After many years of using SO, I’ve started using ChatGPT for all of my programming questions and have not looked back once. For my usual “I know X is possible, but how do I do that in Y language” questions, it’s been a dream using ChatGPT.

In Rust, using the Option and Result types make the general flow of the application much easier to organize, make modular, and reuse.

This was a good blog post. I particularly appreciated the statement about the validate and parse function comparison: “Both of these functions check the same thing, but parseNonEmpty gives the caller access to the information it learned, while validateNonEmpty just throws it away.”

Is that a water dispenser? I need something like that.

That’s something I haven’t heard before about the memory safety. In what ways is it not memory-safe?

Do you happen to know of a few situations where bloom filters are super useful? I need to identify when to use them.

I save “template” SQL queries in a special directory so that I don’t have to google how to do specific things. It’s basically my own personal “examples” folder.

Who is writing SQL in the terminal?

AI is surprisingly helpful with providing a starting point. When you want a helloworld app, an example of how to use some part of a crate, or a code snippet showing how to take what you have and do something unusual with it, AI is super useful.

I would love to be able to get the same quality of AI locally as I do from ChatGPT. If that’s possible, please let me know. I’ve got two 3090s ready to go.

But for now, I’m just enjoying the fact that ChatGPT is free. Once they put up a pay wall, it’s back to suffering (or maybe/probably trying out some open-source models).

This comment would make sense if he hadn’t stated that the PR was politically biased but had instead said that it was unnecessary or that it would be inconsistent with the vast majority of the documentation. I’m just reading what he said. He claimed it was a PR based on politics, not language norms or historical norms. Only certain kinds of conservatives view gender-inclusive language as a political issue.

I appreciate that you don’t want to see this person as a hateful bigot and I don’t think he is either. Most people I’ve encountered that share the same reaction as him have basically been tainted by conservative influences, like media or parents, but they don’t have any real hate for trans people in their hearts. They’ve associated the idea of gender-inclusivity as being political and moved on with their lives, accepting the framing and narratives around the topic.

It’s a reference to this: https://github.com/SerenityOS/serenity/pull/6814#issuecomment-830793992

They have a phobia of making changes that are valid if they perceive the change to be motivated by politics. In the example above, the PR is denied because they have been convinced that the PR is about accommodating trans people. The existence of trans people and accommodating them via grammar is political for certain kinds of conservatives. The irony is that their own political beliefs are affecting their ability to distinguish a valid change from a politically-motivated one.

The bit of Clean Code that I read was unimpressive, but Clean Architecture was amazing. I view that book as required reading for anyone who wants to write code professionally. If Uncle Bob hasn’t realized that his coding style is worse than alternatives, I do not see how a second version of the same bad ideas is going to do well.

What if instead we utilized an algorithm, some code, that would ultimately generate the file? I could imagine a program that generates a number which ultimately is more dense than the program. For example, if we just-so-happened to need a million digits of Pi the program would be shorter than the number. Is there a way to tailor an algorithm to collapse down to any number? As an example, what if we needed a million digits of Pi but the last 10 digits need to be all 9s?

Do you happen to know of any good algorithms or numbers? Pi gets harder to calculate with each digit, so it’s not a great candidate.

“Cleaning up…”

Is there an algorithm or number such that we could basically pirate data from it by saying “start digit 9,031,643,679 with length 5,345,109 is an MP4 of Shrek”? Something that we could calculate in a day or less?

Yeah, if it’s purely a Sqlite implementation detail to create temp files, that’s on them to own and fix. I thoroughly dislike that the files are obscured from users.

Oh, I thought that the temp files were named by the user. If that’s not the case, that these are not databases created specifically by McAfee in the temp directory, then I’m not sure what the appropriate solution should be. Obscuring the file type and how the file is used from users is still a bad practice.

I love how the solution didn’t involve changing the prefix to “mcaffee_”. Now users don’t know who to blame. Great. That’s so nice of them.

Yesterday doing a search using vim for a class that shared a lot of characters at the front with many other classes: /Bas.*Some I could have done a more precise search with better regex, but this was quick, easy, and worked.