If Harris won Nevada, as Ralston predicts, then she can afford to lose Penn as long as she gains either Georgia or N Carolina.
Bear in mind that Biden won 306 ECVs and Trump still dragged the process out into January, with violence on the streets. Harris does only need 270 ECVs to win but if she does land in the 270s, it'll get very ugly before it gets better - assuming that Trump doesn't manage to persuade Congress and/or the courts to fiddle the result for him (which is unlikely but certainly not impossible).
Unless it is 268 each Congress has no power to intervene directly and in 2020 remember even most GOP Senators voted that Biden had won even if most GOP House Reps voted to challenge state results and say Trump had won.
The courts cannot intervene either unless to stop the count but Trump would have to be ahead in that states count anyway at that point
He was the Minister of State for Crime, Policing and Fire before the election - so someone who knows the brief.
and I think once you go beyond the 6 candidates who stood for the leadership, Rishi and Hunt everyone else is going to be in the "who are they" department. The Tories have 120 MPs and many of the ones people know the name of lost their seats..
The big advantage this shadow cabinet has is that it mostly passes the "we told you to sling your hooks" test. Even someone like Mel Stride is a relatively fresh face to the electorate.
The bad news is that it does look awfully light on experience. How many of the bigger names turned offers down, if any? (Tom T isn't on the list, for example). It might well be the best shad cab Badenoch can assemble from the people available, but it doesn't exactly inspire. And if there are some big beasts to return in by-elections (Mordaunt, say), where do they slot in?
He was the Minister of State for Crime, Policing and Fire before the election - so someone who knows the brief.
and I think once you go beyond the 6 candidates who stood for the leadership, Rishi and Hunt everyone else is going to be in the "who are they" department. The Tories have 120 MPs and many of the ones people know the name of lost their seats..
The big advantage this shadow cabinet has is that it mostly passes the "we told you to sling your hooks" test. Even someone like Mel Stride is a relatively fresh face to the electorate.
The bad news is that it does look awfully light on experience. How many of the bigger names turned offers down, if any? (Tom T isn't on the list, for example). It might well be the best shad cab Badenoch can assemble from the people available, but it doesn't exactly inspire. And if there are some big beasts to return in by-elections (Mordaunt, say), where do they slot in?
The next months/couple of years presents an amazing opportunity to a new Thatcher/Disraeli to come from nowhere to revive conservatism if it isn't to be Badenoch; almost no household names are still around.
It may take longer, and the person is even now polishing their interviewing skills for PPE at Brasenose/Balliol.
It is more likely the next Conservative Cabinet Ministers will be in a coalition government with Farage and Reform than leading a Tory majority government unless they can find a leader who can really squeeze Reform back in their box. Farage is making inroads and gains with white working class Leavers who voted for Boris and Labour in July but Badenoch is at most treading water and while keeping 2024 Tory voters not gaining either Reform voters or LD or Labour voters.
Having said that a Labour minority government reliant on LD confidence and supply is also more likely at the moment than another Labour majority if Labour keep plumbing the depths of unpopularity
HYUFD: You're far too good a predictor to fall for all that guff. In 2028, Labour will win with a majority easily big enough to contest the (?) 2032 election against a proper Opposition. And with that strengthened Opposition in that (?) 2032 election, the LDs might well get rid of Labour.
Even if it means a hung Parliament with the Greens
I have just noticed that Google Satellite images are not pure images. They are mucked about with.
Our village has had a number of new developments. To build one of them several houses were demolished for access to the plot and a road built through. On the image the development is there but so are the houses that shouldn't be and the road into the estate just fades away as it enters a non existent garden and house that are on the image.
A revealing contrast between the way the £285 increase in university tuition fees is being accepted to the tantrums the oldies have had over losing their £200 WFA.
It certainly isn't being accepted if the twitter posts I have seen are anything to go by, certainly plenty of students crying betrayal and going Green
Though the government also announced higher maintenance loans to ease the blow
What's the view of HYUFD - whom I rate highly - on the proportion of young people who'll actually vote, given that the next (May) election is in term time?
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
It got it right on that occasion. It gets a lot wrong, frequently. Which is fine as long as you are prepared to use you own judgement as a check and balance. The worrying trend I’m seeing is otherwise intelligent, professional, sceptical folk treating its output as gospel. This is a thing.
If I’m allowed to discuss this without being banned, I imagine you are using ChatGPT or Gemini. The first often gets this wrong the second very often gets this wrong
I’m using Anthropic Claude 3.6. By far the best model and the one closest to true AGI. It is absolutely outstanding - and breathtaking in its capabilities. Its amazing it is not better known
The other day I had a debate with it which changed my world view
But I can’t say any more because I’m not allowed. Ah well. Back to American politics! Also a shower after my gym visit and then a nice gin and tonic
None of them are reliable - for the simple reason thar their source material is the corpus of the internet which is itself strewn with errors, satire and fake news. That is not to say that they cannot be a fun and useful tool. But one would be foolish to trust them without verification.
You can't talk to Leon about this - it's beyond his ken.
PB is literally unaware of the existence of the best models. This is laughable
tbf you just gave the false impression that ChatGPT said there were three rs in raspberry, only admitting it was a different 'AI' when called out on it.
and I'm old enough to remember when you were swearing that ChatGPT was on the brink of the singularity, and the imminent update would demonstrate an AGI a million times superior to human intelligence.
A couple of year later and it can't even tell me how many rs are in the word raspberry. So you can understand why people aren't going to take your word for it.
No wonder they don’t have any fucking kids. Who wants to bring a child into a world where, every time they have a beer they HAVE to order some weird pickled chicken and as they reluctantly eat it someone nearby will sneeze so loud they cough it up in alarm
That’s no life for anyone. Better to go childless into the darkness
He was the Minister of State for Crime, Policing and Fire before the election - so someone who knows the brief.
and I think once you go beyond the 6 candidates who stood for the leadership, Rishi and Hunt everyone else is going to be in the "who are they" department. The Tories have 120 MPs and many of the ones people know the name of lost their seats..
The big advantage this shadow cabinet has is that it mostly passes the "we told you to sling your hooks" test. Even someone like Mel Stride is a relatively fresh face to the electorate.
The bad news is that it does look awfully light on experience. How many of the bigger names turned offers down, if any? (Tom T isn't on the list, for example). It might well be the best shad cab Badenoch can assemble from the people available, but it doesn't exactly inspire. And if there are some big beasts to return in by-elections (Mordaunt, say), where do they slot in?
He was the Minister of State for Crime, Policing and Fire before the election - so someone who knows the brief.
and I think once you go beyond the 6 candidates who stood for the leadership, Rishi and Hunt everyone else is going to be in the "who are they" department. The Tories have 120 MPs and many of the ones people know the name of lost their seats..
The big advantage this shadow cabinet has is that it mostly passes the "we told you to sling your hooks" test. Even someone like Mel Stride is a relatively fresh face to the electorate.
The bad news is that it does look awfully light on experience. How many of the bigger names turned offers down, if any? (Tom T isn't on the list, for example). It might well be the best shad cab Badenoch can assemble from the people available, but it doesn't exactly inspire. And if there are some big beasts to return in by-elections (Mordaunt, say), where do they slot in?
The next months/couple of years presents an amazing opportunity to a new Thatcher/Disraeli to come from nowhere to revive conservatism if it isn't to be Badenoch; almost no household names are still around.
It may take longer, and the person is even now polishing their interviewing skills for PPE at Brasenose/Balliol.
It is more likely the next Conservative Cabinet Ministers will be in a coalition government with Farage and Reform than leading a Tory majority government unless they can find a leader who can really squeeze Reform back in their box. Farage is making inroads and gains with white working class Leavers who voted for Boris and Labour in July but Badenoch is at most treading water and while keeping 2024 Tory voters not gaining either Reform voters or LD or Labour voters.
Having said that a Labour minority government reliant on LD confidence and supply is also more likely at the moment than another Labour majority if Labour keep plumbing the depths of unpopularity
To extend the plumbing analogy, there's a LOT of water to flow past Westminster before the next election.
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
It got it right on that occasion. It gets a lot wrong, frequently. Which is fine as long as you are prepared to use you own judgement as a check and balance. The worrying trend I’m seeing is otherwise intelligent, professional, sceptical folk treating its output as gospel. This is a thing.
It’s a kind of mental laziness. “It does all this typing for me - why should I be responsible for making sure it is right?”
The junior we are having an issue with at work seems to have trouble with the idea that he is responsible for the code he commits - even if typed by an “AI”.
“But it’s quicker!!”
Yes, exactly that. It really is worrying that people think that way – but they do. And it's not just the office juniors...
I’ve seen nothing in the stuff overnight that has changed my mind that this will be a Harris win, with Trump taking NV and AZ and Harris the rest.
Trump should take Nevada, it has the highest unemployment rate in the US and probably Georgia which has above average inflation.
