HackerTrans
TopNewTrendsCommentsPastAskShowJobs

ivoras

no profile record

Submissions

Byol – Bring Your Own LLM (Into an SSH Session with OpenCode)

github.com
1 points·by ivoras·4 個月前·0 comments

LLM_amoeba – Can an AI-powered amoeba survive?

ivoras.github.io
1 points·by ivoras·5 個月前·0 comments

Chrome browser extension for chatting about private pages with local LLMs

github.com
1 points·by ivoras·7 個月前·0 comments

comments

ivoras
·5 年前·discuss
It looks like WordPress is the only reason I've installed MySQL at all in the past decade. It's like WP and MySQL are each other's "killer app" :D
ivoras
·5 年前·discuss
Yeah but we're not really talking about bronze ages or if a species can achieve comparable technology, but whether we can correctly detect intelligence, or even life, around us.

I don't think we should label dolphins as un-intelligent just because their physical environment doesn't allow them to smelt metal.

My point is - I'd like to avoid judging something as (not) alive or intelligent just because it is (or isn't) similar to our daily life.
ivoras
·5 年前·discuss
That's a wrong way to think about it, since all modern "humanoids" are descended from the same ancestor not that long ago.

And if you consider dinosaurs humanoid, then so is that meme cat that can walk on its two back legs.

If there was an apex predator in every tree of life, in the "humanoid" shape, and I'm talking "humanoid ants" here, then it would be reasonable to assume that there's something beneficial in the "humanoid" shape that helps long-term survival.

And that's not even taking the majority of lifeforms into account - those in the seas, and greatly discounts birds. We still can't agree on if dolphins and crows are intelligent.

So, no, chances that the "humanoid" shape is somehow special out there are basically 0, since it's not even special here on Earth. We're an accident. Insects, mushrooms, plants - all those life forms are better adapted for mere survival than us (judging by cumulative mass of live organisms).
ivoras
·5 年前·discuss
First we need to decide how intelligent are the dolphins, then let us look at the stars.

Every single life form on Earth today had exactly the same time to do its evolution in. Us, bats, dolphins, tapeworms, birch trees, amoeba, mushrooms, all of that had exactly the same chance(s) in the same time span.

And we don't even know if and how mushrooms and trees communicate, let alone if they "think", for our near-sighted definition of thoughts. We don't even know what to make of birds, e.g. crows, with respect to the size of their brains and what they can do.

Is our planet, taken as a whole, alive, in some form of the Gaia hypothesis? The correct answers as of this time are either "we don't know" or "it depends on the definition of alive."

That Star Trek trope of "everything in the Universe is just like us with different faces" really needs to be put to rest.

Even H.R. Giger was boring and unimaginative with regards to how he envisioned the Alien. That's clearly a creature influenced by Earth-ism - a quadruped, with a single head and mouth and a flexible spine and claws, it's basically a weird cat.
ivoras
·7 年前·discuss
> Making a transaction this large is a security risk. It effectively establishes a billion dollar bounty for any party who can rewrite enough blocks to erase it

... accomplishing exactly nothing as all that would do would be returning the money to original spenders. Really, the amount of misinformation floating here is unexpected.

> Now that the value resides in a single coin, any subsequent transaction faces the same risk.

Non sequitur. A rollback of such a block would, by significance be somewhere between "mildly annoying" and "introducing suspicion that future transactions could also be rolled back." In any case, coins would end up where they are supposed to, and no theft can be done. There would be increased risk of double spends, but worried parties could always invest in more (legit) hashpower.

> A consortium/trust has formed in which individuals pay into a common pool of money.

Speculation, but reasonable. It makes business sense.

> That money is then protected with a multi signature script (consistent with the P2SH type). Given a threshold of signatures, the money can be spent, subject to other constraints. These will remain unknown until the first payment is made. At that point we'll know the number and identity of all the eligible keys and the threshold needed to make payment.

Speculation, but too wild. Basically wishful thinking.

> Edit: to be clear, the "bounty" I'm talking about can't be directly claimed just by mining some blocks. Instead, it would have to be claimed as part of a double spend of either the transaction in question or as subsequent transaction of the now-enormous output. Most likely, there would be collusion of some kind between a miner and the owner of the keys. I'm not saying this will happen, but the bigger the transaction, the greater the risk.

Again, wild speculation. So the theory is that someone collected $1B in BTC just to pull off a massive double spend stunt? At that level, it's likely that rubber hose cryptanalysis would resolve the issue quite efficiently.