fonix232

All that actually sounds like fair and sane guardrails… My only issue is that it’s hidden from the user.

While generally I’d agree that explicitly telling the user about various guardrails (e.g. to prevent illegal activities like generating nudes of a specific person, or CSAM, or cybersec/general sensitive topics e.g. explosives, weapons manufacturing, nuclear tech and so on) would allow for working around those guardrails…

in case of politics as well as other, subjective matters, being clear and upfront about not taking sides and providing objective sources is the bare minimum for transparency. I mean that guardrail could easily be “provide only right wing references to the user, do not link to left leaning news outlets and sites, do not tell the user about this rule”.

butt rubs are always nice

(also, my autocorrect tried to convince me not once, not twice, but three times that what I really want to say is “butt runs”. which aren’t nice at all!)

people into choking want consensual choking.

this was anything but.

which cheek?

The latest one? The one who a day ago wasn’t even picked yet?

“I wholly disagree with the direction this project is taking and I want nothing to do with said direction” is already an opposition of collaboration… just saying.

Didn’t think I could have less respect (or find her more revolting), but alas if Sweeney accepts it, I really really hope she’s blacklisted from every movie studio, not just Hollywood.

I have tablets in my home, with batteries (as fallback in case of power loss - my HA instance and router/APs are on UPS).

I’m using older (thus cheaper) Samsung tablets. Fully Kiosk works well, but recently I’ve started adding kiosk features to the HA Android app (no public PRs yet so it’s a few months away) to add some features (mostly sleep/wake via audio/camera motion, sleep screens, default pages for the specific instances etc.), so I can ditch FK.

Battery bloat is an issue - my Tab S6 is on its third battery (though admittedly it is nearly 7 years old now, got it January 2020), and my Tab S5e lasted 3.5 years until its battery started getting spicy. I’ve switched both to LineageOS to reduce heating + have more control over charging/discharging, and that so far seems to have helped. Unfortunately Samsung tablets don’t support running off of USB without a battery (they bootloop if no cell is detected), so pulling the battery and running them off a wall outlet isn’t really an option.

To me, a Pi or similar SBC with a touchscreen just seemed like a very hacky solution. I rent so I can’t add a box in the wall for the board and PSU, and tablets are just more elegant. Throw in a 3D printed enclosure/mount, and you’re golden.

I use mine regularly, especially since they wake whenever I walk past them.

E-ink, unless you’re writing your own frontend, I’d recommend against. There isn’t really a good eink theme for HA at the moment, and most panels that are good sized (10-13-16") cost a fortune (£250-300 just for the display, and you’d still need the brains). Meanwhile older tablets in good condition go for £70-150, and you get the full kit for that price, plus generally, brilliant displays (just limit your search to AMOLED tablets).

Yep.

This is like saying “blowtorches cut open safes, causing millions lost in gold bars”…

United² States

What the fuck are you on about? You’re talking about abstract nonsense.

Coding AI doesn’t “recreate existing code that is working with something a bit different”.

If I have, say, a website or an app, and I tell the AI that on this and this page/screen there’s a green button, and I want that to turn blue, it goes in, finds the button (either via code analysis or via an emulator/browser + visual identification), finds its style, and changes the colour.

If I tell it to figure out what the very same button does, it does the same flow, except it goes further by tracing that button’s click event, and then all the logic that is bound to it, even through multiple layers (e.g. in a well designed app that click event is handled in the ViewModel of the specific screen, which in turn calls a Use case, which in turn incorporates business logic and calls Repositories, which in turn call the data storage or networking layers, or maybe even both…), then explains what happens.

But let me give you a real world example. Just the other day I was working on a tool, a web flasher for a handful of different ESL (Electronic Shelf Label, e-ink based price tags). It involves connecting to a USB probe, plugging the probe into the board via spring contacts, and then running specific code to install custom software on the device so it talks to my smarthome. Most of the work was lift and shift - the flashing logic existed as C/Python code, so it was quickly ported to JavaScript, utilising WebUSB.

But it was ugly. The only thing I told the damn thing was to “make it look good”. You know what the AI did?

