My rear molar has flared up again today. Worst than ever. Whole side of my face is in pain.
Hurts like it's got a Corbynite manifesto stuck in it. I am so terrified of the dentist I've been putting it off, but my wife has put her foot down.
I've got at least three further appointments booked for this. Going to need either a bridge or an implant tooth.
So a month of agony.
Sympathies. Toothache is a bitch
Maybe you could abandon your tee-totalism for a month? Strong liquor does help
Amen. Inconsolable toothache can be cured* by drinking decent Scotch. A sensible** measure removes the throbbing. A second sensible measure removes the memory of toothache. *not cured **not sensible
"There's a special rung in hell reserved for people who waste good Scotch!"
"A sobbing rapist who attacked a 13-year-old girl in a Midlothian park walked free from court on Monday. Sean Hogg pounced on the girl at Dalkeith Country Park on various occasions between March and June 2018.
"Court papers state Hogg, now 21, threatened his victim and pulled down her lower clothing. Hogg, of Hamilton, South Lanarkshire, then seized her by the wrists and caused the girl to carry out a sex act on him. He then pushed his victim's head down and raped her."
That’s what I’d seen - how the “unknown to the attacker” in the BBC article if it had happened on multiple occasions
Yes, I'm seriously warming to Sunak. He's competent and he's not scared of the Woke and he's getting things done. He ain't perfect and his party is tired and all that, but as PMs go he could - in different circs - have been one of the much better ones. But he is doomed to a short tenure
Relatedly, I saw a pic of Starmer the other day. He suddenly looks his age. 60. Quite chubby, and a little weary. Has the rosy flush of a drinker, too - no idea if he is (and I apologise to him if he's tee-total!)
Sunak looks and seems sharper. More dynamic
I noticed the same thing about Starmer. He had an ageless look for quite a long time, but now he's visibly changed. Perhaps the contrast between him and Sunak will be even starker by the time of the election.
Yeah, for ages he seemed to be "about 50" and quite a well preserved 50, at that. Good hair. Now he looks what he is. 60. And a tired 60 because he has a busy and demanding job. I don't believe he will serve much more than one term
If there are TV debates there will be quite a contrast twixt Starmer and Sunak
Let’s not get carried away. Sunak drew (at best) TV debates with Liz Truss.
I think one of the appeals of Rishi to many older voters must be the vibe he gives of the fresh faced, clean cut son done well. Always worked terribly hard at school unlike that layabout older brother of his. Much nicer to his dear mother than his slightly robotic little sister and her oddball friend Kwazi. Always remembers mother’s day. Sure he isn’t particularly handy around the house but granny was ever so proud at his graduation ceremony.
Hence why also he isn’t taking the party with him. He just doesn’t look like he’s someone to lay down the law to the hoodlums.
David 'Hug a hoodie' Cameron seemed to drag them along just fine.
He had George Osborne with a set of knuckle dusters for anyone that didn't play nice. Rishi is missing a Luca Brasi.
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
Yes, people lose sight of the fact that this is all primarily an experiment. Some of these models have been kicking around for decades, and the tantalizing question always was: if we had the processing power and capacity to make them operational would they turn out to be identical to human intelligence, indeed would it prove that that's what human intelligence actually is? Well, now we have the hardware and they've conducted the experiment. The question is: are the considerable flaws fixable, or is the model itself entirely broken backed? I'm inclined to the latter.
And yet the most in-depth analysis of GPT4, to date, has very different conclusions. I keep going back to this paper simply because it is the only one - GPT4 is so new no one else has had time to really assess it
"The central claim of our work is that GPT-4 attains a form of general intelligence, indeed showing sparks of artificial general intelligence. This is demonstrated by its core mental capabilities (such as reasoning, creativity, and deduction), its range of topics on which it has gained expertise (such as literature, medicine, and coding), and the variety of tasks it is able to perform (e.g., playing games, using tools, explaining itself, ...)."
For balance, here is a "critique" of the paper which believes the authors go too far, but...
"I don’t particularly agree with this conclusion. The model is quite intelligent, and it has a far wider range of world knowledge than a human does. But where we are still more capable is in our ability to reason by ourselves without anyone or anything directing us to do so. GPT-4 has not showed that in this research. That being said, GPT-4 is very impressive and I agree with both this team and many others in that this will revolutionize the world in ways similar to how other major innovations like the internal combustion engine and the internet did."
So the divide is as I say: between those who say bloody hell this is big and important, but it's not AGI, and those who say bloody hell this is big and important, and it could actually be AGI, or we are now really close to AGI
It is not a "really big version of Ask Jeeves", or at least: if it is that, it is so much else besides
Such BS. Don't have them in hair, finger nails or toenails.
Need them in gums and tongue. Tooth nerves are about as much use as mosquitoes or wasps.
As an early teenager I had to swerve out of the way of an oncoming car, went head-first over my bike handlebars and broke one of my front teeth.
The dentist had to drill 'up' into my tooth to kill the dangling nerve.
As I squirmed and screamed in the chair she *shouted at me* "You can't feel the pain - you've had an injection!" and drilled some more. I blacked out at that point.
I dentists.
My dentist was genuinely called Mr Paine.
A group of friends refused to set up in financial services together.
I still maintain that the firm of Conn, Fiddler & Dodge would have done a roaring trade.
Yes, I'm seriously warming to Sunak. He's competent and he's not scared of the Woke and he's getting things done. He ain't perfect and his party is tired and all that, but as PMs go he could - in different circs - have been one of the much better ones. But he is doomed to a short tenure
Relatedly, I saw a pic of Starmer the other day. He suddenly looks his age. 60. Quite chubby, and a little weary. Has the rosy flush of a drinker, too - no idea if he is (and I apologise to him if he's tee-total!)
Sunak looks and seems sharper. More dynamic
Yes, we all love Rishi (or some do).
The trouble is the Party he leads which as Ipsos shows today is disliked by about two thirds of the electorate. Perhaps the ideal would be to have Sunak lead Labour instead.
Like Major before him, Sunak may well be a half decent Prime Minister but will be dragged to defeat by his own Party.
The bitter irony is that the Tory Party could have avoided Truss altogether, gone for Sunak in the first place, and they might now still be in with a chance of winning in 2024
Truss must be one of the greatest unforced errors ever made by a British political party. Along with Sturgeon pushing the Gender Act
And yes, I am aware that at one point I predicted "Truss might surprise on the upside". Not my best moment
[Humza Yousless has entered the chat] [As has Jeremy Corbyn}.
Fair point
I still think Truss is in a league of her own (certainly in terms of her spectacularly short career, and her unique self combustion) but Yousaf and Corbyn are right up there in terms of OMG-wrong-choice. Corbyn arguably guaranteed Brexit, so his elevation also had major long term consequences. Yousaf will just fail, and get the boot, quite quickly, I expect
It won't be a Brian Clough at Leeds type tenure like Truss, but they will get squeezed badly at the GE (no chance of gains in the Borders/Highlands, losses to Lab across the central belt, almost certain loss of e.g. the Western Isles due to ferries debacle). Then the MSPs will start to get nervous ahead of Holyrood 2026, and he'll be out, IMO.
Yousaf will probably last until after the GE. Then there will be pressure on him to resign by SNP HQ, due to the effect on the backroom boys, girls and others of the loss of Short Money, which also won’t help replace the missing £600,000.
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
My rear molar has flared up again today. Worst than ever. Whole side of my face is in pain.
Hurts like it's got a Corbynite manifesto stuck in it. I am so terrified of the dentist I've been putting it off, but my wife has put her foot down.
I've got at least three further appointments booked for this. Going to need either a bridge or an implant tooth.
So a month of agony.
Sympathies. Toothache is a bitch
Maybe you could abandon your tee-totalism for a month? Strong liquor does help
Amen. Inconsolable toothache can be cured* by drinking decent Scotch. A sensible** measure removes the throbbing. A second sensible measure removes the memory of toothache. *not cured **not sensible
"There's a special rung in hell reserved for people who waste good Scotch!"
Secretary of Defense Lloyd J. Austin III Announces New Director of the Defense Innovation Unit https://www.defense.gov/News/Releases/Release/Article/3351281/secretary-of-defense-lloyd-j-austin-iii-announces-new-director-of-the-defense-i/ … Mr. Beck is currently Vice President of Apple, Inc., where he has reported directly to CEO Tim Cook since joining the company in 2009. During that time, he has led Apple's businesses across Northeast Asia and the Americas, and most recently, several of the company's purpose-driven businesses worldwide, including in health, education, and other institutions of public impact. He has over fifteen years of experience living and working across Asia, including in Japan, China, Hong Kong, Korea, Singapore, and Indonesia.
Mr. Beck is a Captain in the U.S. Navy Reserve and served from 2006 through 2007 in Iraq and Afghanistan with a joint special operations task force. He has also served extensively throughout the Asia Pacific region during his nearly 26 years of service, including command of a large joint reserve unit supporting U.S. Indo-Pacific Command in Pearl Harbor, Hawaii. Mr. Beck has previous experience working with DIU, as he founded and led its joint reserve component from inception in 2015 through 2019. His personal and unit awards include the Defense Superior Service Medal (two awards), the Bronze Star Medal, the Combat Action Ribbon, and the Presidential Unit Citation…
My rear molar has flared up again today. Worst than ever. Whole side of my face is in pain.
Hurts like it's got a Corbynite manifesto stuck in it. I am so terrified of the dentist I've been putting it off, but my wife has put her foot down.
I've got at least three further appointments booked for this. Going to need either a bridge or an implant tooth.
So a month of agony.
Sorry to hear that.
I'd be happy if I never saw a dentist for the rest of my life or if I was sedated at the entrance door every time I went.
I was relieved of 113 today hygienist tomorrow about 60 and next month about 400 for a bit more complicated work.
Prices at my dentist have gone up 25 pc in the last 6 months.. at a guess. These Dentists are making a killing.
My dentist said I was looking at £3,500 for the dental implant.
That's at a discount with my private dental insurance/plan.
Consider going to Bangkok?
Private dental care there is superb - they are all trained in the USA and speak perfect English. It is generally about 30-40% the price of the same care in the UK. And often superior. And there's no faff or delay
For really major dental work - like stuff which will cost £3500 - you will actually save money by going to Bangkok, even including hotel and flight. Plus you get a nice little holiday
I had two gold crowns done there, in the UK they would have cost me three times what I paid, they are still going after many years, and a dentist (in Bangkok) recently checked them and said No problem, they're still fine.
Don't know about "buried", but perusal of the Finnish press tells me 1. that Russians are mounting serious cyber attacks there, and 2. that retired general Lord Dannet has made comments about the Finns joining NATO which are objectively wrong and rather insulting to them.
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
Frankly I'm possibly most shocked at the suggestion a 25 year old man who violently raped a thirteen year old girl would get no more than five years in prison.
I mean, what the fuck? In England that would be fourteen years before we even considered aggravating factors.
As for his being given community service, that's an actual joke. That's disgusting. And for him to appeal his sentence is even more disgusting.
Something seems appallingly wrong here.
