fragile ecosystem. The only benefit of HashiCorp headaches is that you will quickly learn Golang while reading some obscure github.com/hashicorp/blah/blah/file.go :)

A decade of Siri research & development... overshadowed by a single clever GPT-3 prompt. Love it.

Go is the best documented ecosystem I've seen. Since godoc is so widespread, people write doc comments and examples all the time. Every single public project has a doc page at pkg.go.dev too.

The syntax rules and the reference spec of the language is very concise too.

The core code lives in the SerenityOS repo (which is an OS built from scratch too!):

* LibWeb (HTML, CSS, SVG, Web APIs): https://github.com/SerenityOS/serenity/tree/master/Userland/...

* LibJS (JS engine): https://github.com/SerenityOS/serenity/tree/master/Userland/...

* LibWasm (WASM interpreter): https://github.com/SerenityOS/serenity/tree/master/Userland/...

and the UI lives in https://github.com/SerenityOS/ladybird and can be built for Linux/Mac/Android.

Sounds believable to me. Google is really good at data warehouse & database tech (BigQuery, BigTable, Spanner) and access management systems, as well as sturdy custom-built tools for internal use by employees.

Even though it might not be intentional, Apple is lagging at such tooling.

> point-in-time backups through the web UI

Is `fly postgres create --name restoredDb --snapshot-id backupId` that hard that's it's a deal breaker?

> support for Heroku buildpacks

I haven't tried it but there's some buildpack support: https://fly.io/docs/reference/configuration/#builder

Update: it completely breaks with React 18 and strict mode enabled - https://github.com/preactjs/signals/issues/70

The implementation involves unsafe monkey-patching:

1. For React, a hook is injected into every single component (!) regardless of whether it uses signals or not via __SECRET_INTERNALS_DO_NOT_USE_OR_YOU_WILL_BE_FIRED.

2. For React, React.createElement is patched so that it can render signal values as Text nodes.

3. For Preact, they claim it's using a "pluggable renderer", yet they monkey-patch the global shouldComponentUpdate (and they don't call the old one so they break other patchy libraries)

[1] - https://github.com/preactjs/signals/blob/d25e8bac09c94ed3bad...

[2] - https://github.com/preactjs/signals/blob/d25e8bac09c94ed3bad...

[3] - https://github.com/preactjs/signals/blob/d25e8bac09c94ed3bad...

Funnily enough, you also increase the app opening rate - since user sessions for that app itself are persisted!

Let's say I send you tempemail.org via Messenger and you use it to create foo@bar. Then you close Messenger and forget about it but you realize you need foo@bar again. The only way to get it is through the Messenger app :/

I hate that if I send Wordle or something cool to a friend that uses localStorage they lose their progress/settings once they leave the chat app.

So frustrating to even explain to people that this thing they are scrolling isn't their own, Safari/Chrome!

Currently watching the launch event! :)

I took a look at the low level stuff, compiled it and tried breaking it and pushing it to the limits. It's quite stable, something that you would expect from a Python Core developer as a founder.

My only concern is that the actual `edgedb-server` CLI isn't documented at all.

Scott Alexander has a cool theory that unifies Bayesian inference, low-level behaviour like neurotransmitter levels and high-level behaviour like learning, making decisions and different sensory phenomena (some of the posts explain autism and schizophrenia):

* https://slatestarcodex.com/2017/09/05/book-review-surfing-un...

* https://slatestarcodex.com/2016/09/12/its-bayes-all-the-way-...

* https://astralcodexten.substack.com/p/towards-a-bayesian-the...

* https://astralcodexten.substack.com/p/trapped-priors-as-a-ba...

The core concept is called predictive coding, which suggests that the brain is constantly generating and updating a model of the world in a top-down and bottom-up manner: https://en.wikipedia.org/wiki/Predictive_coding

Disclaimer: epistemic status = very speculative.

Take a look at Scott Alexander's article on sleep deprivation and mood disorders: https://astralcodexten.substack.com/p/sleep-is-the-mate-of-d...

It has some theories on a synapse level and some AI/NN analogues regarding the paradox of "good sleep = good mood" and yet "no sleep = relief" for depressed people.

