upvote

points

by jcelerier19 hours ago |
comments
upvoteby wtallis19 hours ago|
next
[-]
I don't usually see Claude leaving comments that are obviously out of date compared to the code. The problem is that the comments are just a dumping grounds for Claude's stream of consciousness, verbosely recording random bits of history and context that are true and at least somewhat relevant, but without cohesively explaining why the code is the way it is. And some sentences in the comments will be beginner-style restating what's obvious from reading the code.
reply
upvoteby meowface15 hours ago|
parent|
next
[-]
Exactly. Comments are basically its scratchpad for whatever it wants. Ask it to change a number in a TOML config and it'll also add 2 new comment lines above it with some Claudeslop about it being changed and what that accomplishes, as a useless note to itself.

I now make sure to do a big decommenting pass before every PR.

reply
upvoteby senordevnyc7 hours ago|
parent|
[-]
This kind of stuff annoys me too, and I have instructions in agent.md to avoid it, which helps.

But I also am starting to just let go and stop caring. It’s not clear to me that it causes problems down the road, it’s easy to strip out en-masse if needed, and in my experience, agents now are really good at read git blame, the commit log, even prior agent transcripts if available to sleuth out when a change was made and why. So yeah, it’s annoying, but the code agents write for me is increasingly never read by a human, so does it matter?

reply
upvoteby hungryhobbit6 hours ago|
parent|
[-]
This. Be honest: who reads the human-written comments? ;)

When I read, I skip most comments, especially the larger "Javadoc" style. My brain sees them colored differently in the editor and it doesn't even take mental effort. Then, when I have a question about the code, I look back up for a relevant comment. That doesn't happen very often.

If Claude writes great code and leaves a garbled Claudese-but-accurate comment ... I can read and comprehend (with like 5x the effort of a human comment) ... that's a small price to pay.

reply
upvoteby sfjailbird9 hours ago|
parent|
prev|
next
[-]
Yes, comments might be my biggest gripe with CC due to what you said. 9 out of 10 of my revisions to Claude's work is deleting or rewriting comments. Such a silly thing for it to fail on.

(I do have detailed instructions for it on how to comment (or not) but it has not fixed this.)

reply
upvoteby Shadowmist7 hours ago|
parent|
prev|
next
[-]
You can only fit so many bits of watermarks in code... but comments on the other hand, the possibilities are endless!
reply
upvoteby stingraycharles19 hours ago|
prev|
[-]
Nah, it’s the typical Claude-isms, and way too much info, info in the wrong places, and putting lots of planning discussions into comments that are completely irrelevant.

It’s always “you explain only what but not why” or “this is way too much prose” or “these comments don’t belong here, they should be inline comments” or “this is completely redundant as it’s already obvious from the code”.

I do find that once I beat it into submission and the codebase is “clean”, the new code it generates gets better and better, which makes sense gives its pattern-prediction nature. But it seems like there is work to do for Anthropic in terms of getting Claude to not confuse code comments with dumping its interactive discussion state into there.

reply