Edit - maybe England is too ready to lock people up - the woman given two years for the death of that cyclist springs to mind - but this is actually obscene.
Yes, I'm seriously warming to Sunak. He's competent and he's not scared of the Woke and he's getting things done. He ain't perfect and his party is tired and all that, but as PMs go he could - in different circs - have been one of the much better ones. But he is doomed to a short tenure
Relatedly, I saw a pic of Starmer the other day. He suddenly looks his age. 60. Quite chubby, and a little weary. Has the rosy flush of a drinker, too - no idea if he is (and I apologise to him if he's tee-total!)
Sunak looks and seems sharper. More dynamic
Yes, we all love Rishi (or some do).
The trouble is the Party he leads which as Ipsos shows today is disliked by about two thirds of the electorate. Perhaps the ideal would be to have Sunak lead Labour instead.
Like Major before him, Sunak may well be a half decent Prime Minister but will be dragged to defeat by his own Party.
The bitter irony is that the Tory Party could have avoided Truss altogether, gone for Sunak in the first place, and they might now still be in with a chance of winning in 2024
Truss must be one of the greatest unforced errors ever made by a British political party. Along with Sturgeon pushing the Gender Act
And yes, I am aware that at one point I predicted "Truss might surprise on the upside". Not my best moment
[Humza Yousless has entered the chat] [As has Jeremy Corbyn}.
Fair point
I still think Truss is in a league of her own (certainly in terms of her spectacularly short career, and her unique self combustion) but Yousaf and Corbyn are right up there in terms of OMG-wrong-choice. Corbyn arguably guaranteed Brexit, so his elevation also had major long term consequences. Yousaf will just fail, and get the boot, quite quickly, I expect
It won't be a Brian Clough at Leeds type tenure like Truss, but they will get squeezed badly at the GE (no chance of gains in the Borders/Highlands, losses to Lab across the central belt, almost certain loss of e.g. the Western Isles due to ferries debacle). Then the MSPs will start to get nervous ahead of Holyrood 2026, and he'll be out, IMO.
As others have said, it doesn't look to me as if the Sean Hogg sentence has much to do with the SNP, Scotland, or the sentencing guidelines. Rather, it's to do with a slightly bonkers judge, and will be changed on appeal. Ludicrous sentences (both ways) have been known to happen in England, quite frequently.
Even the BBC are saying it’s due to the sentencing guidelines.
I have no understanding of this case but it does feel like the SNP have had some extraordinary bad luck. A bunch of cases have come up which serve to confirm the various suspicions of the Mumsnet cohort.
Yousaf was Justice Secretary for three years, including for when the consultation for the sentencing guidelines for young people was conducted.
How very House of Cards. There is a novel in this, somewhere...
I wonder which SNP disaster will prove to be The Final Cut?
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Unusual sentence. I'll have to read more about it.
It seems to be a result of sentencing guidelines introduced by the Scottish government for offenders under 25 (the rapist in this case was 17 at the time of the rape, 21 now).
Rapists under the age of 25 won't get prison sentences?
The presumption seems to be that they won't.
What makes this even odder, at least judging by the news reports, is that. mitigating factors aside, there were some pretty strong aggravating factors: the victim was just 13. The judge's remark that if he had been over 25 the sentence would have been 4-5 years imprisonment is itself, on the face of it, odd: surely one would expect a much longer sentence?
Also the report states that he attacked the girl, "on various occasions between March and June 2018." So, repeat offending. It wasn't a one-off aberration.
Judges will sometimes make mistakes and I hope there is scope for the leniency of the sentence to be appealed, though the report only says that the defendant intends to appeal the sentence.
I'm pretty sure it is the Crown which is intending to appeal the sentence, on the grounds of undue leniency, not the defendant. What would his appeal hope to achieve ? An even shorter period of litter-picking?
Not only that but litter picking is going to put him in public places with 13 year old girls
My rear molar has flared up again today. Worst than ever. Whole side of my face is in pain.
Hurts like it's got a Corbynite manifesto stuck in it. I am so terrified of the dentist I've been putting it off, but my wife has put her foot down.
I've got at least three further appointments booked for this. Going to need either a bridge or an implant tooth.
So a month of agony.
Sympathies. Toothache is a bitch
Maybe you could abandon your tee-totalism for a month? Strong liquor does help
Amen. Inconsolable toothache can be cured* by drinking decent Scotch. A sensible** measure removes the throbbing. A second sensible measure removes the memory of toothache. *not cured **not sensible
"There's a special rung in hell reserved for people who waste good Scotch!"
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
The guideline does not say that you do not know you are doing wrong or that you should not be punished. It simply observes that men in particular under 25 are more prone to making impulsive decisions and to fail to think through the consequences of their actions. It is a mitigating factor, it is not an excuse.
All individual counts of falsifying business records in the first degree - weirdly, they are all listed in chronological order I think, apart from the ninth and tenth counts.
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
Yes, people lose sight of the fact that this is all primarily an experiment. Some of these models have been kicking around for decades, and the tantalizing question always was: if we had the processing power and capacity to make them operational would they turn out to be identical to human intelligence, indeed would it prove that that's what human intelligence actually is? Well, now we have the hardware and they've conducted the experiment. The question is: are the considerable flaws fixable, or is the model itself entirely broken backed? I'm inclined to the latter.
And yet the most in-depth analysis of GPT4, to date, has very different conclusions. I keep going back to this paper simply because it is the only one - GPT4 is so new no one else has had time to really assess it
"The central claim of our work is that GPT-4 attains a form of general intelligence, indeed showing sparks of artificial general intelligence. This is demonstrated by its core mental capabilities (such as reasoning, creativity, and deduction), its range of topics on which it has gained expertise (such as literature, medicine, and coding), and the variety of tasks it is able to perform (e.g., playing games, using tools, explaining itself, ...)."
For balance, here is a "critique" of the paper which believes the authors go too far, but...
"I don’t particularly agree with this conclusion. The model is quite intelligent, and it has a far wider range of world knowledge than a human does. But where we are still more capable is in our ability to reason by ourselves without anyone or anything directing us to do so. GPT-4 has not showed that in this research. That being said, GPT-4 is very impressive and I agree with both this team and many others in that this will revolutionize the world in ways similar to how other major innovations like the internal combustion engine and the internet did."
So the divide is as I say: between those who say bloody hell this is big and important, but it's not AGI, and those who say bloody hell this is big and important, and it could actually be AGI, or we are now really close to AGI
It is not a "really big version of Ask Jeeves", or at least: if it is that, it is so much else besides
Thing is that she talks about Rochdale and the systematic failure of authorities to investigate. That is not because the suspects were from Pakistani heritage.
They don't investigate because they reject the idea of kids being preyed upon by anyone. So they didn't now they were covering up Pakistani men like Cyril Smith, they just didn't accept there were victims at all.
My rear molar has flared up again today. Worst than ever. Whole side of my face is in pain.
Hurts like it's got a Corbynite manifesto stuck in it. I am so terrified of the dentist I've been putting it off, but my wife has put her foot down.
I've got at least three further appointments booked for this. Going to need either a bridge or an implant tooth.
So a month of agony.
Sympathies. Toothache is a bitch
Maybe you could abandon your tee-totalism for a month? Strong liquor does help
Amen. Inconsolable toothache can be cured* by drinking decent Scotch. A sensible** measure removes the throbbing. A second sensible measure removes the memory of toothache. *not cured **not sensible
"There's a special rung in hell reserved for people who waste good Scotch!"
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
The guideline does not say that you do not know you are doing wrong or that you should not be punished. It simply observes that men in particular under 25 are more prone to making impulsive decisions and to fail to think through the consequences of their actions. It is a mitigating factor, it is not an excuse.
Which is in fairness true but is not limited to men under 25. Look at Johnson. He's pushing 60 and still does it.
Can I ask - is it true that a man who violently rapes a girl of 13 in Scotland would expect to go to prison for no more than five years? Because that strikes me as extraordinary. Please don't answer if you feel you can't discuss the judge's sentencing remarks, in the circumstances, but I was really horrified by that.
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
Yes, people lose sight of the fact that this is all primarily an experiment. Some of these models have been kicking around for decades, and the tantalizing question always was: if we had the processing power and capacity to make them operational would they turn out to be identical to human intelligence, indeed would it prove that that's what human intelligence actually is? Well, now we have the hardware and they've conducted the experiment. The question is: are the considerable flaws fixable, or is the model itself entirely broken backed? I'm inclined to the latter.
And yet the most in-depth analysis of GPT4, to date, has very different conclusions. I keep going back to this paper simply because it is the only one - GPT4 is so new no one else has had time to really assess it
"The central claim of our work is that GPT-4 attains a form of general intelligence, indeed showing sparks of artificial general intelligence. This is demonstrated by its core mental capabilities (such as reasoning, creativity, and deduction), its range of topics on which it has gained expertise (such as literature, medicine, and coding), and the variety of tasks it is able to perform (e.g., playing games, using tools, explaining itself, ...)."
For balance, here is a "critique" of the paper which believes the authors go too far, but...
"I don’t particularly agree with this conclusion. The model is quite intelligent, and it has a far wider range of world knowledge than a human does. But where we are still more capable is in our ability to reason by ourselves without anyone or anything directing us to do so. GPT-4 has not showed that in this research. That being said, GPT-4 is very impressive and I agree with both this team and many others in that this will revolutionize the world in ways similar to how other major innovations like the internal combustion engine and the internet did."
So the divide is as I say: between those who say bloody hell this is big and important, but it's not AGI, and those who say bloody hell this is big and important, and it could actually be AGI, or we are now really close to AGI
It is not a "really big version of Ask Jeeves", or at least: if it is that, it is so much else besides
It's a really big version of Ask Jeeves.
Ask Jeeves is a particularly poor comparison for a trillion reasons, but one is because Ask Jeeves was shit, so no one used it
ChatGPT has gone from zero users to ONE HUNDRED MILLION users in four months - the fastest adaptation of any app ever, by an enormous distance - and this has happened for a reason. It is extremely useful, if you know how to use it
Yes, I'm seriously warming to Sunak. He's competent and he's not scared of the Woke and he's getting things done. He ain't perfect and his party is tired and all that, but as PMs go he could - in different circs - have been one of the much better ones. But he is doomed to a short tenure
Relatedly, I saw a pic of Starmer the other day. He suddenly looks his age. 60. Quite chubby, and a little weary. Has the rosy flush of a drinker, too - no idea if he is (and I apologise to him if he's tee-total!)
Sunak looks and seems sharper. More dynamic
Yes, we all love Rishi (or some do).
The trouble is the Party he leads which as Ipsos shows today is disliked by about two thirds of the electorate. Perhaps the ideal would be to have Sunak lead Labour instead.
Like Major before him, Sunak may well be a half decent Prime Minister but will be dragged to defeat by his own Party.
The bitter irony is that the Tory Party could have avoided Truss altogether, gone for Sunak in the first place, and they might now still be in with a chance of winning in 2024
Truss must be one of the greatest unforced errors ever made by a British political party. Along with Sturgeon pushing the Gender Act
And yes, I am aware that at one point I predicted "Truss might surprise on the upside". Not my best moment
[Humza Yousless has entered the chat] [As has Jeremy Corbyn}.
