There was a big decision at the SWICG meeting today to start a Moderation, Trust & Safety Task Force that aims to write a report to improve those aspects of the protocol.
I've volunteered to lead that task force.
@hipsterelectron I do care very much about "tone", because it is how I perceive the trust environment. For example, "ostensibly" is quite a risky word in my opinion since it dismisses expressed intentions in favor of implied ones, which fuels misinterpretation due to incomplete context. I understand Lina to mean that the feedbacks from the maintainers are only "ostensibly" constructive (only my interpretation). It is very difficult to reply to a charge of "ostensibly" indeed.
@hipsterelectron I am sorry. I care about kernel only from a distance indeed, and I have not earned the good will to interject informally, for which I apologize. I see two sides of an important topic are both upset and their mutual misunderstanding are deepening the mutual distrust. I only meant to try to invoke some reflection, but I have only upset you more, for which I also apologize. I know about the difficulties of kernel maintainers (though only shallowly), and I assumed that to be the basis of the whole discussion.
@hipsterelectron You misread the "or else" completely: it's not a description of any human text, but the "hidden lifetime requirements", that is, you either follow these undocumented requirements "or else" it crashes. You accused Lina of misinterpreting based on your own misinterpretation. Chill a bit please.
New blog post / living guide focused on link-time optimisation (LTO)
https://convolv.es/guides/lto/
I recently learned there are new features and choices in the LTO space, so I have started this living guide to cover those features along with others in the future. It's (hopefully) useful to experienced compiler users and compiler hackers, while still being accessible to those who haven't interacted with LTO before.
The ink is dried and the mail has arrived: Ferrocene is ISO 26262 and IEC 61508 qualified!
https://ferrous-systems.com/blog/officially-qualified-ferrocene/
re: Scheme in the browser! Scheme in the BROWSER!!!
But THAT'S NOT ALL FOLKS!
Not only does Spritely's Hoot compiler compile standard scheme to Webassembly, it also ships with a full assembler, disassembler, etc... and even a developer's WASM VM!
Wait, did I say it ships with a VM?
YOU BET I DID
You can actually run WASM code against the developer virtual machine provided by Hoot, including WASM code compiled by Hoot (but not only!) and you can even step through it and so on and so forth. More about this in the next release! But the thrilling thing... the thrilling thing!... is that this is already proving immensely helpful to our own internal team. If you read the blogpost you'll see how @dthompson used the Hoot compiler to test and iterate on the program *as he went*.
That's right. You don't even need to leave the Guile REPL. You can run and test your Webassembly output *directly in the dev tools we provide without switching tools!* (More to say on this in an upcoming release!)
@molly0xfff Somebody on here pointed out that "MFA means it requires one thing that you can lose, and one thing that you can forget" and I keep thinking about that.
STOP COUNTING CVEs
* BUGS WERE NOT SUPPOSED TO BE PLENTIFUL!
* Years of comparing relases by CVE numbers and yet no real-world use found.
* Want to use secure software? We had a tool for that: It was called updating.
* "Yes, please give me the software with the least CVEs". Statements dreamed up by the utterly deranged!
Look at what "security analysts" have been demanding your respect for all this time: low median CVSS score!? least CVEs?
They have played us for absolute fools.
Found a method to see inside some chips, without having to unmount or destroy the chips.
Best part - the method only relies on lightly modded off-the-shelf cameras and lenses.
Read more at https://www.bunniestudios.com/blog/?p=6712
if the EU passes eIDAS 2.0 (article 45), we will be rolling back the clock to Netscape Navigator Export Edition, a reference that approximately 1% of my audience will understand.
https://www.eff.org/deeplinks/2023/11/article-45-will-roll-back-web-security-12-years
love explaining rust's dropck_eyepatch and may_dangle to people because it is genuinely probably the most evil and arcane part of rust that you never ever need to know about and would probably never even observe the consequences of, and even the experts in it will give thousand-yard stares if you ask too much about it
anyway today i learned that some newer revision of dropck_eyepatch introduced the concept of `unsafe impl Drop` which is both extremely funny to read and also objectively correct to require when applying an eyepatch
(https://doc.rust-lang.org/nightly/nomicon/dropck.html if you want some of the horrible details)
I respect VLC replacing the usual "agree"/"disagree" radio buttons underneath the GPL license, which is just an indication of your rights and not a document you need to mark your acceptance of, with a simple "You are now aware of your rights 😇" label
I will not comment on internet’s takes about a project I worked on. I will not comment on internet’s takes about a project I worked on. I will not comment on internet’s takes about a project I worked on. I will not comment on internet’s takes about a project I worked on.
@thephd Thank you for your persistent brilliant work on bringing PL research and best practices to C and Rust. You are the No.1 resource I rely on when it comes to theory behind new language features. It is infuriating to see you suffer this mistreatment and a great pity to see you forced to leave work for Rust. I look forward to whatever you work on next.
he+him cis hetero 🏳️🌈🏳️⚧️/ 日本語 / Wanna-be Haskeller / UBI / Strong Towns / ラブライブ / 京アニ / Hololive / アイドルも声優もプログラム言語もDD / Correctly Useless