Categories
Options
Is Greater Manchester ready for Mayor Galloway? – politicalbetting.com
Is Greater Manchester ready for Mayor Galloway? – politicalbetting.com
NEW: Rochdale's new MP @georgegalloway is considering standing against Andy Burnham to be Greater Manchester mayorQuite the story from @josephtiman for @MENnewsdesk https://t.co/RhMrfR1BVb
0
Comments
Is their no end to the man's vanity?
And what about that Galloway?
https://m.youtube.com/watch?v=Sq1QZB5baNw&pp=ygUJZmlndXJlIDAx
Reform continue their remorseless rise.
https://www.youtube.com/watch?v=9-nX_fIVAOA
I couldn't detect any hint of irony in the denunciation of people with "hate in their hearts and venomous tongues".
Grifter
noun; slang.
1. a person who operates a side show at a circus, fair, etc., especially a gambling attraction.
2. a swindler, dishonest gambler, or the like.
The second is clearly an evolution of the first
I do not see the political side to this. The word is surely more applicable to people in financial services (sorry @kinabalu) than it is to politicians, especially if you see City-style financial dealing as somehow evil (like @kinabalu)
I guess Trump might be metaphorically shoehorned into the first definition - but it is still not an easy fit
Trump is much better characterised as what he is: a ranting egotist and a scheming demagogue. There are plenty of bad labels we can apply to The Donald without having to misapply wrong ones
Con 23.7% (-4%)
Lab 43.3% (-3.7%)
Ref 12.7% (+6.9%)
LD 10.2% (+1.5%)
Green 5.5% (+0.7%)
SNP 2.5% (-0.7%)
The gap between Labour and Conservatives has hardly moved but both have lost votes.
Reform has more than double their vote, with the Lib Dems and Green both slightly up.
The effect of these movements is to make a Conservative drubbing more likely.
Trump University was fairly egregious grifting.
I've seen stills from this video, but not the video. That is INCREDIBLE. That is Arthur C Clarke's "technology sufficiently advanced it is indistinguishable from magic"
I also remember one of the Luddites on here, @Benpointer, airily dismissing AI (about 6 months ago, lol) and saying "get back to me when a robot can stack my dishwasher". That was literally his definition of AI
Well, there is it Ben. A robot that will stack your dishwasher - and also do maths and make music and tell you jokes and answer puzzles and have all the info in the known world inside its net-accessible brain
It would actually get me to vote Labour for only the second time in my life (I voted for Ken over Boris in London 2008).
The 2017 GM mayoral election is actually the only election for which I've been eligible to vote where I haven't. TBF my wife was dealing with a different sort of labour at the time.
I thought the story was mostly fluff and that it would blow over. He'd get some good advice, and craft a decent apology for what he'd said. Maybe even do something to demonstrate that he'd learned from the furore.
But the Tories wobbled for ages before condemning his words. He issued a half-hearted apology that merely apologised for being rude. It turns out that there are other such incidents of him using 'colourful' language at staff meetings. And there's another £5m donation in progress that the party plan to accept, which will mean that he's the source of almost a third of their entire funding.
So I think my initial thoughts were wrong. This isn't fluff. He hasn't apologised for the 'should be shot' comment, and as of this moment shouldn't be considered a fit and proper person to be wielding such a huge influence over our political system.
https://www.salisburyjournal.co.uk/news/24188220.ringwood-shop-worker-new-years-resolution-sex-child/
He's still a dangerous demagogue
"I expect the swing back, I expect a small-mid sized working majority, but I don't see what we should be looking at to put the overall outcome in doubt."
If Starmer fails to win a majority the factors we will point to, which in hindsight will seem obvious, are:
1. Starmer's personal ratings, which are dire, and compare very badly to Blair or Cameron, the only two opposition leaders to become PM in the last 44 years.
2. The small proportion of direct Tory-Labour switchers, compared to the number of don't knows and switchers to Reform.
3. The strength of Tory data-driven, highly-targeted online campaigning, which is very poorly regulated compared to leaflet deliveries and billboard posters.
4. The volatility in the electorate that makes a large campaign turnaround possible (when Starmer flunks it in the campaign).