Harris will be helped in the blue wall though by the fact inflation in Michigan, Wisconsin and Pennsylvania is below the US average and unemployment in Pennsylvania and Wisconsin is also below the US average too
Interesting post. I infer from that that you are forecasting a narrow Harris win? I rate you a lot as a tipster so would be keen to know.
Yes, I bet last night on Harris to win the EC but Trump to win the popular vote.
As I posted earlier the bluewall states doing better economically than the US average may save her but Trump for the same reason has a real chance of winning the popular vote with cost of living an issue in most of the US
Okay, cheers, thanks for the reply. And the very best of luck with your bet.
I have just noticed that Google Satellite images are not pure images. They are mucked about with.
Our village has had a number of new developments. To build one of them several houses were demolished for access to the plot and a road built through. On the image the development is there but so are the houses that shouldn't be and the road into the estate just fades away as it enters a non existent garden and house that are on the image.
The classic one is half-trains; trains which end suddenly at the border between tiles. There's one where there are two half-trains going in different directions.
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
It got it right on that occasion. It gets a lot wrong, frequently. Which is fine as long as you are prepared to use you own judgement as a check and balance. The worrying trend I’m seeing is otherwise intelligent, professional, sceptical folk treating its output as gospel. This is a thing.
If I’m allowed to discuss this without being banned, I imagine you are using ChatGPT or Gemini. The first often gets this wrong the second very often gets this wrong
I’m using Anthropic Claude 3.6. By far the best model and the one closest to true AGI. It is absolutely outstanding - and breathtaking in its capabilities. Its amazing it is not better known
The other day I had a debate with it which changed my world view
But I can’t say any more because I’m not allowed. Ah well. Back to American politics! Also a shower after my gym visit and then a nice gin and tonic
None of them are reliable - for the simple reason thar their source material is the corpus of the internet which is itself strewn with errors, satire and fake news. That is not to say that they cannot be a fun and useful tool. But one would be foolish to trust them without verification.
You can't talk to Leon about this - it's beyond his ken.
To be fair its a bit unfair 'arguing' the topic of AI with someone who is banned from responding.
He's been doing plenty of 'responding' on this thread.
But at least it's about AI, and not his view on white supremacism...
He was the Minister of State for Crime, Policing and Fire before the election - so someone who knows the brief.
and I think once you go beyond the 6 candidates who stood for the leadership, Rishi and Hunt everyone else is going to be in the "who are they" department. The Tories have 120 MPs and many of the ones people know the name of lost their seats..
The big advantage this shadow cabinet has is that it mostly passes the "we told you to sling your hooks" test. Even someone like Mel Stride is a relatively fresh face to the electorate.
The bad news is that it does look awfully light on experience. How many of the bigger names turned offers down, if any? (Tom T isn't on the list, for example). It might well be the best shad cab Badenoch can assemble from the people available, but it doesn't exactly inspire. And if there are some big beasts to return in by-elections (Mordaunt, say), where do they slot in?
He was the Minister of State for Crime, Policing and Fire before the election - so someone who knows the brief.
and I think once you go beyond the 6 candidates who stood for the leadership, Rishi and Hunt everyone else is going to be in the "who are they" department. The Tories have 120 MPs and many of the ones people know the name of lost their seats..
The big advantage this shadow cabinet has is that it mostly passes the "we told you to sling your hooks" test. Even someone like Mel Stride is a relatively fresh face to the electorate.
The bad news is that it does look awfully light on experience. How many of the bigger names turned offers down, if any? (Tom T isn't on the list, for example). It might well be the best shad cab Badenoch can assemble from the people available, but it doesn't exactly inspire. And if there are some big beasts to return in by-elections (Mordaunt, say), where do they slot in?
The next months/couple of years presents an amazing opportunity to a new Thatcher/Disraeli to come from nowhere to revive conservatism if it isn't to be Badenoch; almost no household names are still around.
It may take longer, and the person is even now polishing their interviewing skills for PPE at Brasenose/Balliol.
It is more likely the next Conservative Cabinet Ministers will be in a coalition government with Farage and Reform than leading a Tory majority government unless they can find a leader who can really squeeze Reform back in their box. Farage is making inroads and gains with white working class Leavers who voted for Boris and Labour in July but Badenoch is at most treading water and while keeping 2024 Tory voters not gaining either Reform voters or LD or Labour voters.
Having said that a Labour minority government reliant on LD confidence and supply is also more likely at the moment than another Labour majority if Labour keep plumbing the depths of unpopularity
HYUFD: You're far too good a predictor to fall for all that guff. In 2028, Labour will win with a majority easily big enough to contest the (?) 2032 election against a proper Opposition. And with that strengthened Opposition in that (?) 2032 election, the LDs might well get rid of Labour.
Even if it means a hung Parliament with the Greens
On at least half current polls Labour will already lose its majority
I have just noticed that Google Satellite images are not pure images. They are mucked about with.
Our village has had a number of new developments. To build one of them several houses were demolished for access to the plot and a road built through. On the image the development is there but so are the houses that shouldn't be and the road into the estate just fades away as it enters a non existent garden and house that are on the image.
Various Chinese military installations are cropped ..
Ukraine not too happy either.
https://x.com/politicoeurope/status/1853392764601327685 Google revealed the location of Ukrainian military sites in its latest Maps update, with the tech giant scrambling to remove the pictures, a Kyiv official said yesterday.
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
It got it right on that occasion. It gets a lot wrong, frequently. Which is fine as long as you are prepared to use you own judgement as a check and balance. The worrying trend I’m seeing is otherwise intelligent, professional, sceptical folk treating its output as gospel. This is a thing.
If I’m allowed to discuss this without being banned, I imagine you are using ChatGPT or Gemini. The first often gets this wrong the second very often gets this wrong
I’m using Anthropic Claude 3.6. By far the best model and the one closest to true AGI. It is absolutely outstanding - and breathtaking in its capabilities. Its amazing it is not better known
The other day I had a debate with it which changed my world view
But I can’t say any more because I’m not allowed. Ah well. Back to American politics! Also a shower after my gym visit and then a nice gin and tonic
None of them are reliable - for the simple reason thar their source material is the corpus of the internet which is itself strewn with errors, satire and fake news. That is not to say that they cannot be a fun and useful tool. But one would be foolish to trust them without verification.
You can't talk to Leon about this - it's beyond his ken.
PB is literally unaware of the existence of the best models. This is laughable
tbf you just gave the false impression that ChatGPT said there were three rs in raspberry, only admitting it was a different 'AI' when called out on it.
and I'm old enough to remember when you were swearing that ChatGPT was on the brink of the singularity, and the imminent update would demonstrate an AGI a million times superior to human intelligence.
A couple of year later and it can't even tell me how many rs are in the word raspberry. So you can understand why people aren't going to take your word for it.
*****
I’m not allowed to talk about it so you’ll just have to take it on trust that you’re an idiot
I imagine it won’t require great feats of imagination
Also: sort out your blockquoting, you kartoffel eating halfwit
I have just noticed that Google Satellite images are not pure images. They are mucked about with.
Our village has had a number of new developments. To build one of them several houses were demolished for access to the plot and a road built through. On the image the development is there but so are the houses that shouldn't be and the road into the estate just fades away as it enters a non existent garden and house that are on the image.
The classic one is half-trains; trains which end suddenly at the border between tiles. There's one where there are two half-trains going in different directions.
At one point Yorkshire Wildlife Park had half a lion enclosure.
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
It got it right on that occasion. It gets a lot wrong, frequently. Which is fine as long as you are prepared to use you own judgement as a check and balance. The worrying trend I’m seeing is otherwise intelligent, professional, sceptical folk treating its output as gospel. This is a thing.
If I’m allowed to discuss this without being banned, I imagine you are using ChatGPT or Gemini. The first often gets this wrong the second very often gets this wrong
I’m using Anthropic Claude 3.6. By far the best model and the one closest to true AGI. It is absolutely outstanding - and breathtaking in its capabilities. Its amazing it is not better known
The other day I had a debate with it which changed my world view
But I can’t say any more because I’m not allowed. Ah well. Back to American politics! Also a shower after my gym visit and then a nice gin and tonic
None of them are reliable - for the simple reason thar their source material is the corpus of the internet which is itself strewn with errors, satire and fake news. That is not to say that they cannot be a fun and useful tool. But one would be foolish to trust them without verification.
You can't talk to Leon about this - it's beyond his ken.
PB is literally unaware of the existence of the best models. This is laughable
tbf you just gave the false impression that ChatGPT said there were three rs in raspberry, only admitting it was a different 'AI' when called out on it.
and I'm old enough to remember when you were swearing that ChatGPT was on the brink of the singularity, and the imminent update would demonstrate an AGI a million times superior to human intelligence.
A couple of year later and it can't even tell me how many rs are in the word raspberry. So you can understand why people aren't going to take your word for it.
