I started with C++ and went to Java to .NET to Javascript and now to Terraform.
I know this is all a joke but there’s something definitely different with the ones above and the ones below. There’s a bit of satisfaction you can get sometimes when you’re working with memory directly and getting faster feedback (yes, there’s more math back then and it wasn’t easy to look stuff up, for sure). However, there’s new challenges nowadays … there’s so many layers on top of layers. I feel as though Stack Overflow and ChatGPT are so needed because the error messages and things we give are obfuscated or unclear (not always any library author’s fault as there’s compatibility issues, etc)
We’re doing serverless stuff at my current company and none of our devs run code locally. They have to upload it using CDK or Serverless Framework to run on the cloud. We don’t use SST so we can’t set breakpoints but like that’s a lot of crap inbetween just running your code already. Not even getting into the libraries and transpilers and stuff we use. I spent like a few weeks over Christmas to get our devs to run the code locally. Guess what? None of them use it because they’re so use to uploading it. I was like, "you can put breakpoints in it! you can have nodemon and it instant reloads! nope, none of them care … "
I’ll be the dumb one to ask: what do you mean? Is this that making a mistake that costs a lot is the best teacher, because you only have to mess it up once to learn it forever?
Pretty sure they mean people don’t learn something again when they already learned it. Once you learn how to do something, willingness to learn it again but a different way dries up, and so you stick to bad habits as long as they ‘work’
I started with C++ and went to Java to .NET to Javascript and now to Terraform.
I know this is all a joke but there’s something definitely different with the ones above and the ones below. There’s a bit of satisfaction you can get sometimes when you’re working with memory directly and getting faster feedback (yes, there’s more math back then and it wasn’t easy to look stuff up, for sure). However, there’s new challenges nowadays … there’s so many layers on top of layers. I feel as though Stack Overflow and ChatGPT are so needed because the error messages and things we give are obfuscated or unclear (not always any library author’s fault as there’s compatibility issues, etc)
We’re doing serverless stuff at my current company and none of our devs run code locally. They have to upload it using CDK or Serverless Framework to run on the cloud. We don’t use SST so we can’t set breakpoints but like that’s a lot of crap inbetween just running your code already. Not even getting into the libraries and transpilers and stuff we use. I spent like a few weeks over Christmas to get our devs to run the code locally. Guess what? None of them use it because they’re so use to uploading it. I was like, "you can put breakpoints in it! you can have nodemon and it instant reloads! nope, none of them care … "
First learning is last learning.
Same reason we still do
console.log("FUCK").I’ll be the dumb one to ask: what do you mean? Is this that making a mistake that costs a lot is the best teacher, because you only have to mess it up once to learn it forever?
Pretty sure they mean people don’t learn something again when they already learned it. Once you learn how to do something, willingness to learn it again but a different way dries up, and so you stick to bad habits as long as they ‘work’
It’s a mantra about teaching people and then expecting them to forget it. Doesn’t work. They’ll default to what they already know.
My freshman English teacher got married in October and I called her by her maiden name the entire year.
Like all programming mantras, it’s not universally true, but it’s annoyingly reliable. It reflects the shape of the human brain.