Honestly, I don't want to bang on again, but PB's AI discourse is some sad-ass primary school shit
https://www.trtworld.com/middle-east/israels-netanyahu-approves-rafah-offensive-plan-despite-intl-outcry-17372276
Grifting a'plenty documents by DJT's wiki page:
https://en.wikipedia.org/wiki/Donald_Trump
And you're surely not arguing he's honest ?
So not only do I have to vote Labour I also have to vote for an Evertonian to boot to stop Galloway.
Please don’t run Mr Galloway to stop me being placed in this invidious position.
Sorry to be brutal but that’s the reality.
https://twitter.com/RosieisaHolt/status/1766035642625020203
He's probably even stronger now than last time as well.
I just like to utilise the English language precisely, and I don't think he is a classic"grifter" as the word should be employed. It is a useful and clever word, but it is being stretched in its meaning to basically include "anyone I don't like", that's a shame, to my mind, because then the original precise meaning is lost
And yet.
We all saw May's lead evaporate like an April frost when the sun comes up. We know that campaigns DO see shifts in polling. And we also know that Labour is trying desperately to avoid pledging ANYTHING, and just keep reminding the public - 'Aren't those Tories an absolute shower?!'
And so it still sits there, that 24% profit.
As ever DYOR and only bet what you can afford to lose, but why isn't everyone on Labour Majority?
2. Don't agree. If you look at 1997, direct switching was only at slightly higher levels, and indirect swing (notably abstainers) was at least an equal driver of the overall result.
3. Possibly, but some of the novelty we saw in 2015, has worn off.
4. There has been lots of volatility but, as of this point, we do appear to be in a more settled pattern and Starmer is risk averse. Maybe, though.
Like Isam says, leader change is a possible wildcard, but here In think the Tories are trawling the barely tested and likely inadequate and it is as likely to work against c them as for them.
The grift is absolutely a big part of the reason Trump is in politics. Indeed running against Clinton possibly saved his business at the time.
The addiction to power is also part of it, but that doesn't invalidate the description.
Likewise, and more purely so, RFK Jnr.
Just now Labour MPs prevented debate on a new law to protect children and single sex spaces.
Instead they used parliamentary time to discuss ferret name choices.
https://twitter.com/KemiBadenoch/status/1768647056111861760
(I also note that she's incorrectly using that South African 'just now' phrasing that we were discussing yesterday. Perhaps Kemi should grow up right now.)
https://www.bbc.co.uk/sounds/play/m001wy2d
This is our conversation
I said:
"It’s 8pm in Colombia and I need to eat, but image creation was an emergent property of the early GPTs. The first Dall-e was a remake of GPT3 I think"
A casual remark made as I went to the kitchen to cook up a steak, I didn't check it or anything
He came back, haughtily:
"That's simply not true. I've explained how Dall-E has worked many times before. It's brilliant in its simplicity, and like ChatGPT it uses a neural net.
But Dall-E did not come out of GPT. They are two, entirely different systems."
Why is this interesting? Because he is completely wrong. Dall-E is an evolution of GPTs, indeed GPT3, as I said.
Here's Wikipedia on the subject:
"The first generative pre-trained transformer (GPT) model was initially developed by OpenAI in 2018,[16] using a Transformer architecture. The first iteration, GPT-1,[17] was scaled up to produce GPT-2 in 2019;[18] in 2020, it was scaled up again to produce GPT-3, with 175 billion parameters.[19][5][20]
DALL·E's model is a multimodal implementation of GPT-3[21] with 12 billion parameters[5] which "swaps text for pixels," trained on text–image pairs from the Internet"
https://en.wikipedia.org/wiki/DALL-E
Don't believe Wiki? Then try OpenAI who actually made both these machines:
"DALL·E is a 12-billion parameter version of GPT-3 trained to generate images from text descriptions, using a dataset of text–image pairs"
https://openai.com/research/dall-e
Case closed, m'Lud
What is compelling here is that this is @rcs1000's JOB. He works in software and, I believe, studied Machine Learning at Cambridge. He knows top people in AI. He is very very bright. I owe him a dry martini. But how could he get that so wrong?
