Not made for this, doesn't have good speakers, doesn't have all the surround sound features that the airplay products have. Doesn't have multi user support. These all could be solved by improving the iPad (increasing price drastically), but why do that when you can make a specialized product that is cheaper than an iPad for those who want it, instead of charging everyone for some stuff they won't use?
LLMs are still being detected at the same level since the day coding agents were actually something, hence, it's not improving for 3 years. Why would they be undetectable next month?
Moreover, the parent comment linked to a decent PR presumably human made. If an LLM had written the PR, its description would be 9 paragraphs long and the code changes full of assinine comments.
If the author really did one-shot the PR, then the author must have also put in the extremely detailed and tedious work which would be required to tame LLM output for this specific repo, in which case, they actually put in real effort, applyed expertise and thoughtfulness, which is the whole issue at hand.
You think coding agents or models have not improved in 3 years? From my experience they absolutely have. I've spent a lot of time getting Claude Code to write clean features in a brownfield repo. I've gone from it helping me speed up to now Fable 5.1 doing it perfectly. And there are steps one can take to improve the quality which is to teach it (SKILLs) your important design and coding standards, as well as automatic design or code review to catch edge cases, performance issues, or whatever aspect you find a high priority.
With all due respect (not that I feel like much is due after that response) I am not really sure you know what I am talking about. I'm talking about inference providers, like inference.net, Fireworks, Coreweave, Digital Ocean, etc, to use DeepSeek 4.1 Flash. They didn't create the model, they just are charging you to run inference tasks. That is a different story.
DeepSeek themselves are honest about the fact that they train on inputs by default. You won't hit DeepSeek if you use OpenRouter with ZDR enabled.
I now have two people with sarcastic replies that seem to think I'm talking about OpenAI. We're talking about DeepSeek on third party providers. Given that this thread is neither that long nor that confusing, I'm not sure how people are getting lost this quickly.
I don't really trust OpenAI, although honestly I find it stupid to suggest they'd offer a ZDR policy and violate it. They literally don't have to offer it. People will still pay. Fable doesn't offer ZDR at all and it hasn't stopped people from paying through the nose for it at API pricing.
Off topic, but say you have a program that needs to calculate something but takes a long long time, how to counter hardware failure without needing to calculate everything from the start?
Typically you use redundant computations, which don't have to be done from the start. You could calculate intermediate values and check them before proceeding
Google Maps used to work without it but recently gained a hard dependency on Play services. We could add more shims to make more apps work without it installed but that hasn't been a focus yet.
Makes sense. Any idea if there's any interest in a shim to get notifications working sans Play services?
I haven't looked at anything shim related in quite a while and am not super familiar with the notification stack for Android, so if that's a stupid question I apologize.
2.1MB is a huge most of the world where the internet speeds are not gigabyte etc. Yes, storage wise it's not a lot, but it's extra 5 seconds or so the user needs to wait for the site to load.
There are only a few countries in the world where 2.1 MB extra would add (barely) more than 1 second to site download based on the country’s median internet connection speeds. The vast majority of countries would see less than <100ms from 2.1MB.
And the bigger thing here is that this is all cached after it’s downloaded the first time.
reply