Cwtch partially answers 2,4 and 5 through Cwtch Servers (our untrusted infrastructure (UT)), we have some open questions on resiliency and scaling this kind of design (see: https://openprivacy.ca/blog/2018/06/28/announcing-cwtch/ … and the paper for more details)
-
-
এই থ্রেডটি দেখান
-
We assume that because UT can't manipulate transcript consistency in any kid of targeted way that it i general won't (and that in cases where it does we can detect irrationality and move servers any time
এই থ্রেডটি দেখান -
The idea of a "metadata-resistant consensus algorithm" intrigues me. The group is presumably closed and known so any PBFT based protocol would work assuming you can trust more than two thirds of participants. That's an..interesting...assumption for these kinds of tools.
এই থ্রেডটি দেখান -
By putting consistency within UT you can solve that problem (UT is either correct or appears irrational - it has no data to disrupt communication consistently). But creates practical issues (how long do UT's keep messages, emergent centralization eg. https://fieldnotes.resistant.tech/defensive-decentralization/ … )
এই থ্রেডটি দেখান -
Worth nothing that as with all PBFT algos, 2/3 is necessary to *resolve consensus* not* detect disagreement* i.e a peer can tell if *someone* is lying to them as long as at least 1 other peer is honest. They might not ever be able to work out who, but still could be useful info.
এই থ্রেডটি দেখান -
(Though given Briar's architecture I'm not sure how you would resolve the bounded-delay criteria with the UX of reduced connectivity environments - you would have to add explicit trust somewhere I think)
এই থ্রেডটি দেখান
কথা-বার্তা শেষ
নতুন কথা-বার্তা -
লোড হতে বেশ কিছুক্ষণ সময় নিচ্ছে।
টুইটার তার ক্ষমতার বাইরে চলে গেছে বা কোনো সাময়িক সমস্যার সম্মুখীন হয়েছে আবার চেষ্টা করুন বা আরও তথ্যের জন্য টুইটারের স্থিতি দেখুন।