I Am a Programmer, Not a Rubber-Stamp That Approves Copilot Generated Code

Programmers wrote the StackOverflow answer and wrote that library.

- Check stackoverflow only for very niche issues, never finds what he needs but reaches a solution reading multiple answers and sometimes used to post a better solution for his issue

- Have his own templates if he does repetitive and boring stuff (common), implements the complex logic if any first and get the rest done as fast as possible being mildly disgusted.

- Imports libraries and often take a look at the code noticing stuff that could be improved. Has private forks of some popular opensource libraries that fix issues or improve performance fixing silly errors upstream. Sometimes he is allowed/has time to send the fixes back upstream. When using those libraries sometimes he finds bugs, and the first thing he does is checking the code and try to fix them directly, no tickets to the maintainers, often opens a PR with the fix directly.

I guess it's great for AI companies that they've managed to bait and switch "this will improve your productivity" to "this is how much time you're sinking into this, let's not care about if that was useful"

I heard a rumor recently that AWS are doing this, and managers are evaluated based on what percentage of their direct reports used an LLM (an Amazon-approved model) at least once over a given time period.

I'm pretty sure Cursor also has something similar?

I'm making a comment precisely because it's not obvious when reading the code, and the AI will make up some generic and completely wrong reason.

It is like having an obnoxious co-worker shoving me to the side everytime i type a new line and complete a whole block of code and asking me if it is good without regards to how many times I rejected those changes.

I still use AI, but favor a copy paste flow where I at least need to look at what i am copying and locating the code I am pasting to. At least i am aware of the methods or function names and general code organization.

I also ask for small copy paste changes so that I keep it digestible. A bonus point is that ChatGPT in firefox when the context gets too big, the browser basically slowsdown locks and it works as a form extra sense that the context window is too big and LLM is about to start saying non-sense.

That said AI, is an amazing tool for prototyping and help when out of my domain of expertise.

Every time Visual Studio updates, it’ll turn back on the thing that shoves a ludicrously wrong, won’t even compile, not what I was in the middle of doing line of garbage code in front of my cursor, ready to autocomplete in and waste my time deleting if I touch the wrong key.

This is the thing that Microsoft thinks is important enough to be worth burning goodwill by re-enabling every few weeks, so I’m left to conclude that this is the state of the art.

Thus far I haven’t been impressed enough to make it five lines of typing before having to stop what I’m doing and google how to turn it off again.

The autocomplete, I find it useful. Specially doing menial, very automatic stuff like moving stuff when I refactor long methods. Even the suggestions of comments looks useful. However, the frequency with it jumps it's annoying. It needs to be dialed down somehow (I can only disable it). Plus, it eats the allowed autocomplete quota very quickly.

The "agent" chat. It's like tossing a coin. I find very useful when I need to write a tests for a class that don't have. At least, allows me to avoid writing the boiler player. But usually, I need to fix the mocking setup. Another case when it worked fine, it's when helped me to fix a warning that I had on a few VUE2 components. However, in other instances, I saw miserable falling to write useful code or messing very bad with the code. Our source code is in ISO8859-1 (I asked many times to migrate it to UTF8), and for some reason, sometimes Copilot agent messes the encoding and I need to manually fix all the mess.

So... The agent/chat mode, I think that could be useful, if you know in what cases it would do it ok. The autocomplete is very useful, but needs to be dialed down.

For those on VS, this is how to hide it, if using 17.14 or later,

https://learn.microsoft.com/en-us/visualstudio/ide/copilot-n...

like ... you expect people to actually be committed to "the value of a hard day's work" for its own sake? when owners aren't committed to value of a hard day's worker? and you think that your position is the respectable/wise one? lol

There's a very large number of cases where that's the right choice for the business.

Actual code/projects? Detrimental

[1] E.g. I spent an evening on this: https://github.com/dmitriid/mop

I would guess that interest, passion, and motivation all play a role here. It's kind of like programming itself. If you sit people down and make them program for awhile, some will get good at it and some won't.

And if they really tied their livelihood to working at the same company for next decade because they maxed out their lifestyle relative to the income generated by that company, then that falls all on them and I don't actually feel that bad for them.

Personally I want my MSFT position to increase, so I’m cool with whatever the company does to increase the share price.

It's not just Microsoft. Other smaller employers are aping those guys.

My employer has an utterly ridiculous PowerBI dashboard tracking how much every employee uses LLM-based tools. Make sure to enable the Premium models, because otherwise you won't get credit! There are naughty lists for people whose usage is too low. Luckily the usage goals (for now) aren't very high.

However, they're also getting anal about tracking tasks, and the powers at be have asserted control over all aspects of story creation and work process. There's speculation they're going to start tracking story completion rates, and demanding measured productivity increases.

I know there are people still using PHP 5 and deploying via FTP, but most people moved on to be better professionals and use better tools. Many people are doing this to AI, too, me included.

The problem is that some big companies and influential people treat AI as a silver bullet and convince investors and customers to think the same way. These people aren't thinking about how much AI can help people be productive. They are just thinking about how much revenue it can give until the bubble pops.

I've worked with people using vim who wildly outproduce full teams using IDEs, and I have a strong suspicion that forcing the vim person to use an IDE would lower their productivity, and vice versa

Saying that the people are the problem instead of the tool is a lazy argument IMO. "Its not the companies fault, its the customer"

Useful tools, but I think the idea that they'll replace programmers is (wishful? eek) thinking.

Similarly, many people don't like learning new tools, and don't like changing their behavior. Especially if it's something they enjoy vs something good for the business. It's 2025 and people will have adamantly used vim for 25 years; some people aren't likely to change what they're comfortable with. Regardless of what is good for productivity (which vim may or may not be), developers are picky about their tools, and its hard to convince people to try new things.

I think the assumption that people will choose to boost their own productivity is questionable, especially in the face of their own comfort or enjoyment, and if "the business" must wait for them to explore and discover it on their own time, they risk forgoing profits associated with that employee's work.

If the AI tools actually worked how they are marketed I’d use them because that’s less work for me to have to do. But they don’t.

And even if it was, that's also assuming this benefit would be superior to the benefit of remote work for the individual.