@drq Знаете что? Походу, в последней линейке процов, АМД таки пофиксили свой ссаный контроллер памяти. Ндавно купленный комплект на 5200 запустился 9950X на номинале. #юмор #шоколад #бетон #ritterSport Ritter SPORT 1440 кг БЕТОННАЯ ОГРАДА ПРОИЗВОДИТСЯ В РОССИИ С НАТУРАЛЬНЫМИ ИНГРЕДИЕНТАМИ
Show previous comments
привет, @rf ! мы с @MrClon рады сообщить, что сервисы на anyaforger.art работают в нормальном режиме :wtc_rika_smile: :anya_happy: @wonderfox_dev@suya.place @rf@mastodon.ml @MrClon@lor.sh Я только что понял, что это не "any forager", а "anya forger". How Decentralized Is Bluesky Really? https://dustycloud.org/blog/how-decentralized-is-bluesky/ A technical deep-dive, since people have been asking me for my thoughts. I'll expand a bit on some of the key points here in a thread. 🧵
Show previous comments
" Then along came Google Reader and... friends, if you are reading this and are of a certain age range, there is a strong chance you have feelings just seeing the phrase "Google Reader" mentioned." You got me. Great read, thank you. Thank you for this article. I had a very general understanding of how Bluesky works, you helped me get a more coherent picture. It's even more centralized than I thought it was 🙃
Show previous comments
@cwebber this is the single thing about it that drives me truly up the wall. The only ways in which BS is doing decentralization better is by centralizing. I wrote a whole rant on that thing (<https://wok.oblomov.eu/tecnologia/credible-threat-2/>) @cwebber yeah it's literally just one big server and so they have the reasources to take on millions of users in a week, also easier when there's a single point of entry, unlike having to search for an instance, which already assumes one has a solid knowledge of not only how the fediverse works but which servers are cool and which to avoid, which for me at least took a couple years of trail and error I started playing with DragonflyDB to replace Redis as the app data storage for a Mastodon instance. Redis used 22 GB of memory, DragonflyDB uses 8.5 GB. That's very impressive, probably in part because Mastodon uses a lot of sorted sets (for timelines) and Dragonfly stores those in a much more efficient data structure, resulting in a lot less overhead (see https://www.dragonflydb.io/blog/dragonfly-new-sorted-set). @renchap Wow, that's really impressive! Def need to look into that, we use sorted sets a lot too :) @renchap I appreciate the way dragonflydb provides HA using the operator. Works well enough Есть и другая жизнь — |
It's for that reason that @spritely, while aiming for a *socially collaborative* revolution, is first focusing on a *technical* revolution.
It's too hard to build massively, securely collaborative tools right now. With Spritely's tools, p2p ocap secure tech is the *default output*.
@cwebber I've often framed for myself as: "Tech encodes social relations." Both in the sense that the social relations made possible by tech are constrained by its code, and that code is conditioned by the social relations around its production.
@cwebber One of the things I teach in my information ethics classes (to MBA and BA Management) students is that ethical excellence is a fundamental part of technical excellence.