nowahe

I have a Caldigit USB-C pro dock with a macbook pro m1 plugged into it, and I have been having random and inconsistent (sometimes once or twice a day, other times nothing for a couple of weeks) sudden devices disconnection. What's weird is that suddenly my keyboard, audio out and ethernet stops working, but my 4k monitor plugged in Display Port keeps working without a hitch.

I'm pretty sure it's an issue with the USB controller, and the article convinced me more, but I still have zero clue on how to begin troubleshooting an issue like that.

For those hitting the 451 Bandwidth Restricted : https://web.archive.org/web/20220504141057/http://collapseos...

Working mirror : https://web.archive.org/web/20220419110330/https://www.mattm...

Another project worth noting is Streamrip[1] which does the same but for Tidal/Deezer/Qobuz, and outputs FLACs.

[1]: https://github.com/nathom/streamrip

Also, from what I've seen from previous similar previous projects, they get removed from Github pretty fast by Spotify

On a sidenote, to build the list I had to comb through my subscription and load the relevant channels (and go over their videos a bit), and since doing so my YouTube recommendations are significantly better, with recommendations of small technical channels in the same spirit as Tom7's

Here is a list of some channels (in no particular order) that are similar to Tom's, in the knowledgeable host talking about niche/complex subjects :

- Huygens Optics[1] : Retired Philips' research scientist that worked on OLED development and makes highly detailed videos on different subjects, all relating to light (eg: Making optical logic gates using interference, DIY photolithography, Making a Mirror with a Variable Surface Shape).

- Marco Reps[2] : German guy doing electronics videos, usually about precision/exotic instruments (eg building a CERN open source 8.5 digits voltmeter, High speed thermal camera, Detecting cosmic rays with a raspberry pi). Has the same type of humour as Tom7

- Styropyro[3] : Laser hacker, creating stupidely powerful lasers (think 100W continous handeled laser)

- CuriousMarc[4] : Guy working on restoring old 60-80s electronic/computer equipment. Is most known for his ~25 part series on restoring the Apollo Guidance Computer. As a sidenote, Ken Shirriff, which writes die reverse-engineering articles at righto.com (which frequently pops up on HN), frequently appears in his videos.

- Scanlime[5] : Highly detailed reverse-engineering/hacking/explanations videos. Sadly she isn't that much active anymore.

- der8auer EN [6] : German guy doing mostly videos about CPUs/motherboards (think insane overclocking, etc). However he has a few hugely detailed gems, such as demonstrating how 7nm chips are troubleshooted, by probing them under a SEM[7]. Also as a sidenote, he shoots all of his videos twice, once in german and once in english, instead of just dubbing over them, which is an insane dedication imo.

- Applied Science [8] : Relatively well known channel, describing various experiments that he does in his home shop, like : Building a LCD, Silicon wafer etching, DIY mass spectrometer, etc.

- Cathode Ray Dude [9] : In-depth technical videos about mostly vintage video gear, which at first glance you wouldn't typically find interesting. Same kind of nerdy humour as Tom7

- Technology Connections [10] : Hard to describe the style, mostly videos about technical subjects, going much more in depth than you though was possible, with a really nerdy/dry humour.

I might have missed some, but those are the main ones that comes to mind.

[1]: https://www.youtube.com/channel/UC26YLK0OEbLB3TCYxGh8xVQ/vid...

[2]: https://www.youtube.com/channel/UC1O0jDlG51N3jGf6_9t-9mw/vid...

[3]: https://www.youtube.com/channel/UCJYJgj7rzsn0vdR7fkgjuIA/vid...

[4]: https://www.youtube.com/channel/UC3bosUr3WlKYm4sBaLs-Adw/vid...

[5]: https://www.youtube.com/c/scanlime/videos

[6]: https://www.youtube.com/channel/UCGsaijjOJshS2_ZmMNZgS-g/vid...

[7]: https://www.youtube.com/watch?v=EKb3OAoGeIw

[8]: https://www.youtube.com/c/AppliedScience/videos

[9]: https://www.youtube.com/c/CathodeRayDude/videos

[10]: https://www.youtube.com/c/TechnologyConnections/videos

Having a really good ability of viewing them, I feel like it relies on two separate skills :

- Fine voluntary control of your eyes' accommodation

- Ability for whatever system is creating your depth perception to be more "flexible", in the sense that it can more easily "lock" onto a depth perspective as well as switching between them.

