robryk @robryk@qoto.org
I enjoy things around information theory (and data compression), complexity theory (and cryptography), read hard scifi, currently work on weird ML (we'll see how it goes), am somewhat literal minded and have approximate knowledge of random things. I like when statements have truth values, and when things can be described simply (which is not exactly the same as shortly) and yet have interesting properties.
I live in the largest city of Switzerland (and yet have cow and sheep pastures and a swimmable lake within a few hundred meters of my place :)). I speak Polish, English, German, and can understand simple Swiss German and French.
If in doubt, please err on the side of being direct with me. I very much appreciate when people tell me that I'm being inaccurate. I think that satisfying people's curiosity is the most important thing I could be doing (and usually enjoy doing it). I am normally terse in my writing and would appreciate requests to verbosify.
I appreciate it if my grammar or style is corrected (in any of the languages I use here).
Joined Nov 2020
@eta did you happen to look at what are the provided reasons for both questions?
A problem that occurs is that you sometimes need to operate on those secrets, so they end up in registers or on stack. Then they stay there until the next time that register/stack area is used. Vector registers aren't used that often, so secrets end up staying there longer.
You could try to have a compiler that cares about where data stays behind, but that would make performance worse, so you'll need to mark data for which this is important. That has to be "contagious": you must never be able to silently cast that property away (and should _never_ cast it away). So, you need alternate versions of e.g. libc functions that will operate on "data not to be left afterwards", which implies that it's more of an all-ecosystem evolution than msan support (where you "just" need everything to be compiled with msan, not everything to be duplicated).
robryk
boosted
@samwho Rust is even more aggressive about some of this. You'd have to Pin<> things at a minimum, and that may not be enough. And any code that involves Pin<>s is usually a pain.
Put your secrets in a different address space, ideally in a different processor entirely.
@kravietz Ah, sorry, I was thinking about chamfers and not fillets (3d printing fillets is usually a bad idea, because something somewhere will exceed the maximum overhang angle). For extrusions you can easily get chamfers on the sides along the extrusion length, but getting them on all other edges was always a chore for me that left my list of groups hopelessly nonnavigable, so much that I started wondering whether it'd be better to use solvespace to design 2d shapes, export them as something like svg, and use something like libfive to do the 3d part.
@harce Even better: have something local that will request those pages as `application/ld+json` and render them by itself.
@kravietz Have you found a convenient way to do fillets on all edges of a part?
@koakuma I would suggest luks inside LVM, separately for every volume, using same passwords. That allows you to way more easily choose to have since unenxrypted partition in the future (or even now, depending on how you boot that machine) at a imo small chat if having more places to change the passphrase in.
@danluu Maybe companies choose to engage in PT when they have difficulty hiring, and difficulty hiring can be caused by underpaying?
If we were talking about stable state, this is basically the same as what @mhoye and @danlyke said. However, is there a reason to expect this is a reasonably steady state as opposed to a transition in progress?
I'm also curious about the synchronization mechanism. The mention of a localized flowering in the paper (which I haven't yet read apart from the abstract) seems to suggest that it's not independent for each plant (so either involves communication or synchronization on some external factor).
@futurebird Apollo 7 could be seen as a precursor to that (no explicit strike, but refusal to do things ground control wanted done).
The sense it seems to make is that it will work with an ~arbitrarily shitty Internet connection.
Czy można gdzieś przeczytać dokładniejszą formę tych założeń? (Jestem ciekaw np. dokładniejszej definicji osoby, o której pozyskuje się dane -- ciekaw jestem np. kto byłby takową osobą w przypadku gdy służby szukały właścicieli telefonów obecnych w jakichś dwu miejscach, lub obecnych w jednym a nie w drugim.)
I am very surprised by retention of case files of any kind that's shorter than the time the accused are imprisoned for.
@niconiconi Or 4 lines to write a function, 10 lines to explain its nonstandard calling convention (e.g. passing values via flags) which saves on code size.
I enjoy things around information theory (and data compression), complexity theory (and cryptography), read hard scifi, currently work on weird ML (we'll see how it goes), am somewhat literal minded and have approximate knowledge of random things. I like when statements have truth values, and when things can be described simply (which is not exactly the same as shortly) and yet have interesting properties.
I live in the largest city of Switzerland (and yet have cow and sheep pastures and a swimmable lake within a few hundred meters of my place :)). I speak Polish, English, German, and can understand simple Swiss German and French.
If in doubt, please err on the side of being direct with me. I very much appreciate when people tell me that I'm being inaccurate. I think that satisfying people's curiosity is the most important thing I could be doing (and usually enjoy doing it). I am normally terse in my writing and would appreciate requests to verbosify.
I appreciate it if my grammar or style is corrected (in any of the languages I use here).
Joined Nov 2020
