> I mourn for the generations that will never get to use their brains.
If you aren't using the incredible technology we have around us right now to accelerate your learning then it is us that should be mourning your loss of imagination.
If you aren't using the incredible technology we have around us right now to accelerate your learning then it is us that should be mourning your loss of imagination.
I won't down vote you, because you simply misunderstood me. Of course I'm using this incredible technology to accelerate writing code.
If you never learned the value of learning itself, for its own sake, you have everything to gain from a world in which skills like learning and thinking for yourself are useless. I can hardly blame you for that.
But as someone who has learned by teaching others, people who in turn will learn by teaching others, I don't find your belief in someone else doing the work for you to be particularly sustainable. Or convincing. Regardless of whether I think it's contemptible, lazy, or damaging to your mind, which isn't my problem.
Not GP, but here’s some of the things I have learned in the last couple weeks from LLMs while asking LLMs to review parts of my code and documentation. I wrote the code and documentation without LLMs.
Linux user permissions is a fake because by design, the owner has permissions to write these 3 bits. A good way to reliably deploy a readonly file is owner=root, assign the read permission to the group.
chown command follows links by default, -h switch disables that.
In Linux, resetting environment variable in a long-lived server process doesn’t erase much data because /proc/{pid}/environ is a snapshot taken on process startup.
If my server crashes (or manually killed by a root to test things on staging VM) and Alpine’s supervise-daemon restarts it, the start_pre() function from the OpenRC service script won’t be called again. A fix is use OpenRC to run a launcher shell script as root, in that script manually downgrade the account with `exec su-exec svc:svc /the/server/elf`
This last one is irrelevant to my use cases but still interesting. AES GCM may fail catastrophically when different message encrypted with same password and nonce. For 12 bytes of random nonce, the stuff is only safe up to approximately 4E+9 messages using the same password, beyond that birthday paradox ruins things.
- Kalman Filters
- Bayesian Tracking
- Hybrid Logical Clocks
I was able to ask questions and understand the current state of the art.
This is not very hard, and we all used to know how to do this not 3 years ago.
https://www.youtube.com/watch?v=rdj_Fm4XM60
If a person writes a textbook, a tutorial, a blog post etc. which is that bad it won‘t be recommended by peers and a beginner will either not know it exists, or find see the bad reviews from peers. The beginner will then look for a different textbook, tutorial, blog post etc.
When self learning with AI, you don‘t get that. The best advice a peer can give you, is simply, do not use AI to learn stuff you don‘t already know.
Thats a pretty bold position to take.
I find asking a question about whether an application should be doing what it is doing, and getting someone's braindead response of "Claude says it does this..." shows that people aren't learning. I know what it is doing, I can read an IF statement, I don't need someone to spit out 3 paragraphs of unrelated nonsense.
With Claude people aren't learning. They aren't even reading their copy/pasta response to see if it relates to the question.
Thinking about it now, I think the new thing of copying LLM output is worse for a couple of reasons. Although both are a sort of vampirism that sucks all the oxygen out of a text chain, at least pasting articles implies that one has read that article and agrees. Someone can be held responsible ror the selection. With LLM output, there's an implied shrugging off of any responsibility... it's not "here's something I read and agreed with," it's just "here's what ChatGPT said." There's less implicit effort, and less risk of staking one's credibility on something, because saying "hey, that article is junk science and these studies disprove it" would require a response beyond "well, ChatGPT can be wrong".
Secondly, and maybe more metaphysically or epistemologically, there's a certain decline in cognitive function that I associate with pasting LLM answers that goes beyond what was previously considered a lazy pasting of articles someone had speed-read or skimmed. (It also doesn't help interpersonally that my brother is a speed-reader who can form eloquent opinions on a document within seconds, although apparently it's fabulous in trial).
So you're right to point that out, but it's also a red flag to me that some sort of mass lead/heavy metal poisoning is happening to our society, to see both credulity and shirking of responsibility from one of the smartest people I know.
I'm just rambling now, but I also have a cousin who was an engineer for years on the Supercollider in Texas. Also a very intelligent guy. But during covid, he had retired by then, he started spamming the whole family with emails insisting that covid was actually caused by 5G radio waves and that vaccines were poison. I outright attacked him in a full family-wide email, telling him that I had lost all respect for his once-great mind, which I had formerly looked up to.
Somehow I feel like the lack of analytical discrimination, logic, and above all effort is the common thread here.