verstra @ verstra @programming.dev

Posts

10

Comments

196

Joined

2 yr. ago

Woah, the landing page looks like some unfinished Wordpress template and the forge itself looks 15years old. Nothing wrong, with the former, I just don't like the style.

Camel case?

this_is_not_going_to_be_a_serious_debate

No

This is the way.

Almost completely pure way of storing ideas. With this I mean that you don't store unnecessary data such as "background should be white" or "left page margin is 1.3cm". It's just text. What's important is what it says + minimal markup.

Presentation is left to the reader's client. Do you want dark mode? Get a markdown editor/reader that supports it. Do you want serif font? Again, that's client's choice and not part of the document.

I wish browsers would support markdown out of the box, so you could open https://example.com/some-post.md

You, my friend, should try EdgeDB. A database and an ORM in one.

When you change the data model, you can get to 100%, which you say is impossible for ORMs

Barges into a lemmy thread

Gives a strong, but quite abstract opinion criticizing abstractions

Refuses to elaborate further

I have a convention to correlate the size of variable scope with its name length.

If a variable is used all over the program, it will be named "response". If it is <15 lines, then it can be "res". If it is less than 3 lines, it can be only "r".

This makes reading code a bit simpler, because it makes unimportant, local vars short and unnoticeable.

No biggie, it a nice entry barrier to have, because nowadays, there just too much new frameworks and languages and crypto currencies.

Because when you divide by zero and get a runtime error, the error will point you to location in SQL, not PRQL.

It's like if an error in a C++ program would point you to an offset in a binary and not the location in the source. This has a slight tone of sarcasm, because that's how compiled languages used to work. But after the years, they patched all leaks of their abstraction and now you are dealing just with the new language.

That's because the tooling is not there quite yet. For what you describe, that would have to be implemented by the compiler bindings for your language. And it's not that hard - basically one function. But yeah, not there yet.

That you can easily use JOIN after you've already used LIMIT or GROUP BY.

As a developer of the compiler from prql to sql, I can say that I'm working on a new language because I know SQL quite in-depth.

It seems ok on surface, but there are many inconsistenties that you either learn to live with or don't even know are there until they trip you.

Cool. Any idea how would i use this with rustc?

What's up with the cartoon characters on this blog?

I've seen a few posts and they are generally of decent quality and technical contents, but I'm a bit weirded out by the children's illustrations.

I prefer the pomsky syntax. https://pomsky-lang.org/

No

Because it looks like Python?

I strongly believe that code formatting does not hold any information and that because of it, compilers should reject non-normalized code, saying "run a formatter first". We have the tooling for most of the languages to make this absolutely painless, remove all head scratching of "what the hell is up with this code style" and just focus of the semantics. I believe that you should be allowed to produce code abominations like this code sample and have it auto-formatted into the code style that everyone can easily interpret.