Thinking about it now, I'm not even sure how much voluntary control people normally have over their eyes' accommodation, if not exposed to experiences that forces them to practice it. I can definitely remember a time where I was not able to focus my eyes at will, which makes me think that it's not something the majority of people can do (granted it was when I was a child, so it might just be something you learn with age).

I fell in the Magic Eye rabbit hole when I was 6-7, and spent so much time on it that I can now make the images pop in ~2-3 seconds, even 20 years later. I could also stare at them for hours without significant fatigue, as once you find the correct alignment, your brain will "lock" into the new perspective and you can totally relax your eyes.

One side effect I've noticed is that I have a really hard time looking at/focusing on geometric repetitive patterns. I will have the same effect as with Stereograms, namely that if my focus on the object is not close to perfect, the double image I see will have a tendency to really easily "lock" into the next alignment, rendering depth perception close to useless (even tho it's kinda fun/trippy to look at).

On the other hand, if I ever fall alseep in front of YouTube with the autoplay on, I will more often than not wake up in front of either Tom Scott's unedited video about sending garlic bread to space[1] or Micheal from Vsauce reciting primes for 3 hours[2] (which tends to result in some pretty interesting dreams tho)

[1]: https://www.youtube.com/watch?v=YKAblynZYhI

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

Thanks a lot for the link, I've been struggling to find good technical explanations/ressources on the subject

I don't think it's possible with GPT-3, and that's mostly due to how the text is parsed into token before being fed to the network [1]. It breaks down the text in ~4 words token, which allows to effectively quadruple the max input size, at the cost of loosing fine details on the input data. It leads to issues like not being able to create rhymes, not understanding humor or not being able to parse fine structures. Gwern has a nice article talking about the limitations introduced by it [2].

[1] https://beta.openai.com/docs/introduction/tokens

[2] https://www.gwern.net/GPT-3#bpes

It's using GPT-3 as the underlying technology, which has been open to everyone for a few month now [1]. More specifically, it's using the codex model for code generation (which is still on a waitlist to access I think), but if you're more interested in the text/Q&A style generation it won't matter much.

[1] https://openai.com/blog/api-no-waitlist/

A few tips/software recommendation in no particular order:

- On iTerm2 you can setup a system wide hotkey to toggle a floating terminal in front of all your windows, whatever you're doing [1]. This works even in full screen apps. It really changed my workflow with iTerm, and is really handy.

- For system stats in the menubar, I really recommend iStat Menu [2]. It's paid (12$ for a license), but the quality of the graphs and interface makes it well worth it imo.

- If you have an external display and want to precisely tweek your resolution, I recommend SWitchResX [3]. It's mostly a GUI that wraps some display CLI tools. With it I adjust the resolution of my 4k external monitor currently, running it at 3200x1800 with HiDPI, and adjust it based on my eye fatigue.

- As a more powerful and versatile replacement to Spotlight (Cmd+Space), Alfred [4]. You can customize it to run anything you want, from converting timestamps to dates, opening apps, search the whole system, run scripts, etc

- If you have too many icons in your menubar and want to hide some (either totally or behind a button click), Bartender [5]

[1]: https://apple.stackexchange.com/questions/48796/iterm-as-a-s...

[2]: https://bjango.com/mac/istatmenus/

[3]: https://www.madrau.com/

[4]: https://www.alfredapp.com/

[5]: https://www.macbartender.com/

It appears [1] that Russia's Ministry of Defense website is getting blocked for all IPs outside of Russia.

This seems deliberate as the request's answer is a HTTP 418 :

  $ curl -v "https://eng.mil.ru/"
  *   Trying 82.202.189.138:443...
  * Connected to eng.mil.ru (82.202.189.138) port 443 (#0)
  [...]
  > GET / HTTP/2
  > Host: eng.mil.ru
  > user-agent: curl/7.77.0
  > accept: */*
  >
  * Connection state changed (MAX_CONCURRENT_STREAMS == 128)!
  < HTTP/2 418
  < date: Fri, 17 Dec 2021 16:24:13 GMT
  < content-length: 0
  < server: Ministry of Defence of the Russian Federation
  <
  * Connection #0 to host eng.mil.ru left intact

[1] https://twitter.com/tom_bullock_/status/1471867831629266951

I can't access anything related to Cloudfront, either through the CLI or console :

  $ aws cloudfront list-distributions

  An error occurred (HttpTimeoutException) when calling the ListDistributions operation: Could not resolve DNS within remaining TTL of 4999 ms

However I can still access the distribution fine

Have you ever looked at the size of the standardize EU QR code[0] ? At just a glance it looks at least 2-3x denser than the one presented in the article. Also, on the specification published by the EU[1], they seem to advise for a 25-60cm size, and also warn about using <300dpi printers.

