I finally delivered on the promise of both writing about it and having a place to put all my pdfs.

jimmyhmiller.com/ai-own-your-...

Working on it. I have never organized them like I should and doing that. Will probably be a blog post about it :)

How do you always find such amazing papers? And do you keep a list or collection fill them anywhere?

@spiralganglion.com and @todepond.com are here as well

Super clean implementation! A good amount of separation without too much. I’ll definitely be following along.

For the longest time I shied away from low level programming. I always found it intimidating. It turns out it isn’t any harder than other programming. It’s just that everything written about it assumes too much. So here’s my attempt at an intro to machine code

jimmyhmiller.com/machine-code...

As of now, ai generated content does not have copyright because there is no human author. Given the requirement for a derivative work to be a substantial transformation the bears the personality of the author I don’t think llms work here.

If that’s what you want to guard against but are okay with people mirroring or remixing your content as long as they aren’t making money, your current plan seems fine. The mechanism to protect yourself if people do violate it is a copyright strike and/or legal action.

If you do MIT for the code, people can definitely use it for commercial purposes. If you don’t want people to use the code for commercial purposes, full stop, no open source license does that. AGPL often in practice stops people from using it commercial (because of its virality)

Can software express critical reflections on its own nature in the same way post-modern architecture does?

I don't have the answers, but I wrote a long text with some early thoughts: tomasp.net/architecture/

Without the flash editor, what does it look like in practice to make new flash apps? Just a bunch of actionscript files?

I hooked up my orthoremote to stepping in my debugger in vs code. The whole thing was a simple 50 line node script that connected to the Bluetooth and then issued keyboard commands.

Early in my career, I was convinced there was “good code” and “bad code”. That you could look at something and — without knowing anything about the context — pass judgement on it. These days, my views are much more nuanced. I try to adapt my style to the situation. The code I write, and the process I use, depends a lot more on what the goals are. Am I trying to bang together a quick prototype to learn something? Or am I fixing a bug that might affect hundreds of thousands of users? My approach would be completely different in those two scenarios. Years ago, I read Edward De Bono’s Six Thinking Hats, which describes a framework for problem solving and creative thinking. The idea is that you can “put on a hat” to deliberately adopt a specific mode of thinking. It’s a bit corny but (imo) there’s a useful idea there. Maybe this applies to different styles of programming, too? What “coding hats” do I use?

New blog post: Five coding hats
→ dubroy.com/blog/five-c...

Yeah it's kinda goofy, but maybe also useful?

In writing, there is a distinction between those who love to outline and those who discover their story through writing. We seem lack that distinction in the programming world. So I wanted to introduce it.

jimmyhmiller.github.io/discovery-co...

Representing Type Information in Dynamically Typed Languages by David Gudeman does a fairly good job describing a bunch of different approaches.

citeseerx.ist.psu.edu/document?rep...

Also enjoyed arxiv.org/pdf/2411.16544 for something a bit more recent but not as comprehensive.

Do you have a feature that will let people distinguish them? Do you want such a feature? My language will have them not be equal because 1) all objects have built in structural equality that will look at type as one aspect. 2) I will offer reflection. I don’t know the needs of your language

Thinking in a dynamic context, I’d say it depends on if your language gives you a way to distinguish between the two. If you can check equality, or you can check parent type, etc. If so, then by Leibniz’s law they aren’t identical.

Such a fantastic paper! Always happy to see others enjoy it. So many people miss the richness of Naur’s notion of theory. Glad to see you get it.

One thing we did on the future of coding podcast was reading Ryle’s notion of theory. Theory being about know-how is often missed

I think the most likely answer is you don’t. And that’s why we need to be building operating systems.

Thanks for the encouragement :) Not sure I’ll make a sponsorship for something like this. But I definitely plan on doing more of these through the year. Though everyday was a bit much. Only sponsorship thing I have now is the future of coding patreon

Interview | Future of Coding Get more from Future of Coding on Patreon

FUTURE OF CODING
INTERVIEW

featuring @unkai.bsky.social

patreon.com/posts/118714678

Final Advent of Papers post: Against a Universal Definition of ‘type’.

This paper is fantastic. Both haters and fans of types should read it. If every paper in computer science were as good as this one, this series would have been way easier.

jimmyhmiller.github.io/advent-of-pa...

Thank you so much for sharing that story. I agree. I wish it weren’t either.

Advent of Papers day 23: Do Artifacts Have Politics?

I said no classics, but I made an exception for this one. It is undoubtably a classic, though one I don't think makes HN lists of papers, nor many CS curricula

Your code does in fact have politics.

jimmyhmiller.github.io/advent-of-pa...

Advent of Papers day 22:

jimmyhmiller.github.io/advent-of-pa...

Yes we of course do conceptual engineering. That’s what I’m saying too.

But because we don’t make reify as a first class concept, we don’t have papers/books that directly talk about all the concerns. How do you get people to adapt your concept? How do you preserve it? And many other questions

Yeah I wouldn’t saying naming thing is the same as conceptual engineering. Yes you end up using words to stand for concepts. But conceptual engineering about the concepts themselves. Is model a good concept? What purposes do we want model to serve? It’s not about the name

What is Conceptual Engineering and What Should It Be?

For day 21 of advent of papers we are covering a topic completely ignored by software engineering, but crucial to it, Conceptual Engineering. How do we fix our broken concepts? How do we make new ones?

jimmyhmiller.github.io/advent-of-pa...

I would definitely do that paper for this series! But we already recorded (unreleased) a future of coding episode about it. And I’m trying to avoid that.

Advent of Papers day 20

Amnon H Eden argues that computer science is really three different paradigms. Rationalistic, Technocratic, and Scientific.

This paper argues that we should abandon the first two and "be scientific". Are they right?

jimmyhmiller.github.io/advent-of-pa...