[ Question ]

What is a good moment to start writing?

by toonalfrink1 min read29th May 201917 comments


Writing (communication method)Productivity

I notice that I never have a complete answer to anything.

I may have digged into something way more than my peers, but there are always all kinds of hunches I haven't put into words yet, all kinds of data that I haven't internalized yet, all kinds of unresolved confusions.

I know much more than a year ago. If I explain my ideas now, I'm going to be embarrassed by it next year.

My opposite worry is that I will learn too much. Forget how I learned all this stuff, at which point the inferential distance is too large to overcome.

This worry is not about impressions. It's about not wasting people's time with half-baked ideas. Naively one might just max out the tree of details before reporting them, but the tree is never maxed out. There is always ambiguity somewhere. How much is acceptable to pass on?

New Answer
Ask Related Question
New Comment

6 Answers

What is a good moment to start writing?


Pithy replies aside, I think it's likely the answer you could do with hearing. Echoing what some other people have said, I recommend finding what feels like a super low stakes place to write (personal blog, a shortform feed, private google doc, emails to a friend) and work on writing a lot. I find that having a thought-out channel makes it easier to have thoughts in the first place. If you are frequently in the practice of trying to crystallize ideas, you'll get better at it.

I find that whenever I try to "do a post justice" it kills my ability and motivation. I think what's happening in my head is that I assume a mantle of "well I'm supposed to be a careful thinking, well written, master rationalist, so anything that isn't that can't get out, or else how can I act authoritative?". It helps to remind myself that my intent is not to be an authority, but to express my ideas.

Perpetual draft style might also help with getting over "but what if I'm wrong?".

Also, specifically speaking to:

I know much more than a year ago. If I explain my ideas now, I'm going to be embarrassed by it next year.

I have a lot of notebooks that I've kept over the last 5 years that contain a lot of whatever interesting-to-me thing I was thinking at the time. It's been incredibly useful to have this record! A lot of the breakthroughs, epiphanies, and slow-shifts-in-thought aren't completely unique and different from thoughts I've had in the past. It's been super useful to ask "What has changed since the last time I thought about this?" and actually be able to find a record of how I used to think, and see the difference between now and then.

If you can switch from "I'm embarrassed at what I used to think" to "See how much I've grown!" you might find it much easier to write.


If you have an idea and are inspired to write, write. If you have an idea that you would understand better if you wrote it out, write. If you want to get better at writing, write. Now. Ideally, every day.

That's how one improves at writing. Also understanding. I learn a ton by trying to write stuff out.

Often I then trash what I'd written. Don't be afraid to do that, either. Writing and then discarding is not bad. It's not a waste of time.

If you write what you know now, then a year from now you know more, great! Write it again. Show people how you got from there to here.

When should you post what you've written versus not, once done writing (or at least ready for the editing pass)? That's trickier, but again, I think that the bigger and more common mistake is not posting. Most of the things you're worried might happen aren't even bad.

Consider writing your thoughts in a personal blog, often and without procrastinating, and link it here when you feel it's good enough for others here to read. That's how many regulars here operate.

I've been doing the "shortform-feed" thing that many people have been doing on the site. I've found it to work quite well.


4Wei_Dai2yJust to be sure, you're not encouraging this as a norm for everyone, in other words, people who feel comfortable just posting short articles to their "personal" blog on LW should still do so?
4Raemon2yFWIW I argue with habryka a lot that it's pretty important that personal-blogposts feel like a place you can write off the cuff ideas and that I'm worried that he himself doesn't feel comfortable doing so. (That said, it's still understandable that he doesn't.)
2habryka2yYep, I am only encouraging this as something that might make people feel even more comfortable. Short ideas as top-level posts on your personal blog are definitely encouraged.
3Hazard2yI second having a shortform feed! I try to collect a lot of "minimum intriguing ideas". One way I interact with my shortform is to not work on a post over time. If I have a brief nugget of an idea that seems like it might be self contained or interesting, I try to just force it out. If I get it, great. If I get a mess, I ditch it and don't worry about coming back to finish it later. (Note: this was a comment on how I shortform, not really my advice for toon)

If you don't write in the moment of inspiration it either won't get done or will turn into very boring report by committee style prose. IME.

It's about not wasting people's time with half-baked ideas.

If you waste someone's time, that's not your responsibility, they decided to come over to your blog and lay their eyes on your …research notes, let's say.

Write immediately and write continuously, as you learn. This will (1) give you practice, so that your writing is much better by the time you have something really good that would deserve good writing, and (2) you get feedback from your friends (if you gently beat it out of them), which will make your progress on your ideas faster.

If I explain my ideas now, I'm going to be embarrassed by it next year.

What if you just honestly report on what's going on in your head? Not, "folks, I know the Truth, let me lay it out for you," but "these are the results of my research so far, and these are the half-baked intuitions that I get out of that, at the moment, and this is where I'm planning to go next to sharpen or falsify those intuitions." Sounds reasonable and non-embarrassing.

Also, the more often you bump your thoughts against the harsh judgement of other people, the faster they sharpen up and the less embarrassing they will be in the long run. Something like that.

(Half jokingly) What about using imitation learning followed by reinforcement learning? In other words, start by looking at what kinds of posts your role models write, try to infer when they tend to start writing, do the same thing and then make adjustments as you go using RL. I think this is what I did (subconsciously).

If you want a more explicit kind of answer, I think I need a better idea of what your dilemma is. You wrote:

If I explain my ideas now, I’m going to be embarrassed by it next year.

but also:

This worry is not about impressions.

which confuses me because it seems like worrying about being embarrassed is worrying about impressions?

which confuses me because it seems like worrying about being embarrassed is worrying about impressions?

What I meant to say is that I can tell that my work isn't going to be very good from next year's standards, which are better standards because they're more informed

3Wei_Dai2yWhy is that a problem, unless you're worried about wasting people’s time with half-baked ideas (which you said that you're not)? Surely we should just judge our work by our current standards, otherwise we might as well judge our work by what we expect post-Singularity standards to be and then never publish anything?
1toonalfrink2yBecause some people might already be at this level, and I worry that I'm just adding noise to their signal. Maybe my question is this: given that, every year, I unexpectedly learn important considerations that discredit my old beliefs, how can I tell that my models are further along this process than those written by others?
2Wei_Dai2yI guess you can either write down your current models and ask people privately for feedback, or just talk to people who you think might have better models than you and try to learn from them. Write down your models for public consumption when your private feedback/learning suggests that your models are as good as the state of the art, or at least competitive with what has been publicly written down. Also I'm curious, what important considerations have you learned recently?
0Pattern2yI think if you write, next year's standards will be even better standards because you'll improve. And if you want, you'll be able to write them better, and if not, stuff you write after that will be better.
1 comments, sorted by Highlighting new comments since Today at 2:15 AM

" I know much more than a year ago. If I explain my ideas now, I'm going to be embarrassed by it next year. "

My take is that the ideas are typically not something to be embarrassed about. But perhaps that is viewing idea as something other than "developed theory/model" or "answer".

If you wait until you are sure you won't be embarrassed in the future you may never say very much. (I wonder if your post here is somewhat related to the earlier post about saying it wrong.) And when you do you may not really be adding anything to what has already been said --then again I was told a long time ago that most of us will never say/write anything that new; but we will say/write that in our way so....