It's just in the nature of LLMs that they aren't very good at straight facts. They have no concept of true or false, so it doesn't make sense to use them for such tasks. However, they are incredibly useful for the sort of jobs that require some degree of creativity based on the corpus of human knowledge encompassed by the internet. Suggesting recipes, summarising texts, suggesting coding approaches; that sort of thing.
Edit: The quotes are shot to hell in this. The last paragraph is mine.
Will anything significant be known about results before midnight GMT today?
First State results are likely to be from Florida a couple of hours after that.
It’s usually better to make it an early start rather than a stay up late - unless you’re doing loads of in-play betting of course, in which case you get up at midnight having slept a few hours already!
IIRC four years ago most of the stay-ups went to bed sure Trump had won.
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
That doesn't surprise me. I asked the question because someone on Instagram had noticed it couldn't count the R's in strawberry. It got strawberry right when I asked it - suggesting someone had since fixed it on R's in strawberry - so I asked it about raspberry, where it was still wrong.
So someone has manually fixed it on raspberry now too, or possibly on recognising that sort of question in general.
Think it depends on the state tbh - some eg IN/KY start closing at 6pm local, some eg I think NY & RI close at 8pm local.
But in general the poll closing times are early compared to other weekday elections eg UK, Israel, Netherlands, Ireland, Norway, Denmark, Canada (I think) but not Indonesia which closes at 1pm
Will anything significant be known about results before midnight GMT today?
Probably not - Florida will have only just closed their queues then with voting continuing until everyone in the queue has voted.
I think @rcs1000 said there was likely to be useful information at about 1am from Georgia(?) but I think it will be 3/4am before you get anything concrete.
I suspect I will get up at about 6 (when Mrs Eek gets up) to see what's going on..
Will anything significant be known about results before midnight GMT today?
First State results are likely to be from Florida a couple of hours after that.
It’s usually better to make it an early start rather than a stay up late - unless you’re doing loads of in-play betting of course, in which case you get up at midnight having slept a few hours already!
So if I get up at 4am, say, that might work?
Is there a livestream on YouTube or something? I would prefer US coverage if poss. I don't get Sky.
What if the universe has run out of entropy or something and all the things that are supposed to be one random thing or another are just coming out at 50/50.
Who on God’s green Earth is Chris Phillip? I am a political nerd and I’ve never heard of him!
It's Chris Philp.
P H I L P
Get it right! ;-) He's a South Londoner, of course you don't know him. And a Physics graduate.
In fairness, when HYUFD first mentioned him I thought he'd made a(n uncharacteristic) spelling error. Chris Philp must have gone through his whole life having to spell his name out and correct people. Having an unusual name or address is a pain in the arse; having an unusual name (or address) which looks like a misspelling of a common one must be doubly so.
Will anything significant be known about results before midnight GMT today?
Probably not - Florida will have only just closed their queues then with voting continuing until everyone in the queue has voted.
I think @rcs1000 said there was likely to be useful information at about 1am from Georgia(?) but I think it will be 3/4am before you get anything concrete.
I suspect I will get up at about 6 (when Mrs Eek gets up) to see what's going on..
My analysis of the election has changed. If I had a vote, it would be for Trump. The defining issue for me is Ukraine. I don't think that the democrats can solve this and I think they will end up gradually losing the war in Ukraine, as the conflict continues on its current trajectory with no apparent strategic direction. This could cause the rapid collapse of more countries, and significant consequential damage. My gut feeling is that Trump would be more likely to find a strategic solution to the issue of conflict with Russia. This puts me at odds with almost everyone, including everyone I know in Finland, but it is my assessment of the situation. There is too much familiarity bias and continuity bias which fuels the assumption that Harris would preserve NATO and the European security arrangement. With Trump there is the possibility of an updated solution to the security question, there is a risk that this fails, but the current direction of travel seems to lead to failure anyway. I have this view even accepting that Trump is a significant threat to democracy itself. But I was influenced a lot by Niall Ferguson's recent comments.
No there isn’t - Trump will simply cut off the supply of arms to the Ukraine and Russia will take over. Then wait a few years to rebuild their supplies before moving on to the next part of “mother Russia” that was stolen from Moscow
It's one of those takes that assumes he must intend something than the very obvious interpretation of what he says, hinging everything on that 'possibility'.
Or we could look at what he actually says and believe him, and figure out what actions likely follow that.
Quite an interesting comment from Farage that Harris should pardon Trump, if she wins, so as to create national unity.
Suggests the Trump team have given some thought to what happens if he loses. The deal would he be, he goes quietly and tells his supporters not to cause any trouble in exchange for a pardon (presumably from Biden so Harris isn't sullied by it)
The interesting part there is Republicans
If he loses, the non-MAGA branch have a chance to bury him and try to reclaim their party
What if the universe has run out of entropy or something and all the things that are supposed to be one random thing or another are just coming out at 50/50.
Which universe?
I would say just pick one at random but that may no longer be possible
Who on God’s green Earth is Chris Phillip? I am a political nerd and I’ve never heard of him!
You will have. He was always the go to gob on a stick to defend the indefensible during the last year of the last Government. Not very bright is an understatement. Some of her other appointees aren't too bad mind.
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
It got it right on that occasion. It gets a lot wrong, frequently. Which is fine as long as you are prepared to use you own judgement as a check and balance. The worrying trend I’m seeing is otherwise intelligent, professional, sceptical folk treating its output as gospel. This is a thing.
If I’m allowed to discuss this without being banned, I imagine you are using ChatGPT or Gemini. The first often gets this wrong the second very often gets this wrong
I’m using Anthropic Claude 3.6. By far the best model and the one closest to true AGI. It is absolutely outstanding - and breathtaking in its capabilities. Its amazing it is not better known
The other day I had a debate with it which changed my world view
But I can’t say any more because I’m not allowed. Ah well. Back to American politics! Also a shower after my gym visit and then a nice gin and tonic
None of them are reliable - for the simple reason thar their source material is the corpus of the internet which is itself strewn with errors, satire and fake news. That is not to say that they cannot be a fun and useful tool. But one would be foolish to trust them without verification.
You can't talk to Leon about this - it's beyond his ken.
PB is literally unaware of the existence of the best models. This is laughable
I might be a little out of date but Claudia Schiffer for me.
I have just noticed that Google Satellite images are not pure images. They are mucked about with.
Our village has had a number of new developments. To build one of them several houses were demolished for access to the plot and a road built through. On the image the development is there but so are the houses that shouldn't be and the road into the estate just fades away as it enters a non existent garden and house that are on the image.
Various Chinese military installations are cropped ..
Ukraine not too happy either.
https://x.com/politicoeurope/status/1853392764601327685 Google revealed the location of Ukrainian military sites in its latest Maps update, with the tech giant scrambling to remove the pictures, a Kyiv official said yesterday.
https://darksitefinder.com/map/ is better if you want to find military installations (Satellite, opacity of dark sky overlay to 0%)
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
That doesn't surprise me. I asked the question because someone on Instagram had noticed it couldn't count the R's in strawberry. It got strawberry right when I asked it - suggesting someone had since fixed it on R's in strawberry - so I asked it about raspberry, where it was still wrong.
So someone has manually fixed it on raspberry now too, or possibly on recognising that sort of question in general.
Think it depends on the state tbh - some eg IN/KY start closing at 6pm local, some eg I think NY & RI close at 8pm local.
But in general the poll closing times are early compared to other weekday elections eg UK, Israel, Netherlands, Ireland, Norway, Denmark, Canada (I think) but not Indonesia which closes at 1pm
I met up with a former colleague of mine from Vodafone last week.
They are very excited about the merger, thinking it will deliver benefits to not only Vodafone and Three but O2 too, so it will impact the majority of mobile customers. It is a true re-balancing of the market.
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
It got it right on that occasion. It gets a lot wrong, frequently. Which is fine as long as you are prepared to use you own judgement as a check and balance. The worrying trend I’m seeing is otherwise intelligent, professional, sceptical folk treating its output as gospel. This is a thing.
If I’m allowed to discuss this without being banned, I imagine you are using ChatGPT or Gemini. The first often gets this wrong the second very often gets this wrong
I’m using Anthropic Claude 3.6. By far the best model and the one closest to true AGI. It is absolutely outstanding - and breathtaking in its capabilities. Its amazing it is not better known
The other day I had a debate with it which changed my world view
But I can’t say any more because I’m not allowed. Ah well. Back to American politics! Also a shower after my gym visit and then a nice gin and tonic
None of them are reliable - for the simple reason thar their source material is the corpus of the internet which is itself strewn with errors, satire and fake news. That is not to say that they cannot be a fun and useful tool. But one would be foolish to trust them without verification.
You can't talk to Leon about this - it's beyond his ken.
PB is literally unaware of the existence of the best models. This is laughable
The travel hack speaks. I'd take your word over travel as you seem to have been almost everywhere. I don't believe you are an expert on AI any more than you are on viral evolution. Happily there are people on PB who are experts on computing, coding and relevant fields.
Who on God’s green Earth is Chris Phillip? I am a political nerd and I’ve never heard of him!
