appcustodian2

it's web scale!

You're not wrong, but you could ask your favorite LLM to interact with the LG API's to switch the input for you. At least for my case, it was ~10 minutes of work to develop a small tray app that looks for a specific USB device and switches the TV input if it gets inserted. I don't really even touch the remote anymore

i'm sure that will last

Just in case it didn't occur to you already, you can just build whatever coding agent you want. They're pretty simple

Not just that but it will probably break later and ruin everything

N model is basically just N-1 model with revised context window handling and more compute thrown at it

I would think that it is still capable of exploring the codebase and reading other related files like any other coding agent already does.

source on the 8096 tokens number? i'm vaguely aware that some previous models attended more to the beginning and end of conversations which doesn't seem to fit a simple contiguous "attention window" within the greater context but would love to know more

You think people open sourced things mostly because of license obligations?

how do you know when a page is "settled"?

Don't worry we'll all be taking the Claude certification courses soon enough

How do you figure anything out? You go use it, a lot.

It's extremely fair to compare depreciation against MSRP. What's not fair is to say that they were being "sold new at massive discounts" when in reality it's an asterisk-ridden rebate process that applied to one model year under specific circumstances. That article was spam when it was written, can you provide a first party source for these massive discounts?

I won't argue whether their "human-like" marketing is dumb but I will argue that whatever LLM's are doing is plenty sufficient to find the vast majority of vulnerabilities. Don't tell my employer I said that though

P2P “Share/Collaborate” mode: the UI text and toasts (“Share”, “Close connect”, “connected”, “No connected.”) plus e.rtc.user strongly suggest a feature where someone “hosts” a DB and others connect directly to run queries/see results live.

is your comment supposed to be pessimistic or optimistic? my interpretation of your post is that we should be investing more into AI because eventually we will achieve human-like AI running on 20 watts of power. we know it's possible because as you said, our brains are doing it.

Yes as I mentioned I use often-recommended knives (victorinox, shun) and have them occasionally sharpened professionally and at least in my case the ultrasonic knife appears to solve some very real problems that knife maintenance cannot.

just want to point out that maintaining a knife is a whole hobby that requires a time investment of learning the skill and storage of additional tools and materials. i'm surprised at all the apparent knife enthusiast posts trashing this device. I take my victorinox (which is absolutely nothing special and surprises me that it costs $60+ dollars) to the farmers market for sharpening but sharpness isn't even the problem. Potatoes in particular stick to the blade like a strong magnet and it takes me 5x longer to prep. I enjoy cooking but not chopping endless veggies and i'm hoping this thing can carry some of that weight without looking like i'm using an oversized electric toothbrush.

> All of the approaches of driving the browser outside of the browser is going to be slow

Why? I would think any cross-process communication through the CDP websocket would have imperceptible overhead compared to what already takes long in the browser: a ton of HTTP I/O

What is Karma? What are you executing in the browser?

spoken like someone who has never had to operate on someone else's machine that they provisioned for you on an isolated network