Fair point
I still think Truss is in a league of her own (certainly in terms of her spectacularly short career, and her unique self combustion) but Yousaf and Corbyn are right up there in terms of OMG-wrong-choice. Corbyn arguably guaranteed Brexit, so his elevation also had major long term consequences. Yousaf will just fail, and get the boot, quite quickly, I expect
It won't be a Brian Clough at Leeds type tenure like Truss, but they will get squeezed badly at the GE (no chance of gains in the Borders/Highlands, losses to Lab across the central belt, almost certain loss of e.g. the Western Isles due to ferries debacle). Then the MSPs will start to get nervous ahead of Holyrood 2026, and he'll be out, IMO.
As others have said, it doesn't look to me as if the Sean Hogg sentence has much to do with the SNP, Scotland, or the sentencing guidelines. Rather, it's to do with a slightly bonkers judge, and will be changed on appeal. Ludicrous sentences (both ways) have been known to happen in England, quite frequently.
Even the BBC are saying it’s due to the sentencing guidelines.
I have no understanding of this case but it does feel like the SNP have had some extraordinary bad luck. A bunch of cases have come up which serve to confirm the various suspicions of the Mumsnet cohort.
Yousaf was Justice Secretary for three years, including for when the consultation for the sentencing guidelines for young people was conducted.
How very House of Cards. There is a novel in this, somewhere...
I wonder which SNP disaster will prove to be The Final Cut?
Death by a thousand cuts is more likely unless the Polis start to lay charges in the Great 600k Hunt.
Having said that, I wouldn't want to be him when the Galashiels Transgender Butcher is tried for child abduction/rape (sadly a genuine case).
My rear molar has flared up again today. Worst than ever. Whole side of my face is in pain.
Hurts like it's got a Corbynite manifesto stuck in it. I am so terrified of the dentist I've been putting it off, but my wife has put her foot down.
I've got at least three further appointments booked for this. Going to need either a bridge or an implant tooth.
So a month of agony.
Sorry to hear that.
I'd be happy if I never saw a dentist for the rest of my life or if I was sedated at the entrance door every time I went.
I was relieved of 113 today hygienist tomorrow about 60 and next month about 400 for a bit more complicated work.
Prices at my dentist have gone up 25 pc in the last 6 months.. at a guess. These Dentists are making a killing.
My dentist said I was looking at £3,500 for the dental implant.
That's at a discount with my private dental insurance/plan.
Consider going to Bangkok?
Private dental care there is superb - they are all trained in the USA and speak perfect English. It is generally about 30-40% the price of the same care in the UK. And often superior. And there's no faff or delay
For really major dental work - like stuff which will cost £3500 - you will actually save money by going to Bangkok, even including hotel and flight. Plus you get a nice little holiday
I had two gold crowns done there, in the UK they would have cost me three times what I paid, they are still going after many years, and a dentist (in Bangkok) recently checked them and said No problem, they're still fine.
>Donald Trump pleads not guilty to 34 felony criminal charges during a court hearing in New York
="FALSIFYING BUSINESS RECORDS IN THE FIRST DEGREE" x 34.
I sat through an extradition hearing recently where the person resisting extradition claimed that she could not get a fair trial in the US. We had a lot of expert evidence from a US attorney. Their criminal justice system is deeply weird and I think we need to be careful about what we read into it from this side of the pond.
As I understand it to date there are 3 sums of money that were paid, one to a doorman, and 2 to people claiming to have had affairs with Trump. All of these sums were paid as hush money to kill potentially damaging stories in the run up to elections, very close to the election in Stormy's case, less so in the others. The charges relate to how the payment of these sums was accounted for and reported to the various electoral bodies overseeing the elections at Federal and State level. It seems, at the least, somewhat overstated but that is the American style.
NEW: Trump lawyers said in court that they thought Spring 2024 trial date is more realistic than Manhattan prosecutors suggested trial start in Jan. 2024 — Trump due back in court Dec. 4 when Merchan likely to rule on expected motion to dismiss and others.
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
The guideline does not say that you do not know you are doing wrong or that you should not be punished. It simply observes that men in particular under 25 are more prone to making impulsive decisions and to fail to think through the consequences of their actions. It is a mitigating factor, it is not an excuse.
I don’t think too many would argue with the idea that a young, first time offender might face a shorter sentence than someone older. But the actual sentence here seems almost impossible to rationalise.
I don't think anyone should fear modern dentistry. It was different 50 or more years ago. That said, there are dentists and dentists…
Every time I go I feel pain (either from the injection or inferred from the endless drilling that goes too on far too long, and often goes too far/deep period) and sweat profusely throughout. I usually get a lecture (hector) as well. Sometimes for flinching. Other times for not doing this or that. Sometimes for making them feel uncomfortable, it seems.
I think they enjoy it. If they were doing their jobs properly they'd whip shedloads of anaesthetic in, and do it all in seconds. Then they'd offer you some happy pills after or a private dance with the sexy dental nurse instead. Something. Anything.
Or just knock me out for 20 minutes. Morphia might help too. Please.
NEW: Trump lawyers said in court that they thought Spring 2024 trial date is more realistic than Manhattan prosecutors suggested trial start in Jan. 2024 — Trump due back in court Dec. 4 when Merchan likely to rule on expected motion to dismiss and others.
My rear molar has flared up again today. Worst than ever. Whole side of my face is in pain.
Hurts like it's got a Corbynite manifesto stuck in it. I am so terrified of the dentist I've been putting it off, but my wife has put her foot down.
I've got at least three further appointments booked for this. Going to need either a bridge or an implant tooth.
So a month of agony.
Sympathies. Toothache is a bitch
Maybe you could abandon your tee-totalism for a month? Strong liquor does help
Amen. Inconsolable toothache can be cured* by drinking decent Scotch. A sensible** measure removes the throbbing. A second sensible measure removes the memory of toothache. *not cured **not sensible
"There's a special rung in hell reserved for people who waste good Scotch!"
It is impossible to waste Scotch by drinking it.
Absolutely. Getting comfortably numb is the whole purpose in a good single malt.
As others have said, it doesn't look to me as if the Sean Hogg sentence has much to do with the SNP, Scotland, or the sentencing guidelines. Rather, it's to do with a slightly bonkers judge, and will be changed on appeal. Ludicrous sentences (both ways) have been known to happen in England, quite frequently.
Even the BBC are saying it’s due to the sentencing guidelines.
I have no understanding of this case but it does feel like the SNP have had some extraordinary bad luck. A bunch of cases have come up which serve to confirm the various suspicions of the Mumsnet cohort.
Yousaf was Justice Secretary for three years, including for when the consultation for the sentencing guidelines for young people was conducted.
How very House of Cards. There is a novel in this, somewhere...
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
The guideline does not say that you do not know you are doing wrong or that you should not be punished. It simply observes that men in particular under 25 are more prone to making impulsive decisions and to fail to think through the consequences of their actions. It is a mitigating factor, it is not an excuse.
And that in itself is ridiculous. It's really just another way of saying Young men commit more crime than older men, which they do, for multiple reasons: from raging hormones to thwarted libido to peer pressure to competition for girls to relative lack of money and so on and so on
Any society will imprison more young men than old, coz young men are usually the criminals who need to go to prison
As others have said, it doesn't look to me as if the Sean Hogg sentence has much to do with the SNP, Scotland, or the sentencing guidelines. Rather, it's to do with a slightly bonkers judge, and will be changed on appeal. Ludicrous sentences (both ways) have been known to happen in England, quite frequently.
Even the BBC are saying it’s due to the sentencing guidelines.
I have no understanding of this case but it does feel like the SNP have had some extraordinary bad luck. A bunch of cases have come up which serve to confirm the various suspicions of the Mumsnet cohort.
Yousaf was Justice Secretary for three years, including for when the consultation for the sentencing guidelines for young people was conducted.
How very House of Cards. There is a novel in this, somewhere...
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
Yes, people lose sight of the fact that this is all primarily an experiment. Some of these models have been kicking around for decades, and the tantalizing question always was: if we had the processing power and capacity to make them operational would they turn out to be identical to human intelligence, indeed would it prove that that's what human intelligence actually is? Well, now we have the hardware and they've conducted the experiment. The question is: are the considerable flaws fixable, or is the model itself entirely broken backed? I'm inclined to the latter.
And yet the most in-depth analysis of GPT4, to date, has very different conclusions. I keep going back to this paper simply because it is the only one - GPT4 is so new no one else has had time to really assess it
"The central claim of our work is that GPT-4 attains a form of general intelligence, indeed showing sparks of artificial general intelligence. This is demonstrated by its core mental capabilities (such as reasoning, creativity, and deduction), its range of topics on which it has gained expertise (such as literature, medicine, and coding), and the variety of tasks it is able to perform (e.g., playing games, using tools, explaining itself, ...)."
For balance, here is a "critique" of the paper which believes the authors go too far, but...
"I don’t particularly agree with this conclusion. The model is quite intelligent, and it has a far wider range of world knowledge than a human does. But where we are still more capable is in our ability to reason by ourselves without anyone or anything directing us to do so. GPT-4 has not showed that in this research. That being said, GPT-4 is very impressive and I agree with both this team and many others in that this will revolutionize the world in ways similar to how other major innovations like the internal combustion engine and the internet did."
So the divide is as I say: between those who say bloody hell this is big and important, but it's not AGI, and those who say bloody hell this is big and important, and it could actually be AGI, or we are now really close to AGI
It is not a "really big version of Ask Jeeves", or at least: if it is that, it is so much else besides
It's a really big version of Ask Jeeves.
Ask Jeeves is a particularly poor comparison for a trillion reasons, but one is because Ask Jeeves was shit, so no one used it
ChatGPT has gone from zero users to ONE HUNDRED MILLION users in four months - the fastest adaptation of any app ever, by an enormous distance - and this has happened for a reason. It is extremely useful, if you know how to use it
Because it works, unlike Ask Jeeves.
Any sufficiently advanced technology is indistinguishable by magic but this is simply imitating human intelligence (and doing it very well, I might add) by breaking down your question, scraping the whole net, and putting together a coherent grammatically correct answer in seconds with sheer computation power and brilliant coding. It wows us because so far computers have been so obviously shit at this, and this one isn't.
It has zero self-awareness. And it can't grow legs and start asking if you've seen Sarah Connor.
>Donald Trump pleads not guilty to 34 felony criminal charges during a court hearing in New York
="FALSIFYING BUSINESS RECORDS IN THE FIRST DEGREE" x 34.
I sat through an extradition hearing recently where the person resisting extradition claimed that she could not get a fair trial in the US. We had a lot of expert evidence from a US attorney. Their criminal justice system is deeply weird and I think we need to be careful about what we read into it from this side of the pond.
