What's the round trip latency on this? ask question -> response. Do you parse the question word by word and feed into llm or wait for the whole question before feeling into llm?
Was this good/bad? Idk Room for improvement? Maybe? Clearer direction with the benefit of hindsight? Maybe. but the majority of the sentiment in the responses is coming from people not type rated in a 737.
Software engineers are the translators. We (as a metaphorical community are in denial). Read the 100s of comments on this post: either (AI code is wrong or this isn’t correct or it’s not effective or some other justification). At the end of the day it’s really hard to accept the change.
True, recently I started feeling that part of what I had been doing is simply translating natural language to programming languages. Though coding also involves things such as algorithm and data structures, context and background knowledge, but these can all be done in natural language. Once the natural language description is given, what remains is only translation. LLMs have good knowledge of almost anything, though they are currently weak on inference or derivation, but this already makes them good on the two end of software engineering -- context knowledge and translation.
I understand those people who hate LLMs for coding, I partly share the feeling, because I enjoy typing on a keyboard, editing part of the code, reading the characters, I am an Emacs user. If LLMs can do the work, even if just save the typing and editing part, some of the fun has been eliminated for me.
Think about chess and Go, though AI can easily beats human now, people are still playing it. For programming, if one day AI can do 80% of the programming work, I guess only few of programmers today can keep doing it as a job. Just like few people can play chess and Go as their job.
I don't think that's a fair characterisation. I gave these tools the benefit of the doubt and many genuine tries, and still do occasionally. My judgment that they're unfit for the task is based on finding that they objectively don't help me work. There are studies showing that LLM use slows experienced developers down, even as they believe the opposite. I used to be the only one in my immediate work environment who thought LLMs are of jack use in my development work. Over the past year, most devs who used LLMs heavily have now switched over to my view.
Translation is natural language in, natural language out. That has very little to do with programming. Dijkstra knew back in the 70s that the very concept of natural language programming was bullshit, and that fancy new tech is never going to change the fundamental issues with the idea. In my opinion, his argument is as valid as ever.
Edit: I wasn't trying to imply that natural to natural language translation is trivial. The best LLM tools can be pretty shit at that even today, too. Tried using localised Microsoft developer documentation lately. It's unintelligible.
Thanks for you opinion. I agree with your opinion that current LLMs are less helpful to senior/experienced developers. Of course LLMs can't solve complicated, advanced problems, otherwise we would see great advance in science and technology now.
However, there are a lot of trivial work in software development. For example, the behavior of a button, or editable text. Same to any kind of art, most developers in the industry are doing trivial or ordinary work, and few are solving challenging problems. What I mean is that those who do trivial work will be culled, or liberated, from doing that as a job, while the elite could remain there to solve unique, challenging problems.
Recently I have been working with coding agent to fix bugs and implement new features. I implemented my solution first and let coding agent come up with its own solution. Most of the time, the coding agent came up a better solution than mine. Most of the time, its solution was not worse than mine, and I would keep most of its code and only do some minor improvement. Most probably, I am not an experienced/advanced developer, since I seldom deal with advanced math/algorithms/architecture in my day-to-day work. Thus, it would be natural that we have different feeling to the same thing.
I am interested in Dijkstra's idea about natural language programming you mentioned, is that EWD667?
1) not all coding is the same. You might be working on a production system. I might need a proof of concept
2) not everyone's use of the coding agents is the same
3) developer time, especially good developer time has a cost too
I would like to see an article that frames the tradeoffs of AI assisted coding. Specifically without assigning value judgments (ie goodness or badness). Really hard when your identity is built around writing code.
Letting an agent figure out how to compile old projects is magical. What used to be multiple days of slog is now “compile this, make changes and download tools as needed” with 10 mins of git review to make sure it didn’t do anything stupid.
Senior engineer: "you shouldn't have to use llm to do simple things"
Their boss: "hey I need you to write this thing in Go for performance reasons"
Senior engineer: "you shouldn't make me learn new languages, js is plenty performant for this task just install a new server. why are they making us use all these new tools? everything is new again, blah blah excuses excuses"
vs.
Boss: "hey can you write this in go"
Vibe Coder: "write this loop to ignore the first semicolon in this response", and then proceeds to move on with life.
It seems like when people evaluate the merits of what a code generation ai agent should and shouldn't do, they leave out a lot of their implicit assumptions about how project should work and their own theory of mind around coding.
"I don't know Claude did that" isn't a bad thing in and of itself... If someone is spending a bunch of time on code that Claude could have done and easily verified it was correct, they are going to move slower and produce less useful things of value than someone who cares about reading every line of code.
Any situation where you’re submitting under your signature code to production without knowing what it does should be at the very least cause for a talk.
The policeman says to the judge, on the stand "I don't know why my sworn affidavit says that, your honor. But I can write twice as many affidavits now so it's all for the best."
> If someone is spending a bunch of time on code that Claude could have done and easily verified it was correct, they are going to move slower and produce less useful things of value.
This is the future fellow greybeards. We will be shunned for being try-hards, and when it turns out we were right?... Well no one likes a know it all.
If it’s easily verified as correct, then you should have verified its correctness before bringing it to someone more senior and asking for their help, and so you should be able to explain why it is there when they ask.
If you don't understand you code how you can be sure it's correct? You actually are pushing it into your colleagues who will verify and fix the code later.
The only thing that changed with AI is that the narrative went from "you can't always know perfectly what every part of the program does" to "don't even try".
But the LLM writes the tests too. Probably even made some private methods public so it can test them, because you asked it to write a comprehensive test suite. It’s a beautiful self-jerk circle of shitty code based on wrong assumptions proven by correct tests testing shitty code based on wrong assumptions…
Or the wonderful loop of "These test steps are still failing, let's try something simpler and remove the failing tests. Great! The tests are passing now!"
Sad reality is test engineers headcount over last years was cut even more than developers. Most companies see testing as obstacle and unnecessary costs and has no will to invest into testing.
usually cynicism is a coping mechanism for underlying/subconscious pain... If you resolve that, then there is room for happier/positive thoughts to emerge. However, it's just like any other addiction (alcohol), it's not resolved in a day and takes hard work.
Much more comfortable to think; (everything sucks, those in power are out to get me) than (wow I really had a high hope for what my life would be, and this sure isn't it).
To achieve its goal I would be worried that it dropped the verbatim function signature. Claude still figured it out, but for more obscure stuff that could be an issue.
Following up — I just pushed a fix for this. This latest version significantly improves how references like protocol conformances and default implementations are rendered.