Fun fact: staying up for 24h+ causes a relief of symptoms in >50% of depressed individuals. However, almost all relapse the moment they sleep again (even if it's a 'micro-sleep').

https://en.wikipedia.org/wiki/Wake_therapy#Response_rate_to_...

You had a depressive episode, and you recovered.

It's not getting older, but the fact that time heals most brains just like exercise and antidepressants do. About 70% recover within 12 months [1]

[1] - https://pubmed.ncbi.nlm.nih.gov/12204924/

Edit:

OP said "Was depressed that [...]" and also mentions death of a loved one. Even if I'm wrong, personally I really wish someone mentioned MDD as a possibility to me earlier.

I felt similarly, and I thought all my focus and motivation problems were because I was lazy; but it was quite the opposite, I was fighting depression very hard.

I lost so many years of my life reading about death coping skills, optimizing diet and exercise, learning how to be less lazy (?) when it had absolutely nothing to do with any of those.

I don't get it? Most of the "limitation" examples are on par with humans, both linguistically and artistically.

So much so that it feels like a DALL-E ad :)

1. Set realistic goals and keep in mind a lot of social stuff that people pull off isn't effortless, they put A TON of work in.

2. Confidence boosts: gym, clothing style, telling friends about your vulnerabilities and feelings (there's no better feeling than being told by a friend that what you worried is something they really never saw or cared about), inviting people to things.

3. Too much blushing and physical vulnerability may be fixable with exposure therapy or may require medication. It may also be a result of Asperger's. Don't be afraid of seeking medical help for stuff that's out of your control.

4. Realise that ALL HUMANS are broken in some fundamental way and that includes the person next to you. So don't be afraid to interact with them and do not assume they'll judge you harshly.

Higher parameter count and less face samples aren't mutually exclusive.

You can take a huge GAN and then use a single image of a face as a guide for navigating the latent space.

You can replace "out of their control" with "almost zero chance to go in the else{} branch".

And "can choose to" with "can go to else{} if rx is true" - and social interaction and reward/punishment can make rx == true.

... because it's a good idea?

If a person with Tourette's said profanities to you during your Zoom meeting, would you hold them equally responsible as a person not having it? No, because it's a symptom and you assume it is out of their control and they didn't have a malicious intent. If both persons said "sorry it wasn't in my control", you'd think

1) the person A didn't do bad, because they have a tic and they definitely don't have negative feelings towards me

2) the person B has an anger management issues and may or may not have negative feelings towards me, maybe he's bad

We assume intent is malicious when a person has the capability to stop themselves from doing something negative towards us and yet they "chose" not to.

A CPU that doesn't work when power is off could be a good CPU, but a CPU that doesn't work when power is on is a faulty CPU.

It's nice, but it doesn't cover any topic of Theoretical Computer Science [1]. It's just the math required for undergrad general CS courses.

[1] - https://en.wikipedia.org/wiki/Theoretical_computer_science

> how much freedom you have, how much protection you have, etc. and then ask yourself : where does it come from

Game theoretic behaviour in society that has advanced beyond zero sum.

Same goes for reasonable approach to FOSS, like Marcan is doing himself vs the cult and the arbitrary zero tolerance rules.

The repo/company you linked is about utilizing zero knowledge proofs which is literally the best we've got in regards to decentralization.

If anything, that gives him credentials to talk about it?

I'd rather have a non-portable SQL feature than an app-level "portable" implementation that almost certainly is slower and non-transactional.

Migrating to another DB will always take work, no matter how much you try to ignore flavor-specific features. Query planning, encoding and charsets, locking abilities tend to be very different. A query can run fine in MySQL and cause a deadlock in Postgres even though it's syntactically valid in both.

Should be enough from a FS perspective yes, although:

(1) the WSL distros don't have systemd

(2) they ship with a custom kernel that doesn't support custom kernel modules

First point is trivial to fix but (2) may be painful. You could even share the desktop environment as WSLg uses X11/Wayland.

edit: apparently compiling modules is (kinda) working with wsl2: https://github.com/microsoft/WSL/issues/4425