Michael Honey

Imagining Palmer having a debate with Charlene McLennan and coming off second-best greens.org.au/tas/person/c...

Adelaide University cancels literary festival event with UN Gaza investigator Francesca Albanese Festival organisers criticise the university for last-minute booking cancellation of event headlined by special rapporteur for Palestinian territories

Another Australian institution self-destructing rather than let someone talk about genocide, this time it’s the good Albanese www.theguardian.com/australia-ne...

operation eschatological jihad

nudibranch boss

you can’t reason with fanatics, they think they have the imprimatur of heaven

Can confirm The First Gulf War was utterly without merit

faction

BYD Atto. Sells well in Australia, I think BYD currently banned in the US because they‘re justifiably afraid BYD will eat market share

we got the juice here in Tassie now, can confirm the wet stuff is enjoyable

The world could be such a nice place if we allowed it. It's all so goddamn unnecessary. There's no need for any of it. It's so beautiful here. It should be so cool to be alive

This Is All So Goddamn Unnecessary Iran, and Courage, and The Waste Of It All

www.firewalledmedia.com/p/this-is-al...

ignorant people crave simple binaries

you‘re right that the changes have seemed incremental, yet the latest models have reached a capability threshold that makes them dramatically more useful. my work recently has felt like a joyful collaboration with a capable partner

LLMs are also way more broadly trained than humans: they know all the languages, have read everything. You could imagine slimmed-down specialist AIs that have skilfully forgotten 90% of what they know in order to work efficiently in a limited domain

because they are, at base, vain and insecure. They don’t want to have good personalities: they want to look good in the mirror

@verybadwizards.bsky.social relevant to your interests

laundry list of ai-related opinions:
1) llms do work and sometimes writes good code
2) ai is not llms. ai is neural networks. "training neural networks" works so well it's baffling
3) "the singularity" is an easy concept and if you scoff at it you're being tribalistic or deliberately obtuse

I appreciate your long-term commitment to our enjoyment

when our society eventually grows out of our militaristic toddlerhood, it’ll be fun converting aircraft carriers into mobile, self-sufficient islands

flight decks permacultured with food crops, cruising the oceans, cleaning up floating garbage, towing icebergs to communities who need fresh water

biblically accurate ancestor / cartesian soul-nexus

microdosing polyamory by having 3 different LLM subscriptions

making good charts is a beautiful collaboration between humans and computer, each doing what it does well

2026 and the conservatives and centrists focus on virtue signalling and ideology, while the progressive parties actually want to get things done and improve the material and environmental conditions of our lives

You’re doing important work here

a reasonable worry. I’m also wary of turning measures into targets: I think of this as a way to understand the “quantity” of transformation, but there could be others. Some transformations are reversible, some are qualitatively more profound than others

Thanks! I don’t know if it’s been used elsewhere: I was thinking about Hamming distance or some other edit-distance concept, but realised that some quite big transformations are easy to compute. For me “compute distance” captures the idea that what’s important is the complexity of transformation

I love a good reconceptualisation

meatmind concept-savants, harvesters of meaning-fog

In the digital world we might invoke something like “compute distance”: how much computer is required to get from the current set of bits to the desired one. In the built world there might be a corresponding repair/reuse/manufacture effort. ”Good” tech makes this distance small

“That which cannot be repaired is already broken”

wiki.xxiivv.com/site/permaco...