It's no crime. We all make errors, I know I do. But it is still noteworthy that on the subject of AI the clever people at PB are consistently clueless and unable to extrapolate or comprehend what is going on, and some of the discourse is cringeworthy
But I've said I shall shut the feck up about AI, at least for today, I know it annoys people - and so I will. Coda
I also have to go and look at Bolivar's deathplace. Hasta Luego, amigos
My nomineee is a YT channel that's started poping up on my feed, namely "American Express Club" which clearly IMHO has ZERO connection with the real American Express.
"American Express Club" is currently pumping out at an amazing rate poor quality videos (apparently captured by old-school home VCR machine) of "New Perry Mason" made-for-TV movies first broadcast 1973-74.
So note anything a weeeeeee bit OFF about THIS typical blub, that "AEC" is using to promote it's offerings:
"Perry Mason - Collection 45 - Best Crime Movie TV Series Premiere 2024"
> NOT "Perry Mason" the original classic late 1950s - early 1960s TV show (based on novels of Earl Stanley Gardner) but rather the rather crappy 1970s "New Perry Mason".
> Episodes of "NPM" premiered in 1973-74 and NOT 2024.
> Visual stills (or whatever they call 'em) used to advertise this on YT feature AI-enhanced (I think) still from the ORIGINAL "Perry Mason", they are definitely NOT from "New Perry Mason".
Total freaking fraudistry.
Trump is no classier, but operates on a much larger scale.
White House Counsel letter today to Speaker Johnson:
“It is clear the House Republican impeachment is over. It is obviously time to move on.”
“There is too much important work to be done for the American people to continue wasting time on this charade.”
https://twitter.com/IanSams46/status/1768573442310181026
Full text at the link.
He's also a clever and open minded journalist, and sometimes bang on the money
A peculiar case. For me he is worth watching, but do it with great skepticism
He is a very well rounded villain.
The soliciting of political donations for his personal use is unquestionably grift, though.
But agree for a large number of reasons there is no way that Galloway will come close to winning GM mayor (and he wouldn't want the job anyway).
Perhaps he was previously a grifter but, on becoming President, surely he graduated to become a scammer?
https://www.theguardian.com/education/2024/mar/13/how-difficult-is-it-to-bake-a-potato-head-hits-out-at-school-caterers-southampton
https://www.theguardian.com/education/2024/mar/15/i-cant-make-them-eat-it-teachers-and-parents-share-concerns-over-school-lunches-in-england
The thing that underlies both Dall.E and GPT-X is a domain agnostic transformer. When it is trained on a big old set of data, it produces a model (e.g. Dall.E or GPT-3) that takes streams of input data and transforms it into streams of output data. The crucial thing is that it does not care *what* the streams of input data might be.
But training the model is not a hands off affair. It involves a lot of tweaking and special casing.
The initially slightly surprising thing is that when the input was changed from pure text to text and streams of pixel data, it produced a model whose output that was "quite reasonable" streams of text and output pixel data (once a bit of jiggery pokery was applied to make the pixel data a valid image).
Likewise with video - it was assumed that too much jiggery pokery would be needed to make that work reasonably well. But it turns out you can get ok results for short clips with only "a lot" of effort rather than insurmountable effort.
The confusion arises because in their marketing material, OpenAI sometimes refer to an actual trained model as GPT-X, sometimes the general domain agnostic transformer, and sometimes the product(s) that wrap up the trained models.
https://www.politico.com/news/magazine/2024/03/15/kyrsten-sinema-super-pac-00147081
..For years, Sinema was on the receiving end of a relatively unusual political-money phenomenon in the capital’s politics industry: the single-target PAC, an outfit geared towards creating precisely the outcome that became real when the senator announced her exit.
For better or worse, it is a model that probably won’t stay rare for long. And whatever you think of Sinema, the effort against her is also likely to speed up some of the most brutal trends in politics, another way for deep-pocketed donors to further wage permanent war on rivals who might not always make such obvious targets.
Other political committees might beat up on a senator in the name of an issue or to help a particular rival. The Replace Sinema super PAC, by contrast, existed solely to run robust oppo research on, buy ads against, pitch unflattering media stories about and otherwise hound, harry and hector one solitary elected official: Sinema..
