Vulkan?
I'm not too familiar with either, but this article goes into more detail: A Comparison of Modern Graphics APIs
Ollama uses the Metal API on Apple Silicon Macs for GPU acceleration.
How does one measure code quality? I'm a big advocate of linting, and have used rules including cyclomatic complexity, but is that, or tools such as SonarQube, an effective measure of quality? You can code that passes those checks, but what if it doesn't address the acceptance criteria - is it still quality code then?
What I got from the article is an example of how generative AI can fix a bug, if you provide it with a reproducing case. Yet funnily enough, the AI introduced a bug in the first place by using an older version of a dependency.
The author of the article is Dan Abramov, the co-creator of Redux and a prominent React contributor. Putting aside what you may think of vibe coding, there is little doubt that he is an experienced developer, that knows what he is doing.
A good companion piece to this article, is the Dead Framework Theory article, which discusses AI coding tools bolstering React's dominance.
React is very much the dominant framework, and that dominance is being bolstered by AI coding tools. I wonder if an AI coding tool would even suggest Preact as an alternative to React? 🤔
The author does make some good points about colours as visual cues, instead of just making things look colourful. I have to admit prior to reading this post, I always picked my themes on aesthetics, but it has made me think about colour as utility.
Preact is not new, it's been around for about a decade.
Exactly but generative AI has exacerbated the problem
What is new is the scale of the problem being created as lightning-speed code generators spew reams of unread code into millions of projects
My understanding is that an example of a hypothesis, is that users want a feature. The experiment is putting that feature in front of users, or performing user research, which which then allows you to validate if a hypothesis is true or not.
How strange. It was definitely working when I shared it.