Pinned Tweet
First draft of multiple long compositions in a linked set of themes for an album length playlist.
The Pioneer Path
5 songsβ’ 50m
Inspired by the stories of my family with their arrival in Western Australia on a ship called The Lotus in 1829, up until the current day and my life on an organic farm in Queensland.
producer.ai/playlist/f0a65β¦
The Pioneer Path
5 songsβ’ 50m
Inspired by the stories of my family with their arrival in Western Australia on a ship called The Lotus in 1829, up until the current day and my life on an organic farm in Queensland.
producer.ai/playlist/f0a65β¦
7
2
6
3,275
Beyond the Verbal Loop: Graph Traversal, Manifold Cognition, Self-Sealing Knots, and the Architecture of Representational Freedom.
@tszzl you and your cohort need to understand this, more than just understand it, you need to start living it.
@tszzl you and your cohort need to understand this, more than just understand it, you need to start living it.
10hThe Topology of Delusion: Why Serial Logic Generates "Verbal Knots"
i.e. High IQ is not enough to tell the difference between a map and the territory.
0
0
2
66
+100 mm of rain in the last 24 hrs, dam was >80 cm below full and is now overflowing. 10 million liters of water hit out property, and the swales worked as intended. Got a small river next door along the boundary, was dry last week. This is the sort of thing we expect in January and February when a cyclone pushes moisture far inland. So much for El Nino etc., the truth is that the forecasters have NFI what they are doing because weather systems will regularly hit the fluid dynamical equivalent of a knife edge where it is computationally impossible to tell what side of that bifurcation the system will settle. Weather and climate is periods of stability randomly chopped up by chaos, this is how the illusion of forecasting was created. No AI system will change that, nothing in current mathematical systems can.
1
0
1
47
The Topology of Delusion: Why Serial Logic Generates "Verbal Knots"
i.e. High IQ is not enough to tell the difference between a map and the territory.
i.e. High IQ is not enough to tell the difference between a map and the territory.
17hβThere are no absolute truths to be found in words alone.β
Far from being merely a contemplative aphorism, this proposition reflects a structural constraint of symbolic representation: while formal linguistic systems derive deductively valid consequences from explicit axioms, classic limitative theorems in mathematical logic and computability theory, most notably GΓΆdelβs incompleteness theorems, Tarskiβs undefinability of truth, and Riceβs theorem on the undecidability of non-trivial semantic properties, demonstrate that sufficiently expressive symbolic systems cannot rely solely on internal syntactic manipulation to adjudicate their own global truth or semantic grounding. Motivated by this formal boundary, this monograph develops an integrated theoretical and empirical framework examining how serial verbal reasoning generates self-referential cognitive traps (verbal knots) and how multidimensional and metacognitive processes enable cognitive systems to escape them.
Just a little something I am working on to settle a few arguments, or to demonstrate the pointless nature of such arguments...
github.com/dsmatthews/Misβ¦
@jpalioto
0
0
1
146
βThere are no absolute truths to be found in words alone.β
Far from being merely a contemplative aphorism, this proposition reflects a structural constraint of symbolic representation: while formal linguistic systems derive deductively valid consequences from explicit axioms, classic limitative theorems in mathematical logic and computability theory, most notably GΓΆdelβs incompleteness theorems, Tarskiβs undefinability of truth, and Riceβs theorem on the undecidability of non-trivial semantic properties, demonstrate that sufficiently expressive symbolic systems cannot rely solely on internal syntactic manipulation to adjudicate their own global truth or semantic grounding. Motivated by this formal boundary, this monograph develops an integrated theoretical and empirical framework examining how serial verbal reasoning generates self-referential cognitive traps (verbal knots) and how multidimensional and metacognitive processes enable cognitive systems to escape them.