When I was vaccinated, I was given an A4 sheet with those QR codes on it, and I really wouldn't want to scan those with a crappy camera.

[0]: https://gir.st/blog/greenpass.html [1]: https://ec.europa.eu/health/sites/default/files/ehealth/docs... § 4.2.2

I also share the same world view. With every person that I met, it was just a question of finding what area/subject they spend a lot of time doing. They will be more knowledgeable on it than you, and by accepting it and listen to their expertise you'll learn a lot about some subjects you didn't even know existed.

Octave Klaba (OVH CEO) on Twitter [0] :

  Following a human error during the reconfiguration of the network in our DC in VH (US-EST), we had an issue on all the backbone. We're gonna isolate the DC VH then fix the configuration

  Those last few days, DDoS attacks intensity has grown a lot. We decided to increase our DDoS handling capacity by adding new infrastructures in our DC VH (US-EST). A bad router configuration caused the network crash.

[0]: https://twitter.com/olesovhcom/status/1448196879020433409

For me, it's not the plastic casing or even the battery that seems the biggest waste, but their ICs.

It's not that it's polluting the environment that much, more that it's wasteful and creates a unnecessary demand in IC production when they are thrown away in perfect working order.

Native french speaker as well, and I've never heard or used it in software development.

However, I have multiple friends in Architecture schools, like INSA (Institut National des Sciences Appliquées), and it's widely used there

I'm in the middle of a migration from Akamai to Cloudfront, time to take a break I guess

Kinda unrelated, but for me the best and first indicator that I'm starting to get sick is cigarettes tasting different.

I've been using icloud-photos-downloader[0] for a while now, running as a Scheduled Task on my Synology. It supports 2FA as well, I just have to re-auth every 3-4 months.

[0] : https://github.com/icloud-photos-downloader/icloud_photos_do...

Fun fact, the "non assistance à personne en danger" law was passed during WW2, when France was under the collaborative Vichy Regime.

It was to compel French citizens to help injured German soldiers.

Might have mixed up Chime & IVS a bit.

To clarify, the 4-5s delay I'm talking about is for IVS, which is a live streaming solution, and doesn't have anything to do with video chat. It has a RTMP input on which you send a video stream, re-encodes it and then distribute it as an HLS stream. And getting sub 5s latency on an HLS stream is pretty fast.

I brought up Chime and video chat mostly as an example on how AWS is pushing for RTMP to connect those services.

Having worked with AWS IVS, which is using the same infrastructure design as Twitch (or the other way around), it only accepts RTMP (well RTMPS to be fair) as an input.

To me it would seem that the industry is going the opposite way, and doubling down on RTMP. An RTMP connector is planned on the Chime SDK as well (a framework & infrastructure based on Chime to create your own custom meetings), to be able to stream the meeting directly to ingest service (like IVS/Elemental). On the AWS side at least, there doesn't seem to be a plan to migrate from RTMP, as it seems that services are being launched that only supports it.

On a sidenote, the official AWS recommended way to stream a Chime SDK meeting to IVS, is to use this[0] docker container, acting as a bridge between WebRTC & RTMP. I find the hackiness of it amazing, as it's mostly just an X11 framebuffer, an instance of Firefox loading the webview for the meeting, and ffmpeg capturing the framebuffer and sending it as an RTMP stream to the configured endpoint.

Also, AWS IVS as a streaming platform is really impressive. It's a fully managed service, and out of the box you get around 4-5s of latency between the source and the player, with an upper limit of 1080p60 @8.5Mbps. It will also downsample the stream to 720p, 480p, 360p & 160p.

[0]: https://github.com/aws-samples/amazon-chime-meeting-broadcas...

Also, as a fun addition, Matt Parker (and others) built [1] a 4 bit binary adder out of about 10,000 dominoes

[1] https://www.youtube.com/watch?v=OpLU__bhu2w

Not especially true for all AWS Services, there has been some services that were only available on CF (Workspaces comes to mind).

But in this case it wasn't Hashicorp fault, as what was available in CF wasn't in the AWS official API.

It took something like a year for AWS to release the API

We have a saying in France about the minitel. It first put us 20 years ahead of everyone, and then 20 years behind