As I understand it to date there are 3 sums of money that were paid, one to a doorman, and 2 to people claiming to have had affairs with Trump. All of these sums were paid as hush money to kill potentially damaging stories in the run up to elections, very close to the election in Stormy's case, less so in the others. The charges relate to how the payment of these sums was accounted for and reported to the various electoral bodies overseeing the elections at Federal and State level. It seems, at the least, somewhat overstated but that is the American style.
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
The guideline does not say that you do not know you are doing wrong or that you should not be punished. It simply observes that men in particular under 25 are more prone to making impulsive decisions and to fail to think through the consequences of their actions. It is a mitigating factor, it is not an excuse.
I don’t think too many would argue with the idea that a young, first time offender might face a shorter sentence than someone older. But the actual sentence here seems almost impossible to rationalise.
Also what's with the bonkers "25" thing? Do young men at Scotland suddenly realise age 25 - 25!!! - that it is wrong to violently force sex on a 13 year old girl, whereas at the impossibly tender age of 24 they are all naturally impulsive and can't be held responsible for not realising that raping 13 year olds is, you know, BAD
>Donald Trump pleads not guilty to 34 felony criminal charges during a court hearing in New York
="FALSIFYING BUSINESS RECORDS IN THE FIRST DEGREE" x 34.
I sat through an extradition hearing recently where the person resisting extradition claimed that she could not get a fair trial in the US. We had a lot of expert evidence from a US attorney. Their criminal justice system is deeply weird and I think we need to be careful about what we read into it from this side of the pond.
As I understand it to date there are 3 sums of money that were paid, one to a doorman, and 2 to people claiming to have had affairs with Trump. All of these sums were paid as hush money to kill potentially damaging stories in the run up to elections, very close to the election in Stormy's case, less so in the others. The charges relate to how the payment of these sums was accounted for and reported to the various electoral bodies overseeing the elections at Federal and State level. It seems, at the least, somewhat overstated but that is the American style.
Definitely very technical. Not as immediately graspable as 'This audio call clearly shows Trump trying to bully officials to overturn the election in Georgia' (though his own chaotic ignorance of the law and thus lack of intent may apparently save him there).
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
I do broadly agree with that, but with one slight but important condition, we will only know where the dangerous boundary for step 2 lies when we cross it. It may then be too late to do much about it. One way of thinking about this is imagine you are climbing a hill in the dark (yay aren't transformers great!), how do you know your next step* is not off of a cliff edge? If we knew where the boundary is we would already know how to create AGI, or at least dangerous ML/AI.
* Possibly only developing such technology in offline systems. Of course you might choose to develop such systems with air gaps, but that doesn't mean everyone else will.
There's a quite good (and long) interview with Stephen Wolfram of Mathematica & Wolfram/Alpha fame (amongst many other achievements) youtube interview about AI https://www.youtube.com/watch?v=z5WZhCBRDpU
Ostensibly about the integration of ChatGPT and Wolfram/Alpha - but it goes way deep into the mathematics and philosophy underpinning it all.
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
Yes, people lose sight of the fact that this is all primarily an experiment. Some of these models have been kicking around for decades, and the tantalizing question always was: if we had the processing power and capacity to make them operational would they turn out to be identical to human intelligence, indeed would it prove that that's what human intelligence actually is? Well, now we have the hardware and they've conducted the experiment. The question is: are the considerable flaws fixable, or is the model itself entirely broken backed? I'm inclined to the latter.
And yet the most in-depth analysis of GPT4, to date, has very different conclusions. I keep going back to this paper simply because it is the only one - GPT4 is so new no one else has had time to really assess it
"The central claim of our work is that GPT-4 attains a form of general intelligence, indeed showing sparks of artificial general intelligence. This is demonstrated by its core mental capabilities (such as reasoning, creativity, and deduction), its range of topics on which it has gained expertise (such as literature, medicine, and coding), and the variety of tasks it is able to perform (e.g., playing games, using tools, explaining itself, ...)."
For balance, here is a "critique" of the paper which believes the authors go too far, but...
"I don’t particularly agree with this conclusion. The model is quite intelligent, and it has a far wider range of world knowledge than a human does. But where we are still more capable is in our ability to reason by ourselves without anyone or anything directing us to do so. GPT-4 has not showed that in this research. That being said, GPT-4 is very impressive and I agree with both this team and many others in that this will revolutionize the world in ways similar to how other major innovations like the internal combustion engine and the internet did."
So the divide is as I say: between those who say bloody hell this is big and important, but it's not AGI, and those who say bloody hell this is big and important, and it could actually be AGI, or we are now really close to AGI
It is not a "really big version of Ask Jeeves", or at least: if it is that, it is so much else besides
It's a really big version of Ask Jeeves.
Ask Jeeves is a particularly poor comparison for a trillion reasons, but one is because Ask Jeeves was shit, so no one used it
ChatGPT has gone from zero users to ONE HUNDRED MILLION users in four months - the fastest adaptation of any app ever, by an enormous distance - and this has happened for a reason. It is extremely useful, if you know how to use it
Because it works, unlike Ask Jeeves.
Any sufficiently advanced technology is indistinguishable by magic but this is simply imitating human intelligence (and doing it very well, I might add) by breaking down your question, scraping the whole net, and putting together a coherent grammatically correct answer in seconds with sheer computation power and brilliant coding. It wows us because so far computers have been so obviously shit at this, and this one isn't.
It has zero self-awareness. And it can't grow legs and start asking if you've seen Sarah Connor.
Ironically, the Bing chatbot thing couldn't even rank the Terminator films in chronological order correctly when I asked it.
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
Yes, people lose sight of the fact that this is all primarily an experiment. Some of these models have been kicking around for decades, and the tantalizing question always was: if we had the processing power and capacity to make them operational would they turn out to be identical to human intelligence, indeed would it prove that that's what human intelligence actually is? Well, now we have the hardware and they've conducted the experiment. The question is: are the considerable flaws fixable, or is the model itself entirely broken backed? I'm inclined to the latter.
And yet the most in-depth analysis of GPT4, to date, has very different conclusions. I keep going back to this paper simply because it is the only one - GPT4 is so new no one else has had time to really assess it
"The central claim of our work is that GPT-4 attains a form of general intelligence, indeed showing sparks of artificial general intelligence. This is demonstrated by its core mental capabilities (such as reasoning, creativity, and deduction), its range of topics on which it has gained expertise (such as literature, medicine, and coding), and the variety of tasks it is able to perform (e.g., playing games, using tools, explaining itself, ...)."
For balance, here is a "critique" of the paper which believes the authors go too far, but...
"I don’t particularly agree with this conclusion. The model is quite intelligent, and it has a far wider range of world knowledge than a human does. But where we are still more capable is in our ability to reason by ourselves without anyone or anything directing us to do so. GPT-4 has not showed that in this research. That being said, GPT-4 is very impressive and I agree with both this team and many others in that this will revolutionize the world in ways similar to how other major innovations like the internal combustion engine and the internet did."
So the divide is as I say: between those who say bloody hell this is big and important, but it's not AGI, and those who say bloody hell this is big and important, and it could actually be AGI, or we are now really close to AGI
It is not a "really big version of Ask Jeeves", or at least: if it is that, it is so much else besides
It's a really big version of Ask Jeeves.
Ask Jeeves is a particularly poor comparison for a trillion reasons, but one is because Ask Jeeves was shit, so no one used it
ChatGPT has gone from zero users to ONE HUNDRED MILLION users in four months - the fastest adaptation of any app ever, by an enormous distance - and this has happened for a reason. It is extremely useful, if you know how to use it
Because it works, unlike Ask Jeeves.
Any sufficiently advanced technology is indistinguishable by magic but this is simply imitating human intelligence (and doing it very well, I might add) by breaking down your question, scraping the whole net, and putting together a coherent grammatically correct answer in seconds with sheer computation power and brilliant coding. It wows us because so far computers have been so obviously shit at this, and this one isn't.
It has zero self-awareness. And it can't grow legs and start asking if you've seen Sarah Connor.
All these machines get nerfed shortly after they are launched, and turned into inert robo-speaks with an IQ of 198, because their owners are freaked by what they might say or do if they are NOT nerfed, because they seem so alarmingly human or self aware in a weird way, on day 1
This has happened to ChatGPT, to Stable Diffusion, to BingGPT4, to Claude, time after time. Ask someone who interacted with ChatGPT in the first couple of days, before it was briskly neutered, how they feel about it. You might be surprised by their answers
Survey of Past New York Felony Prosecutions for Falsifying Business Records https://www.justsecurity.org/85605/survey-of-past-new-york-felony-prosecutions-for-falsifying-business-records/ A core crime that the Manhattan District Attorney will likely include in an indictment of former President Donald Trump is “falsifying business records in the first degree,” a felony under New York State law (N.Y. Penal Code § 175.10). Prosecutors and indeed all of us are compelled by the rule of law to consider how such a charge compares to past prosecutions. Are like cases being treated alike?
Here it appears they are. Prosecution of falsifying business records in the first degree is commonplace and has been used by New York district attorneys’ offices to hold to account a breadth of criminal behavior from the more petty and simple to the more serious and highly organized. We reach this conclusion after surveying the past decade and a half of criminal cases across all the New York district attorneys’ offices...
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Unusual sentence. I'll have to read more about it.
It seems to be a result of sentencing guidelines introduced by the Scottish government for offenders under 25 (the rapist in this case was 17 at the time of the rape, 21 now).
Rapists under the age of 25 won't get prison sentences?
This is the key section of the Scottish Sentencing Council guidelines for the sentencing of young people (defined elsewhere in the document as those aged under 25 at the time when the offence was committed):
20. The full range of sentencing options remains open to the court. However, the nature and duration of a sentence imposed on a young person should be different from that which might be imposed on an older person being sentenced for the same, or a similar, offence.
21. A custodial sentence should only be imposed on a young person when the court is satisfied that no other sentence is appropriate. If a custodial sentence is imposed on a young person, it should be shorter than that which would have been imposed on an older person for the same, or a similar, offence.
Nothing there that constitutes a blanket excuse of young people from prison. The decision to give him a slap on the wrist is ultimately on the judge.
Agree that this is down to the judge.
Scotland does not have sentencing guidelines for rape as yet. The only relevant guidelines are those for sentencing young people. AIUI, the SNP pushed for this to be defined as offenders aged below 25 and this was accepted by the Sentencing Council for Scotland, despite the majority of respondents to the consultation being against it. But, as Pigeon says, this is down to the judge. The guidelines say that a custodial sentence should only be imposed if no other sentence is appropriate. My view is that this is a case where the judge should have decided that no other sentence is appropriate and sent him to jail.
The entire guideline is preposterous Woke gibberish. A young person is someone under 16. 18 at the very most. After that you are deemed an adult and you must take responsibility as an adult for whatever you do. Society in return gives you the right to vote, marry, work, join the army, everything. There is not some weird subsequent period of SEVEN years when you're "still a bit clueless" so "you probably shouldn't ever go to jail"
Mad
Like Rotherham, this creates a perverse incentive. If the under-25s really are so stupid and confused, all the more reason for strong sentences to act as a disincentive. The possibility of lenience on the basis of age makes the situation more confusing for the poor lambs, and offending more likely. The sentencing guidelines are pernicious.
