LOL, hearing that the first "search" results for the most popular AI coding engines are malware. (both organic search and the hidden "sponsored search" results). Quite the security kerfuffle at work over this. 🙄 #ai #vibecoding #cybersecurity
"cut and paste this into your command line" from random websites ALWAYS goes well. /s #cybersecurity
@ai6yr curl pipe to bash for the win!
@Viss Yeah, this is why I manually install any Linux projects instead of that if I can, I don't trust anyone's shell script to be clean!!
@ai6yr sooo muuuuuch stuuuuuf is installed that way. ruby, homebrew, rust, ai shit
@Viss @ai6yr you both have formal training in CS/coding right? Don't they cover best practices or is reality at most companies just get it done quick security problems are overblown?
@CliffsEsport @ai6yr ehm. i took pascal and borland c++ in highschool, but dropped out of college outright when they changed that shit to msvc++. went straight into sysadmin. taught myself perl on the job, then bash, then python. i have no official formal dev training and i dont call myself a dev. im ok in bash but python i always need a crutch and im barely functional
@CliffsEsport @ai6yr also, "best practices" is corpospeak for 'i dont actually know the answer so i copied my classmates homework', so its prolly not the actual best way to do stuff
@Viss @CliffsEsport LOL most schools don't cover best practices, that's way too practical.... Not sure on the state of the art, but traditionally, lots of theory and things like algorithms, etc. and precious little practical knowledge. For awhile everyone was only learning Java in school (which was terrible). Not sure what the language of the moment is....
@ai6yr @CliffsEsport yeah i bailed before they went to java and went sysadmin mode
@Viss @CliffsEsport Just did a quick search, depending on what school, C++ and python. So, they finally figured out Java sucks lol.
@Viss @CliffsEsport I was finding people who (attempted) to learn how to program in Java could not program. The language was too cumbersome....
@ai6yr @CliffsEsport just the mechanics of making anything in java at all. such a fucking nightmare. and android is the same because its all fucking java under the hood
@Viss @ai6yr So as someone who only had the time/money/energy budget to get homelab going in last year what modern language do you recommend really learning? I understand enough of the basics of CS that I can piece together what I need to do for the little bits in past, but have been thinking C though think I'd prefer lower level personally.

@CliffsEsport @Viss Well, for me, python is the easiest to implement and pick up to build things. The main upside and downside of python is the availability of libraries. (able to find libraries for many things). That's an upside, BUT, the downside, is the supply chain for python libraries is ALSO not very well vetted, and there are most certainly bits of malware or Bitcoin mining or other crap you do NOT want on your systems that show up...

As a basic language (without loading libraries, willy nilly), it's probably the best. (Disclaimer: that's the tech stack I am using most now)

@ai6yr @CliffsEsport @Viss

Depending on your specialty/interest:

  • Bash
  • Python
  • C#
  • C++

Couldn't hurt to give you a foundation of understanding for how programming operates and works. File & logic handling, errors, logging, networking, etc.

There are useful tool 'challenges' to build your own toolkit. Stuff that you will likely use anyhow, that can get you started.

@jackryder @ai6yr @Viss Yeah C family and Bash probably make most sense. Does it really matter which C I start with?

@CliffsEsport @ai6yr @Viss

tl;dr - no. It is turtles all the way down.

Honestly the hardest choices are the ones that don't move us forward. Picking a random language and diving into it will serve you equally well. There is a literal tidal wave of languages available, each designed to solve a set of problems the author felt needed to be solved.

Personally? I tend to apply the language to the problem. If you are playing with web tools, stick to web-friendly environments.

Find the environments that interest you, and then look at what moves them. Becoming an expert in JavaScript is not as useful to Data Scientists, but the mechanics are similar enough to work off of.

If you are building towards a team, quite literally look at their githubs and figure out the languages they seem to focus in. Those are likely popular in the social group you are trying to work in.

@jackryder @ai6yr @Viss I tend to be very cross domainm my main goals are building my knowledge, and maybe also make some money. The way my brain works it cares more about learning than making money.

@CliffsEsport @ai6yr @Viss Oh, well then that's perfect.

Grab one of the languages listed above and find a project to build in it. Start small but meaningful to you.

Example: You have mentioned homelab and understanding CS and C. Create a repo and build something for your homelab using one of the C-languages.

Web tool: C#
CLI: C
CLI: C++

Keep the tool simple and productive. Work through the hows and the why's of the stuff you are typing in.