Radan

AI industry horrified to face largest copyright class action ever certified Copyright class actions could financially ruin AI industry, trade groups say.

"Copyright class actions could financially ruin AI industry"

Rooting for the authors! These tech companies need to be told that they can't just steal others work and profit off it

arstechnica.com/tech-policy/...

Digging into exactly how Hotwire Native integrates with Turbo Drive to be able to convert web app page navigation into native looking screens navigation: radanskoric.com/articles/tur...

Rails Blocks - Rails UI Components using Tailwind CSS and Stimulus Rails UI components for building modern web applications, simply copy and paste the code into your app.

Really nice to see a new library of frontend components based on Stimulus. And it's looking very polished: railsblocks.com

Didn't even think about that. Talk about double standards.

Absolutely! I think another related difference between LLMs and Humans (as well as Humans that convert experience to knowledge better or worse) has a lot to do with building mental models. Experiences that are not intentionally converted into mental models tend to be shallow.

Ok, I'll be honest. I was a bit skeptical about the value of the addition of "it" as an implicit keyword for the block parameter. Until now. Now I'm sold on it. :) #ruby

Master Hotwire: Master Hotwire to Build Modern Web Apps with Rails Simplicity “Master Hotwire” is structured to help you build a strong mental model of Hotwire. It combines explanations of functionality with hands-on coding. The book guides you through building a collaborative ...

I've added an all new section on Hotwire Native to "Master Hotwire". I've put over 70h into making it the fastest way for experienced Rails developers to get up to speed with Hotwire Native. But I'll leave the old price until the end of the week: masterhotwire.com

That's a great take. The wording confused me at first, because for humans it's impossible to gather a lot of experiences without at least some level of understanding. But since LLMs are not brains (despite frequent anthropomorphization) they have no problem with experience without understanding.

I'm trying to shape a thought on LLMs, wondering what others think. Here's what I got: "Understanding" was always more valuable than just "knowing". And LLMs give "knowledge" for free but don't "understand". This must skyrocket the value of "understanding". So this is where human effort should go.

How would he know that he's at the edge of Quantum Physics?? :D

Next Tuesday I'll add a whole new Hotwire Native section to masterhotwire.com. It's a deep but concise intro written specifically for experienced Rails developers. I'll up the price a few days later. Also, a discount will go out to my mailing list on Tuesday. #hotwire

Finished going over all 170 comments from beta readers of second part of "Master Hotwire". I'm so much happier with it now! Getting other people to critique your work is always sooo useful, from code review to beta reader feedback.

People outside tech starting to realize how quality software and typing/generating code fast are not correlated

I'm following your writing and it often makes me think in stimulating ways. :) I think the interest will eventually swing back. AI prompt guidelines are increasingly emphasising being precise. Eventually the industry has to remember that the whole point of programming languages is to be precise.

Excellent example! And there are others in their niches (R, Prolog, Mathematica ...). And the children of LISP are extremely malleable. I'd argue Ruby is the most expressive while still easy to read and general purpose. And I don't think we're anywhere near done with the evolution of languages ...

If you can read mathematical notation, it's usually the shortest way to express a mathematical concept. Certainly shorter than English. Every time I see an AI prompt be much shorter than the code it makes me wonder what we're doing wrong that the code can't be more expressive?!?

I guess tomorrow you'll bring up axiom of choice? :D

That far I still felt comfortable. I started to feel uneasy about there being infinitely many uncountable infinities because every power set (e.g. set of subsets) has higher cardinality than the set. It all makes clear mathematical sense but at this point my mind went: "C'mon, really?"

You're right, I always viewed it as a generalisation of explaining set sizes to a toddler: "See, if you put each thing from this bag next to a thing from the other bag, everything has its pair and the bags are empty ... no, don't eat that" A bijection is a really elegant way of generalising that.

Been using it for a while, pretty happy with it. 🙌

I also like to use `bundle open` (bundler.io/man/bundle-o...). It opens the gem folder for the current Gemfile install. It also makes it easy to drop `debugger` in the gem source, restart rails, rerun request, and now you've got a debugger session triggered inside the gem.

This is such a bad idea on so many levels, it's hard to begin. The whole point of asking submitters to type out a detailed report is to have them think it through. The actual issue is more of a byproduct than a goal in itself. Well done Github, you're encouraging elimination of the valuable part.

Render the “download now”button as you planned but also attach an “autoclick” stimulus controller that clicks its element on connect.

If you put in the effort to make the code easy to understand, Ruby is there to help and support you all the way.

Does anyone know why pretty much all LLMs like to use em dashes so much? They'll eager-use them near-everywhere. It's weird. Where in the training data could it be coming from?

Turns out, creating a simple AI coding agent is not very hard, especially in Ruby: radanskoric.com/articles/cod...

Multitasking was always a bad idea. At best humans are faking it, or, in the case of me personally, failing miserably. And in the age of AI it's full expiry date is around the corner: anything simple enough that humans can multitask it will soon be automated. Best to focus on the deep, human work.

I love reading these honest behind the scenes stories, thanks! :)