Whatitat90
Joined 8 years ago
Joined 8 years ago
And mpv with "hwdec=vaapi" in ~/.config/mpv/mpv.conf
In private window it just works but still I consider the design of checking login on a page that doesn't require it quite bad.
Comparison of all these tools to GnuPG is valid and it clearly shows not only implementation problems but design ones as well in gpg.
What I fear is future riddled with all these incompatible tools. Even if they're written by brilliant engineers and cryptographers they are not standards (e.g. IETF standards). Why is that important? For example rewriting libsignal from scratch (for example to publish it under permissive licenses) can be problematic [0].
> DNS-Based Authentication of Named Entities (DANE) Bindings for OpenPGP
WKD has some benefits over OPENPGPKEY - it keeps the request confidential (as WKD uses plain HTTPS). WKD is just easier to get right, that's why it's more broadly supported. GnuPG, that supports both of them, defaults to WKD. If OPENPGPKEY request is made it seems GnuPG doesn't even validate DNSSEC signatures: https://lists.gnupg.org/pipermail/gnupg-users/2011-December/...
That's why most recent versions of GnuPG automatically create keys with expiry set to 2 years.
Interestingly I just read about this in a Lawrence M. Krauss book [0] where it was described that David Hilbert insisted on hiring Noether by the university but was overruled by the male majority that didn't like the idea that a female would teach male students. Hilbert commented that "this is university, not a bath-house!" [1].
[0]: I think it was "The Greatest Story Ever Told—So Far"
[1]: I'm writing that from memory so excuse some inaccuracies.
Here's the source: https://github.com/wiktor-k/openpgp-proofs#openpgp-proofs