It works pretty well for me, but doesn't do nearly what I'd expect.

EG I can talk to it like I would chatgpt and it works well. But I can't be like "hey I want to get dinner with my wife on our anniversary, please book the best available option in my city for fine dining"

It's still way better than Siri, which feels like a voice CLI to me (same as Alexa, which is very low quality IME)

I don’t think I’d want to talk to a voice assistant like that. Maybe it’s a generational thing? Things like that are ambiguous enough discussing them with human beings and a big part of things like voice assistants is understanding how it’s going to interpret and execute a response based on what I say to it.

Yeah. Just like we do for smoking. I don't think I get your point. Are you agreeing with the prior commenter?

That sounds like a pain in the ass though. Explaining to my dad how to use Downloader on a FireTV is a lot easier than "hey dude buy a raspberry pi"

If there's a demand, someone will sell them ready to use, with preloaded microSD card, case, and cables.

There are a lot of Chinese devices out there with official android TV that do what firestick does and better with less ads and maybe even less tracking. People wanting to pirate will move to those. Raspberry pi is not cheap or the solution I agree.

I mean even among android tv sticks alone there are plenty of no name chinese options out there, these things are dirt cheap to make and are basically commodities.

any you would recommend that would compete with a fire-stick 4k Max level of hardware? and is open software, but that would also allow the installation of other media apps, like netflix, HBO max, etc?

Xiaomi has a really good one I've been using for years at this point. The 4K one. I'm sure there are more reputable chinese brands with alternatives as well!

I’m not in this market I have a plex server so idk.

My guess is that PT was chosen to highlight the fact that this happened in the middle of the night for most of the responding ops folks.

(I don't know anything here, just spitballing why that choice would be made)

Their headquarters is in Seattle (Pacific Time.) But yeah, I hate time zones.

It makes me think that a lot of the folks commenting on this stuff haven't actually used the tooling.

Agreed, it's generally quite accurate. I find for hectic meetings, it can get some things wrong... But the notes are generally still higher quality than human generated notes.

Is it perfect? No. Is it good enough? IMO absolutely.

Similar to many other things, the key is that you don't just blindly trust it. Have the LLM take notes and summarize, and then _proofread_ them, just as you would if you were writing them yourself...

I think the cost of inaccuracy is very a important factor in if it works for a specific use case. Meeting notes probably don't have much cost of inaccuracy. Medical records on the other hand...

Absolutely, I was just using this example in response to someone who specifically mentioned meeting notes. That’s an area where LLMs are a clear benefit ime.

I'm confused about Meta AI in general. It's _horrible_ compared to every other LLM I use. Customer ingress is weird to me too - do they expect people to use Facebook chat (Messenger) to talk to Meta AI mainly? I've tried it on messenger, the website, and have run llama locally.

My (completely uninformed, spitballing) thinking is that Facebook doesn't care that much about AI for end users. THe benefit here is for their ads business, etc.

Unclear if they have been successful at all so far.

Too much training on facebook and insta shitposts.

Counterpoint: I'm fine paying for convenience.

Wikipedia should still exist, but I'm fine giving OpenAI 20 bucks a month to make my life easier.

On the flipside, then you have to maintain instances of everything.

For most of what I run these days, I'd rather just have someone else run and administer my database. Same with my load balancers. And my Kubernetes cluster. I don't really care if there is an outage every 2 years.

Quite expensive to build though. Many of these companies don't have the sharpest engineers building multi-cloud.

IMO, going multi AZ or multi-cloud adds a good amount of complexity.

TBH I don't care if last.fm doesn't work for 8 hours a year, that isn't a big deal. My bank? Yeah that should work.

You're running local LLMs on an ipad?

as someone else mentioned, lots of apps to run local models. I also wrote my own app, in Apple’s app store, that uses Apple’s system model.

There are a number of apps that do this.