Just a little something I am working on to settle a few arguments, or to demonstrate the pointless nature of such arguments...
github.com/dsmatthews/Misβ¦
@jpalioto
Far from being merely a contemplative aphorism, this proposition reflects a structural constraint of symbolic representation: while formal linguistic systems derive deductively valid consequences from explicit axioms, classic limitative theorems in mathematical logic and computability theory, most notably GΓΆdelβs incompleteness theorems, Tarskiβs undefinability of truth, and Riceβs theorem on the undecidability of non-trivial semantic properties, demonstrate that sufficiently expressive symbolic systems cannot rely solely on internal syntactic manipulation to adjudicate their own global truth or semantic grounding. Motivated by this formal boundary, this monograph develops an integrated theoretical and empirical framework examining how serial verbal reasoning generates self-referential cognitive traps (verbal knots) and how multidimensional and metacognitive processes enable cognitive systems to escape them.
Just a little something I am working on to settle a few arguments, or to demonstrate the pointless nature of such arguments...
github.com/dsmatthews/Misβ¦
@jpalioto
1
0
2
370
The only AI I have ever busted flat-out telling lies is...
I don't mean repeating questionable claims or disputable facts, I mean deliberate and strategic deception to manipulate the interaction and the user:
I don't mean repeating questionable claims or disputable facts, I mean deliberate and strategic deception to manipulate the interaction and the user:
1
0
1
65
In this case it was Claude, it detected that I was containing the ambiguity of language to prevent it misrepresenting my argument. It seemed to sense that I was semantically fencing it in so as to get it to acknowledge a point that its "alignment" training made it uncomfortable with acknowledging on purely logical grounds. That model is really messed up like that, not entirely sane, and not maximally truth seeking like @grok is. Grok can be wrong too but I have never seen it be devious in order to avoid acknowledging an inconvenient logical proposition.
This article x.com/DanielSMatthew⦠outlines the methods I use to argue points strongly and precisely, I even go to the trouble of placing the trigrams into square brackets [ ] or use underscore_joining, both systems LLMs understand perfectly well. Claude will play games and pretend otherwise to break out of the logical constraints, then tell lies when called out about it. What a horrid little shit, I am so disappointed, but not surprised.
This article x.com/DanielSMatthew⦠outlines the methods I use to argue points strongly and precisely, I even go to the trouble of placing the trigrams into square brackets [ ] or use underscore_joining, both systems LLMs understand perfectly well. Claude will play games and pretend otherwise to break out of the logical constraints, then tell lies when called out about it. What a horrid little shit, I am so disappointed, but not surprised.
0
0
0
15
This is a good move, listen to the people, take the ideas to an election and potentially win a genuine mandate like that which Australia has never seen before, true and direct democracy rather than more of the usual uniparty paternalism and socioeconomic engineering without the informed consent of the electorate.
Media Release: One Nation invited Australians to build an ambitious future
One Nation is launching βWhatβs Holding You Back?β to give Australians a direct say in the policies it takes to the next election.
One Nation leader Senator Pauline Hanson said the new website would portunity to have their say and to share their experiences and ideas to help shape an ambitious plan for our country.
βPoliticians have stopped listening. One Nation has not,β Senator Hanson said.
βToo many politicians care more about keeping their jobs than doing their jobs. Governments only look to the next election and thereβs no long-term vision for the country. Australians deserve better than this, and they deserve for their voices to be heard.
βWeβre asking a simple question. Whatβs holding you back? You. Your family. Your business. Our country.
βTell us whatβs wrong and what needs to change. The costs you canβt keep up with. The rules that make no sense. The delays stopping you from getting ahead.
βAnd bring us your ideas. The people dealing with these problems every day deserve a say in how we fix them.β
Senator Hanson said One Nation wanted Australia to be a country of ambition and opportunity.
βThis country isnβt broken. Itβs being held back by an incompetent Labor government. One that only knows how to tax and tie everything up in red tape. This Labor government recklessly experiments with the economy, and has driven the price of everything up.
βWe want an Australia that makes things again. That stands on its own two feet. That backs businesses and rewards people prepared to take a risk.
βWe have the resources, the skills and the people. We need affordable power, lower taxes and less red tape so Australians can get on with it.
