Rev Dr Chris
@chrisgoringe.bsky.social
Scientist. Ordained minister. 🏳️🌈🏳️⚧️❤️.
Everyone is welcome. Especially 🐕.
Everyone is welcome. Especially 🐕.
"It features a man who turns into a cormorant"
Because of course it does
Because of course it does
January 19, 2026 at 12:55 AM
"It features a man who turns into a cormorant"
Because of course it does
Because of course it does
My favourite of all your books :)
January 19, 2026 at 12:47 AM
My favourite of all your books :)
The videos are enough to bring a civil case - burden of proof is lower (balance of probabilities rather than beyond reasonable doubt)
That medical care was not provided (indeed was denied) is itself enough to bring a case.
That medical care was not provided (indeed was denied) is itself enough to bring a case.
January 17, 2026 at 11:00 AM
The videos are enough to bring a civil case - burden of proof is lower (balance of probabilities rather than beyond reasonable doubt)
That medical care was not provided (indeed was denied) is itself enough to bring a case.
That medical care was not provided (indeed was denied) is itself enough to bring a case.
Just using “don’t shoot the messenger” when we’re talking about someone actually getting shot.
Not a big thing, but I think it adds to the trivialising of gun violence.
Not a big thing, but I think it adds to the trivialising of gun violence.
January 17, 2026 at 4:22 AM
Just using “don’t shoot the messenger” when we’re talking about someone actually getting shot.
Not a big thing, but I think it adds to the trivialising of gun violence.
Not a big thing, but I think it adds to the trivialising of gun violence.
Maybe not the best choice of phrase in the circumstances…
January 17, 2026 at 4:18 AM
Maybe not the best choice of phrase in the circumstances…
You don’t think her wife will bring a civil suit?
January 17, 2026 at 4:16 AM
You don’t think her wife will bring a civil suit?
Assuming pre-teens don’t have accounts?
That doesn’t pass the laugh test.
That doesn’t pass the laugh test.
January 16, 2026 at 9:12 AM
Assuming pre-teens don’t have accounts?
That doesn’t pass the laugh test.
That doesn’t pass the laugh test.
So sorry.
No wise words to add 😞
No wise words to add 😞
January 2, 2026 at 10:05 PM
So sorry.
No wise words to add 😞
No wise words to add 😞
(And if you haven’t seen/heard an ACD in defensive aggression mode, it’s not for the faint hearted)
December 28, 2025 at 3:47 PM
(And if you haven’t seen/heard an ACD in defensive aggression mode, it’s not for the faint hearted)
What do you have against elephants?
December 28, 2025 at 12:33 PM
What do you have against elephants?
We rescued a traumatised deaf cattle dog three years ago. Took three weeks before she would offer her back to us, or not bark in panic whenever someone entered the room (even if they left just moments before).
Best terrible decision ever.
Best terrible decision ever.
December 28, 2025 at 12:22 PM
We rescued a traumatised deaf cattle dog three years ago. Took three weeks before she would offer her back to us, or not bark in panic whenever someone entered the room (even if they left just moments before).
Best terrible decision ever.
Best terrible decision ever.
But you *should* be right, surely that counts for something?
December 28, 2025 at 2:50 AM
But you *should* be right, surely that counts for something?
Originally written in Greek, not Latin
December 22, 2025 at 12:48 PM
Originally written in Greek, not Latin
Anyway, it’s midnight here in Sydney - should be in bed, not having interesting discussions on Bluesky. Cheers!
December 19, 2025 at 12:51 PM
Anyway, it’s midnight here in Sydney - should be in bed, not having interesting discussions on Bluesky. Cheers!
Yeah, that would suck. A good argument that I’m fighting the wrong battle. 😛
December 19, 2025 at 12:48 PM
Yeah, that would suck. A good argument that I’m fighting the wrong battle. 😛
I get the desire to distinguish, as I conceded in another reply.
IMO the problem is that the problem domain of LLMs is language, which means hucksters can create an illusion of intelligence.
And tools in other domains get dragged down when people realise the claims made for LLMs are 💩.
IMO the problem is that the problem domain of LLMs is language, which means hucksters can create an illusion of intelligence.
And tools in other domains get dragged down when people realise the claims made for LLMs are 💩.
December 19, 2025 at 12:45 PM
I get the desire to distinguish, as I conceded in another reply.
