actually agree, I copied the summary but it's more sensational than it should be.. I'm excited but I've just gone from 1% believing it to 5%, it's far from confirmed.
considering what it would mean I'm still super excited however.. but I'll edit the summary
Berkley one is simulation, one is a chinese lab that seems to have replicated some of the actual material and confirmed some properties of superconductivity.
According to the simulation apparently it only works if the copper atoms end up in an unlikely place in the crystal, so fabrication might be unreliable. I recall someone in another thread saying the authors themselves had around a 10% success rate. So other efforts to replicate are likely to see more failures until the fabrication is better understood. Makes sense I guess.
The fact that another team saw something I think is really hopeful, even if it's hard to produce and poorly understood, if there is a room-temp effect then it's only a matter of time before it's studied properly and understood. From there hopefully a reliable method of fabrication can be published.
If there is no effect replicated anywhere then the paper is fraud, but if there's any effect replicated then it's just a matter of study and engineering to figure out what it is. It would be interesting even if it isn't exactly room temp superconductivity.
yeah I'd prefer a PR that removes the filename and actually builds so it could be merged.. the emojis might be better as a comment. although json doesn't support comments iirc so maybe just a thoughtful commit message.
Ya if it’s not federated already, I’m thinking the link will just remain as-is?
yeah that seems entirely reasonable. it'll still be a big improvement in most cases. also even if it only happens as it federates initially it'll probably still catch many cases. is it possible to do a follow-up check only for links that weren't updated the first time? if you could store a cool-off and maximum number of tries that would probably keep it light and as functional as possible.
when serving posts with un-processed links that haven't been updated, and it hasn't been checked in the last 3 hours, re-check for a federated article and update. do this for up to 24 hours after the comment is posted, after that just give up? that gives it a few opportunities but doesn't continue to waste resources if it's unlikely the link will be resolved.
I'm not sure if it's easy to add columns to track those attempts though.
oh that's awesome! yeah i can see it being an issue especially if that post isn't already federated. would be a great quality of life improvement though.
ok sounds like it's not a feature - the link format i suggested includes both the instance ID and the instance url, so it would be possible for your home server to use that to preprocess to a useable link. I wonder if there's a suggestions box 🙂
ahhh that makes sense. i was remembering microsoft getting broken up/fined and wondered how an "everything company" could possibly avoid that. turns out you have to have a monopoly to start with, which might have been an issue before he tanked twitter. so i guess genius move, 5d chess etc
I think my home instance blocks porn-only instances, which is fine with me. I can open an account on the porn instance if I want, it's not the same kind of browsing.
dbilitated @ dbilitated @aussie.zone Posts 16Comments 318Joined 2 yr. ago
he probably should have expected that