musesum

What I hope to do with refactoring is to distill namespace and common patterns into a DSL. I am very curious about what tradeoffs you found.

Am waiting for the next Mac Studio to come out to experiment with the "AI for everything" approach. Most likely, the open source distilled models will lower quality. So, another "price vs quality" tradeoff. Still, will be fun to code like I'm at a foundation lab.

Good idea. Am still a bit shy around token budget spend.

I wrote something similar in a Claude Code instructions.md: "minimize cyclomatic complexity" What happened next? It generated an 8 line wrapper function called only once from a different file. So, I told it to inline that logic in the caller. The result? One. Line. Of. Code.

So, I asked it to modify its instructions.md file to not repeat that mistake. The result was the new line "Avoid single-use wrapper functions; inline logic at the call site unless reused"

instructions.md is the new intern.

I wrote something similar in a Claude Code instructions.md: "minimize cyclomatic complexity" What happened next? It generated an 8 line wrapper function called only once from a different file. So, I told it to inline that logic in the caller. The result? One. Line. Of. Code.

So, I asked it to modify its instructions.md file to not repeat that mistake. The result was the new line "Avoid single-use wrapper functions; inline logic at the call site unless reused"

instruction.md is the new intern.

I wonder. I follow @DamiLeeArch on YouTube. She talks about architecture in the built environment. Ostensibly, built for Humans. At least for now.

Has California Law changed, since 2015?

My company was acquired by another company that had in-house lawyers. California Law states that you have rights to your own IP, when produced on your hardware and on your own time. So, I was careful to air gap all my work on a separate computer. Meanwhile, the acquiring company ask me to sign an employment agreement. Its terms restated the California law in very ambiguous terms. I couldn't tell if I was declaring rights to my own IP or signing away my rights. So, I asked them: "which is it?" Their replay was "Yes." ... I was an employee for a day.

Long ago, a proud moment for me was writing -20K lines of code in a month.

A few years ago, I tried that again with a remote team of 20 coders. I failed; I couldn't keep up with the barrage of pull requests.

Today, pair programming with Claude Code and GPT feels more like the later.

I think there is an opportunity here for smart refactoring. But, needs a larger context. I tried this on some legacy code with Cursor and Claude opus 4.1, but a million tokens is not enough. I dunno; maybe translate between a private and shared LLM. Has anyone tried this?

So, aside from trust the biggest barrier is lack of adaptability?

My chain of thought:

1) Aaron Marcus - who found optimal menu count to be 5 +/- 2

2) Magic number 7 +/- 2

3) Fitt's Law selectivity (bigger is easier)

4) Shared layout for mobile + desktop

5) I hate short form

6) Is 5) a non-sequitur?

7) No! I now have the attention span of a goldfish.

8) Maybe I should read a book

Personal context is tacit experience. Cultural context is aggregate.

Wittgenstein stated that the differences between personal and cultural language makes it impossible to agree on anything foundational to philosophy.

Godel did something similar to discrete structures by hacking self reference -- a kind of recursive self reference with no exit state.

I think pair programming with an LLM is interesting. You get to compare personal context with aggregate context. Plus, the external feedback helps me break out of excessive personal self reference.

> In 2023, the company said it pays out nearly 70% to the industry

Does not stop there; there are tiers of middlemen: https://youtu.be/kVY7-Ti77UQ?si=2neBSw4yJL7qGdxX&t=441

Apple also squandered maps ... until they didn't.

I look forward to the James Hoffmann review, should it ever come about. Am tempted to buy as a piece of functional art that I can drink. Currently, the 9Barista serves that purpose. Every. Morning. Bought it after the Hoffmann review. Good luck!

I was going to submit the same link. Like Benn Jordan, I was an AutoPilot enthusiast. Was going to buy a Model Y with FSD, until a drive from CA to AZ nearly drove me off the road.

https://news.ycombinator.com/item?id=31706955#31708672

Someone suggested that companies with a board of directors are the first AGI.

Somehow OpenAI reminds me of a paper by Kenneth Colby, called "Artificial Paranoia"

[*] https://www.sciencedirect.com/science/article/abs/pii/000437...

ALS is a configuration. Blindness is a configuration. Conventional UIs are built for a different configuration. Accesible UI usually overloads the conventional. If you start from scatch, sometimes you find something new. If you're lucky, the new thing scales to other configurations.

For example, I wrote a NLP parser for a calendar app, at Tempo.AI. It was much more efficient than the visual interface. And thus, it was accessible. But, it didn't use the accessible idiom. Instead, it was universally more efficient, whether you are blind or not.

A good example is a wheelchair accessible doorway. One method is to have a button at wheelchair height. The other method is to have the door open with an electronic eye. The first is Accessible. The second is Universal. Doesn't matter if you are in a wheelchair or not. It's a throughput multiplier.

I've been working on the menuing side [1] based on crossing Fitt's Law with Huffman trees. But, don't know the constraints for ALS.

Hopefully, whomever takes this on doesn't take the standard Accessibility approach, which is adding an extra layer of complexity on an existing UI.

A good friend, Gordon Fuller, found out he was going blind. So, he co-founded one of the first VR startups in the 90's. Why? For wayfinding.

What we came up with is a concept of Universal design. Start over from first principles. Seeing Gordon use an Accessible UI is painful to watch, it takes three times as many steps to navigate and confirm. So, what is the factor? 0.3 X?

Imagine if we could refactor all apps with a LLM, and then couple it with an auto compete menu. Within that menu is personal history of all your past transversals.

What would be the result? A 10X? Would my sister in a wheelchair be able to use it? Would love to find out!

[1] https://github.com/musesum/DeepMenu

Was tested while attending an electronic music class. Professor displayed everyone's result, so we could compare. His chart was the worst, as a cautionary tale. Too many live performance in front of the PA.

well, I used to have super hearing. Had a hard time being around CRT displays because the flyback transformers were really really loud. Now, with age, I get a phantom limb tone of 16KHz. [1]

Coincidentally, the peak notch in the OP was the 1st subharmonic of that.

[1] https://en.wikipedia.org/wiki/Flyback_transformer

My old boss invested in Kiteship. The Oracle Americas Cup team deployed one for a day -- but, not during competition. I suspect it messed with the competition's head for a brief moment. He did mention container ships. Feel free to reach out and I'll introduce.

Would love to see a directed Graph generated from a LLM prompt. So, instead of a NoFlo source text, a NoFlo-style graph.

Coincidentally, I'm writing a flow based patchbay for scalars, called "MuFlo"[1]. Kinda low level. I wonder how MuFlo and NoFlo might coexist.

[1] https://github.com/musesum/MuFlo

[edit] -Ironically +Coincidentally

Seems like a tradeoff between fixed assets versus variable assets. Writing code that doesn't change for years is a fixed asset. UX is a fixed asset. Somewhat like the construction cost of putting up a building. Maintaining the code and data is a variable asset. Somewhat like maintaining the building. Variable costs scale with the number of users.

With OSS, fixed assets are often overvalued. All of the R&D to get the UX and algorithms working well is easily copied. So, monetization relies on variable costs. Frequent updates. Data hosting. Ironically, buggy and obscure code helps monetization.

I almost became a developer of a successful database product. I attended one of their conferences. That's when I learned that the company made half of its money from consulting. It rammed home a joke I heard from an employee at Lotus Development: job security through obscurity. I left the conference dissuaded.

I recall a game developer had a deferred software release. You could download last year's game engine. But, to have access to the current engine (and thus influence its direction), you had to pay.

Let's see: spend x to train something that would cost .0001 x to copy. Tis a quandry.

Plus: a new tool for state actors to spread disinformation in perfectly convincing English. What could go wrong?

For a shaggy-dog-story from the previous century, check out The Silver Eggheads, by Fritz Leiber.

To be honest, I wasn't smart enough to wrap my head around Schank's semantic graphs. However, was/am inspired by this: "best way to make a machine think is to first make it teach" [1]

[1] https://www.wired.com/1994/08/schank/

Whaaa???!! I guess if you can think of it, it will probably exist at some point.

Will Wright taught me about "turtleducks"

Met him at the Computer History Museum, maybe around 2007. He had a branching storyboard-style presentation with 100's of slides. In the middle of the talk, he asked the audience about which direction they wanted to explore. Based on a show of hands, he proceeded to branch off into a few dozen more slides. Pretty epic.

Anyway, Turtleducks: we met later and I demoed a visual music synth. Was wondering if it would fit somehow into Sim-City. He said: "Well, what I have is a turtle and what you have is a duck. You can't just put a duck on a turtle and get a turtleduck." [Am paraphrasing.]

Ironically, a year later, he introduce Spore -- thus, solving the turtleduck problem. At least, in the turtle/duck domain.

A friend once owned Mike Markula's old home. He claimed that is where the Apple garage was. He sold the house for $1M, which was high for the time. The owners then razed the house (and the garage) to build something bigger. But, then came the dot com crash and new owners went bankrupt. So, the alleged "Apple Garage" became an empty lot.

Dunno, if this is where the real Apple garage was; maybe he embellished?

Extreme commitment makes sense if you believe in the mission. Such as saving the planet with EVs. Not sure what Twitter's altruistic mission is. Giving up family life to enable trolls may not seen like a good trade.