188 karma · joined November 14, 2019
I wonder if a potential application of LLMS could be: have two experts have a really interesting but dense conversation with each other, and then translate the conversation into simpler language with interjections for explanations.
It may not be enjoyable for the most general audience, but it would scratch an itch for some of us.
Or... maybe you have be both discerning and naive in regards to it for the illusion to work.
I was going to mention how much I loved that game, until I realized I played UT99. Time sure does fly...
The first time I played I thought king would take my pieces with some probability if they weren't guarded. I spent way, way more effort than was necessary playing with higher stakes.
Everyone else presuming you need a list to keep track of them is missing the point - spontaneous and immediate action with minimal commitment.
I have a similar profile, along with seizures in childhood that led to repeated head injuries. Cause and effect was always an open question for me.
My version of the cursor kept a continuous scan velocity horizontally across the screen, and you controlled the y-axis via pitch. Button presses were (intentional) collisions.
What I was trying to do is get you to explain what you mean by this:
> If our code ran in real-time to show us a representation of the final configuration, and we could trace how each final configuration value was generated, then it wouldn't be a problem. [...] But no systems are designed with this capability, even though it is quite trivial to do. Configuration is always an after-thought.
I want to migrate to an actual nested data-structure using (maybe) JSON - and these engineers absolutely will not write code, so config-as-code is a no-go, in addition to the disadvantage you mentioned.
My next thought was that there should be a better way to show the configuration, and allow that configuration to be modified. I was thinking maybe some sort of visual UI which where the user can navigate a representation of the final product, select a part and modify a parameter that way.
Is that along the lines of your suggestion? If not will you please expand a little? Configuration is the absolute core of this application.
I extended the 20 hour practice idea from Josh Kaufman [2]
I highly recommend Sam Harris (Waking Up), Alan Watts, Laozi for mindfulness / meditation, as well as Iain McGilchrist (The Master & His Emissary), Mihály Csíkszentmihályi (Flow) and David Foster Wallace (Infinite Jest) for metaphorical inspiration. Learning to play at life in any moment helps to eliminate tedium & boredom from your vocabulary.
[1] https://imgur.com/a/pMudls5 [2] https://www.youtube.com/watch?v=5MgBikgcWnY
2. How do I avoid going all in? Going all in is my default. I treat the 20 hour mark not only as a goal, but as a stopping point. I must make the conscious decision to draw another meter in order to continue, after considering how my time is being spent - otherwise I can become obsessed and lose balance in my life.
Note that your phrasing is self-defeating. You are telling yourself a story which appears maladaptive. Just listen to yourself: 4 months of valuable experience was an "evil" that you had to "knock out" to the "detriment" of "actual" practice.
You identified a weakness: You knew that you needed to develop your ear, and make sense of the territory. You might have collected a hundred brief sound clips and challenged yourself to identify the sub-genre. You might have randomly selected a genre and challenged yourself to produce a very brief (even minimal) piece that expresses its essence. You might have taken a short composition and, for any given genre, challenged yourself to reinterpret your own song in that style.
Here's an idea. Take 5 different genres, and challenge yourself to making a short piece that gradually and naturally transitions between them. Optionally ignore all other constraints, like global structure, general appeal, consistent aesthetics, etc.
You did 4 months of practice which was in all likelihood less than ideal, somewhat inefficient. So what? That's a positive, not a negative. Next time, try to incorporate "doing the thing" into your regimen and you won't feel like you've lost momentum.
The structure of the skill is the curriculum, so to speak - but the skill itself is an action. When someone is learning music theory, they are really just learning what the structure of the skill is. If they knew the series of actions they needed to take, the goals they needed to set for themselves in practice, and the relationship between those competencies - then they would have no need of theory. The theory informs us of what the structure is.
In most skills and hobbies, you don't need much theory. This is because people have created drills and identified the fundamentals over time. Everything has been organized for you.
That's the structure. Here's the interesting thing - you don't need much structure at all. If the structure is a tree or graph representing the fundamentals you need to master, all you need to know is the next node - the next skill you will focus on.
That's where practice comes in. Practice should be your primary focus, not your secondary focus. Some fraction of that should be "free-play", or curiosity driven exploration & synthesis. You add that in because it's critical for learning, and because it keeps things feeling fresh - but the real value-add is to learn to enjoy the practice. If you "play" at practicing, then you're golden.
Practice though, must be deliberate. You don't just "make music". You have identified something you want to work on, and you attempt it through repetition. You compare your effort to the end-result (using feedback), and you repeat. This loop never ends. If you feel you've mastered something, you move on to the next thing. If you don't know what the next thing is, then you turn to theory in order to identify it.
That's all the balance you need. Learning about a thing is not doing the thing. Do the thing. Identify what you want to work on, draw a little progress meter on piece of paper, and start doing it. Do it for 20 hours, filling up the meter. At the end of 20 hours, you have a choice: draw another meter for another task, or stop doing the thing.
The problem you face is thinking that motivation is your problem. Your problem is that you are not doing the thing. How you feel when you are not doing the thing is irrelevant. How you feel after you've done the thing - and you notice that it's not what you wanted to it be - is irrelevant. You will never be at the goal, that's the whole point of practice.
But playing a game and learning about a game are two separate activities. The key is to spend your time actually playing the game. Playing the game consistently will feed the drive to learn more about it, and if both of those things feel like a chore, then why are you doing it in the first place?
I can say with absolute conviction that it isn't the rind giving you trouble. I can pick every ripe Hass avocado out of a bin of 50, and never once be wrong. That said, there are a couple of stores in my current area that have garbage Hass avocados. The ones I buy from this store are consistently gross and fibrous, even when ripe, and have a very short window. I boycott these stores. They either have a second rate supplier, or else somehow ruin the product in transport.
By the way, you need a gentle touch and the ability very slowly, smoothly increase pressure. You pick up the avocados that are not yet ripe, the moment one is on the threshold, such that there is a nearly imperceptible give which does not bounce back, you buy the avocado. It should not bruise the fruit, and you won't be able to detect it when peeled. After this check, you can visually tell when it has fully ripened, ~1-2 days later. For what it's worth, I've never successfully ripened one AFTER cutting into it, and I've never kept an uncut avocado for more than 4 days.
I've heard that much of the ruins from Ancient Greeze were likely brightly colored and painted. I can't help but wonder if the Venus I mentioned had been painted as well. It's possible that they did have detailed faces, and that they simply weren't sculpted.
I'm the furthest thing from an expert of course.
I'm interested in learning more about the self-hosting option. I take it everything is still closed source? Can you share anything about your stack? The front-end is great! What's the scaffold? Is it JavaScript soup to nuts, or Python, or what?
As I move through the document, I already have the location of the cursor origin saved in a register and can jump there, or any of several other locations with a single keypress.
I've always considered the mouse & menu system to be an emergency backup for beginners who have hit a wall. They will probably never be first class citizens.
What editor do you use?
My question is not particular to Excel of course, but you seem like an organized person so I figured I'd ask.