Yeah my thoughts exactly. Definitely slop. I have no objection to using AI to help writing. I just don't want to read the same sloppy cliches again and again and again. The short sentences. The Bigger Picture. Here's the rub. It's not just A, it's B.
It's like those cliche titles - for fun and profit, the unreasonable effectiveness of, all you need is, etc. etc. but throughout the prose. Stop it guys!
Sure. Short sentences like "It shouldn’t be.", "I’ve moved on.", "Ollama didn’t.", etc.
Not-this-but-that like "The local LLM ecosystem doesn’t need Ollama. It needs llama.cpp."
Weird signposting: "Benchmarks tell the story."
Heres-the-rub conclusion: "The Bigger Picture"
Starting every title with "The ...".
It's definitely largely human-written, but there are enough slop-isms to make it annoying to read. And of course it's totally possible for a human to write an an AI style, but that doesn't make it any less annoying.
Totally impossible. Closed source software often contains IP licensed from other entities. Just because a company folds doesn't mean they can violate licensing agreements.
> Just because a company folds doesn't mean they can violate licensing agreements.
It does if that's the law. Every jurisdiction routinely overrules contracts as unenforceable on the basis of some overriding law, so it wouldn't even really be that unusual. Whether it's a good idea or not is another question and one that depends almost entirely on second, third and higher order effects.
There probably is a world where all software is libre software and we still see similar rates of development, but it's not at all clear how you could get there. Especially not if you cared about the damage caused by upending the business models of a significant fraction of the world economy.
The biggest blocker there is probably whatever remaining creditors to the company when it goes under then have claims on remaining assets like the software.
One solution would be putting something in the tax code such that donating the code to an open source foundation gives a bigger benefit than simply writing it off as a total loss and destroying it.
I wonder if the YUV conversion could be offloaded somehow to the ARM inside the Hollywood or somehow using a shader (or equivalent) if the graphics were accelerated - though maybe this is way way too much.
I considered this! There were a lot of things I wanted to try but didn't want the timeline of this project to blow up any more than it already had. Now that I've done the hard part of writing about it and publishing it, I can revisit some of these ideas :)
At the top could have been a link to equivalent llamacpp workflows to ollamas.
I wish the op had gone back and written this as a human, I agree with not using Ollama but don't like reading slop.
reply