Can you give a quick concrete example of a success using Bing in this way? I've written machine learning code of my own but am fairly old school about search engine use and my experience with LLMs is largely trying to get them to say the nigger word etc.
How do you know what it tells you is correct if you don't already know the subject?
The nice thing about code is that you can test it and the computer will usually quickly shit its pants if it's wrong
Bing has inline citations, so you can check its work, so to speak. Generally you can tell if something's off or not, or use it to springboard further research etc. One nice thing I've used it for is the prompt "Find resources about...", which will usually get you 5 or so real links to technical documents or recommendations for books to 'take out from the library'.
I feel like the only people that think AI can actually replace programmers dont understand anything about ai or programming. But I will say AI is really fucking useful for regex cause I aint memorizing that fucking shit and its great for learning cause its gives answers that are relevant to your specific project.
I think this becomes a problem along the lines of the classic "Just because it compiles doesn't mean it works properly." phenomenon.
y by using the chain rule without good reason. I'm confident most calc students wouldn't catch stuff like this; what makes us stupid code monkeys any different when even the best of us overlook simple shit?It's really not that hard.
My biggest issue with regex is that every single thing uses subtly different syntax, so you get to play a fun game of "figure out which special characters need backslashes and which character classes are supported with which syntax" every time you switch context.
Which is why any time I have to use regular expressions I just pull up copilot, ask it to give me a regular expression that formats a string as whatever or tests for whatever and it always gives me the right regex. In fact, I bet I can make a tool that does it no ai necessary because fuck memorizing and typing out that god damn gibberish.
Very fair point, and a gripe of mine as well. Putting aside all of Perl's additions that are often missing from other languages, I love GNU, but this Basic vs Extended shit infuriates me to no end. I argue the "Basic" version is a bigger pain in the ass to work with, considering how often groups are used in basic patterns.
(regexp #rx"(\\\t)*")This amounts to typing a few extra backslashes to make use of normal regex characters (e.g.
(.+)). Modern languages tend to replace the direct need for things like assertions and the fancier forms of capturing, so you can get by most of the time using the very basics that should work across most implementations. Hardly a reason to forgo learning something so incredibly powerful and quick to use once you're familiar with it.
Well, Im not opposed to memorizing the basics, but sometimes it just makes more sense to look it up.
My impression of G4G after mentoring a lot of college students is that came to be and exists primarily for students, especially foreign students, to cheat through a CS degree, kind of like a glorified Chegg. That it provides any sort of real assistance to others doing non-college related shit may be a (positive) side effect.
This man has clearly never written old school PHP or Javascript.
That everyone doesn't just use PCREs at this point (including and especially Python), is a fucking travesty.
Something something Chomsky heirarchy
Any language based on Meta Language is extremely good for writing interpreters. I would genuinely suggest Haskell, it's ungodly fast if you put the time and effort into learning the language + the quirks of how it compiles (like when to use foldl vs foldr vs foldl' vs foldr'). Basically you can unintentionally end up using a ton of memory on accident if you don't understand what the compiler is doing. Richard Bird's books are a great resource for this. But once you get it right it's fucking miraculous how fast it is. I love writing toy languages and at this point would only choose Haskell for that job. If you're the kind of person who would like to write a toy language you're also probably the type of person who will find Haskell fun to write.
and its name is POSIX
niggerlicious