It's Chris Philp.
P H I L P
Get it right! ;-) He's a South Londoner, of course you don't know him. And a Physics graduate.
In fairness, when HYUFD first mentioned him I thought he'd made a(n uncharacteristic) spelling error. Chris Philp must have gone through his whole life having to spell his name out and correct people. Having an unusual name or address is a pain in the arse; having an unusual name (or address) which looks like a misspelling of a common one must be doubly so.
I've often thought that, when he's been on the TV. At least her's got a first name which isn't (often anyway) a surname.
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
It got it right on that occasion. It gets a lot wrong, frequently. Which is fine as long as you are prepared to use you own judgement as a check and balance. The worrying trend I’m seeing is otherwise intelligent, professional, sceptical folk treating its output as gospel. This is a thing.
If I’m allowed to discuss this without being banned, I imagine you are using ChatGPT or Gemini. The first often gets this wrong the second very often gets this wrong
I’m using Anthropic Claude 3.6. By far the best model and the one closest to true AGI. It is absolutely outstanding - and breathtaking in its capabilities. Its amazing it is not better known
The other day I had a debate with it which changed my world view
But I can’t say any more because I’m not allowed. Ah well. Back to American politics! Also a shower after my gym visit and then a nice gin and tonic
None of them are reliable - for the simple reason thar their source material is the corpus of the internet which is itself strewn with errors, satire and fake news. That is not to say that they cannot be a fun and useful tool. But one would be foolish to trust them without verification.
You can't talk to Leon about this - it's beyond his ken.
PB is literally unaware of the existence of the best models. This is laughable
I might be a little out of date but Claudia Schiffer for me.
I see that Margoe Robbie has had a child. Several PBers have opined that they would be 'up' for it - so has anyone recently become a father?
Will anything significant be known about results before midnight GMT today?
First State results are likely to be from Florida a couple of hours after that.
It’s usually better to make it an early start rather than a stay up late - unless you’re doing loads of in-play betting of course, in which case you get up at midnight having slept a few hours already!
So if I get up at 4am, say, that might work?
Is there a livestream on YouTube or something? I would prefer US coverage if poss. I don't get Sky.
There should be live streams from all the major networks, and plenty of independents.
My plan at the moment is to switch between CNN and The Daily Wire on the computer, depending on who’s happiest at any given moment, with the AP results page and PB on the ipad.
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
It got it right on that occasion. It gets a lot wrong, frequently. Which is fine as long as you are prepared to use you own judgement as a check and balance. The worrying trend I’m seeing is otherwise intelligent, professional, sceptical folk treating its output as gospel. This is a thing.
If I’m allowed to discuss this without being banned, I imagine you are using ChatGPT or Gemini. The first often gets this wrong the second very often gets this wrong
I’m using Anthropic Claude 3.6. By far the best model and the one closest to true AGI. It is absolutely outstanding - and breathtaking in its capabilities. Its amazing it is not better known
The other day I had a debate with it which changed my world view
But I can’t say any more because I’m not allowed. Ah well. Back to American politics! Also a shower after my gym visit and then a nice gin and tonic
None of them are reliable - for the simple reason thar their source material is the corpus of the internet which is itself strewn with errors, satire and fake news. That is not to say that they cannot be a fun and useful tool. But one would be foolish to trust them without verification.
You can't talk to Leon about this - it's beyond his ken.
PB is literally unaware of the existence of the best models. This is laughable
I might be a little out of date but Claudia Schiffer for me.
I see that Margoe Robbie has had a child. Several PBers have opined that they would be 'up' for it - so has anyone recently become a father?
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
It got it right on that occasion. It gets a lot wrong, frequently. Which is fine as long as you are prepared to use you own judgement as a check and balance. The worrying trend I’m seeing is otherwise intelligent, professional, sceptical folk treating its output as gospel. This is a thing.
If I’m allowed to discuss this without being banned, I imagine you are using ChatGPT or Gemini. The first often gets this wrong the second very often gets this wrong
I’m using Anthropic Claude 3.6. By far the best model and the one closest to true AGI. It is absolutely outstanding - and breathtaking in its capabilities. Its amazing it is not better known
The other day I had a debate with it which changed my world view
But I can’t say any more because I’m not allowed. Ah well. Back to American politics! Also a shower after my gym visit and then a nice gin and tonic
None of them are reliable - for the simple reason thar their source material is the corpus of the internet which is itself strewn with errors, satire and fake news. That is not to say that they cannot be a fun and useful tool. But one would be foolish to trust them without verification.
You can't talk to Leon about this - it's beyond his ken.
PB is literally unaware of the existence of the best models. This is laughable
I might be a little out of date but Claudia Schiffer for me.
I see that Margoe Robbie has had a child. Several PBers have opined that they would be 'up' for it - so has anyone recently become a father?
I have assured her I wouldn't reveal any of our secrets.
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
It got it right on that occasion. It gets a lot wrong, frequently. Which is fine as long as you are prepared to use you own judgement as a check and balance. The worrying trend I’m seeing is otherwise intelligent, professional, sceptical folk treating its output as gospel. This is a thing.
If I’m allowed to discuss this without being banned, I imagine you are using ChatGPT or Gemini. The first often gets this wrong the second very often gets this wrong
I’m using Anthropic Claude 3.6. By far the best model and the one closest to true AGI. It is absolutely outstanding - and breathtaking in its capabilities. Its amazing it is not better known
The other day I had a debate with it which changed my world view
But I can’t say any more because I’m not allowed. Ah well. Back to American politics! Also a shower after my gym visit and then a nice gin and tonic
None of them are reliable - for the simple reason thar their source material is the corpus of the internet which is itself strewn with errors, satire and fake news. That is not to say that they cannot be a fun and useful tool. But one would be foolish to trust them without verification.
You can't talk to Leon about this - it's beyond his ken.
PB is literally unaware of the existence of the best models. This is laughable
I might be a little out of date but Claudia Schiffer for me.
I see that Margoe Robbie has had a child. Several PBers have opined that they would be 'up' for it - so has anyone recently become a father?
I have assured her I wouldn't reveal any of our secrets.
To avoid any confusion that's Margoe Robbie from Bridgend not Margot Robbie from Melbourne.
Re Georgia, and the view that more women out is a positive for Harris. It is worth remembering that Georgia was the site of a particularly high-profile case of an illegal immigrant killing a female student and one that generated a lot of anger.
It might not be abortion that is driving these female voters out.
Yes there’s been a lot of commentary on the Republican side about Laken Riley, and sadly a number of other similar cases that have been mentioned at rallies.
Here’s Megyn Kelly at Trump’s rally last night, endorsing Trump and mentioning both the sad cases of murder and sexual assault by illegals, but also the controversial issue of women’s sports which might also drive turnout among younger women. It’s not necessarily just abortion behind the differential female turnout. https://x.com/megynkellyshow/status/1853643950512316882
But if ALL the polling can agree on one thing this election, it is that there is a significant pro-Harris slant by women voters. Both in enthusiams and actual voting intention.
And if you want a demonstration of that energy - just look at the votes ALREADY cast in Georgia.
"The polls" have Georgia either too close to call or call it for Trump. Bullshit, I say.
Here's the maths to support my contention that Harris wins Georgia handily. Before today's voting, she already has. It requires only two data points: the actual votes cast as per the official Georgia website - and the latest and last ABC voting intention poll on the split in how men and women will vote.
"The final ABC News/Ipsos poll before Election Day, released on Sunday, found the gender gap among all likely voters to be 16 points. Harris had a 11-point advantage among women, 53% to 42%, while Trump had a 5-point advantage among men, 50% to 45%."
I make no apology for reposting this. It's a betting site. The data is telling us the result, as 56% of the electorate have already voted. Absent a huge surge in voting, there's only 10% or so going to vote today in Georgia.
So here are the likely votes cast already in Georgia, as best we can tell.
Using the latest ABC gender splits, Harris leads by 11% in the 56% of female votes we know have been cast.
ABC says Trump leads by 5% in that 43.8% of male votes we know have been cast.
2.257 million women have early voted (fact). ABC says that breaks down:
1.765 million men have early voted (fact). ABC says that breaks down:
Dem @ 47.5 = 0.838m male Democrat votes cast Rep @ 52.5 = 0.926m male Republican votes cast
Democrat total votes cast: 1.252 + 0.838 = 2.090 million Republican total votes cast 1.004 + 0.926 = 1.930 million
So ABC polling on the gender split says there is a 160,000 Dem firewall after 56% of the electorate have already early voted.
Reminder: Biden won Georgia in 2020 by just 11,779 votes
And in case you think it's just Georgia, that's a fluke - it's not:
Gender split in Pennsylvania is again huge in early voting. As of November 1st:
"More than 1.6 million commonwealth voters have cast early mail-in ballots for the Nov. 5 presidential election. An Inquirer analysis of early ballot returns found that women composed 56% of the early mail-ins, with men trailing at 43%, according to voting data obtained from the Pennsylvania Department of State this week."