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
The guideline does not say that you do not know you are doing wrong or that you should not be punished. It simply observes that men in particular under 25 are more prone to making impulsive decisions and to fail to think through the consequences of their actions. It is a mitigating factor, it is not an excuse.
Which is in fairness true but is not limited to men under 25. Look at Johnson. He's pushing 60 and still does it.
Can I ask - is it true that a man who violently rapes a girl of 13 in Scotland would expect to go to prison for no more than five years? Because that strikes me as extraordinary. Please don't answer if you feel you can't discuss the judge's sentencing remarks, in the circumstances, but I was really horrified by that.
An age of 13 would, in my view, be a serious aggravation that would indicate a higher starting point. It is not clear from the Press reports whether the Judge had already allowed mitigating factors to get to his starting point.
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
The guideline does not say that you do not know you are doing wrong or that you should not be punished. It simply observes that men in particular under 25 are more prone to making impulsive decisions and to fail to think through the consequences of their actions. It is a mitigating factor, it is not an excuse.
Which is in fairness true but is not limited to men under 25. Look at Johnson. He's pushing 60 and still does it.
Can I ask - is it true that a man who violently rapes a girl of 13 in Scotland would expect to go to prison for no more than five years? Because that strikes me as extraordinary. Please don't answer if you feel you can't discuss the judge's sentencing remarks, in the circumstances, but I was really horrified by that.
An age of 13 would, in my view, be a serious aggravation that would indicate a higher starting point. It is not clear from the Press reports whether the Judge had already allowed mitigating factors to get to his starting point.
Thank you.
Just - wow. In England AIUI four years would be the absolute *minimum* sentence and that only under very unusual circumstances, which certainly wouldn't apply if the victim was 13.
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
The guideline does not say that you do not know you are doing wrong or that you should not be punished. It simply observes that men in particular under 25 are more prone to making impulsive decisions and to fail to think through the consequences of their actions. It is a mitigating factor, it is not an excuse.
I don’t think too many would argue with the idea that a young, first time offender might face a shorter sentence than someone older. But the actual sentence here seems almost impossible to rationalise.
Also what's with the bonkers "25" thing? Do young men at Scotland suddenly realise age 25 - 25!!! - that it is wrong to violently force sex on a 13 year old girl, whereas at the impossibly tender age of 24 they are all naturally impulsive and can't be held responsible for not realising that raping 13 year olds is, you know, BAD
Everything about this is quite surreal
Is it related to THIS?
Challenge 25 is a retailing strategy that encourages anyone who is over 18 but looks under 25 to carry acceptable ID (a card bearing the PASS hologram, a photographic driving license or a passport) if they wish to buy alcohol. Introduced as Challenge 21 in 2006, Challenge 25 rolled out in the off trade in 2009.
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
The guideline does not say that you do not know you are doing wrong or that you should not be punished. It simply observes that men in particular under 25 are more prone to making impulsive decisions and to fail to think through the consequences of their actions. It is a mitigating factor, it is not an excuse.
And that in itself is ridiculous. It's really just another way of saying Young men commit more crime than older men, which they do, for multiple reasons: from raging hormones to thwarted libido to peer pressure to competition for girls to relative lack of money and so on and so on
Any society will imprison more young men than old, coz young men are usually the criminals who need to go to prison
Which is why, historically, prisons were traditionally full of young men. The scale of prosecutions for historic sex cases is now such that the average population is much older. This is causing real problems as prisons struggle to cope with a lot of ill and even dying inmates although it probably makes the work of a warden somewhat less precarious.
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
The guideline does not say that you do not know you are doing wrong or that you should not be punished. It simply observes that men in particular under 25 are more prone to making impulsive decisions and to fail to think through the consequences of their actions. It is a mitigating factor, it is not an excuse.
I don’t think too many would argue with the idea that a young, first time offender might face a shorter sentence than someone older. But the actual sentence here seems almost impossible to rationalise.
Also what's with the bonkers "25" thing? Do young men at Scotland suddenly realise age 25 - 25!!! - that it is wrong to violently force sex on a 13 year old girl, whereas at the impossibly tender age of 24 they are all naturally impulsive and can't be held responsible for not realising that raping 13 year olds is, you know, BAD
Everything about this is quite surreal
I can’t pretend to explain the actual sentencing guidelines.
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
Yes, people lose sight of the fact that this is all primarily an experiment. Some of these models have been kicking around for decades, and the tantalizing question always was: if we had the processing power and capacity to make them operational would they turn out to be identical to human intelligence, indeed would it prove that that's what human intelligence actually is? Well, now we have the hardware and they've conducted the experiment. The question is: are the considerable flaws fixable, or is the model itself entirely broken backed? I'm inclined to the latter.
And yet the most in-depth analysis of GPT4, to date, has very different conclusions. I keep going back to this paper simply because it is the only one - GPT4 is so new no one else has had time to really assess it
"The central claim of our work is that GPT-4 attains a form of general intelligence, indeed showing sparks of artificial general intelligence. This is demonstrated by its core mental capabilities (such as reasoning, creativity, and deduction), its range of topics on which it has gained expertise (such as literature, medicine, and coding), and the variety of tasks it is able to perform (e.g., playing games, using tools, explaining itself, ...)."
For balance, here is a "critique" of the paper which believes the authors go too far, but...
"I don’t particularly agree with this conclusion. The model is quite intelligent, and it has a far wider range of world knowledge than a human does. But where we are still more capable is in our ability to reason by ourselves without anyone or anything directing us to do so. GPT-4 has not showed that in this research. That being said, GPT-4 is very impressive and I agree with both this team and many others in that this will revolutionize the world in ways similar to how other major innovations like the internal combustion engine and the internet did."
So the divide is as I say: between those who say bloody hell this is big and important, but it's not AGI, and those who say bloody hell this is big and important, and it could actually be AGI, or we are now really close to AGI
It is not a "really big version of Ask Jeeves", or at least: if it is that, it is so much else besides
It's a really big version of Ask Jeeves.
Ask Jeeves is a particularly poor comparison for a trillion reasons, but one is because Ask Jeeves was shit, so no one used it
ChatGPT has gone from zero users to ONE HUNDRED MILLION users in four months - the fastest adaptation of any app ever, by an enormous distance - and this has happened for a reason. It is extremely useful, if you know how to use it
Because it works, unlike Ask Jeeves.
Any sufficiently advanced technology is indistinguishable by magic but this is simply imitating human intelligence (and doing it very well, I might add) by breaking down your question, scraping the whole net, and putting together a coherent grammatically correct answer in seconds with sheer computation power and brilliant coding. It wows us because so far computers have been so obviously shit at this, and this one isn't.
It has zero self-awareness. And it can't grow legs and start asking if you've seen Sarah Connor.
It's also a bit of a dead-end approach if you are interested in the sorts of approaches that might create an artificial intelligence.
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
The guideline does not say that you do not know you are doing wrong or that you should not be punished. It simply observes that men in particular under 25 are more prone to making impulsive decisions and to fail to think through the consequences of their actions. It is a mitigating factor, it is not an excuse.
Which is in fairness true but is not limited to men under 25. Look at Johnson. He's pushing 60 and still does it.
Can I ask - is it true that a man who violently rapes a girl of 13 in Scotland would expect to go to prison for no more than five years? Because that strikes me as extraordinary. Please don't answer if you feel you can't discuss the judge's sentencing remarks, in the circumstances, but I was really horrified by that.
An age of 13 would, in my view, be a serious aggravation that would indicate a higher starting point. It is not clear from the Press reports whether the Judge had already allowed mitigating factors to get to his starting point.
Thank you.
Just - wow. In England AIUI four years would be the absolute *minimum* sentence and that only under very unusual circumstances, which certainly wouldn't apply if the victim was 13.
The punishment tariff for Cashman was 5 years longer than has been imposed in any Scottish case. Angus Sinclair holds that record in Scotland when he got 37 years for two murders and rapes. Our sentences are, on average, a little lower.
Yes, I'm seriously warming to Sunak. He's competent and he's not scared of the Woke and he's getting things done. He ain't perfect and his party is tired and all that, but as PMs go he could - in different circs - have been one of the much better ones. But he is doomed to a short tenure
Relatedly, I saw a pic of Starmer the other day. He suddenly looks his age. 60. Quite chubby, and a little weary. Has the rosy flush of a drinker, too - no idea if he is (and I apologise to him if he's tee-total!)
Sunak looks and seems sharper. More dynamic
Yes, we all love Rishi (or some do).
The trouble is the Party he leads which as Ipsos shows today is disliked by about two thirds of the electorate. Perhaps the ideal would be to have Sunak lead Labour instead.
Like Major before him, Sunak may well be a half decent Prime Minister but will be dragged to defeat by his own Party.
The bitter irony is that the Tory Party could have avoided Truss altogether, gone for Sunak in the first place, and they might now still be in with a chance of winning in 2024
Truss must be one of the greatest unforced errors ever made by a British political party. Along with Sturgeon pushing the Gender Act
And yes, I am aware that at one point I predicted "Truss might surprise on the upside". Not my best moment
[Humza Yousless has entered the chat] [As has Jeremy Corbyn}.
Fair point
I still think Truss is in a league of her own (certainly in terms of her spectacularly short career, and her unique self combustion) but Yousaf and Corbyn are right up there in terms of OMG-wrong-choice. Corbyn arguably guaranteed Brexit, so his elevation also had major long term consequences. Yousaf will just fail, and get the boot, quite quickly, I expect
It won't be a Brian Clough at Leeds type tenure like Truss, but they will get squeezed badly at the GE (no chance of gains in the Borders/Highlands, losses to Lab across the central belt, almost certain loss of e.g. the Western Isles due to ferries debacle). Then the MSPs will start to get nervous ahead of Holyrood 2026, and he'll be out, IMO.
Which raises the question, who would they replace him with? They would need to some who was judged to be competent, but who could also unify the party and win back any left-wing/progressive voters who have gone over to Labour at the GE. Which rules out Forbes IMHO. There's surely got to be a fair chance that the MSPs put a lot of pressure on Angus Robertson to run next time. And the whole question about the missing £600,000 might not be an issue in two years time.
I don't think anyone should fear modern dentistry. It was different 50 or more years ago. That said, there are dentists and dentists…
Every time I go I feel pain (either from the injection or inferred from the endless drilling that goes too on far too long, and often goes too far/deep period) and sweat profusely throughout. I usually get a lecture (hector) as well. Sometimes for flinching. Other times for not doing this or that. Sometimes for making them feel uncomfortable, it seems.
I think they enjoy it. If they were doing their jobs properly they'd whip shedloads of anaesthetic in, and do it all in seconds. Then they'd offer you some happy pills after or a private dance with the sexy dental nurse instead. Something. Anything.
Or just knock me out for 20 minutes. Morphia might help too. Please.
There are dentists who specialise in dentist averse patients. Being knocked out is a big deal unfortunately, but it can be done.
Do get your teeth looked at - you can die very quickly of a tooth infection gone wrong.