IMO the problem is that the problem domain of LLMs is language, which means hucksters can create an illusion of intelligence.
And tools in other domains get dragged down when people realise the claims made for LLMs are 💩.
IMO the problem is that the problem domain of LLMs is language, which means hucksters can create an illusion of intelligence.
And tools in other domains get dragged down when people realise the claims made for LLMs are 💩.
Fair argument, I guess. I’d prefer to lose the term AI completely, but maybe letting “AI” carry the stink and using “ML” is wise 😃
December 19, 2025 at 12:26 PM
Fair argument, I guess. I’d prefer to lose the term AI completely, but maybe letting “AI” carry the stink and using “ML” is wise 😃
The architecture of LLMs, image analysers, folding etc. are essentially the same: multi layer transforms with non-linear activation, trained by differential back propagation of a loss function, that predict an output based on an input.
Function differs, but base tech is the same.
None really AI!
Function differs, but base tech is the same.
None really AI!
December 19, 2025 at 12:23 PM
The architecture of LLMs, image analysers, folding etc. are essentially the same: multi layer transforms with non-linear activation, trained by differential back propagation of a loss function, that predict an output based on an input.
Function differs, but base tech is the same.
None really AI!
Function differs, but base tech is the same.
None really AI!
Thank you for including the adjective "generative" there. AI has already contributed incredibly to medical science (protein folding, anyone?) and has clear potential in (e.g.) pathology screening.
Generative AI? Yeah nah.
Generative AI? Yeah nah.
December 19, 2025 at 5:52 AM
Thank you for including the adjective "generative" there. AI has already contributed incredibly to medical science (protein folding, anyone?) and has clear potential in (e.g.) pathology screening.
Generative AI? Yeah nah.
Generative AI? Yeah nah.
Your initial comment didn’t say that it was the American Salvation Army you were describing. I added a comment so that other readers who aren’t in the USA would have a more complete picture.
Did my comment hurt any one?
Did my comment hurt any one?
December 18, 2025 at 6:06 AM
Your initial comment didn’t say that it was the American Salvation Army you were describing. I added a comment so that other readers who aren’t in the USA would have a more complete picture.
Did my comment hurt any one?
Did my comment hurt any one?
No, but some people who read what you post do :).
December 18, 2025 at 5:40 AM
No, but some people who read what you post do :).
counterpoint - our dog sheds all the time, and our robot vacuum cleaner is a total lifesaver, collecting a full load of hair every night.
But it's not a Roomba, so....
But it's not a Roomba, so....
December 18, 2025 at 5:36 AM
counterpoint - our dog sheds all the time, and our robot vacuum cleaner is a total lifesaver, collecting a full load of hair every night.
But it's not a Roomba, so....
But it's not a Roomba, so....
A brief reminder that while this is absolutely true of the Salvation Army in the USA, it is not true globally.
In Australia the Salvation Army have been awarded rainbow tick accreditation, having radically changed their approach to full LGBTIQA+ inclusion.
In Australia the Salvation Army have been awarded rainbow tick accreditation, having radically changed their approach to full LGBTIQA+ inclusion.
December 18, 2025 at 5:18 AM
A brief reminder that while this is absolutely true of the Salvation Army in the USA, it is not true globally.
In Australia the Salvation Army have been awarded rainbow tick accreditation, having radically changed their approach to full LGBTIQA+ inclusion.
In Australia the Salvation Army have been awarded rainbow tick accreditation, having radically changed their approach to full LGBTIQA+ inclusion.
Really?
The AI biochemists trained on public protein folding data, solving one of the hardest problems in the field, opening up a huge range of medical research possibilities, is unethical?
And the little self-learning game AI that I wrote, that uses no external data? That is unethical?
Why?
The AI biochemists trained on public protein folding data, solving one of the hardest problems in the field, opening up a huge range of medical research possibilities, is unethical?
And the little self-learning game AI that I wrote, that uses no external data? That is unethical?
Why?
December 17, 2025 at 2:01 AM
Really?
The AI biochemists trained on public protein folding data, solving one of the hardest problems in the field, opening up a huge range of medical research possibilities, is unethical?
And the little self-learning game AI that I wrote, that uses no external data? That is unethical?
Why?
The AI biochemists trained on public protein folding data, solving one of the hardest problems in the field, opening up a huge range of medical research possibilities, is unethical?
And the little self-learning game AI that I wrote, that uses no external data? That is unethical?
Why?