aarchi

This looks like a great service. There doesn't look like there's a way to post a bounty for a repo that's not already signed up though.

Great outlook on working culture and hours:

> We work less, by keeping our needs small. We both cannot deal with working within rigid schedules, and we both know that we find our best selves when working close to nature.

I see: Donald J. Trump's term ended on 2021-01-11 19:40:41.

I think this article is missing a decimal point: the archive is 12.4GB, not 124GB.

It would work great for LEGO mosaics as well

This was fun to play. I set it up today and we all enjoyed it. We noticed that it helps to jump forwards in the beginning until the cacti speed up, otherwise you'll usually land on the cactus. It's much harder to achieve a comparable score than on chrome://dino though. The only obstacles are cacti in this version, but I think ducking under birds would be far too hard.

Is there a service that provides crowd-sourced data on airplane reliability by registration number? The registration number is probably only revealed just before boarding though, so that would limit its utility.

> we could see the registration number: G-EZRU. The exact same plane that had flown out of Geneva, had both its pressure systems fail, had been known to have other problems in its service history

A FUSE filesystem that fetches objects on demand seems like it would be great when working in a large repo such as the Linux Kernel or Chromium, especially when only a small subset of files are needed. It would also be useful to monitor the fs for file changes since `git status` scans the entire tree. Does anything like this exist?

nth-ancestor(n) and upward(arg) look handy. They're uBO extensions.

https://github.com/gorhill/uBlock/wiki/Procedural-cosmetic-f...

More generally, is there a known correlation between kernel panics and exploits, especially on macOS?

> Almisshal’s device shows what appears to be an unusual number of kernel panics (phone crashes) between January and July 2020. While some of the panics may be benign, they may also indicate earlier attempts to exploit vulnerabilities against his device.

The gron tool mentioned by the author is incredibly useful for grepping json.

https://github.com/tomnomnom/gron

A particularly useful one from that list:

prepone - to reschedule to an earlier time

The page on defective verbs is also interesting.

https://en.wikipedia.org/wiki/Defective_verb

This attack and Van Eck phreaking are why Edward Snowden, while typing passwords and other sensitive information, would pull a blanket over himself and his laptop.

I have several highly-redundant NTFS backups that I'd like to compress into a read-only fs. Can DwarFS preserve all NTFS metadata?

Is it possible to rebuild a DwarFS fs to incorporate changes from an overlay fs without decompressing, then recompressing?

It seems feasible that a second DwarFS fs could be built from an overlay/DwarFS, then delete the original overlay/DwarFS fs. That would require 2N storage as the new DwarFS is being built. Is it possible to patch an existing DwarFS?

By overlay, are you referring to overlayfs [0]?

https://wiki.archlinux.org/index.php/Overlay_filesystem

Why do you say that it's undesirable for your registrar to handle mail? I've considered using Gandi's mail with my domains.

Radio Spätkauf's podcast series (in English) “How to F#€k Up an Airport” investigates the BER mismanagement over time and is quite entertaining and well informed.

https://www.radiospaetkauf.com/ber/

MLIR and Verona are both ambitious state-of-the-art compilers projects and it's interesting to read about the intersection of the two.

I've been planning a language project to be implemented in MLIR, once I have some more free time. The dialect mixing is an attractive approach that allows for progressive IR lowering and I'm excited to use it.

See here for discussion on Wasmtime team moving from Mozilla to Fastly: https://news.ycombinator.com/item?id=24897641

> If you leave your love of programming and the feeling of the computer revolution romance on autopilot, the industry will burn through you. It’s your responsibility to know exactly why you’re doing the specific job you’re doing, what your long-term goal is, how to develop a credible plan to achieve it, and what you need to do (and not to do) to keep your underground spring of creativity alive. Don’t outsource this to anyone else, and don’t neglect it. No one can do that job except you, and the job needs to get done. Unless you want to spend the rest of your life growing hydroponic tomatoes.

This is something that I think about often, but it's difficult to find a path that is creativity-sustaining. In practice it hasn't worked for me. Alternatively, I have considered seeking tech jobs that are not demanding and have fixed hours, leaving time to explore passions outside of work.

The feature that has kept me from switching to Firefox is history archiving. I like to keep a record of my browsing history using Chromium's “History Trends Unlimited” extension to easily find articles previously read. However, Firefox appears to only record the most recent site access date and there's no similar extension. Does anyone know of an alternative?

Do you have a source for this?

Yes, Chromium has this same behavior.

ungoogled-chromium is a project that removes Google integration from Chromium. Here is the patch they use to remove this special treatment of Google sites:

https://github.com/Eloston/ungoogled-chromium/blob/master/pa...

This same blog has a post on the idiosyncrasies of the C preprocessor. Reminds me of several convoluted macros I assembled to stress test a C/C++ static analyzer at my last job. My favorite that I wrote was the following, combining line continuations, trigraphs, digraphs, universal character names, and block comments. Particularly insidious is the line continuation between / and * in the block comment. It is equivalent to: #define 🇩🇪() "de"

   ??/
  %: \
  /??/
  */*\
  **/\
  def\
  ine\
   \U\
  0??/
  001\
  F??/
  1E9\
  ??/\
  U00\
  01F\
  1EA\
  (??/
  ) "\
  de"\

This example also exposes undefined behavior from the C++ standard: "if a splice results in a character sequence that matches the syntax of a universal-character-name, the behavior is undefined" [0].

[0]: https://eel.is/c++draft/lex.phases#1.2

I designed Nebula [0], a LLVM front end for Whitespace, written in Go. It's the first ahead-of-time compiler for Whitespace. It features its own bespoke intermediate representation in static single assignment form for simpler optimization. Whitespace instructions are stack-oriented like Forth, thus most instructions modify the stack. Nebula lowers these stack-oriented instructions to registers in SSA form such that the stack is modified only once per basic block. It was trivial to add support for Brainfuck, lowering to the same IR.

I have record of almost 100 Whitespace interpreters [1], so I am confident that Nebula is the first AOT compiler for Whitespace. Another project is CensoredUsername's Whitespace JIT compiler in Rust that assembles code on the fly and deoptimizes on error.

[0]: https://github.com/andrewarchi/nebula

[1]: https://github.com/andrewarchi/ws-collection

In fact, Edwin Brady wrote a Whitespace interpreter in Idris: https://github.com/edwinb/WS-idr

Has anyone used Darcs [0]? Its design based around a patch theory [1] looks very compelling to me, but I haven't tried it myself.

[0]: http://darcs.net/ [1]: https://en.wikibooks.org/wiki/Understanding_Darcs/Patch_theo...

Neue Haas Grotesk was revived to restore the small details lost from Helvetica, so if you aren't very particular you may as well use the easily-available Helvetica. There's plenty of more interesting typefaces than Neue Haas Grotesk or Helvetica though.

My favorite sans-serif font is Meta by Erik Spiekermann which is described as the anthesis of Helvetica. It appeals to my exact nature and has humanist touches like the bend at the top of ascenders (b, h, k, l). Fira Sans is a free alternative that is derived from it and works well when Meta is not an option. For code samples, Fira Code pairs well with it.

https://en.wikipedia.org/wiki/FF_Meta

PDF files embed the vectors outlines for the letters used in the text and it's trivial to extract those into a font file. However, as the mathematical representation of the typeface is under copyright, this conversion does not make it free because the vector curves remain the same. The process would also likely lose the fine tuning of a professional typeface like the kerning, so it would not be desirable anyways.