A tree can be seen as a formal language. Look into L-systems.
If you generalize what a symbol is (the rgb value of a pixel) you can write a grammar that ends producing a list of pixels. You can then place it in a 2d matrix and you have an image.
I guess a better approach would be wave function colapse, but seems to me like it could be formally described as a grammar (CS or CF, dunno, would have to look into it)
Didn’t you go into Turing machines and the Halting problem from that?
That was my intro into computation: regex, automatas, state machines, stack state machines, formal languages, grammars, Turing machines, Hanting Problem, P NP.
I’m really sceptic about that kind of metrics because many of them take carbon offsets into account, and carbon offsets are mostly greenwashing.
Power mix in the world right now is over 50% coal and gas, and only hydro is over a 10%. This is worldwide, so mix varies depending on where you are.
In the end EVs are no making a dent in power demand. They are increasing it. The percentage of fossil fuels is maybe going down but total fossil fuel consumption is increasing as our demand does. Green energy is only taking some of the slack from the increase.
EVs will be remembered as the thing we did to keep using cars and feeling good about it.
Are they? Because unless you live in some green energy paradise, most EV are charged using coal plants.
I’m sorry but framework and library in this post are going to be used loosely, because even React, Vue, Angular, Svelte, etc devs use the terms loosely.
React is mostly a UI library like you would find in most native app development. Of them all them JS frameworks/libraries is one of the less opinionated and with less batteries included. By design it does not does everything. Most other frameworks do way more.
It lets you define custom components. The components can have properties that their parent component defines and internal state. If the state or the properties change the component gets redrawn (magically). There are some lifetime functionalities (things to do on first render for example) and performance improving stuff (memoization) but mostly that’s it.
All the other features you talk about are third party libraries or frameworks that can operate with react or are build on top of and cover the bases, like routing, fetching, caches, server side rendering, styling utility libraries, component libraries, animation libraries, global state management, etc.
The big difference with the vanilla way is that the approach is mostly declarative. The runtime takes charge of updating the DOM when your components state or properties change.
You take a big performance hit, and an even bigger bundle size one, but the speed of development and huge ecosystem of readymade solutions can be really important for some use cases.
Other frameworks take different approaches to solve the same problems:
Yeah, writing prompts it’s the long term goal, programming will be obsolete.
Nobody that can write a problem in a structured language, taking edge cases into account, will be able to write a prompt for a LLM.
Prompt writers will be the useful professionals, because NO big tech company is trying to make it obsolete making AI ubiquitous and transparent, aiming it to work for natural language requests made by normal users or simply from context clues. /s
Prompt engineering it’s the griftiest side of the latest AI summer. Look a who is selling the courses. The same people that sold crypto courses, metaverse courses, Amazon dropship store courses…