Campaigning against an elected representative you want to oust is a reasonable activity - but coupled with the unlimited expenditure available without oversight or real limits in the US, it morphs into something much more destructive of democratic politics.
Eg Nigel Farage. For years a genuine fighter for a genuine cause (Brexit) he genuinely believed in. Mr Genuine. Surprised that wasn't his nickname.
But now (let's assume for the sake of this illustration) knocking on, Brexit done, no longer interested in changing the world, instead content to exploit his personal brand for pure gratification and profit (GBNews, AskFarage, TalkingPints, BrexitCords, StripalongaNigel etc), this is a different NF, clearly, and what is it that he has become?
Yep. You got it.
Seattle Times is informing me, that based on 2023 population estimates released by US Census Bureau, King County WA has just passed Queens County NY (aka NYC Borough of Queens) as USA's 12th-largest county.
Queens now 13th place, where King was last year.
HOWEVER, note that Kings County NY (aka NYC Borough of Brooklyn) is still larger in 9th place.
Here is current Top 12 list
1. Los Angeles Co CA
2. Cook Co IL (Chicago)
3. Harris Co TX (Houston)
4. Maricopa Co AZ (Phoenix)
5. San Diego Co CA
6. Orange Co CA (Anaheim-Irvine-Santa Ana)
7. Miami-Dade Co FL
8. Dallas Co TX
9. Kings Co NY (Brooklyn)
10. Riverside Co CA
11. Clark Co NV (Las Vegas)
12. King Co WA (Seattle)
Even more confusingly, you can use data produced by a GPT-type model to train a different GPT-type model (often focused on a simpler task).
Go back to the original statements, they are blunt. I said Dall-e came out of GPT, he said
"Dall-E did not come out of GPT. They are two, entirely different systems"
He is wrong
Dall-e is an evolution of "image GPT" - ie, GPT
They first experimented with GPT2
"Notably, we achieved our results by directly applying the GPT-2 language model to image generation."
https://openai.com/research/image-gpt
(look how shite the images are, that was just four years ago: scary)
Then they finessed it and span Dall-e out of GPT3, as they clearly say in their OWN blog
"DALL·E is a 12-billion parameter version of GPT-3 trained to generate images from text descriptions, using a dataset of text–image pairs"
https://openai.com/research/dall-e
Cooroboration:
"How does Dall-E work?
Dall-E works by using a number of technologies including natural language processing (NLP), large language models (LLMs) and diffusion processing.
Dall-E was built using a subset of the GPT-3 LLM. Instead of the full 175 billion parameters that GPT-3 provides, Dall-E uses only 12 billion parameters in an approach that was designed to be optimized for image generation. Just like the GPT-3 LLM, Dall-E also makes use of a transformer neural network -- also simply referred to as a transformer -- to enable the model to create and understand connections between different concepts."
https://www.techtarget.com/searchenterpriseai/definition/Dall-E
They are NOT two entirely different systems, it is not true to say they have no relationship. That is simply untrue. A non true statement. An error of fact. Wrong
And now, I will make good on my promise and STFU
"They are NOT two entirely different systems" is only correct if "entirely" is taken to mean "share no commonality at all", but they are two entirely different models built using (evolutions of) the same underlying technology.
English has the richest vocabulary of any language on earth: exploit it!
Why are so many paedophiles claiming to be trans?
Robert Smithson said:
"Dall-E did not come out of GPT. They are two, entirely different systems"
Whereas OpenAI, who actually made Dall-e and GPT (unlike Robert Smithson) say this:
"DALL·E is a 12-billion parameter version of GPT-3 trained to generate images from text descriptions, using a dataset of text–image pairs"
Now that really is it. Finito. Manana. Kiss kiss
Which means the Tory position actually looks better than last year. RefCon combined is 2.7% higher than a year ago while LLG is 1.5% lower.
We did have another writer, @Mysticrose?
I miss her. She was fun. She was NOT me, btw
It's just when you come over all faux pedant.
https://theathletic.com/5338544/2024/03/14/carragher-cbs-abdo-football-coverage/
A campaign he's almost certain to lose could be quite entertaining.
The Lost Art of Scottish Hospitality
Chapter one: Greeting Guests