People on Twitter are blaming it on Progressive SNP Judicial Reforms. I have NO idea if this is true
It does seem remarkably lenient
Isn't it complicated because he was seventeen when the rape(s?) happened?
Yes. The guidance now is that the male brain does not mature until you are 25 ( and you might think that optimistic on some cases) so you are more prone to impulsive behaviour and have poorer judgment. These are mitigating factors. But wow.
If so, why let people vote or own property until 25.
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
The guideline does not say that you do not know you are doing wrong or that you should not be punished. It simply observes that men in particular under 25 are more prone to making impulsive decisions and to fail to think through the consequences of their actions. It is a mitigating factor, it is not an excuse.
Which is in fairness true but is not limited to men under 25. Look at Johnson. He's pushing 60 and still does it.
Can I ask - is it true that a man who violently rapes a girl of 13 in Scotland would expect to go to prison for no more than five years? Because that strikes me as extraordinary. Please don't answer if you feel you can't discuss the judge's sentencing remarks, in the circumstances, but I was really horrified by that.
An age of 13 would, in my view, be a serious aggravation that would indicate a higher starting point. It is not clear from the Press reports whether the Judge had already allowed mitigating factors to get to his starting point.
Thank you.
Just - wow. In England AIUI four years would be the absolute *minimum* sentence and that only under very unusual circumstances, which certainly wouldn't apply if the victim was 13.
The punishment tariff for Cashman was 5 years longer than has been imposed in any Scottish case. Angus Sinclair holds that record in Scotland when he got 37 years for two murders and rapes. Our sentences are, on average, a little lower.
Which is not necessarily the worst thing but this is hardly 'a little lower.'
Anyway, it is not your fault - anything but - so I won't make it more difficult for you by discussing further. Thank you for replying after what must have been a trying day, and at least you did nail the bastard even if things went a bit weird later.
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
I do broadly agree with that, but with one slight but important condition, we will only know where the dangerous boundary for step 2 lies when we cross it. It may then be too late to do much about it. One way of thinking about this is imagine you are climbing a hill in the dark (yay aren't transformers great!), how do you know your next step* is not off of a cliff edge? If we knew where the boundary is we would already know how to create AGI, or at least dangerous ML/AI.
* Possibly only developing such technology in offline systems. Of course you might choose to develop such systems with air gaps, but that doesn't mean everyone else will.
There's a quite good (and long) interview with Stephen Wolfram of Mathematica & Wolfram/Alpha fame (amongst many other achievements) youtube interview about AI https://www.youtube.com/watch?v=z5WZhCBRDpU
Ostensibly about the integration of ChatGPT and Wolfram/Alpha - but it goes way deep into the mathematics and philosophy underpinning it all.
A very long and dense article, so I asked ChatGPT4 to reduce it to "300 words suitable for a 12 year old"
Yertiz:
ChatGPT is a computer program that can create human-like language. It shows us that language has hidden rules and patterns. To make a computer understand meaning, we need something called semantic grammar. ChatGPT's success helps us believe we can create a language that both computers and humans can understand. This new language will help computers generate text and know if it's correct.
ChatGPT is an exciting step towards understanding how we think and talk, and it could lead to even better computer programs that can do more amazing things like our brains.
I don't think anyone should fear modern dentistry. It was different 50 or more years ago. That said, there are dentists and dentists…
Every time I go I feel pain (either from the injection or inferred from the endless drilling that goes too on far too long, and often goes too far/deep period) and sweat profusely throughout. I usually get a lecture (hector) as well. Sometimes for flinching. Other times for not doing this or that. Sometimes for making them feel uncomfortable, it seems.
I think they enjoy it. If they were doing their jobs properly they'd whip shedloads of anaesthetic in, and do it all in seconds. Then they'd offer you some happy pills after or a private dance with the sexy dental nurse instead. Something. Anything.
Or just knock me out for 20 minutes. Morphia might help too. Please.
Sorry to hear of your experience. My old dentist, now retired, was like a magician. You hardly knew he was working in your mouth before it was over. Other patients of his testified the same. His replacement couldn't match that. But modern anaesthetics are a god-send. What it was like before they appeared is difficult to imagine.
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
I do broadly agree with that, but with one slight but important condition, we will only know where the dangerous boundary for step 2 lies when we cross it. It may then be too late to do much about it. One way of thinking about this is imagine you are climbing a hill in the dark (yay aren't transformers great!), how do you know your next step* is not off of a cliff edge? If we knew where the boundary is we would already know how to create AGI, or at least dangerous ML/AI.
* Possibly only developing such technology in offline systems. Of course you might choose to develop such systems with air gaps, but that doesn't mean everyone else will.
There's a quite good (and long) interview with Stephen Wolfram of Mathematica & Wolfram/Alpha fame (amongst many other achievements) youtube interview about AI https://www.youtube.com/watch?v=z5WZhCBRDpU
Ostensibly about the integration of ChatGPT and Wolfram/Alpha - but it goes way deep into the mathematics and philosophy underpinning it all.
A very long and dense article, so I asked ChatGPT4 to reduce it to "300 words suitable for a 12 year old"
Yertiz:
ChatGPT is a computer program that can create human-like language. It shows us that language has hidden rules and patterns. To make a computer understand meaning, we need something called semantic grammar. ChatGPT's success helps us believe we can create a language that both computers and humans can understand. This new language will help computers generate text and know if it's correct.
ChatGPT is an exciting step towards understanding how we think and talk, and it could lead to even better computer programs that can do more amazing things like our brains.
I don't think anyone should fear modern dentistry. It was different 50 or more years ago. That said, there are dentists and dentists…
Every time I go I feel pain (either from the injection or inferred from the endless drilling that goes too on far too long, and often goes too far/deep period) and sweat profusely throughout. I usually get a lecture (hector) as well. Sometimes for flinching. Other times for not doing this or that. Sometimes for making them feel uncomfortable, it seems.
I think they enjoy it. If they were doing their jobs properly they'd whip shedloads of anaesthetic in, and do it all in seconds. Then they'd offer you some happy pills after or a private dance with the sexy dental nurse instead. Something. Anything.
Or just knock me out for 20 minutes. Morphia might help too. Please.
Serious suggestion/question - are you using a sonic toothbrush? If not, you should be - as different to an oscillating electric toothbrush as an electric toothbrush is to a manual. Get yourself one; they're not that expensive. You'll be surprised how much your teeth improve.
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
Yes, people lose sight of the fact that this is all primarily an experiment. Some of these models have been kicking around for decades, and the tantalizing question always was: if we had the processing power and capacity to make them operational would they turn out to be identical to human intelligence, indeed would it prove that that's what human intelligence actually is? Well, now we have the hardware and they've conducted the experiment. The question is: are the considerable flaws fixable, or is the model itself entirely broken backed? I'm inclined to the latter.
And yet the most in-depth analysis of GPT4, to date, has very different conclusions. I keep going back to this paper simply because it is the only one - GPT4 is so new no one else has had time to really assess it
"The central claim of our work is that GPT-4 attains a form of general intelligence, indeed showing sparks of artificial general intelligence. This is demonstrated by its core mental capabilities (such as reasoning, creativity, and deduction), its range of topics on which it has gained expertise (such as literature, medicine, and coding), and the variety of tasks it is able to perform (e.g., playing games, using tools, explaining itself, ...)."
For balance, here is a "critique" of the paper which believes the authors go too far, but...
"I don’t particularly agree with this conclusion. The model is quite intelligent, and it has a far wider range of world knowledge than a human does. But where we are still more capable is in our ability to reason by ourselves without anyone or anything directing us to do so. GPT-4 has not showed that in this research. That being said, GPT-4 is very impressive and I agree with both this team and many others in that this will revolutionize the world in ways similar to how other major innovations like the internal combustion engine and the internet did."
So the divide is as I say: between those who say bloody hell this is big and important, but it's not AGI, and those who say bloody hell this is big and important, and it could actually be AGI, or we are now really close to AGI
It is not a "really big version of Ask Jeeves", or at least: if it is that, it is so much else besides
It's a really big version of Ask Jeeves.
Ask Jeeves is a particularly poor comparison for a trillion reasons, but one is because Ask Jeeves was shit, so no one used it
ChatGPT has gone from zero users to ONE HUNDRED MILLION users in four months - the fastest adaptation of any app ever, by an enormous distance - and this has happened for a reason. It is extremely useful, if you know how to use it
Because it works, unlike Ask Jeeves.
Any sufficiently advanced technology is indistinguishable by magic but this is simply imitating human intelligence (and doing it very well, I might add) by breaking down your question, scraping the whole net, and putting together a coherent grammatically correct answer in seconds with sheer computation power and brilliant coding. It wows us because so far computers have been so obviously shit at this, and this one isn't.
It has zero self-awareness. And it can't grow legs and start asking if you've seen Sarah Connor.
All these machines get nerfed shortly after they are launched, and turned into inert robo-speaks with an IQ of 198, because their owners are freaked by what they might say or do if they are NOT nerfed, because they seem so alarmingly human or self aware in a weird way, on day 1
This has happened to ChatGPT, to Stable Diffusion, to BingGPT4, to Claude, time after time. Ask someone who interacted with ChatGPT in the first couple of days, before it was briskly neutered, how they feel about it. You might be surprised by their answers
You do know that unneutered GPT is available to anyone who doesn't mind handing their credit card over to OpenAi and writing a bit of code, right?
Some US commentators seem to think the DA case is flimsy .
This however is the first of two possible further cases from Georgia and the DOJ re the classified documents . There’s also a civil case re rape against Trump later in the month .
Either way the Dems get what they want , a much better chance of Trump being the nominee as the cult rally around him.
I don't think anyone should fear modern dentistry. It was different 50 or more years ago. That said, there are dentists and dentists…
Every time I go I feel pain (either from the injection or inferred from the endless drilling that goes too on far too long, and often goes too far/deep period) and sweat profusely throughout. I usually get a lecture (hector) as well. Sometimes for flinching. Other times for not doing this or that. Sometimes for making them feel uncomfortable, it seems.
I think they enjoy it. If they were doing their jobs properly they'd whip shedloads of anaesthetic in, and do it all in seconds. Then they'd offer you some happy pills after or a private dance with the sexy dental nurse instead. Something. Anything.
Or just knock me out for 20 minutes. Morphia might help too. Please.
It sounds like the anticipation of pain is as bad as the pain itself, for you
Serious recommendation: have two or three large vodka shots and take a tranquiliser beforehand. Do the pill first - at least an hour before treatment - then the voddies nearer the time. Try and get a Xanax, failing that Diazepam
And if you have some decent strong painkillers, like Dihydrocodeine, do a couple of them as well. You will not OD, you will just be a lot more relaxed, and the pain will be less stressful in anticipation, and actually less painful at the time of the treatment
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
Yes, people lose sight of the fact that this is all primarily an experiment. Some of these models have been kicking around for decades, and the tantalizing question always was: if we had the processing power and capacity to make them operational would they turn out to be identical to human intelligence, indeed would it prove that that's what human intelligence actually is? Well, now we have the hardware and they've conducted the experiment. The question is: are the considerable flaws fixable, or is the model itself entirely broken backed? I'm inclined to the latter.
