PSA: The LessWrong Feedback Service
At the bottom of the LessWrong post editor, if you have at least 100 global karma, you may have noticed this button.
Many people click the button, and are jumpscared when it starts an Intercom chat with a professional editor (me), asking what sort of feedback they’d like.
So, that’s what it does. It’s a summon Justis button.
Why summon Justis?
To get feedback on your post, of just about any sort. Typo fixes, grammar checks, sanity checks, clarity checks, fit for LessWrong, the works. If you use the LessWrong editor (as opposed to the Markdown editor) I can leave comments and suggestions directly inline. I also provide detailed narrative feedback (unless you explicitly don’t want this) in the Intercom chat itself.
The feedback is totally without pressure. You can throw it all away, or just keep the bits you like. Or use it all!
In any case, I aim to provide feedback within about a day. So you won’t have to wait long.
Why Justis in particular?
Here’s my editing website, if you want some credentials. The shorter version is that I’ve been editing in this corner of the internet for about a decade, I have a penchant for giving detailed feedback (indeed, I once ran a literary review that also gave prompt and detailed feedback to every submission), and I’m numerate enough that I can follow and usefully engage with (most) technical submissions.
I think I’m a pretty good editor, too.
Am I doing it right?
Probably! At current margins, users worry too much about this. Most commonly:
How often can I request feedback?
There’s no hard limit. We’ll tell you if it’s getting to be too much, but even if that does happen (in four years, it’s happened exactly once), we’ll just ask you to slow down.
Relatedly, it’s totally normal to use the feature for an entire sequence, either as the posts are drafted or in a batch.
Can I use the feature for linkposts/crossposts?
Yep! As long as you also intend to post the finished product to LessWrong in some form, that’s fine. If you have some specific reason not to crosspost, but some piece of writing is very obviously LessWrong relevant (e.g. it’s an op ed or formal academic paper), you can also ask, and we’ll decide whether to give feedback on a case by case basis.
What if I click the button by mistake?
Happens all the time. Just say so! I won’t read your draft.
Should I credit you?
Up to you! If you don’t, I won’t mention that I provided feedback to that post. It makes me happy when people want to, but it’s definitely not a requirement. Most people don’t.
Couldn’t I just use an LLM?
You could! In fact, I tend to run posts on my blog by multiple LLMs. But they’re pretty bad at catching subtler inconsistencies, and their “rewording for clarity” suggestions often change your meaning, and (almost) always steer your prose toward their style. I think good human editors are still better, but I would think that!
Since the LessWrong service is free, if I were you I’d try both, and use whatever combination you end up preferring.
Why does Justis do this?
For money, and also for love of the game. The service is free to users, but subsidized by LessWrong. It is pretty fun to be a (contract) professional blog post critic. Believe in your dreams.
Also, while I have your attention, I’ll be at LessOnline 2025. If you’ve enjoyed my feedback before, I hope to meet you there!
- PSA: EA Forum Feedback Service by (EA Forum; 15 May 2025 20:14 UTC; 45 points)
- Using the LessWrong Editor for a Static Blog by (5 Sep 2025 18:00 UTC; 15 points)
- Contain and verify: The endgame of US-China AI competition by (22 May 2025 8:13 UTC; 6 points)
- 's comment on CstineSublime’s Shortform by (23 Jul 2025 2:16 UTC; 5 points)
Justis has been very helpful as a copy-editor for a bunch of Redwood content over the last 18 months!
Cool service/feature, but would it be worth defusing the “jumpscare” with an interstitial that explains the function of the button? At least the first time any given user clicks it.
Justis upped the clarity of my writing by 10% with his constant reminders to clarify what I mean by “this”, “that”, and “things”.
Haha, yup. I have a Shoulder Justis now that frequently reminds me
of thisto disambiguate words like “this” and “that”, which I’m grateful for.This is really cool, and the PSA was helpful since I had no idea. Kudos to you and the LW team!
This seems to imply there are better and worse times of day to try and get immediate feedback. If so, what are they?
Your odds are probably best in the evening in Eastern US time, and worst in the middle of the night in the same time zone.
Justis has been helpful as a copy-editor for some AIsafety.info content recently. Will be
abusingusing his services more.Recently had Justis give me feedback on a post. A bunch of failure modes I tend to fall into were pointed out to me. So, not only was this one post improved, but all the writing I do from now on will be too.
FWIW I ran a draft post through this service once, asked some questions along the way, and got exactly the kind of feedback I was looking for. Some of the feedback resulted in a minor cascade of realizations about what edits I’d want to make and how I’d want to think about things, and then some other stuff happened that shoved the task priority down indefinitely, so for now the draft is still sitting there unpublished—but consider this an endorsement!
I had no idea about this; thanks for flagging it! And thanks LW for providing this service
I would be great to have automated feedback on the epistemics of a piece of text. An LLM that can read text and identify reasoning errors or add appropriate qualifiers. As a browser plugin, it would also be helpful when reading news articles. Perhaps it can be done by using the Constitutional AI methodology and using Rationality: From A-Z(or something similar) as the constitution.