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
My state of understanding (which I should improve, but it's too late today) was that store-and-forward in briar is intended to work for intermediate peers that are in physical proximity, as at least a measure to combat DoS.
I guess I don't see how a store and forward (that I can imagine) that could be usable over the wide Internet would help with metadata protection. You ~can't use peers unrelated to sender or receiver for storing messages (because this invites trivially exploitable DoS issues), and using peers related to either side moves the potentially-exposed metadata from "who's talking to whom" to "who uses whom as an intermediate peer", which admittedly does seem somewhat better (but not as much as I thought you'd want the situation to improve).
@rysiek @briar @delta @signalapp
I get the complications around everything related to iOS (or rather: get them vaguely and shy away from learning about their precise nature).
That said, I'm confused about the other part:
On the "why is this even necessary" side: if we use Tor we have some sort of poor protection of metadata. To get a better protection you need store-and-forward via peers that do not happen to also be your contacts. I thought previously that the main point of s&f in briar is to make it work over bluetooth (made into a mesh by having devices move around), not to operate it over Internet. Was I wrong?
@rysiek @briar @delta @signalapp
BTW. This complicated thing about metadata is the main reason why I have misgivings about the "stories" in signal: they would create conversations between each of their users and ~all their contacts, which would show more of the social graph to anyone with ability to monitor what signal the service is doing.
@rysiek @briar @delta @signalapp
Yeah, it's complicated and I got really annoyed when people claimed that sealed sender made that metadata unavailable to signal the service...
Which part is not going to happen anytime soon?
robryk
boosted
So, don't tell anybody, but we're going to announce Palm Pilot emulation in the browser at Internet Archive in the next week or so. I've still got to get descriptions in for the hundreds of apps that are uploaded, and add a few more classics. But if you want to help with that, or just let me know how it's going, ping me here.
But don't tell anybody! It's a secret!
To be fair, signal also leaks metadata of the form of which-ip-talked-to-which-ip to signal the service (but signal the service promises to throw it away immediately). This case is both worse (because there are more entities involved) and better (because metadata is leaked only to entities somehow related to the sender and recipient as opposed to some single entity that cannot be avoided).
@brianleroux The most interesting thing that's somewhat buried in that article is the element correspondence that can be achieved via view-transition-name.
I wonder whether it's (a) useful (b) feasible to achieve e.g. element-wise correspondence of a non-statically-bounded number of e.g. entries in a list.
@Scriv @joningold @Jollyspaniard
Hm~ in a way you could think of all games through the lens of deity-and-human:
There are people who think that descriptions of voice of gods from ~ancient Greece were describing what we'd call now our "internal narration". https://slatestarcodex.com/2020/06/01/book-review-origin-of-consciousness-in-the-breakdown-of-the-bicameral-mind/ is a sort-of book review of a book that touches on that topic.
@Scriv @joningold @Jollyspaniard
It sounds like you are describing a setup where the player is a deity/voice-in-the-head of the main character. I vaguely remember hearing about games that tried that; do you think anyone pulled that off in a convincing fashion?
@foone Many forms that are intended to collect addresses, but were designed in US, try very hard to ask for the equivalent of state in any other country. In many cases it's amusing when they try to use it as a mailing address later (it usually works, but ends up being addressed in a way that is very atypical).
@js@mstdn.io Do you mean c-s specifically or the whole spec?
@js@mstdn.io Frankly, I prefer the client-server ActivityPub API. It does leave media handling to be defined by implementers, which is not great, but I'd prefer if people implemented c-s ActivityPub and server-specific media handling over everything being server specific.
Specific reasons why I like it more is that ActivityPub allows one to express things that cannot be expressed in MastodonAPI for silly arbitrary reasons (for example, boosts with non-public targets; Mastodon claims that it has per-post privacy model, but boosts belie that: they all share their privacy settings with ~the profile).
@grrrr_shark https://www.tcs.ch/de/
Not a very good fit, but I don't know of anything closer to an "automobilists club" here.
@grrrr_shark I wonder if people who were in a similar situation talked with e.g. TCS (and thus ended up with them knowing about previous instances/useful approaches/the lawyer who was involved previously/...).
@gabrielesvelto Tangentially: I would really appreciate it if Firefox didn't look at how low the _system_ is on memory, but (on OSes where this is a thing) look at the cgroup/other resource allocation process group it's in. (e.g. I run Firefox in a cgroup with a memory limit, so that I can leave some memory aside for everything else in a guaranteed fashion.)
defederation, usa, cops or feds
@moonbolt What happened?
@gabrielesvelto How does this relate to unload-tabs-on-memory-pressure mechanism? (I would naively expect that to always run out of unloadable tabs first; is that the case?)
I seems that the answer is (or at least was a year ago) no: https://socialhub.activitypub.rocks/t/activitypub-client-to-server-faq/1941
Weirdly enough Pleroma is claimed to implement it.
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