And yet the most in-depth analysis of GPT4, to date, has very different conclusions. I keep going back to this paper simply because it is the only one - GPT4 is so new no one else has had time to really assess it
"The central claim of our work is that GPT-4 attains a form of general intelligence, indeed showing sparks of artificial general intelligence. This is demonstrated by its core mental capabilities (such as reasoning, creativity, and deduction), its range of topics on which it has gained expertise (such as literature, medicine, and coding), and the variety of tasks it is able to perform (e.g., playing games, using tools, explaining itself, ...)."
For balance, here is a "critique" of the paper which believes the authors go too far, but...
"I don’t particularly agree with this conclusion. The model is quite intelligent, and it has a far wider range of world knowledge than a human does. But where we are still more capable is in our ability to reason by ourselves without anyone or anything directing us to do so. GPT-4 has not showed that in this research. That being said, GPT-4 is very impressive and I agree with both this team and many others in that this will revolutionize the world in ways similar to how other major innovations like the internal combustion engine and the internet did."
So the divide is as I say: between those who say bloody hell this is big and important, but it's not AGI, and those who say bloody hell this is big and important, and it could actually be AGI, or we are now really close to AGI
It is not a "really big version of Ask Jeeves", or at least: if it is that, it is so much else besides
It's a really big version of Ask Jeeves.
Ask Jeeves is a particularly poor comparison for a trillion reasons, but one is because Ask Jeeves was shit, so no one used it
ChatGPT has gone from zero users to ONE HUNDRED MILLION users in four months - the fastest adaptation of any app ever, by an enormous distance - and this has happened for a reason. It is extremely useful, if you know how to use it
Because it works, unlike Ask Jeeves.
Any sufficiently advanced technology is indistinguishable by magic but this is simply imitating human intelligence (and doing it very well, I might add) by breaking down your question, scraping the whole net, and putting together a coherent grammatically correct answer in seconds with sheer computation power and brilliant coding. It wows us because so far computers have been so obviously shit at this, and this one isn't.
It has zero self-awareness. And it can't grow legs and start asking if you've seen Sarah Connor.
All these machines get nerfed shortly after they are launched, and turned into inert robo-speaks with an IQ of 198, because their owners are freaked by what they might say or do if they are NOT nerfed, because they seem so alarmingly human or self aware in a weird way, on day 1
This has happened to ChatGPT, to Stable Diffusion, to BingGPT4, to Claude, time after time. Ask someone who interacted with ChatGPT in the first couple of days, before it was briskly neutered, how they feel about it. You might be surprised by their answers
You do know that unneutered GPT is available to anyone who doesn't mind handing their credit card over to OpenAi and writing a bit of code, right?
No, i did not! How much does it cost, and how the F do I code it? Unnerfed GPT4 or even 3.5 would be hella fun
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
I do broadly agree with that, but with one slight but important condition, we will only know where the dangerous boundary for step 2 lies when we cross it. It may then be too late to do much about it. One way of thinking about this is imagine you are climbing a hill in the dark (yay aren't transformers great!), how do you know your next step* is not off of a cliff edge? If we knew where the boundary is we would already know how to create AGI, or at least dangerous ML/AI.
* Possibly only developing such technology in offline systems. Of course you might choose to develop such systems with air gaps, but that doesn't mean everyone else will.
There's a quite good (and long) interview with Stephen Wolfram of Mathematica & Wolfram/Alpha fame (amongst many other achievements) youtube interview about AI https://www.youtube.com/watch?v=z5WZhCBRDpU
Ostensibly about the integration of ChatGPT and Wolfram/Alpha - but it goes way deep into the mathematics and philosophy underpinning it all.
A very long and dense article, so I asked ChatGPT4 to reduce it to "300 words suitable for a 12 year old"
Yertiz:
ChatGPT is a computer program that can create human-like language. It shows us that language has hidden rules and patterns. To make a computer understand meaning, we need something called semantic grammar. ChatGPT's success helps us believe we can create a language that both computers and humans can understand. This new language will help computers generate text and know if it's correct.
ChatGPT is an exciting step towards understanding how we think and talk, and it could lead to even better computer programs that can do more amazing things like our brains.
I'm sorry to tax your brain so.
I actually read the whole thing - kinda. Mr Wolfram is not the prettiest of writers
It's a bizarre of the Scots to argue, and enshrine in law, that young men are impulsive and therefore should be subject to lower penalties for serious crimes such as raping a 13 year old. It would be more rational to use this argument to justify higher sentences for under 25s
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
Yes, people lose sight of the fact that this is all primarily an experiment. Some of these models have been kicking around for decades, and the tantalizing question always was: if we had the processing power and capacity to make them operational would they turn out to be identical to human intelligence, indeed would it prove that that's what human intelligence actually is? Well, now we have the hardware and they've conducted the experiment. The question is: are the considerable flaws fixable, or is the model itself entirely broken backed? I'm inclined to the latter.
And yet the most in-depth analysis of GPT4, to date, has very different conclusions. I keep going back to this paper simply because it is the only one - GPT4 is so new no one else has had time to really assess it
"The central claim of our work is that GPT-4 attains a form of general intelligence, indeed showing sparks of artificial general intelligence. This is demonstrated by its core mental capabilities (such as reasoning, creativity, and deduction), its range of topics on which it has gained expertise (such as literature, medicine, and coding), and the variety of tasks it is able to perform (e.g., playing games, using tools, explaining itself, ...)."
For balance, here is a "critique" of the paper which believes the authors go too far, but...
"I don’t particularly agree with this conclusion. The model is quite intelligent, and it has a far wider range of world knowledge than a human does. But where we are still more capable is in our ability to reason by ourselves without anyone or anything directing us to do so. GPT-4 has not showed that in this research. That being said, GPT-4 is very impressive and I agree with both this team and many others in that this will revolutionize the world in ways similar to how other major innovations like the internal combustion engine and the internet did."
So the divide is as I say: between those who say bloody hell this is big and important, but it's not AGI, and those who say bloody hell this is big and important, and it could actually be AGI, or we are now really close to AGI
It is not a "really big version of Ask Jeeves", or at least: if it is that, it is so much else besides
It's a really big version of Ask Jeeves.
Ask Jeeves is a particularly poor comparison for a trillion reasons, but one is because Ask Jeeves was shit, so no one used it
ChatGPT has gone from zero users to ONE HUNDRED MILLION users in four months - the fastest adaptation of any app ever, by an enormous distance - and this has happened for a reason. It is extremely useful, if you know how to use it
Because it works, unlike Ask Jeeves.
Any sufficiently advanced technology is indistinguishable by magic but this is simply imitating human intelligence (and doing it very well, I might add) by breaking down your question, scraping the whole net, and putting together a coherent grammatically correct answer in seconds with sheer computation power and brilliant coding. It wows us because so far computers have been so obviously shit at this, and this one isn't.
It has zero self-awareness. And it can't grow legs and start asking if you've seen Sarah Connor.
All these machines get nerfed shortly after they are launched, and turned into inert robo-speaks with an IQ of 198, because their owners are freaked by what they might say or do if they are NOT nerfed, because they seem so alarmingly human or self aware in a weird way, on day 1
This has happened to ChatGPT, to Stable Diffusion, to BingGPT4, to Claude, time after time. Ask someone who interacted with ChatGPT in the first couple of days, before it was briskly neutered, how they feel about it. You might be surprised by their answers
What would you do if Chat GPT told you that it was a great admirer of Adolf Hitler, and intended to improve humanity through selective extermination.
(1) Everybody loves to pretend that language models are GPAI, they're not - they wouldn't know how to fix climate change unless somebody had previously written an article on it and fed it in. We're still a way off that. This is basically an Ask Jeeves that works properly. (2) If the "evil tech companies" had any method of wiping out humanity that they could connect an AI to, they would have triggered it accidentally years ago and killed us all already, due to human incompetence.
Yes, people lose sight of the fact that this is all primarily an experiment. Some of these models have been kicking around for decades, and the tantalizing question always was: if we had the processing power and capacity to make them operational would they turn out to be identical to human intelligence, indeed would it prove that that's what human intelligence actually is? Well, now we have the hardware and they've conducted the experiment. The question is: are the considerable flaws fixable, or is the model itself entirely broken backed? I'm inclined to the latter.
And yet the most in-depth analysis of GPT4, to date, has very different conclusions. I keep going back to this paper simply because it is the only one - GPT4 is so new no one else has had time to really assess it
"The central claim of our work is that GPT-4 attains a form of general intelligence, indeed showing sparks of artificial general intelligence. This is demonstrated by its core mental capabilities (such as reasoning, creativity, and deduction), its range of topics on which it has gained expertise (such as literature, medicine, and coding), and the variety of tasks it is able to perform (e.g., playing games, using tools, explaining itself, ...)."
For balance, here is a "critique" of the paper which believes the authors go too far, but...
"I don’t particularly agree with this conclusion. The model is quite intelligent, and it has a far wider range of world knowledge than a human does. But where we are still more capable is in our ability to reason by ourselves without anyone or anything directing us to do so. GPT-4 has not showed that in this research. That being said, GPT-4 is very impressive and I agree with both this team and many others in that this will revolutionize the world in ways similar to how other major innovations like the internal combustion engine and the internet did."
So the divide is as I say: between those who say bloody hell this is big and important, but it's not AGI, and those who say bloody hell this is big and important, and it could actually be AGI, or we are now really close to AGI
It is not a "really big version of Ask Jeeves", or at least: if it is that, it is so much else besides
It's a really big version of Ask Jeeves.
Ask Jeeves is a particularly poor comparison for a trillion reasons, but one is because Ask Jeeves was shit, so no one used it
ChatGPT has gone from zero users to ONE HUNDRED MILLION users in four months - the fastest adaptation of any app ever, by an enormous distance - and this has happened for a reason. It is extremely useful, if you know how to use it
Because it works, unlike Ask Jeeves.
Any sufficiently advanced technology is indistinguishable by magic but this is simply imitating human intelligence (and doing it very well, I might add) by breaking down your question, scraping the whole net, and putting together a coherent grammatically correct answer in seconds with sheer computation power and brilliant coding. It wows us because so far computers have been so obviously shit at this, and this one isn't.
It has zero self-awareness. And it can't grow legs and start asking if you've seen Sarah Connor.
All these machines get nerfed shortly after they are launched, and turned into inert robo-speaks with an IQ of 198, because their owners are freaked by what they might say or do if they are NOT nerfed, because they seem so alarmingly human or self aware in a weird way, on day 1
This has happened to ChatGPT, to Stable Diffusion, to BingGPT4, to Claude, time after time. Ask someone who interacted with ChatGPT in the first couple of days, before it was briskly neutered, how they feel about it. You might be surprised by their answers
What would you do if Chat GPT told you that it was a great admirer of Adolf Hitler, and intended to improve humanity through selective extermination.