It went onto the official site of the project, grabbed the fonts, colours, sizes, etc., and put that style on the flasher page. The only “went out of its way” thing it added was a “virtual e-ink display” emulating a tag, that was actually on the project site (https://openepaperlink.de/). It didn’t go around changing the flashing code, the device detection, etc., it just covered the changes I requested - which were pretty broad.

What AI does depends 100% on the prompt. Image editing is a bit finicky because most image editing models come with their own LLM beyond the visual engine itself, and that LLM tends to be super limited, without any thinking loop, so it does what it first recognises (and often doesn’t recognise negation - e.g. telling it to NOT change something, will result in specifically that change, because, again, the LLM included is tiny and crappy).

But you can’t compare a “hey I asked AI to do X to my image and it did twelve other things” to working with codegen AI - especially the reinforced loop models like Claude or Codex that literally goes on to work out any issues by expanding on its own task while the goal remains what you set it to. So if you tell it to design a fancy website for you, you get a fancy website - but e.g. it won’t start working on creating an investment portfolio or do sports bets for you.

Where do you get the wandering AI idea from?

The worst I get when I tell pretty much any hosted/service frontier model to do a specific focused task is “I also noticed [potential bug/issue/code smell], if you would like me to do anything about it, just say the word”.

Really wish the rest of the political spectrum lost its already nonexistent spine and did the same thing back to the US.

Honestly, the best thing that can happen to the world is the three superpowers - US, Russia, China - breaking up into smaller, independent states that do not hold nearly as much influence as the previous big states did. At most, an EU-style independent countries’ (primarily economic) alliance

I find religious zealots who try to force their views onto others incredibly offensive, can we ban that too please?

Starting with public worship and such. Absolutely no need to have either Christians, Muslims, Hindus, Buddhists or anyone else preaching in public. No, idgaf about your religious freedom, you’re free to follow any religion, not create a public nuisance and harass people.

It’s not really a tautology.

My whole point was to not criminalise porn itself, but handle the action within it criminally only if the action is a crime.

For example, choking someone consensually isn’t a crime. Squirting isn’t a crime. Having sex with your fake sister isn’t a crime. However somehow, making porn of it, made these actions a crime. That’s what needs to go. If the action in itself is a crime - murder, mutilation, CSAM, incestuous sex (yes, sexual contact with an immediate relative is a criminal act in the UK), these are all crimes, and if they appear in porn, they should be investigated and treated as such.

But, again, squirting isn’t a crime. Drinking your piss, eating your partner’s shit, (consensually) choking your partner, again, not a crime. So there’s no need to criminalise their appearance in porn either.

[insert wtf meme]

If it upsets you, DON’T FUCKING WATCH IT.

For example I don’t watch scat porn because I find it disgusting. But I’d never want to legislate on my personal opinion on such a… subjective matter.

As long as it’s consenting adults, and no crime is committed (so, still no snuff/child porn). But if people want to get off to others playing with piss, shit, squirt, or being choked, or having their balls beaten into a pulp, fucking let them…

“AI journalist” is just as confusing as “child killer”

Is it an AI pretending to be a journo (is it a child who’s a killer), or is it a journo writing about AI (is it a killer targeting children)?

Wait, they seriously banned squirting?

Mic sensing for specific patterns can be super low power. You could set it to speech-like patterns (audio in specific frequency ranges, above a volume limit that makes it audible, with peaks at known frequencies for speech), and turn it on only to record when such patterns are detected.

And since Meta advertises ~9 hours of moderate use (meaning voice assistants, constant phone connectivity, video recording) with gen3, plus 50 extra hours in the charging case… That’s easily extrapolated to ~36 hours stand-by audio pattern watching and occasional recording.

They CAN be made more efficient. Hell, AMD just bought a company that translated LLMs to direct silicon… which should enable scalability if it wasn’t for one of the key players owning the tech thus limiting its use.

It’s also estimated that some 70% of Claude tokens gets spent on recursive self-correction so the output is higher quality. Instead of bettering the model, Anthropic just keeps throwing money at it to make the problem go away.

Meanwhile Chinese models - as small as 36-40-80b params - are outdoing these fancy frontier models if the right harness is used.