The Wednesday Wellness threads are meant to encourage users to ask for and provide advice and motivation to improve their lives. It isn't intended as a 'containment thread' and any content which could go here could instead be posted in its own thread. You could post:
-
Requests for advice and / or encouragement. On basically any topic and for any scale of problem.
-
Updates to let us know how you are doing. This provides valuable feedback on past advice / encouragement and will hopefully make people feel a little more motivated to follow through. If you want to be reminded to post your update, see the post titled 'update reminders', below.
-
Advice. This can be in response to a request for advice or just something that you think could be generally useful for many people here.
-
Encouragement. Probably best directed at specific users, but if you feel like just encouraging people in general I don't think anyone is going to object. I don't think I really need to say this, but just to be clear; encouragement should have a generally positive tone and not shame people (if people feel that shame might be an effective tool for motivating people, please discuss this so we can form a group consensus on how to use it rather than just trying it).
Jump in the discussion.
No email address required.
Notes -
My apologies. I was very annoyed, for what I hope were understandable reasons. I'm happy to accept feedback when it's not framed as a personal attack alongside, IMO, very poor justification. I'm happy to hear what you have to say!
Hmm.
The thing is, markdown is cool and incredibly powerful. LLM chatbots like ChatGPT (that aren't base models), are under heavy selection pressure to conform to human preferences. That means a convergence to certain norms, because the average user or RLHF monkey prefers! Headlines, emphasis, bullet points, em-dashes — they're all useful. They make text more legible and help it flow better.
In other words, I've come to appreciate the benefits to writing in a certain structure. I personally prefer it, and I think the majority do (by revealed preference) and it strikes some people as AI-like. The last bit is an unfortunate side effect.
(I would say a bigger influence is Scott. I'm a fanboy, and his advice is solid)
I'm not sure what you mean by a change in tone or word choice, though I make an intentional effort to be less acerbic these days.
However:
I do use AI, sometimes! I've never tried to hide it, or deny its influence when anyone asks. That does not mean that any of my posts are writtrn by AI. I use LLMs for research, fact checking, proof-reading and editorial purposes.
That usually entails writing a draft, then submitting it into an LLM for advice or critique, which I may or may not use.
I think this is entirely above board, and I champion its use. It is categorically not the same as throwing a prompt into a box and then getting the AI to do the heavy lifting. The AI is an editor, not a ghostwriter.
Precisely the opposite. My style has changed, for what I think is the better. I'd hope so, given that I must have written like 1-5 million words in between, including a novel. It has also become more LLM-like, but that is because I like some of the things LLMs do, and not because I'm replaced by an LLM. Case in point, I've never had anyone accuse me of including unsourced or inaccurate information, even when they're criticizing my style, because it's a point of pride that I always review anything an LLM tells me.
When I said:
I mean that that specific comment had zero AI in it, and is of a style that strikes me as self_made_human from a few years back, as raw as it gets. It was quickly jotted off, with none of the usual edits, revisions or edit passes I make a point of doing manually. It is as me as it gets, and wouldn't be out of place three years back. It lacks the effort and polish I aspire to today.
Hell, I was doubly mad because I made an intentional effort not to succumb to just asking him to check ChatGPT (which would have given him excellent advice on a topic as done to death as this one) since he clearly wanted a more personal touch. I didn't even ask ChatGPT to write boilerplate that I could have theoretically co-opted as my own. I saw the comment, noticed, hey, I'm actually studying NICE guidance on initiating and managing antidepressant usage, and decided to just scribble down my understanding of best practice. I am, after all, mostly a shrink, even if I'm got more shrinking to do.
So, here I am, providing what I hope is accurate and helpful advice, the old-fashioned way, and someone comes along and starts shit. I might be a moderator, but I have my limits. Anyone calling me a "slopmonger" can fuck right off. As this current example of discourse demonstrates, I am more than happy to be civil and take pains to explain myself if the other person extends me the same courtesy. I appreciate that you have.
https://www.themotte.org/post/2368/culture-war-roundup-for-the-week/354239?context=8#context
Here is a thread outlining my stance towards prior accusations of AI usage, where I am perfectly happy to acknowledge that I have used it (when I've actually used it). You'll notice that I've spent a great deal of time explaining the same thing to jkf in good faith, in an attempt to convince him of the merits of my stance. That hasn't worked, and I am offended by new accusations when the evidence on display is very clearly not AI. It's like someone going around with a loudspeaker telling people I'm a sex offender, when the rap was for public urination while drunk. Even if it was technically correct (it wasn't here), I have little energy to spare to have this argument again.
Alternatively, this:
https://www.themotte.org/post/2368/culture-war-roundup-for-the-week/354252?context=8#context
I don't think opening the moderation log is an abuse of mod power in any meaningful sense. Moderation actions are public, anyone can see them on the sidebar. The panel only shows me the ones linked to a specific user. I didn't slap him with a ban, or start a fight. Moderators are only human, mea culpa.
If he's going to call me a slopmonger, when I think I've got more than enough evidence of engagement (presumably high quality, though everyone is at liberty to form their own opinions, I'm not your dad, I think), then I feel within my rights to point out that he has almost nothing to his name, and what he does is negative. It's genuinely impressive to have been here so long and still achieve so little. Both lurking moar or engaging less are valid options.
And I hope that I have demonstrated, to your satisfaction, that I am usually open to criticism, and have, in fact, had this same conversation with him in the past.
I'm not even against using a LLM to refine your writing. I wish I had so I wouldn't have made that annoying set of typos.
I do think that particular bit of criticism was poorly formed, and I would have been very annoyed as well. I didn't understand of all posts to try to call out like that you would choose that one. Somehow I thought it would be unhelpful to leave it as if it was just one user trolling you though.
Ironically, maybe that is what I've been noticing. I think, I'm probably unreasonably annoyed by that cloying droll persona that they give the average chatbot. I suppose, liking a slightly facetious and combative tone is slightly pathological on my part.
Funny story. Do you know why I made that effort?
A guilty pleasure of mine is to copy and paste entire pages of my profile into an LLM and ask for a summary/user profile (without telling them I'm the user in question). When I first started, maybe a year or so back, I noticed that the models would regularly call me acerbic and prone to cutting humor, even when they happily acknowledged the positives.
I thought about it, and decided, huh, it might be worth an effort to intentionally tone it down myself. If it's not obvious, I adore Scott, and he is probably so mild-mannered that his toddlers walk all over him.
(Oh, wait.)
So I decided, hey, it's worth trying to be nicer, even though I do not suffer fools gladly. Or perhaps I'm getting old, and realizing that yelling at people on the internet is of little utility and only raises my blood pressure.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link