The latest figures are 1.77m early votes, but that 56-43 margin is still there.
A 13% split. If we again take the last ABC split of women voters favouring Harris by 11%, that is another huge firewall banked by Harris.
I'm calling it at....4.12 Eastern Time on the 5th November:
KAMALA HARRIS WILL BE THE NEXT PRESIDENT OF THE UNITED STATES OF AMERICA.
I would be happier if I were a Democratic organiser in Georgia, than a Republican. But, there are two issues:
1. If turnout matches 2020, there are 750,000 votes still to come on the day. We don't know how those people will break. They might very well overturn a 166,000 vote firewall, among early voters.
2. We don't actually know how women in Georgia split. The split might be better than an 11% margin for Harris, or it might be worse. That's a major source of uncertainty.
You say it's not about race, but I would counter that race is always the elephant in the room in Southern State elections, some of which are close to ethnic headcounts.
Companies that poll Georgia might in fact, be quite correct.
Requires some special pleading for Harris to lose Georgia. It would have to be in a world all of its own.
That 750,000 still to vote would have to split better than 455k-295k Republican for Trump to win. Nowhere - but nowhere - is he getting splits above 60:40 in national polling. He can't get above 48%.
It's possible that turnout goes through the roof. But then ask yourself which is more likely - a wave of late-breaking Trump voters coming to the polls - or a wave of late-breaking Harris supporters boosting turnout? The polling is saying the late breakers are going decisively to Harris. I know which I think far more likely...
The voting intentions of Georgians who vote in person, on the day, may very well be quite different from those who have voted early. In fact we know quite well that they were in 2020.
What you're doing is taking the raw voting numbers, and then doing what no pollster would do. You're assuming that only one weighting (by sex) matters. Whereas income, religion, education, political affiliation race etc. are just as important influences on voting intention as sex.
I'd say it's fairly common thinking amongst educated here that having fewer/no children is absolutely fine and would do a great service to the planet, so its not really a problem.
This ignores the very serious social, political and economic consequences if it continues unchecked - which could ultimately lead to states falling apart into (and not excluding at the extremes) bankruptcy, disorder and anarchy.
What you probably want is some sort of long-term population stabilisation.
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
It got it right on that occasion. It gets a lot wrong, frequently. Which is fine as long as you are prepared to use you own judgement as a check and balance. The worrying trend I’m seeing is otherwise intelligent, professional, sceptical folk treating its output as gospel. This is a thing.
If I’m allowed to discuss this without being banned, I imagine you are using ChatGPT or Gemini. The first often gets this wrong the second very often gets this wrong
I’m using Anthropic Claude 3.6. By far the best model and the one closest to true AGI. It is absolutely outstanding - and breathtaking in its capabilities. Its amazing it is not better known
The other day I had a debate with it which changed my world view
But I can’t say any more because I’m not allowed. Ah well. Back to American politics! Also a shower after my gym visit and then a nice gin and tonic
Just tried Anthropic Claude 3.5
Unfortunately it has m1r purl and m1l purl reversed. This means it's giving instructions for a right-leaning increase instead of a left-leaning one.
It's actually a kinda impressive mistake. I can't quite work out how it's done it. All the training data will have it the right way round. But it's very wrong.
A revealing contrast between the way the £285 increase in university tuition fees is being accepted to the tantrums the oldies have had over losing their £200 WFA.
£285 for each of 3 years - a total of £855 which will be needed to be paid back over decades, if ever.
Versus a minimum of £200 a year over 10, 15 or 20 years which has to be found immediately each year rather than being deferred perhaps indefinitely.
Edit - I actually agree with withdrawing the WFA in principle although I disagree with the way in which who should still get it is assessed. But the point is that the potential finacial impact of the two decisions are very different.
But its not just the present oldies who will lose the WFA, its all the future oldies as well.
The present oldies may have already had 20+ more years of WFA than I'm ever going to get.
The WFA is more than covered by the VERY generous pension increase next year which was 8.5% thanks to the triple lock rather than the 2.6% it would otherwise have been.
Th WFA was £200 that extra 5.9% on the pension is worth over £600...
No-one mentions the weather when discussing the WFA. Colder winters are more expensive than milder ones.
Re Georgia, and the view that more women out is a positive for Harris. It is worth remembering that Georgia was the site of a particularly high-profile case of an illegal immigrant killing a female student and one that generated a lot of anger.
It might not be abortion that is driving these female voters out.
Yes there’s been a lot of commentary on the Republican side about Laken Riley, and sadly a number of other similar cases that have been mentioned at rallies.
Here’s Megyn Kelly at Trump’s rally last night, endorsing Trump and mentioning both the sad cases of murder and sexual assault by illegals, but also the controversial issue of women’s sports which might also drive turnout among younger women. It’s not necessarily just abortion behind the differential female turnout. https://x.com/megynkellyshow/status/1853643950512316882
But if ALL the polling can agree on one thing this election, it is that there is a significant pro-Harris slant by women voters. Both in enthusiams and actual voting intention.
And if you want a demonstration of that energy - just look at the votes ALREADY cast in Georgia.
"The polls" have Georgia either too close to call or call it for Trump. Bullshit, I say.
Here's the maths to support my contention that Harris wins Georgia handily. Before today's voting, she already has. It requires only two data points: the actual votes cast as per the official Georgia website - and the latest and last ABC voting intention poll on the split in how men and women will vote.
"The final ABC News/Ipsos poll before Election Day, released on Sunday, found the gender gap among all likely voters to be 16 points. Harris had a 11-point advantage among women, 53% to 42%, while Trump had a 5-point advantage among men, 50% to 45%."
I make no apology for reposting this. It's a betting site. The data is telling us the result, as 56% of the electorate have already voted. Absent a huge surge in voting, there's only 10% or so going to vote today in Georgia.
So here are the likely votes cast already in Georgia, as best we can tell.
Using the latest ABC gender splits, Harris leads by 11% in the 56% of female votes we know have been cast.
ABC says Trump leads by 5% in that 43.8% of male votes we know have been cast.
2.257 million women have early voted (fact). ABC says that breaks down:
1.765 million men have early voted (fact). ABC says that breaks down:
Dem @ 47.5 = 0.838m male Democrat votes cast Rep @ 52.5 = 0.926m male Republican votes cast
Democrat total votes cast: 1.252 + 0.838 = 2.090 million Republican total votes cast 1.004 + 0.926 = 1.930 million
So ABC polling on the gender split says there is a 160,000 Dem firewall after 56% of the electorate have already early voted.
Reminder: Biden won Georgia in 2020 by just 11,779 votes
And in case you think it's just Georgia, that's a fluke - it's not:
Gender split in Pennsylvania is again huge in early voting. As of November 1st:
"More than 1.6 million commonwealth voters have cast early mail-in ballots for the Nov. 5 presidential election. An Inquirer analysis of early ballot returns found that women composed 56% of the early mail-ins, with men trailing at 43%, according to voting data obtained from the Pennsylvania Department of State this week."
The no booze without food thing does get really old very quickly, in Korea
Also it’s near impossible to buy Tabasco in central Seoul
These tiny things begin to grate
What's the point in being rich if you can't order both - then not touch the food?
I’ve found a really cool little Taiwanese sort-of izakaya and I’m having Taiwanese food ordered from a special ipad and using my iPhone to translate with photos
A revealing contrast between the way the £285 increase in university tuition fees is being accepted to the tantrums the oldies have had over losing their £200 WFA.
£285 for each of 3 years - a total of £855 which will be needed to be paid back over decades, if ever.
Versus a minimum of £200 a year over 10, 15 or 20 years which has to be found immediately each year rather than being deferred perhaps indefinitely.
Edit - I actually agree with withdrawing the WFA in principle although I disagree with the way in which who should still get it is assessed. But the point is that the potential finacial impact of the two decisions are very different.
But its not just the present oldies who will lose the WFA, its all the future oldies as well.
The present oldies may have already had 20+ more years of WFA than I'm ever going to get.
The WFA is more than covered by the VERY generous pension increase next year which was 8.5% thanks to the triple lock rather than the 2.6% it would otherwise have been.
Th WFA was £200 that extra 5.9% on the pension is worth over £600...
No-one mentions the weather when discussing the WFA. Colder winters are more expensive than milder ones.
Re Georgia, and the view that more women out is a positive for Harris. It is worth remembering that Georgia was the site of a particularly high-profile case of an illegal immigrant killing a female student and one that generated a lot of anger.
It might not be abortion that is driving these female voters out.
Yes there’s been a lot of commentary on the Republican side about Laken Riley, and sadly a number of other similar cases that have been mentioned at rallies.
Here’s Megyn Kelly at Trump’s rally last night, endorsing Trump and mentioning both the sad cases of murder and sexual assault by illegals, but also the controversial issue of women’s sports which might also drive turnout among younger women. It’s not necessarily just abortion behind the differential female turnout. https://x.com/megynkellyshow/status/1853643950512316882
But if ALL the polling can agree on one thing this election, it is that there is a significant pro-Harris slant by women voters. Both in enthusiams and actual voting intention.