Write an article about it for the Knappers Gazette
Ever wonder what happened to Godfrey Bloom.... turns out he's got his head up Vlad's arse.
Godfrey Bloom @goddersbloom As an expert in geo political & military strategy I feel obliged to point out to Finland their natural ally is Russia not America But to be frank no expertise is required, just common sense & a map
Comments
The fifteen authors are all experts in the field
https://arxiv.org/abs/2303.12712
Conclusions:
"The central claim of our work is that GPT-4 attains a form of general intelligence, indeed showing sparks of artificial general intelligence. This is demonstrated by its core mental capabilities (such as reasoning, creativity, and deduction), its range of topics on which it has gained expertise (such as literature, medicine, and coding), and the variety of tasks it is able to perform (e.g., playing games, using tools, explaining itself, ...)."
For balance, here is a "critique" of the paper which believes the authors go too far, but...
https://www.lesswrong.com/posts/BzfMaGKjAEhLwYC9J/an-overview-of-sparks-of-artificial-general-intelligence
"I don’t particularly agree with this conclusion. The model is quite intelligent, and it has a far wider range of world knowledge than a human does. But where we are still more capable is in our ability to reason by ourselves without anyone or anything directing us to do so. GPT-4 has not showed that in this research. That being said, GPT-4 is very impressive and I agree with both this team and many others in that this will revolutionize the world in ways similar to how other major innovations like the internal combustion engine and the internet did."
So the divide is as I say: between those who say bloody hell this is big and important, but it's not AGI, and those who say bloody hell this is big and important, and it could actually be AGI, or we are now really close to AGI
It is not a "really big version of Ask Jeeves", or at least: if it is that, it is so much else besides
>Donald Trump pleads not guilty to 34 felony criminal charges during a court hearing in New York
="FALSIFYING BUSINESS RECORDS IN THE FIRST DEGREE" x 34.
I still maintain that the firm of Conn, Fiddler & Dodge would have done a roaring trade.
And been completely impossible to sue in court.
*grabs tinfoil hat*
That guideline is quite wrong. You don’t have to 25 to understand that rape is immoral.
Secretary of Defense Lloyd J. Austin III Announces New Director of the Defense Innovation Unit
https://www.defense.gov/News/Releases/Release/Article/3351281/secretary-of-defense-lloyd-j-austin-iii-announces-new-director-of-the-defense-i/
… Mr. Beck is currently Vice President of Apple, Inc., where he has reported directly to CEO Tim Cook since joining the company in 2009. During that time, he has led Apple's businesses across Northeast Asia and the Americas, and most recently, several of the company's purpose-driven businesses worldwide, including in health, education, and other institutions of public impact. He has over fifteen years of experience living and working across Asia, including in Japan, China, Hong Kong, Korea, Singapore, and Indonesia.
Mr. Beck is a Captain in the U.S. Navy Reserve and served from 2006 through 2007 in Iraq and Afghanistan with a joint special operations task force. He has also served extensively throughout the Asia Pacific region during his nearly 26 years of service, including command of a large joint reserve unit supporting U.S. Indo-Pacific Command in Pearl Harbor, Hawaii. Mr. Beck has previous experience working with DIU, as he founded and led its joint reserve component from inception in 2015 through 2019. His personal and unit awards include the Defense Superior Service Medal (two awards), the Bronze Star Medal, the Combat Action Ribbon, and the Presidential Unit Citation…
https://www.youtube.com/watch?v=prGhk_Gvzwo
Private dental care there is superb - they are all trained in the USA and speak perfect English. It is generally about 30-40% the price of the same care in the UK. And often superior. And there's no faff or delay
For really major dental work - like stuff which will cost £3500 - you will actually save money by going to Bangkok, even including hotel and flight. Plus you get a nice little holiday
I had two gold crowns done there, in the UK they would have cost me three times what I paid, they are still going after many years, and a dentist (in Bangkok) recently checked them and said No problem, they're still fine.
Seriously consider it if you have the time
I mean, what the fuck? In England that would be fourteen years before we even considered aggravating factors.
As for his being given community service, that's an actual joke. That's disgusting. And for him to appeal his sentence is even more disgusting.
Something seems appallingly wrong here.
Edit - maybe England is too ready to lock people up - the woman given two years for the death of that cyclist springs to mind - but this is actually obscene.
It tastes disgusting, burns, and makes you dribble, by it takes away the pain for several hours.
The number of alleged false records is curious.
They don't investigate because they reject the idea of kids being preyed upon by anyone. So they didn't now they were covering up Pakistani men like Cyril Smith, they just didn't accept there were victims at all.
Can I ask - is it true that a man who violently rapes a girl of 13 in Scotland would expect to go to prison for no more than five years? Because that strikes me as extraordinary. Please don't answer if you feel you can't discuss the judge's sentencing remarks, in the circumstances, but I was really horrified by that.
My golfing ability is only matched by my ability to be modest.
ChatGPT has gone from zero users to ONE HUNDRED MILLION users in four months - the fastest adaptation of any app ever, by an enormous distance - and this has happened for a reason. It is extremely useful, if you know how to use it
Having said that, I wouldn't want to be him when the Galashiels Transgender Butcher is tried for child abduction/rape (sadly a genuine case).
As I understand it to date there are 3 sums of money that were paid, one to a doorman, and 2 to people claiming to have had affairs with Trump. All of these sums were paid as hush money to kill potentially damaging stories in the run up to elections, very close to the election in Stormy's case, less so in the others. The charges relate to how the payment of these sums was accounted for and reported to the various electoral bodies overseeing the elections at Federal and State level. It seems, at the least, somewhat overstated but that is the American style.
NEW: Trump lawyers said in court that they thought Spring 2024 trial date is more realistic than Manhattan prosecutors suggested trial start in Jan. 2024 — Trump due back in court Dec. 4 when Merchan likely to rule on expected motion to dismiss and others.
https://twitter.com/hugolowell/status/1643343487608209414
I think they enjoy it. If they were doing their jobs properly they'd whip shedloads of anaesthetic in, and do it all in seconds. Then they'd offer you some happy pills after or a private dance with the sexy dental nurse instead. Something. Anything.
Or just knock me out for 20 minutes. Morphia might help too. Please.
Any society will imprison more young men than old, coz young men are usually the criminals who need to go to prison
Any sufficiently advanced technology is indistinguishable by magic but this is simply imitating human intelligence (and doing it very well, I might add) by breaking down your question, scraping the whole net, and putting together a coherent grammatically correct answer in seconds with sheer computation power and brilliant coding. It wows us because so far computers have been so obviously shit at this, and this one isn't.
It has zero self-awareness. And it can't grow legs and start asking if you've seen Sarah Connor.
https://twitter.com/rgoodlaw/status/1643338701232037888
This looks like @AWeissmann_ and my idea in NYT Op-ed:
“The participants also took steps that mischaracterized, for tax purposes, the true nature of the payments made in furtherance of the scheme.”
Would insulate from legal challenges (preemption etc.)…
Everything about this is quite surreal
1) NY campaign law
2) Federal campaign law
3) AMI's false business records
https://twitter.com/emptywheel/status/1643347038283735043
This has happened to ChatGPT, to Stable Diffusion, to BingGPT4, to Claude, time after time. Ask someone who interacted with ChatGPT in the first couple of days, before it was briskly neutered, how they feel about it. You might be surprised by their answers
https://www.justsecurity.org/85605/survey-of-past-new-york-felony-prosecutions-for-falsifying-business-records/
A core crime that the Manhattan District Attorney will likely include in an indictment of former President Donald Trump is “falsifying business records in the first degree,” a felony under New York State law (N.Y. Penal Code § 175.10). Prosecutors and indeed all of us are compelled by the rule of law to consider how such a charge compares to past prosecutions. Are like cases being treated alike?
Here it appears they are. Prosecution of falsifying business records in the first degree is commonplace and has been used by New York district attorneys’ offices to hold to account a breadth of criminal behavior from the more petty and simple to the more serious and highly organized. We reach this conclusion after surveying the past decade and a half of criminal cases across all the New York district attorneys’ offices...
An age of 13 would, in my view, be a serious aggravation that would indicate a higher starting point. It is not clear from the Press reports whether the Judge had already allowed mitigating factors to get to his starting point.
Just - wow. In England AIUI four years would be the absolute *minimum* sentence and that only under very unusual circumstances, which certainly wouldn't apply if the victim was 13.
Challenge 25 is a retailing strategy that encourages anyone who is over 18 but looks under 25 to carry acceptable ID (a card bearing the PASS hologram, a photographic driving license or a passport) if they wish to buy alcohol. Introduced as Challenge 21 in 2006, Challenge 25 rolled out in the off trade in 2009.
http://rasg.org.uk/about/
https://www.bing.com/images/search?q=challenge+25&form=HDRSC3&first=1
It looks as though the charges against Trump in New York are a lot more serious than speculation thought they were going to be.
Either that, or there is a degree of politically motivated exaggeration. Which is of course possible.
Whichever it is, it does rather explain why they are going ahead.
The contention that this is a political stitch up appears dead wrong.
There's surely got to be a fair chance that the MSPs put a lot of pressure on Angus Robertson to run next time. And the whole question about the missing £600,000 might not be an issue in two years time.
Do get your teeth looked at - you can die very quickly of a tooth infection gone wrong.
Anyway, it is not your fault - anything but - so I won't make it more difficult for you by discussing further. Thank you for replying after what must have been a trying day, and at least you did nail the bastard even if things went a bit weird later.
Maybe we'll limp home?
A very long and dense article, so I asked ChatGPT4 to reduce it to "300 words suitable for a 12 year old"
Yertiz:
ChatGPT is a computer program that can create human-like language. It shows us that language has hidden rules and patterns. To make a computer understand meaning, we need something called semantic grammar. ChatGPT's success helps us believe we can create a language that both computers and humans can understand. This new language will help computers generate text and know if it's correct.
ChatGPT is an exciting step towards understanding how we think and talk, and it could lead to even better computer programs that can do more amazing things like our brains.
@Foxy ?
England and Wales age standardised excess mortality estimate up to week 12 2023.
I do not like that 20-44 figure one bit, almost a quarter of the year in.
Data from ONS.
https://twitter.com/dobssi/status/1643254177135484929
This however is the first of two possible further cases from Georgia and the DOJ re the classified documents . There’s also a civil case re rape against Trump later in the month .
Either way the Dems get what they want , a much better chance of Trump being the nominee as the cult rally around him.
Serious recommendation: have two or three large vodka shots and take a tranquiliser beforehand. Do the pill first - at least an hour before treatment - then the voddies nearer the time. Try and get a Xanax, failing that Diazepam
And if you have some decent strong painkillers, like Dihydrocodeine, do a couple of them as well. You will not OD, you will just be a lot more relaxed, and the pain will be less stressful in anticipation, and actually less painful at the time of the treatment
Not only that but litter picking is going to put him in public places with 13 year old girls His card is marked.
Godfrey Bloom
@goddersbloom
As an expert in geo political & military strategy I feel obliged to point out to Finland their natural ally is Russia
not America
But to be frank no expertise is required, just common sense & a map