Our great sponsors
-
InfluxDB
Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.
Hey thanks for taking the time to critique!
I actually do have an ANTLR file that is about 90% of the way there ( https://github.com/kstenerud/concise-encoding/tree/master/an... ), so I could use those as a basis...
One thing I'm not sure about is how to define a BNF rule that says for example: "An identifier is a series of characters from unicode categories Cf, L, M, N, and these specific symbol characters". BNF feels very ASCII-centric...
Most of my ideas are, to put it bluntly, utterly shite.
That said, usually as I explore an idea it begins to slowly crystalize in to something interesting. I've built furniture, created recipes, written software, designed games, made toys, invented interesting gadgets, published stories and articles, and shot videos, and in all of those cases the resulting artifacts looked nothing like the original concept. There might be a sliver of the original idea, but it has usually drifted far from the initial idea.
I'm currently building an electrocs workbench, I say building, but I mean "thinking about it a lot." I've purchased some cabinet hardware, I've got a stack of boxes for internal storage, but the workbench itself sits as nothing more than a CAD drawing as I tweak how the workbench & cabinet will work. "How deep should this drawer be? How wide should I make that swing out tray?"
I am currently kicking around an ML idea to do with some data at work, I'm doing it as a "fun side project" in my own time, my PM keeps trying to "task it and track it." Don't do that, I need to keep kicking the idea around for a while. When I'm ready, we'll stick a pin in it, nobody is waiting on this, nobody outside of me and the PM knows about it, why add artificial urgency and process to something that's fun?
As a personal side-project I'm building a task list. The idea has evolved so far from "I want to build a task list" that it is barely recognizable. I wrote a lot of the lab notes from that project here at https://github.com/justinlloyd/retro-chores, and without expecting anyone to read through those lab notes, there are places where I write "I'm going to solve it this way" and then a few paragraphs later I write "nope, actually we're going in this direction now" because a little research showed some new evidence.
My lab notes are very free-form, the way to kill creativity and an idea is to impose structure on it. As a creative exercise for problem solving when working on an invention or gadget, is to take a part or component with me, and go for a walk, and mull over how the component works, how it fits in to the design, what issues I will run it to with it. The component acts as a McGuffin, to keep my thoughts focused on the project as I walk.