HN user

mrybczyn

272 karma

Shepherd of electrons.

Posts0
Comments132
View on HN
No posts found.

because arc agi involves de novo reasoning over a restricted and (hopefully) unpretrained territory, in 2d space. not many people use LLMs as more than a better wikipedia,stack overflow, or autocomplete....

this exact 'cottage industry' you speak of is what existed in north america and started little conpanies like Apple, etc.

The sad fact that we lost all of these because of the entire electronics supply and design chain moving to Taiwan and China, is why we are where we are. These barriers might bring some back, who knows.

Ultimately global open borders, for goods and services, had their own issues. For example open competition between free market economies and centrally planned economies creates rather obvious advantages of scale that are skewed...

At the 1940s Manhattan project, back when computer meant a job: "person who computes mathematical statements", major advancements were made in the integration of hyperbolic PDEs, by substituting electro-mechanical and then vacuum-tube machines to do the job. You know, those hard-wired vacuum tube monsters like ENIAC.

You could argue that the First useful thing electronic computers did was integration...

https://www.tandfonline.com/doi/full/10.1080/00295450.2021.1...

Eh? By all indications compute is now evolving SLOWER than ever. Moore's Law is dead, Dennard scaling is over, the latest fab nodes are evolutionary rather than revolutionary.

This isn't the 80s when compute doubled every 9 months, mostly on clock scaling.

Thanks for all your work buro9! I've been an lfgss user for 15 years. This closure as a result of bureaucratic overreach is a great cultural loss to the world (I'm in Canada). The zany antics and banter of the London biking community provided me, and my contacts with which I have shared, many interesting thoughts, opinions, points of view, and memes, from the unique and authentic London local point of view.

LFGSS is more culturally relevant than the BBC!

Of course governments and regulations will fail realize what they have till it's gone.

- Pave paradise, put up a parking lot.

We just have to get creative, or discover some new physics :-)

In addition to the ark-ship colony, or the cryosleep slow ship:

1. Assuming it's a stream of robotic probes doing flybys, without decelerating, we have the Breakthrough Starshot approach. Maybe there's a way to use the target system's sun for solar sail braking? Send smart enough robots that have agency, that can do the exploring for us.

2. For human travel - it could just be a bunch of frozen embryos with a robotic nursemaid, accelerated via external propulsion and decelerated via nukes / high-g aerobraking... (Raised by Wolves had a cool introduction like this in the first episode - then went quickly downhill)

3. ...

Yes it was a cautionary tale, but not about the technology, but about the role of empathy and the human spirit.

The actual technology of Gattaca has already started 50 years ago with the first "test tube babies" and IVF treatments. You get a "optical selection" of the most well developed blastocysts, checking for symmetry, growth rates, and other biomarkers, without even resorting to DNA sequencing selection.

It will absolutely make the world a better place long term - all of our children can be genetically "the best of us", according to various criteria. I'm all for it.

A fantastic view on the nature of inference and the physical world! What I appreciate most about our LLM adventures in the past 5 years, is that it's finally:

1. Plausible to discuss, and find hints of what "AI" might mean - ("AGI" or "Strong AI").

2. Dig up all those crusty and dusty notions of philosophers long dead, and some nerding out reads from my teenage years. And try to apply them to an interesting system to study empirically, instead of just humanistic pondering.

3. Same with psychology! Now that our pattern matching systems are starting to pass various turing tests, what does that mean for "sentience" and "sapience" and should we all become armchair psychologists as we teach our systems to understand and act in the world? Recall the first part of Oddyssey 2010; where the AI researcher is more of a child psychologist...

Lastly - LLMs and "glimpses of AI" are something new! Not the same old same old recycled tech and popculture. Truly something new under the sun. Good times.

Bots, so many bots 2 years ago

Passport might be a bit onerous - it's expensive and painful process and many don't need it.

But it's a hilarious sign of worldwide government incompetence that social insurance or other citizen identification cards are not standard, free, and uniquely identifiable and usable for online ID purposes (presumably via some sort of verification service / PGP).

Government = people and laws. Government cannot even reliably ID people online. You had one job...

Yes, openAI is dumping the market with chat-gpt 3.5. Vulture capital behaviour at its finest, and I'm sure government regulations will definitely catch on to this in 20 or 30 years...

It's cheaper than the ELECTRICITY cost of running a llama-70 on your own M1.Max (very energy efficient chip) assuming free hardware.

I guess they are also getting a pretty good cache hit rate - there are only so many questions people ask at scale. But still, it's dumping.