dabbz

Sounds good: http://simbook.56k.net/

I mean, there’s more privacy protection on payment data than there is for email address and browsing habits. Plus most companies just push out payment handling to stripe. So most websites have little idea who actually transacted with them to begin with (email dependent).

This bothers me so much with the existing offerings. I start with the chat interface then as soon as I want to get technical/run scripts/automation, I have to copy the context into a fresh code session. So cumbersome.

Metronome has been experimenting with real-time payment on usage based billing for tokens. That technology leverages micro payments using crypto. It's cool but all it's doing is moving the accumulated balance from a provider onto your crypto wallet.

I mean one of those makes them look better than the other. That makes it tricky to be able to tell what is truth vs half truth vs a lie. You're right though, based off of what they claim, their great success is the cause of their failure.

How hard is it to up scale their servers though...

I was going to interject that very sentiment. The upfront costs of running any model of value to the average consumer is so large to be basically non-existent. We need hardware to become commoditized and capable of handling LLMs at the scale necessary to enjoy these agentic workflows.

Hey everyone! I found the COBOL developer!

Your analogy falls apart somewhere for me but I can't pinpoint the exact correction to make your analogy work for me. Either way I agree. If they don't like only being able to buy digital PS games via Sony's platform, go to one where they can.

That's not the fight I'd personally focus on. I'd personally focus on them revoking digital access after you've "purchased it". Then again, I don't have a game console and I don't "buy" digital media. I either rent via streaming subscription or own the physical.

Weird conclusion to draw from my statement. Feels intentionally hostile of an assessment about me, but I digress.

Machines don't pay my bills though. If the content was free to share widely, then yea, who cares how it's accessed.

I see a lot of discourse about it being fast-to-deprecation. But I see it a different way personally.

Modern LLMs are trying to do more with less. Focus on doing the right thing the first time. Even if we squeeze dumb LLMs, the significantly faster speed means quicker iterations. So a bad decision doesn't cost the time and inference costs that it cost before. It theoretically changes the scale of errant token spend.

I compare it to the 1 thousand monkeys on a typewriter. In this case it's 1,000 monkeys with stale training data of everything ever written and the ability to search the web.

Far better than the band 999mb, they don't do well in the gig economy.

This is my thought as well. Models have to be intentional about which tokens they burn because there's a real lag time. If you can just fork out 10 different reasoning sessions at once with no regard for token waste/lag, you can compensate a smaller model with just doing more at once with it. No idea if this is reasonably true though.

My company moved to GH Actions 6 months ago even after I pushed back with a "Are you sure given their reliability issues of late?"

Now I'm stuck twiddling my thumbs with PR checks stuck/failing...

There's a PLIST entry possibly relevant to it in `~/Library/Preferences/com.apple.networkserviceproxy.plist`

I couldn't figure out how to get any changes to reflect but maybe it's a good pointer towards a solution?

EDIT: Reading into this, looks like they're very intentional about not allowing arbitrary enable/disable calls to it, likely to prevent a hostile entity from switching it off on unsuspecting users. The systems setting calls an XPC into a daemon that makes a web request, then validates the response from the API to update the PLIST structure.

My least favorite ones are the ones that fit in the rails to the left and right of the content that scroll with you. You can't click anywhere without an accidental ad click. Using any device without ublock reminds me just how much better life is with it...

True, we have way too many thematic versions of Monopoly. I'm still missing so many copies.

Honestly I was more thinking about TVs than monitors but I often forget Dell makes good monitors.

It's getting really hard to find a solid, reputable brand of screen manufacturers that don't pull this kind of crap. I swore off Samsung screens after their popup ads in phones and TVs. LG was my next choice but this is deterring me even further.

So OP needs to rename their slideshow tool to Claris?

I have been using Plex since 2014, paid for the lifetime pass back then. It is by-far the most used piece of software I've ever relied on (excluding any general purpose software like OS/Browser).

I have my family setup on Plex, my automations, my apps, my preferences, etc. I'm so comfortable in Plex that until they do something catastrophic to my experience, I can't find the motivation to train my mother to use another app.

This reminds me immensely of DARK Basic built by The Game Creators. It was how I got my start into programming back in the day.

I've found personally it's better to use AI to build a deterministic script for calculations like that. (anything that manipulates data should be a script not an AI).

I suspect we'll see a hybrid before an all or nothing. Local models for computer control or delegating, online models for things that need strong reasoning, planning, and knowledge access. Again, I'd be more than happy to be wrong. I just see models growing faster than the hardware can.

I agree that would improve transparency but it'd also lock us and Anthropic into a situation where changing that would require new plans, prices, notifications, legal, etc. I think the fact that we don't see tokens shows just how subsidized the plan is. Which is mostly the point. They want us getting used to just using the product, not tracking tokens (for whatever behavior that builds later on).

I mean if you want to spend $200, and get a guaranteed amount of tokens in return you should be using the API.

I agree there should be more transparency what 20x gets you that 5x doesn't.

Though I also imagine it's a tough problem when you could be using Dispatch, Cowork, design, chat, code, etc. All of which use different contexts, models, and resources. I'd argue they actually need to either simplify their offerings, or charge more for upsells (charge for cloud-based agents, upcharge for design outcomes, etc).

I mean you got me there. There will be places who do have the means to build up massive GPU servers. There's just a lot more to it and I don't know if we're going to pinpoint an exact catch-all moat.

I feel like the models have no moat paradigm died when a single model expanded past the memory of single GPU slices. The moat is hosting the model. Even paying a server host to run a rack of GPUs has immense upstart cost, and then you're still struggling to compete on the add-ons of the things on top of the model (prompts, validation loops, etc). You can only throw so much money at a problem.

Literally this. Every response initially was "we don't see that and we let it sit for weeks". Now there's just no response from devs. I was unable to browse past more than 1 or 2 pages before memory ballooned to 30gb.

Software has been riding on the backs of the insane hardware growth curves for the last 20 years. I miss the days of reading about how software engineers had to delete standard C libraries in build time to shave extra memory so they can stream more of the level in.

I also fully acknowledge that change starts with me, unfortunately those changes don't pay the bills.