Notices by Don Romano (alt) (thor@noagendasocial.com), page 36

@philcolbourn I would prefer to have a linter, but I'm not sure if I literally need to run bits of code inside the editor. I mean, hot reloading is pretty close to that already...

@philcolbourn What I'm trying to do now is just learn ClojureScript with Figwheel, which will hot-reload your code. The Lisp plugin I found for VSCode wants me to set up nREPL and that's where I got into deep water.

@philcolbourn With Common Lisp, they kind of just immediately start talking about how to set up Emacs with Slime and connecting that to the running REPL. It's heavily implied that not doing so will set you back somehow, because you hardly ever see anyone advocating not doing that.

@philcolbourn I'm left with the feeling that I'm not truly getting the Lisp experience unless my editor can talk to the REPL, and that I'm missing out if I just edit the source code and run it.

@philcolbourn I'm using Ubuntu and Visual Studio Code. Almost every guide I can find touts the virtues of running the REPL inside of the editor, and sends you down the path of getting that to work.

@philcolbourn A more accurate statement would perhaps be that Lisp source code closely resembles its own AST.

You could tell me that I could just run the REPL and play with that, but that's not what the READMEs tell me to do. They tell me to set up a project folder and run the REPL in an editor, and that's where it gets cumbersome.

The very first ClojureScript extension I could find for Visual Studio Code wanted me to set up nREPL and that's apparently not supported by the standard project templates, and I don't know how they work yet, so I don't understand the instructions for modifying them.

I have tried to get started with Lisp multiple times now.

I tried getting into Common Lisp earlier, but it took me half a day just to set up the tools. It felt extremely clumsy, and I lost interest when I realised that I'd also have to learn Emacs.

Today, I tried setting up ClojureScript and it had me trying to piece together four different tool kits and their ecosystems in the first 30 minutes.

Lisp itself looks elegant, but these heavy and cumbersome tools are a showstopper for me.

@philcolbourn Yes, but isn't every language a serialised AST? If your language can't be deserialised into some sort of memory data structure, how do you write an interpreter or compiler for it?

The extremely regular syntax of Lisp makes it easy to learn and easy to parse. Unfortunately, it also makes it hard to read. Needing to read full words in order to tell if you're inside a block, a data structure or a list of function arguments is not a good thing. The ease of skimming source code is about usability, not cosmetics.

@trickster @lynnesbian ARENT•YOV•FORCETTINC•THT•CERTAIN•LETTERS•VVERE•ONL•INVENTED•LTR•AND•ALSO•THEY•LKED•TO•ABBREV?

I have never really figured out the right moment to hit on someone I'm attracted to. The right moment seems to be never.

In terms of being hit on, I think that's only happened to me once, and naturally, I screwed it up. Apparently, I'm not very attractive, because other men seem to get hit on more often.

This is why I'm single.

It would be interesting to see the results of a large study where people are asked questions like:

"Lopax was born in the blue-skinned country of Pertania but migrated to the green-skinned country of Cepland when she was 10. She speaks fluent Ceplandic, retains her Pertanian religion and dress, but has mostly adopted Ceplandic attitudes to sex and marriage. Is Lopax Pertanese or Ceplandic? What if she was a mix and had turquoise skin?"

@bifpowell You could go pretty far down this path, down to modeling the sag on the power supply rails when the amp is played loud, and the thermal noise in the system.

@bifpowell To turn this into a practical digital amplifier, you'd then need to boil this down to a graph of DSP algorithms described by various transfer functions. For the common analog filters, the transfer functions are known, and all you have to do is plug in coefficients for stuff like resistance and capacitance, and you will have a filter that sounds like the analog one.

@bifpowell Yeah, so different amps have different sounds even if you play them through the same cabinet and dial the gain and the EQ to sound similar.

If you took apart those amps, looked at their topologies, made SPICE simulations of them, and fed the guitar audio through that, it would probably be hard to hear a difference between those and their real counterparts.

You'd need fairly accurate SPICE models, complete with flaws, for the various components, though.

For example, I want to make audio gear, but who would buy it? It's such a mature market. If you manage to find an unfilled niche, it will be a weirdly specific one, and you're left with a product that will never catch on.

It's kind of sad that having a job usually means performing tasks you don't like in return for a bribe. You usually can't make a living on what you do on your own terms. Some hobbies resemble jobs, but once you try to turn a hobby into a business, the creative freedom that made you like it is gone — traded for whatever the customer is wanting — typically something you have zero enthusiasm for. It's a shame that the stuff we love to make is never the stuff other people want.