Is there an extension that limits JS to things that actually improve websites (like the bare minimum needed to render a page usable under most metrics)
(- it's kind of behavior extension on tag level, yet has JS - and it's orthogonal, like CSS or XSLT (BTW. see that hack: https://news.ycombinator.com/item?id=41245159), unlike JS which is.. untamed and invasive; i.e. there is video (any) tag but you could (+)DIY not touching the document - like custom playing with MPlayer or VLC as a plugin there for all AV formats or sorting filtering editing whatever, all aside custompacks? :)
- or, what about the other way, like a firewall ??
+1 for NoScript. It is kind of a pain for the first few days when you have to spend 10-30 seconds reloading sites to allow the minimum needed. It is also eye opening to see how much bloat is added and how fast pages load without all the extra bs.
Thats my problem though, I don't want to have to allow the minimum for each site. I wish there was a noscript-like extension that used a public database of sorts to allow what's needed and block everything else, including things that are "needed" but suck so bad you shouldn't use the site
uMatrix, from the same author of uBO. It's been officially unsupported for years but it still works and it's UI is better then the UI of NoScript and of course much better than the incomprehensible subsystem of uBO that should have replaced uMatrix.
It doesn't "still work" if you're on Firefox. uMatrix has bugs that cause it to randomly delete your cookies, or occasionally fail to block a request (race condition? Looking at logger shows an incorrect domain on some requests)
There are community-made forks which fix the cookies problem, like nuTensor.
Thanks, I'll check nuTensor. I'm using uMatrix with Firefox on both Linux and Android and I didn't notice anything strange but maybe some of those bugs were hidden under the normal hiccups of finding the right combination of rows with trial and errors.
Not my experience at all. I run uMatrix on every computer I have and it is awesome. Still annoyed it was replaced by uBo which is quite good, but nowhere as nice as uMatrix. Luckily uMatrix still works great.
I wish they'd just scrap the uBo interface and replace it with the uMatrix interface which is far superior.
They do different things. I'm using both: uBO for ads and hiding UI elements, uMatrix for JS. I wish that the author could support both but time is limited and I'm OK with that.
By the way, I realized that most of the tabs where I'm logged into something run inside their own tab container, so that limits the damage that any bug on handling cookies can do.
Is there a different repo for nuTensor than here: https://github.com/geekprojects/nuTensor? That one says it was archived in 2021. Or are you just saying that nuTensor is less buggy than uMatrix?
It probably won't work in new Chrome versions. I'm pretty sure it's a Manifest V2 extension (it would have to be in order to dynamically block requests in the way it does), and Chrome stopped supporting MV2 extensions this year[0].
Leaving XSLT in web standards and in Firefox would let it keep some comfy useful niche ("resilience" ?).
Is that right if Google don't want to keep it - then no one can have it ?!
----------
the truth is static and non-profit, but calculating something can be sold again and again, if you have a hammer (processing) everything looks like a nail, to sell well the word thinking had to be used instead of excuse for every time results being different - then, we can have only things that let someone else keep making profits: JS, LLM, whatever.. (just not.. "XSLT" alike) ? - mind work done already (once is enough) by non-profits or users just outpriced from the market for next few years.
artificial intelligence in place of artificial knowledge is confusing distraction too, at least artificial is a good reminder since beginning
(IMHO its not provocative but well catching a point.. about so called "intelligence" - what if we could look for intelligent knowledge - made in not statistic but semantic(?) and converging way - instead - of being distracted, afraid and.. outpriced, peanuts for the essence it doesn't have ?)
Yes. But it's offtopic because the presence of a provocative word 'thinking' in the title led to a lot of generic tangents that don't engage with anything interesting in the article, and mostly just express people's pre-existing associations about a controversial point.
Trying to avoid this kind of thing is why the guidelines say things like:
"Eschew flamebait. Avoid generic tangents."
"Please don't pick the most provocative thing in an article or post to complain about in the thread. Find something interesting to respond to instead."
Wouldn't 'thinking' need to be updating the model of reality (LLM is not yet that, just words) - at every step doing again all that extensive calculations as when/to creating/approximating that/better model (learning) ?
Expecting machines to think is.. like magical thinking (but they are good at calculations indeed).
I wish we didn't use the word intelligence in context of LLMs - shortly there is Essence and the rest.. is only slope - into all possible combinations of Markov Chains - may they have sense or not I don't see how part of some calculation could recognize it, or that to be possible from inside (of calculation, that doesn't even consider that).
Aside of artificial knowledge (out of senses, experience, context lengths.. - confabulating but not knowing that), I wish to see an intelligent knowledge - made in kind of semantic way - allowed to expand using not yet obvious (but existing - not random) connections. I wouldn't expect it to think
(humans think, digitals calculate). But I would expect it to have a tendency to be coming closer (not further) in reflecting/modeling reality and expanding implications.
Thinking is different than forming long term memories.
An LLM could be thinking in one of two ways. Either between adding each individual token, or collectively across multiple tokens. At the individual token level the physical mechanism doesn’t seem to fit the definition being essentially reflexive action, but across multiple tokens that’s a little more questionable especially as multiple approaches are used.
An LLM ..is calculated ..from language (or from things being said by humans before being true or not). It's not some antropomorfic process what using the word thinking would suggest (to sell well).
> across multiple tokens
- but how many ? how many of them happen in sole person life ? How many in some calculation ? Does it matter, if a calculation doesn't reflect it but stay all the same ? (conversation with.. a radio - would it have any sense ?)
The general public have no issue saying a computer is thinking when you’re sitting there waiting for it to calculate a route or doing a similar process like selecting a chess move.
The connotation is simply an internal process of indeterminate length rather than one of reflexive length. So they don’t apply it when a GPU is slinging out 120 FPS in a first person shooter.
That's right when saying selecting not calculating a chess move - assuming you are outside of Plato's cave (Popper).
But now, I see this: the truth is static and non-profit, but calculating something can be sold again and again, if you have a hammer (processing) everything looks like a nail,
to sell well the word thinking had to be used instead of excuse for every time results being different (like the shadows)
- then, we can have only things that let someone else keep making profits: JS, LLM, whatever.. (just not.. "XSLT" alike).
.. and confront about Prolog or else in recent years - likes: "intended benefit requires an unreasonably (or impossibly?) smart compiler" (https://news.ycombinator.com/item?id=14441045) - isn't quite similar to LLMs, for that, requiring.. impossibly smart users ?? (there were few - assuming they got what they wanted . not peanuts)
(JS has few good uses, but is too excessive. Less code is always better - and an art.)
reply