And if you want a demonstration of that energy - just look at the votes ALREADY cast in Georgia.
"The polls" have Georgia either too close to call or call it for Trump. Bullshit, I say.
Here's the maths to support my contention that Harris wins Georgia handily. Before today's voting, she already has. It requires only two data points: the actual votes cast as per the official Georgia website - and the latest and last ABC voting intention poll on the split in how men and women will vote.
"The final ABC News/Ipsos poll before Election Day, released on Sunday, found the gender gap among all likely voters to be 16 points. Harris had a 11-point advantage among women, 53% to 42%, while Trump had a 5-point advantage among men, 50% to 45%."
I make no apology for reposting this. It's a betting site. The data is telling us the result, as 56% of the electorate have already voted. Absent a huge surge in voting, there's only 10% or so going to vote today in Georgia.
So here are the likely votes cast already in Georgia, as best we can tell.
Using the latest ABC gender splits, Harris leads by 11% in the 56% of female votes we know have been cast.
ABC says Trump leads by 5% in that 43.8% of male votes we know have been cast.
2.257 million women have early voted (fact). ABC says that breaks down:
1.765 million men have early voted (fact). ABC says that breaks down:
Dem @ 47.5 = 0.838m male Democrat votes cast Rep @ 52.5 = 0.926m male Republican votes cast
Democrat total votes cast: 1.252 + 0.838 = 2.090 million Republican total votes cast 1.004 + 0.926 = 1.930 million
So ABC polling on the gender split says there is a 160,000 Dem firewall after 56% of the electorate have already early voted.
Reminder: Biden won Georgia in 2020 by just 11,779 votes
And in case you think it's just Georgia, that's a fluke - it's not:
Gender split in Pennsylvania is again huge in early voting. As of November 1st:
"More than 1.6 million commonwealth voters have cast early mail-in ballots for the Nov. 5 presidential election. An Inquirer analysis of early ballot returns found that women composed 56% of the early mail-ins, with men trailing at 43%, according to voting data obtained from the Pennsylvania Department of State this week."
The no booze without food thing does get really old very quickly, in Korea
Also it’s near impossible to buy Tabasco in central Seoul
These tiny things begin to grate
What's the point in being rich if you can't order both - then not touch the food?
I’ve found a really cool little Taiwanese sort-of izakaya and I’m having Taiwanese food ordered from a special ipad and using my iPhone to translate with photos
The no booze without food thing does get really old very quickly, in Korea
Also it’s near impossible to buy Tabasco in central Seoul
These tiny things begin to grate
What's the point in being rich if you can't order both - then not touch the food?
I’ve found a really cool little Taiwanese sort-of izakaya and I’m having Taiwanese food ordered from a special ipad and using my iPhone to translate with photos
It’s all tremendous fun and Korea is brilliant
Never been yet. It's on the list.
Although it look like Mauritania first...
Are you serious? Is it a birder destination or something?
The no booze without food thing does get really old very quickly, in Korea
Also it’s near impossible to buy Tabasco in central Seoul
These tiny things begin to grate
What's the point in being rich if you can't order both - then not touch the food?
I’ve found a really cool little Taiwanese sort-of izakaya and I’m having Taiwanese food ordered from a special ipad and using my iPhone to translate with photos
The question I posed was: "How many letter r's in the word raspberry?"
It answered two.
I just tried this. Five seconds ago
Answer:
“Let me count the r's explicitly:
r(1)aspber(2)r(3)y
There are 3 r's in "raspberry".”
It got it right on that occasion. It gets a lot wrong, frequently. Which is fine as long as you are prepared to use you own judgement as a check and balance. The worrying trend I’m seeing is otherwise intelligent, professional, sceptical folk treating its output as gospel. This is a thing.
If I’m allowed to discuss this without being banned, I imagine you are using ChatGPT or Gemini. The first often gets this wrong the second very often gets this wrong
I’m using Anthropic Claude 3.6. By far the best model and the one closest to true AGI. It is absolutely outstanding - and breathtaking in its capabilities. Its amazing it is not better known
The other day I had a debate with it which changed my world view
But I can’t say any more because I’m not allowed. Ah well. Back to American politics! Also a shower after my gym visit and then a nice gin and tonic
None of them are reliable - for the simple reason thar their source material is the corpus of the internet which is itself strewn with errors, satire and fake news. That is not to say that they cannot be a fun and useful tool. But one would be foolish to trust them without verification.
You can't talk to Leon about this - it's beyond his ken.
To be fair its a bit unfair 'arguing' the topic of AI with someone who is banned from responding.
He responded to me, after I responded to someone else.
Are we all to stop talking about AI just so he doesn't respond?
Re Georgia, and the view that more women out is a positive for Harris. It is worth remembering that Georgia was the site of a particularly high-profile case of an illegal immigrant killing a female student and one that generated a lot of anger.
It might not be abortion that is driving these female voters out.
Yes there’s been a lot of commentary on the Republican side about Laken Riley, and sadly a number of other similar cases that have been mentioned at rallies.
Here’s Megyn Kelly at Trump’s rally last night, endorsing Trump and mentioning both the sad cases of murder and sexual assault by illegals, but also the controversial issue of women’s sports which might also drive turnout among younger women. It’s not necessarily just abortion behind the differential female turnout. https://x.com/megynkellyshow/status/1853643950512316882
But if ALL the polling can agree on one thing this election, it is that there is a significant pro-Harris slant by women voters. Both in enthusiams and actual voting intention.
And if you want a demonstration of that energy - just look at the votes ALREADY cast in Georgia.
"The polls" have Georgia either too close to call or call it for Trump. Bullshit, I say.
Here's the maths to support my contention that Harris wins Georgia handily. Before today's voting, she already has. It requires only two data points: the actual votes cast as per the official Georgia website - and the latest and last ABC voting intention poll on the split in how men and women will vote.
"The final ABC News/Ipsos poll before Election Day, released on Sunday, found the gender gap among all likely voters to be 16 points. Harris had a 11-point advantage among women, 53% to 42%, while Trump had a 5-point advantage among men, 50% to 45%."
I make no apology for reposting this. It's a betting site. The data is telling us the result, as 56% of the electorate have already voted. Absent a huge surge in voting, there's only 10% or so going to vote today in Georgia.
So here are the likely votes cast already in Georgia, as best we can tell.
Using the latest ABC gender splits, Harris leads by 11% in the 56% of female votes we know have been cast.
ABC says Trump leads by 5% in that 43.8% of male votes we know have been cast.
2.257 million women have early voted (fact). ABC says that breaks down:
1.765 million men have early voted (fact). ABC says that breaks down:
Dem @ 47.5 = 0.838m male Democrat votes cast Rep @ 52.5 = 0.926m male Republican votes cast
Democrat total votes cast: 1.252 + 0.838 = 2.090 million Republican total votes cast 1.004 + 0.926 = 1.930 million
So ABC polling on the gender split says there is a 160,000 Dem firewall after 56% of the electorate have already early voted.
Reminder: Biden won Georgia in 2020 by just 11,779 votes
And in case you think it's just Georgia, that's a fluke - it's not:
Gender split in Pennsylvania is again huge in early voting. As of November 1st:
"More than 1.6 million commonwealth voters have cast early mail-in ballots for the Nov. 5 presidential election. An Inquirer analysis of early ballot returns found that women composed 56% of the early mail-ins, with men trailing at 43%, according to voting data obtained from the Pennsylvania Department of State this week."
The latest figures are 1.77m early votes, but that 56-43 margin is still there.
A 13% split. If we again take the last ABC split of women voters favouring Harris by 11%, that is another huge firewall banked by Harris.
I'm calling it at....4.12 Eastern Time on the 5th November:
KAMALA HARRIS WILL BE THE NEXT PRESIDENT OF THE UNITED STATES OF AMERICA.
@MarqueeMark your enthusiasm is infectious but from a betting point of view I must confess I take all your posts with a very large pinch of salt!
We need cold hard objectivity over the next 24 hours. Not to grasp the best bits of news we want to hear.
My post is hard-nosed objectivity! Based on two data points: the very real actual polling numbers from the official Georgia website, showing how far ahead women voters are with 56% of all those entitled to vote having done so (so 100% objectivity); and the last ABC poll which showed how the female voters were breaking decisivley to Harris (with males breaking less solidly for Trump) - which you could argue with but it doesn't seem out of line with other pollsters you could choose.
You really don't need other data points to go "bloody hell!"
Our solar panels had their worst week for energy production last week since January, and as I look out from our balcony at the vast wind farm in the Irish sea, the blades are barely turning
A revealing contrast between the way the £285 increase in university tuition fees is being accepted to the tantrums the oldies have had over losing their £200 WFA.
£285 for each of 3 years - a total of £855 which will be needed to be paid back over decades, if ever.
Versus a minimum of £200 a year over 10, 15 or 20 years which has to be found immediately each year rather than being deferred perhaps indefinitely.
Edit - I actually agree with withdrawing the WFA in principle although I disagree with the way in which who should still get it is assessed. But the point is that the potential finacial impact of the two decisions are very different.
But its not just the present oldies who will lose the WFA, its all the future oldies as well.
The present oldies may have already had 20+ more years of WFA than I'm ever going to get.
The WFA is more than covered by the VERY generous pension increase next year which was 8.5% thanks to the triple lock rather than the 2.6% it would otherwise have been.
Th WFA was £200 that extra 5.9% on the pension is worth over £600...
No-one mentions the weather when discussing the WFA. Colder winters are more expensive than milder ones.
We've had a spate of mild winters recently.
We have! And are told continually beforehand this will be the coldest winter. Like the Telegraph pre budget information. Garbage.
A revealing contrast between the way the £285 increase in university tuition fees is being accepted to the tantrums the oldies have had over losing their £200 WFA.
£285 for each of 3 years - a total of £855 which will be needed to be paid back over decades, if ever.
Versus a minimum of £200 a year over 10, 15 or 20 years which has to be found immediately each year rather than being deferred perhaps indefinitely.
Edit - I actually agree with withdrawing the WFA in principle although I disagree with the way in which who should still get it is assessed. But the point is that the potential finacial impact of the two decisions are very different.
But its not just the present oldies who will lose the WFA, its all the future oldies as well.
The present oldies may have already had 20+ more years of WFA than I'm ever going to get.
The WFA is more than covered by the VERY generous pension increase next year which was 8.5% thanks to the triple lock rather than the 2.6% it would otherwise have been.
Th WFA was £200 that extra 5.9% on the pension is worth over £600...
No-one mentions the weather when discussing the WFA. Colder winters are more expensive than milder ones.
We've had a spate of mild winters recently.
Starmer better pray there isn't another "Beast from the East" in the coming winters.
But hey, they are only Tory pensioners that die...
Re Georgia, and the view that more women out is a positive for Harris. It is worth remembering that Georgia was the site of a particularly high-profile case of an illegal immigrant killing a female student and one that generated a lot of anger.
It might not be abortion that is driving these female voters out.
Yes there’s been a lot of commentary on the Republican side about Laken Riley, and sadly a number of other similar cases that have been mentioned at rallies.
Here’s Megyn Kelly at Trump’s rally last night, endorsing Trump and mentioning both the sad cases of murder and sexual assault by illegals, but also the controversial issue of women’s sports which might also drive turnout among younger women. It’s not necessarily just abortion behind the differential female turnout. https://x.com/megynkellyshow/status/1853643950512316882
But if ALL the polling can agree on one thing this election, it is that there is a significant pro-Harris slant by women voters. Both in enthusiams and actual voting intention.
And if you want a demonstration of that energy - just look at the votes ALREADY cast in Georgia.
"The polls" have Georgia either too close to call or call it for Trump. Bullshit, I say.
Here's the maths to support my contention that Harris wins Georgia handily. Before today's voting, she already has. It requires only two data points: the actual votes cast as per the official Georgia website - and the latest and last ABC voting intention poll on the split in how men and women will vote.
"The final ABC News/Ipsos poll before Election Day, released on Sunday, found the gender gap among all likely voters to be 16 points. Harris had a 11-point advantage among women, 53% to 42%, while Trump had a 5-point advantage among men, 50% to 45%."
I make no apology for reposting this. It's a betting site. The data is telling us the result, as 56% of the electorate have already voted. Absent a huge surge in voting, there's only 10% or so going to vote today in Georgia.
So here are the likely votes cast already in Georgia, as best we can tell.
Using the latest ABC gender splits, Harris leads by 11% in the 56% of female votes we know have been cast.
ABC says Trump leads by 5% in that 43.8% of male votes we know have been cast.
2.257 million women have early voted (fact). ABC says that breaks down:
1.765 million men have early voted (fact). ABC says that breaks down:
Dem @ 47.5 = 0.838m male Democrat votes cast Rep @ 52.5 = 0.926m male Republican votes cast
Democrat total votes cast: 1.252 + 0.838 = 2.090 million Republican total votes cast 1.004 + 0.926 = 1.930 million
So ABC polling on the gender split says there is a 160,000 Dem firewall after 56% of the electorate have already early voted.
Reminder: Biden won Georgia in 2020 by just 11,779 votes
And in case you think it's just Georgia, that's a fluke - it's not:
Gender split in Pennsylvania is again huge in early voting. As of November 1st:
"More than 1.6 million commonwealth voters have cast early mail-in ballots for the Nov. 5 presidential election. An Inquirer analysis of early ballot returns found that women composed 56% of the early mail-ins, with men trailing at 43%, according to voting data obtained from the Pennsylvania Department of State this week."
The latest figures are 1.77m early votes, but that 56-43 margin is still there.
A 13% split. If we again take the last ABC split of women voters favouring Harris by 11%, that is another huge firewall banked by Harris.
I'm calling it at....4.12 Eastern Time on the 5th November:
KAMALA HARRIS WILL BE THE NEXT PRESIDENT OF THE UNITED STATES OF AMERICA.
@MarqueeMark your enthusiasm is infectious but from a betting point of view I must confess I take all your posts with a very large pinch of salt!
We need cold hard objectivity over the next 24 hours. Not to grasp the best bits of news we want to hear.
My post is hard-nosed objectivity! Based on two data points: the very real actual polling numbers from the official Georgia website, showing how far ahead women voters are with 56% of all those entitled to vote having done so (so 100% objectivity); and the last ABC poll which showed how the female voters were breaking decisivley to Harris (with males beaking less solidly for Trump) - which you could argue with but it doesn't seem out of line with other pollsters you could choose.
You really don't need other data points to go "bloody hell!"
I hope you are not wishcasting like all our Trumpian friends, who claim not to be wishcasting.
Re Georgia, and the view that more women out is a positive for Harris. It is worth remembering that Georgia was the site of a particularly high-profile case of an illegal immigrant killing a female student and one that generated a lot of anger.
It might not be abortion that is driving these female voters out.
Yes there’s been a lot of commentary on the Republican side about Laken Riley, and sadly a number of other similar cases that have been mentioned at rallies.
Here’s Megyn Kelly at Trump’s rally last night, endorsing Trump and mentioning both the sad cases of murder and sexual assault by illegals, but also the controversial issue of women’s sports which might also drive turnout among younger women. It’s not necessarily just abortion behind the differential female turnout. https://x.com/megynkellyshow/status/1853643950512316882
But if ALL the polling can agree on one thing this election, it is that there is a significant pro-Harris slant by women voters. Both in enthusiams and actual voting intention.
And if you want a demonstration of that energy - just look at the votes ALREADY cast in Georgia.
"The polls" have Georgia either too close to call or call it for Trump. Bullshit, I say.
Here's the maths to support my contention that Harris wins Georgia handily. Before today's voting, she already has. It requires only two data points: the actual votes cast as per the official Georgia website - and the latest and last ABC voting intention poll on the split in how men and women will vote.
"The final ABC News/Ipsos poll before Election Day, released on Sunday, found the gender gap among all likely voters to be 16 points. Harris had a 11-point advantage among women, 53% to 42%, while Trump had a 5-point advantage among men, 50% to 45%."
I make no apology for reposting this. It's a betting site. The data is telling us the result, as 56% of the electorate have already voted. Absent a huge surge in voting, there's only 10% or so going to vote today in Georgia.
So here are the likely votes cast already in Georgia, as best we can tell.
Using the latest ABC gender splits, Harris leads by 11% in the 56% of female votes we know have been cast.
ABC says Trump leads by 5% in that 43.8% of male votes we know have been cast.
2.257 million women have early voted (fact). ABC says that breaks down:
1.765 million men have early voted (fact). ABC says that breaks down:
Dem @ 47.5 = 0.838m male Democrat votes cast Rep @ 52.5 = 0.926m male Republican votes cast
Democrat total votes cast: 1.252 + 0.838 = 2.090 million Republican total votes cast 1.004 + 0.926 = 1.930 million
So ABC polling on the gender split says there is a 160,000 Dem firewall after 56% of the electorate have already early voted.
Reminder: Biden won Georgia in 2020 by just 11,779 votes
And in case you think it's just Georgia, that's a fluke - it's not:
Gender split in Pennsylvania is again huge in early voting. As of November 1st:
"More than 1.6 million commonwealth voters have cast early mail-in ballots for the Nov. 5 presidential election. An Inquirer analysis of early ballot returns found that women composed 56% of the early mail-ins, with men trailing at 43%, according to voting data obtained from the Pennsylvania Department of State this week."
The latest figures are 1.77m early votes, but that 56-43 margin is still there.
A 13% split. If we again take the last ABC split of women voters favouring Harris by 11%, that is another huge firewall banked by Harris.
