Everythingβs political, but some topics are more politicized than others
26.02.2026 08:35 β π 1 π 0 π¬ 0 π 0Everythingβs political, but some topics are more politicized than others
26.02.2026 08:35 β π 1 π 0 π¬ 0 π 0let it be known that my achilles heel is cadbury mini eggs
25.02.2026 20:25 β π 4 π 0 π¬ 1 π 0See bsky.app/profile/davi...
25.02.2026 20:05 β π 0 π 0 π¬ 0 π 0GPT is v unreliable for citations, I hear Perplexity is much better but havenβt tried myself
25.02.2026 10:26 β π 0 π 0 π¬ 0 π 0For sure, to be clear I donβt think itβs mysterious why this is happening, thereβs numerous very obvious factors.
25.02.2026 08:34 β π 2 π 0 π¬ 1 π 1Climate change is of course well trodden here. If I make a video warning about raising temperatures, one half of the US will actively avoid it, and to some extent the opposite is true (but less poignantly). As an educator I want to challenge someoneβs thinking, not just reinforce their biases.
25.02.2026 08:33 β π 2 π 0 π¬ 2 π 0Although I totally understand how and why this is happening, itβs also a real challenge for open dialogue and science communicators. The social media feedback engine risks reinforcing leanings into ideologies on this topic. Soft priors become delta functions.
25.02.2026 08:29 β π 4 π 0 π¬ 1 π 0One thing Iβm seeing is that AI appears to be perceived by many as an increasingly political topic, whereas a decade ago it was largely one of technology alone. Both statements of critique and praise for such systems are becoming seen as latently partisan, much like climate science experienced.
25.02.2026 08:25 β π 11 π 1 π¬ 3 π 0Descartes!
23.02.2026 16:09 β π 2 π 0 π¬ 0 π 0Very interesting Astrophysics colloquium given today by Prof David Kipping (@davidkipping.bsky.social) about the role of Bayesianism and Anthropics in searches for life. Excited to see the paper published! Much to think about when it comes to finding life with future space telescopes.
23.02.2026 14:59 β π 23 π 5 π¬ 1 π 0Right now itβs pretty quiet so booking isnβt a stress, but I wish Iβd had a bit longer to look around if anything, many houses close for public at 4pm
21.02.2026 19:37 β π 1 π 0 π¬ 0 π 0Pompeii really is amazing
20.02.2026 16:56 β π 18 π 0 π¬ 1 π 0None of us can prove that anyone else is truly thinking. Iβm not sure what you find concerning about what I wrote here.
20.02.2026 10:05 β π 0 π 0 π¬ 1 π 0
βI think, therefore I am.β
The only thing we can be sure of.
So what does that mean about our thinking machines?
βI am somebodyβ shaped how millions of us saw ourselves as kids, including me. That message mattered. It still does.
Rest in power, Jesse Jackson
ππ½
Jesse Jackson 1941-2026 A photo of Jackson in his later years wearing a Rainbow/PUSH Coalition pin.
The US civil rights leader Jesse Jackson, a pioneer of progressive Democratic politics, close ally of Martin Luther King Jr, and two-time candidate for the presidential nomination, has died at 84.
See his life in pictures: bit.ly/3OxOT3f
Claude Code is highly addictive. I find myself constantly walking back to my laptop to check on builds and agents. Is this the new social media?
16.02.2026 11:41 β π 6 π 0 π¬ 0 π 0It feels like all shows and movies coming out these days are extremely derivative, ruin source material and have similar look/feel anywayβ¦.
15.02.2026 12:37 β π 1 π 0 π¬ 1 π 0Quite possible. When TV first arrived, I think there was similarly a lot of resistance about what would happen to radio. But radio still exists and it adapted, into a more niche role. Itβs going to be interesting to see how this wave of tech shifts around many industries.
15.02.2026 12:23 β π 0 π 0 π¬ 0 π 0Of course, there are legal and permission issues here. In the same way the JJ Abrams reboot required permission from the IP owners (Paramount), any sequels (be it AI or not) require such endorsement, but thatβs a separate issue to capability which my original message was about.
15.02.2026 09:56 β π 0 π 0 π¬ 1 π 0So if an entity, be it human or artificial, committed enormous intellectual labor and effort to produce something it was proud of and was perceived as βperfectβ by all (see my original carefully worded message!) then dismissing that as βslopβ is just prejudice.
15.02.2026 09:54 β π 0 π 0 π¬ 1 π 0I think weβre going to soon need to rethink our relationship with ever more sophisticated models. Bostrom raised some great points with me (youtu.be/RSMnOFgzfbQ?...) about how we should I be careful not to dehumanize advanced agents, essentially that treating them like shit may come back to haunt us
15.02.2026 09:51 β π 0 π 0 π¬ 2 π 0I just want to get to the point where I can prompt βmake TNG season 8β and itβs perfect
14.02.2026 20:49 β π 11 π 0 π¬ 4 π 0One of the most stupid things Iβve ever seen www.nytimes.com/2026/02/12/c...
13.02.2026 13:08 β π 9 π 0 π¬ 1 π 0
#31 Joshua Winn - Exoplanet New Discoveries, History and Future
youtu.be/ISZHVwY5YjE
Am I the only one who names all my devices after Tolkien Elves? Whatβs your naming system?
05.02.2026 15:36 β π 6 π 0 π¬ 5 π 0Fair point. My thinking here, building from the MCMC example, is that you donβt truly understand these models until you build one yourself. And science, ultimately, is also about understanding (more than optimization).
03.02.2026 13:38 β π 0 π 0 π¬ 1 π 0Cool! What text did you follow? I am particularly interested in astronomy classes to clarify here.
03.02.2026 13:15 β π 0 π 0 π¬ 2 π 0