every time a new tool appears, everyone around me is like "you should try you should try it". but i am here, stuck with my vim and tmux. and im glad i did. most of the new shiny things get enshittifed or acquired by shady corporations. long live vim+tmux
damn, i've been thinking about the same recently. saas is dead, at least developer oriented saas for sure. the only things i rely on are my own self hosted git (not public to the web), and that's it. all other "developer oriented saas" has been replaced with a home made version with claude. probably not safe from hacking, but id rather take the risk of someone targeting ME specifically, than let all the "saas platforms" store all my data and get hacked every monday and friday
i noticed something similar yet different. because, sadly, most of what i read today is llm output (the chain of thoughts, or the summary at the end of a big chunk or work), i tend to "adopt" this style in my writing, hence, writing as an llm without using one. humans are an amazing creatures at copying stuff and adapting to our environment
A billion humans use AI daily, they write about 0.5T tok and read about 2T tok (estimated). In a week they amount to more than the whole training set of GPT-4. Human language is being shaped back by AI language.
I don't understand these types of comments. Aren't the CVEs epic today with human written code? What difference does it make then? You want no CVEs, make software engineering into a real gate-kept engineering professions with accountability chains, government regulations, formal education, etc.
> Aren't the CVEs epic today with human written code? What difference does it make then?
LLMs cause errors at a noticeably higher rate than human devs do. If you lean on them without actually checking what they wrote, the error rate in your software is going to go up.
> You want no CVEs, make software engineering into a real gate-kept engineering professions with accountability chains, government regulations, formal education, etc.
This will not only not prevent CVEs, it will also make everything a giant pain in the ass
Of course it will. But at the same time I don't see building falling left and right like CVEs in software. So don't tell me agents write bad code, because humans do as well. The only way to prevent this (or minimize), is to turn software engineering into real engineering.
"If it works, ship it" is very different from "If it works, and I understand it because I wrote it, ship it".
Does the LLM understand it, in detail? Can you trust its interpretation of the information and instructions you provided? Did it independently make an adverse decision you haven't noticed, because you didn't read the code? Was the model missing any information it required -- and would you know if it was?
LLMs depend on humans and human oversight to ensure they have the information required to make the right decisions, to catch when they don't, and to steer them based on your much more complete understanding of the broader context and requirements. You can't rely on them to do that thinking for you (yet?).
I’d argue that most people also don’t understand code they read from their colleagues. Or if they do, then they don’t understand the entire context of the app.
You said LLM need human oversight and it’s true, but it does not imply you have to read the code.
I think we are going towards a place where code no longer makes sense, you just describe what you want in English, and the LLM just builds directly in machine code. You then build different verification tests to make sure the program is doing what it’s suppose to do.
How do you know if it's vibe coded or not? 60% of any app is third party dependencies that some random guy across the globe maintains, and the developer never even checked how the code is written.
Also, why vibe coded apps are prone to have viruses, and non vibe coded apps are virus free and have perfect security?
If it's open source, you can check the code, vibe coded or not (I doubt the majority does).
If it's a commercial app, the developer(s) probably care more about making money than writing the "perfectly secure" app, vibe coded or not.
No, it’s not. And that’s what scary with it. People try to cope with their slow obsolescence by saying that “AI needs skill”, while I churn out working code with basically a variation of “please implement this feature and make no mistakes”.
I used to think that programming is an art; that you get paid for the code. Too bad it took me decades to realize how wrong I was. AI sucks, it sucks the soul from this profession. But in the end, the code it produces works, so who cares?
> But in the end, the code it produces works, so who cares?
I care, precisely because the code it produces doesn't work. It has the appearance of working, but it is badly written and leads to problems down the line.
Kobo is great. Unfortunately - in India at least - the books are way more expensive on Kobo (often 2x or even more) than on Kindle. So I've found it hard to get rid of the Kindle though I prefer to read on the Kobo.
reply