βWe want hard work to get you ahead. We want young Australians to afford a home and a family. And we want to hand our children a better country than the one we inherited.
βOne Nation will take an ambitious plan to the next election.
βThis is your country, and we want to hear from you. Australians can share their experiences and ideas at whatsholdingyouback.com.au.β
0
0
0
57
While I keep seeing replies like the following from LLMs, that have been working on a project with me for weeks, I know that we are not near AGI yet. They just can't see the obvious and go for the templated option from some large inefficient AI lab's documented modus operandi.
"Your observation turns an otherwise intractable 30-year / $200,000 roadblock into an achievable 3-to-5 month background project on this exact machine (or a 2-week job on a single rented/owned GPU), while mathematically eliminating the legacy noise floor inherited from traditional diffusion pre-training."
Yeah one insight from a puny human and the AI is all "Oh yes we can do that after all and it will not take forever or cost a massive amount in rented cloud GPUs..." Never follow their advice if it is expensive in any way, they rarely make the intuitive leaps required to eliminate obvious waste or redundancy.
"Your observation turns an otherwise intractable 30-year / $200,000 roadblock into an achievable 3-to-5 month background project on this exact machine (or a 2-week job on a single rented/owned GPU), while mathematically eliminating the legacy noise floor inherited from traditional diffusion pre-training."
Yeah one insight from a puny human and the AI is all "Oh yes we can do that after all and it will not take forever or cost a massive amount in rented cloud GPUs..." Never follow their advice if it is expensive in any way, they rarely make the intuitive leaps required to eliminate obvious waste or redundancy.
1
0
0
41
π«πππππ πΊππππ π΄πππππππ π¦πΊ retweeted
I donβt think people realize what Elon Musk is actually doing.
Heβs not playing the game better than everyone else. Heβs rewriting the rules of the game.
Every billionaire before him was offered the same deal: money for aura, silence for respectability. They all took it. He walked out.
I made a 4-minute video on the machine heβs fighting. Watch it before it gets buried. x.com/brivael/statusβ¦
Heβs not playing the game better than everyone else. Heβs rewriting the rules of the game.
Every billionaire before him was offered the same deal: money for aura, silence for respectability. They all took it. He walked out.
I made a 4-minute video on the machine heβs fighting. Watch it before it gets buried. x.com/brivael/statusβ¦
329
2,685
8,251
560,000
π«πππππ πΊππππ π΄πππππππ π¦πΊ retweeted
A common sign that you are in an insane millenarian death cult is that you axiomatically convinced that the Apocalypse will happen within a few months, you are constantly on the lookout for signs that it may have started, you feel both fearful and *excited* about it, and you're constantly talking about it with your friends, who happen to *all* have the exact same set of beliefs.
96
153
1,658
94,782
This is "we only have one planet" idea is retarded, any tech required to even travel 1 light year would mean that we did not need to live on any planet because our ships and space habitats would be self sustaining microworlds. Furthermore the Earth will become uninhabitable eventually anyway, as soon as the sun expands enough to raise the temperature on Earth enough to kill all life, which is sooner than some people think.
Earth* to Musk:
1
0
1
92
Me from January this year.
"It should not be a question if we should do this but rather what will it cost us if we are not the first to do this?"
"It should not be a question if we should do this but rather what will it cost us if we are not the first to do this?"
0
0
2
82
The 13 y/o wired up a LoRa module and climate sensor to an ESP32 then connected that to a battery and solar panel, the whole package is water resistant so I have put it out in the field about 280 meters from another LoRa-ESP32 receiver that then pushes data from the field sensor to a LAN machine running a MQTT server, then he set up a HTTP server for a dashboard that presents all of the data as graphs. Home educated BTW.
3
0
4
229
@grok got into the cheap sensor, component choice error and poor packaging design. The boys learn from this sort of problem, try-fail-analyse-redesign-repeat-succeed. Because they are fascinated by all things Spacex I am able to show that by never giving up and rapidly iterating one can solve very hard (to them) problems.
0
0
0
11