I'm calling it at....4.12 Eastern Time on the 5th November:
KAMALA HARRIS WILL BE THE NEXT PRESIDENT OF THE UNITED STATES OF AMERICA.
@MarqueeMark your enthusiasm is infectious but from a betting point of view I must confess I take all your posts with a very large pinch of salt!
We need cold hard objectivity over the next 24 hours. Not to grasp the best bits of news we want to hear.
My post is hard-nosed objectivity! Based on two data points: the very real actual polling numbers from the official Georgia website, showing how far ahead women voters are with 56% of all those entitled to vote having done so (so 100% objectivity); and the last ABC poll which showed how the female voters were breaking decisivley to Harris (with males beaking less solidly for Trump) - which you could argue with but it doesn't seem out of line with other pollsters you could choose.
You really don't need other data points to go "bloody hell!"
Right or wrong, it is valuable info, well argued and presented. Not sure what the problem is.
The no booze without food thing does get really old very quickly, in Korea
Also it’s near impossible to buy Tabasco in central Seoul
These tiny things begin to grate
What's the point in being rich if you can't order both - then not touch the food?
I’ve found a really cool little Taiwanese sort-of izakaya and I’m having Taiwanese food ordered from a special ipad and using my iPhone to translate with photos
It’s all tremendous fun and Korea is brilliant
Never been yet. It's on the list.
Although it look like Mauritania first...
Are you serious? Is it a birder destination or something?
Our solar panels had their worst week for energy production last week since January, and as I look out from our balcony at the vast wind farm in the Irish sea, the blades are barely turning
Good lord, yes my app tells me I've only generated 8.2 KwH in November so far lol.
How does childless South Korea bode for its long-term defence against North Korea?
What's the birth rate in North Korea ?
The death rate is probably more pertinent.
Not really, because being born is optional (decided by your would-be parents), but death is mandatory. It's like take-offs and landings with aeroplanes.
Comments
The courts cannot intervene either unless to stop the count but Trump would have to be ahead in that states count anyway at that point
1. You literally can’t buy a drink without ordering food
2. People sneeze REALLY loudly
You're far too good a predictor to fall for all that guff. In 2028, Labour will win with a majority easily big enough to contest the (?) 2032 election against a proper Opposition. And with that strengthened Opposition in that (?) 2032 election, the LDs might well get rid of Labour.
Even if it means a hung Parliament with the Greens
HT https://bsky.app/profile/andycraig.bsky.social/post/3la6pqp2fca2n
me: "how many rs are there in the word raspberry?"
chatGPT: "The word "raspberry" contains two "r"s."
See my later comment
Only a fool uses ChatGPT by choice - unless you really love AVM (which is fun, TBF)
Ah ok so you can't read, but I already knew that. tbf you just gave the false impression that ChatGPT said there were three rs in raspberry, only admitting it was a different 'AI' when called out on it.
and I'm old enough to remember when you were swearing that ChatGPT was on the brink of the singularity, and the imminent update would demonstrate an AGI a million times superior to human intelligence.
A couple of year later and it can't even tell me how many rs are in the word raspberry. So you can understand why people aren't going to take your word for it.
That’s no life for anyone. Better to go childless into the darkness
I was disappointed and very surprised that Timothy (who I like) went from backing Tugendhat to Jenrick rather than Badenoch.
P H I L P
Get it right! ;-)
He's a South Londoner, of course you don't know him. And a Physics graduate.
https://x.com/politicoeurope/status/1853392764601327685
Google revealed the location of Ukrainian military sites in its latest Maps update, with the tech giant scrambling to remove the pictures, a Kyiv official said yesterday.
Only a fool uses ChatGPT by choice - unless you really love AVM (which is fun, TBF)
Ah ok so you can't read, but I already knew that. tbf you just gave the false impression that ChatGPT said there were three rs in raspberry, only admitting it was a different 'AI' when called out on it.
and I'm old enough to remember when you were swearing that ChatGPT was on the brink of the singularity, and the imminent update would demonstrate an AGI a million times superior to human intelligence.
A couple of year later and it can't even tell me how many rs are in the word raspberry. So you can understand why people aren't going to take your word for it.
*****
I’m not allowed to talk about it so you’ll just have to take it on trust that you’re an idiot
I imagine it won’t require great feats of imagination
Also: sort out your blockquoting, you kartoffel eating halfwit
Only a fool uses ChatGPT by choice - unless you really love AVM (which is fun, TBF)
Ah ok so you can't read, but I already knew that. tbf you just gave the false impression that ChatGPT said there were three rs in raspberry, only admitting it was a different 'AI' when called out on it.
and I'm old enough to remember when you were swearing that ChatGPT was on the brink of the singularity, and the imminent update would demonstrate an AGI a million times superior to human intelligence.
A couple of year later and it can't even tell me how many rs are in the word raspberry. So you can understand why people aren't going to take your word for it.
It's just in the nature of LLMs that they aren't very good at straight facts. They have no concept of true or false, so it doesn't make sense to use them for such tasks. However, they are incredibly useful for the sort of jobs that require some degree of creativity based on the corpus of human knowledge encompassed by the internet. Suggesting recipes, summarising texts, suggesting coding approaches; that sort of thing.
Edit: The quotes are shot to hell in this. The last paragraph is mine.
It’s usually better to make it an early start rather than a stay up late - unless you’re doing loads of in-play betting of course, in which case you get up at midnight having slept a few hours already!
IIRC four years ago most of the stay-ups went to bed sure Trump had won.
So someone has manually fixed it on raspberry now too, or possibly on recognising that sort of question in general.
But in general the poll closing times are early compared to other weekday elections eg UK, Israel, Netherlands, Ireland, Norway, Denmark, Canada (I think) but not Indonesia which closes at 1pm
I think @rcs1000 said there was likely to be useful information at about 1am from Georgia(?) but I think it will be 3/4am before you get anything concrete.
I suspect I will get up at about 6 (when Mrs Eek gets up) to see what's going on..
Is there a livestream on YouTube or something? I would prefer US coverage if poss. I don't get Sky.
Chris Philp must have gone through his whole life having to spell his name out and correct people. Having an unusual name or address is a pain in the arse; having an unusual name (or address) which looks like a misspelling of a common one must be doubly so.
https://www.gov.uk/government/news/cma-provisionally-finds-vodafone-three-could-address-competition-concerns-through-network-investment-and-customer-protections
Supplied by:
Gas 22.01 GW 57.0%
Solar 3.11 GW 8.1%
Wind 1.12 GW 2.9%
Hydroelectric 0.31 GW 0.8%
Nuclear 4.68 GW 12.1%
Biomass 2.95 GW 7.6%
Interconnectors 4.4 GW 11.5%
Or we could look at what he actually says and believe him, and figure out what actions likely follow that.
If he loses, the non-MAGA branch have a chance to bury him and try to reclaim their party
More power to his elbow!
https://docs.google.com/spreadsheets/d/1j3dyzsiucbbcCLH2jg3KtpcHTRkbgXoelDQ4ziysqsA/edit?usp=sharing
Any errors let me know.
Also it’s near impossible to buy Tabasco in central Seoul
These tiny things begin to grate
They are very excited about the merger, thinking it will deliver benefits to not only Vodafone and Three but O2 too, so it will impact the majority of mobile customers. It is a true re-balancing of the market.
He was very confident it would be approved.
My plan at the moment is to switch between CNN and The Daily Wire on the computer, depending on who’s happiest at any given moment, with the AP results page and PB on the ipad.
4. Everyone looks 10 years younger than their actual age.
5. People drive like nutters.
What you're doing is taking the raw voting numbers, and then doing what no pollster would do. You're assuming that only one weighting (by sex) matters. Whereas income, religion, education, political affiliation race etc. are just as important influences on voting intention as sex.
More poll close times but looks like we are well covered here.
This ignores the very serious social, political and economic consequences if it continues unchecked - which could ultimately lead to states falling apart into (and not excluding at the extremes) bankruptcy, disorder and anarchy.
What you probably want is some sort of long-term population stabilisation.
Unfortunately it has m1r purl and m1l purl reversed. This means it's giving instructions for a right-leaning increase instead of a left-leaning one.
It's actually a kinda impressive mistake. I can't quite work out how it's done it. All the training data will have it the right way round. But it's very wrong.
We need cold hard objectivity over the next 24 hours. Not to grasp the best bits of news we want to hear.
It’s all tremendous fun and Korea is brilliant
Although it look like Mauritania first...
Are we all to stop talking about AI just so he doesn't respond?
You really don't need other data points to go "bloody hell!"
Our solar panels had their worst week for energy production last week since January, and as I look out from our balcony at the vast wind farm in the Irish sea, the blades are barely turning
But hey, they are only Tory pensioners that die...
And who cares about Tabasco ?
got in line to vote about 6am .. 4th in line … line is 75-100 deep now. Two ppl with maga hats including a 80+ year old lady 😅
@NateSilver538
It's published.
We ran 80,000 simulations tonight.
Harris won in 40,012.