Hacker Newsnew | past | comments | ask | show | jobs | submit | psychoslave's commentslogin

Freedom of one starts where it confirms freedom of others.

Of course everybody is going to find a point when freedom of speech have to be limited. Otherwise, anyone can justify that cutting the head of their neighbour with a katana while dancing is part of an artistic performance, and absolute free speech is only possible if all artistic expression is given complete license. Those who pretend otherwise will have no ground to defend themselves on legal basis from being wiped out of existence by the very same logic.


Over the last week I went with a bigger dig on using agent mode et work, and my experiment align with this observation.

The first thing that surprising to me is how much the default tuning are leaned toward laudative stances, the user is always absolutely right, what was done is solving everything expected. But actually no, not a single actual check was done, a tone of code was produced but the goal is not at all achieved and of course many regressions now lure in the code base, when it's not straight breaking everything (which is at least less insidious).

The thing that is surprising to me, is that it can easily drop thousands of lines of tests, and then it can be forced to loop over these tests until it succeed. In my experiments it still drop far too much noise code, but at least the burden of checking if it looks like it makes any sense is drastically reduced.


That's my observation too.

And I have been trying to improve the framework and abstractions/types to reduce the lines of code required for LLMs to create features in my web app.

Did the LLM really needed to spit 1k lines for this feature? Could I create abstractions to make it feasible in under 300 lines?

Of course there's cost and diminishing returns to abstractions so there are tradeoffs.


Like, "Your Brain Hallucinates Your Conscious Reality" as exposed by Anil Seth[1]? Found that one while searching for something like "the illusion of the self" a few years ago.

It’s also easy to find this treated in various philosophy/religion through time and space. And anyway as consciousness is eager to project whatever looks like a possible fit, elements of suggesting prior arts can be inferred back as far as traces can be found.

[1] https://www.youtube.com/watch?v=lyu7v7nWzfo


One system to rule them all, blame any issue encountered on lake of full adoption, and label them as defamation. What could go wrong?

How is ASCII not a standard?

And writing system don't fall out of nowhere, especially something as baroque as English which is all but phonetic.


ACII is a standard, but it's also standard. To a really high degree. 99% of webpages are UTF-8.

You could make a similar argument about, say, H.264, but the dominance is not as compelling and drops massively if you account for different container formats.


I wonder what proportion of people find things like TikTok, YouTube shorts, and even Twitter for the text counterpart, absolutely repulsive. It's not even disdain as in "I'm too good for this", more like some people can't stand the view of a spider I guess.

And other things like HN can definitely hook my mind.


Seems fair in the frame of what is responded.

But there is a concern which goes out of the "they" here. Actually, "they" could just as well not exist, and all narrative in the article be some LLM hallucination, we are still training ourself how we respond to this or that behavior we can observe and influence how we will act in the future.

If we go with the easy path labeling people as root cause, that's the habit we are forging for ourself. We are missing the opportunity to hone our sense of nuance and critical thought about the wider context which might be a better starting point to tackle the underlying issue.

Of course, name and shame is still there in the rhetorical toolbox, and everyone and their dog is able to use it even when rage and despair is all that stay in control of one mouth. Using it with relevant parcimony however is not going to happen from mere reactive habits.


Social fame is fundamentally unscalable, as it operates in limited room on the scene and even less in the few spot lights.

Benefits we can get from collective works, including scientific endeavors, are indefinitely large, as in far more important than what can be held in the head of any individual.

Incitives are just irrelevant as far as global social good is concerned.


That's framing the topic completely out of the issue with global impacts of humanity on ecosystemic sustainability, including biodiversity.

Less commut and more collective transportation is going to be far more significant in term of global impact, whatever the engine type.


You can do both! Better trains and more EVs replacing gas cars can be done simultaneously!

You forget the most important aspect of policy: it can't cost a single dime, and everyone must lie about that. Read the first sentence of the article:

"When California neighborhoods increased their number of zero-emissions vehicles"

Obviously neighborhoods/cities/states didn't increase anything. It was just rich people living there buying fancy cars. Of course, this needs to be described as a great accomplishment of local government.

And nowhere in the article is the obvious solution even suggested: advancing electric car technology so they're cheaper than ICE cars. And I don't mean charging extra tax while cutting public transport to make sure poor people don't go anywhere anymore, I mean fixing the technology so everyone has transport, for less money.


California government has a great claim to advancing the state of the art in EVs (and hybrids and just ICE before that).

Some people credit Tesla with kick starting the EV revolution. Californian governance kick started Tesla.

Their EV efforts go back to the ZEV mandate in 1990.


> obvious solution

Shouldn't the obvious solution be based on observable reality? Which is that there is no technology in sight that will make EVs cheaper to build than ICEs. Otherwise you are praying for a miracle, and that's not a sound policy.


Technological advance can be modeled like anything else. Everything about plug-in EVs is cheaper than ICE cars, except the battery. So you can model exactly what you need to get the same as you're currently seeing with solar panels. You can calculate at exactly what point they'll take over aviation and so on.

I mean, this isn't even a very hard thing to model.


> I mean, this isn't even a very hard thing to model.

Could you please explain in more detail what exactly do you want to model here? Above, you mentioned "advancing electric car technology so they're cheaper than ICE cars". Now as we both know the issue is with the battery, do you just want cars with battery so small that the car is cheaper than an ICE but nobody wants it? Because there is no need to model that, it has been tried and failed.

If you mean modeling battery technology that's not yet available in EVs, good luck with that. There are better batteries available than in mass-market vehicles, but they are not cheaper; cheaper technologies are not as good. Sure, in 10 years the batteries will be much better overall, but we don't really have the luxury of waiting until the technology gets perfect and then scaling that, do we.


Decent public transport makes all the difference. Luckily we have good transport here in the Netherlands and I haven't needed a car in 10 years. Also, the trains here have been running 100% on renewable energy since 2017.

Is it just a frontend CLI calling remote external logic for the bulk of operations, or does it come with everything needed to run lovely offline? Does it provide weights under FLOW license? Does it document the whole build process and how to redo and go further on your own?

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: