Habitual coder, guitarist, builder, pedant.
Posts
For some reason I instinctively open apps on my Home Screen with blue icons. I used to open Facebook all the time and it’s so inane these days I eventually deleted the app. For some reason I now open LinkedIn all the time instead, which is just as inane most of the time (it’s miraculous that so many people are AI experts just as the AI bubble reaches peak frothiness, isn’t it?).
Someone make a good and honourable app with a blue icon please.
So I got it into my head that I want a telecaster deluxe with Lollar Deluxe Foil (basically gold foil humbuckers in a wide-range humbucker form factor) and then I was like, why not a Bigsby, and now I’ve spent hours trying to work out what bridge to use for this purpose.
I think the Deluxe has strat string spacing and neck and takes a strat Hardtail six saddle bridge normally, and a Bigsby B5 needs a low rear bridge.
I discovered Tailscale recently (late to the party). It’s solved a bunch of security and remote access issues for me. Combined with Nginx Proxy Manager, PiHole (dhcp and local dns) and LetsEncrypt I have a good setup to expose Home Assistant, Next Cloud and various internal administration tools.
I bought a Raspberry Pi Zero 2 W and an Ethernet adapter and stuck it on the back of my router (powered via its USB port), just running Tailscale as a cheap and hopefully reliable entry point into my home network, even if other services (on my NAS) go down.
Does anyone have an above-average understanding of music theory, (ideally jazz) harmony, and guitar practice, that would be willing to review and give me some feedback on an app I’ve built?
I’ve created an open source app that includes a music theory reference for guitar, chord scales, chord substitutions, harmonic analysis of simple progressions and a system to manage practice routines. I think it’s starting to be pretty good but I would like some help validating what the content is sound and the functionality is useful.
I don’t intend to monetise or advertise it. It’s for my own use mainly. But it might be useful to others and I don’t want to share it if it contains material errors.
I have been working on a DIY reactive load box with IR speaker simulation for playing a tube amp silently - akin to a UA Ox Box or Two Notes Captor X. I think it’s finally done! It was a lot of fun to make.
All the hardware and software are open source and there are extensive build instructions.
Demo: https://www.youtube.com/watch?v=RQHTY69k00g
RE: @simon@fedi.simonwillison.net
I really like Simon’s writings on AI in software engineering. He is always grounded and balanced and he writes well. This introduction (and the set of practices underneath) is well worth bookmarking.
I have learned to capture Impulse Responses from speaker cabs. Well, using my Ox Box as a speaker for now.
Wave Arts IR Capture does a great job and is free. Previous attempt using Reaper and ReaVerb ends in a UI WTF.
Needed a clean amp for the sweep signal and got a tiny TDA2030A based solid state amp from Amazon and soldered some sockets to it.
No enclosure so I put it in a cardboard box.
Actually sounds decent!
Signal chain is USB audio out -> mini amp -> Ox speaker input -> ox line out -> USB audio in. Hardest part was getting levels to match sensibly across the chain and not clip the tiny amp.
There is a lot of AI scepticism in my timeline in general, and around software engineering in particular. Much more than there is AI chest-beating (good!)
And the concerns are real and valid. But I also think there are things to be excited about. I have been able to build products (digital guitar effects in this case) that previously would have been well beyond my grasp. And they turn out to be really good. And it is enormously fun.
Yes you can generate slop with no care for quality. But the code I get out of my use of AI is not that. Nor is it easy to wield these tools - it is a new and complementary skill. And the practices and tools are evolving very rapidly. It’s confusing but sometimes exhilarating.
Nothing is universally good or universally bad. I do think as an industry, software engineering is at a precipice as profound as the birth of agile or the arrival of apps. And the future can still be shaped by those of us who care about quality and value and integrity.
A man owned a lake and wanted to skate on it when frozen over in the winter. Unfortunately it was covered in snow and twigs and bird droppings. He didn’t want to pay someone to sweep it. So he invented CURLING and got people to pay to play a game of stone and brooms on his lake instead.