I actually did this based on writing advice from Eliezer—he talks about putting examples before summary.
I do agree that a short intro would be an improvement—I had one written that my proofreader didn’t like so I ended up just posting without since my title is already a bit of a summary.
In general, “start with examples” is often good advice, and I sometimes do it myself. But I think that it’s somewhat problematic here, since the examples are somewhat long and, furthermore, the way they’re structured makes it somewhat unclear what the reader should be focusing on.
Actually, “these examples could be clarified” is probably more useful advice for me to give than “put a summary first”, given that they’d feel a little confusing even if you did have the summary first. I’ll elaborate.
(Obvious caveat: the follow is just my impression, other people may have found this clearer.)
The first example is clear enough until the end of the first paragraph. Then the second paragraph starts with
To contrast, with animal training, you reinforce behavior you want in the animal, and interrupt, redirect, or completely ignore (ie: no shaming or guilting) behavior you don’t want.
We start out with “to contrast”, but for me at least, it wasn’t clear what the two things that were being contrasted were. Now that I know the right interpretation, I think the intended contrast was something like “eliminating guilt when you’re running a guilt-based system hurts, versus, eliminating guilt while also providing an alternative positive motivator helps”.
But notice that as written, the second paragraph doesn’t actually talk about systems that also eliminate guilt, but rather about systems which don’t increase guilt (ie. systems that do not include a guilt-increasing component). And on top of that, there is the mention of “[m]editation does not use reward during the meditative process”, which makes it sound like meditation also didn’t have any particular positive motivators either. You do mention that people often do add extra positive rituals to meditation, but it currently sounds like a sidenote that admits the existence of an exception to the general rule, rather than the general rule itself.
So, at least to me, by the second paragraph my idea of the contrast you were trying to make was something like “eliminating guilt when you’re running a guilt-based system hurts, versus, there exist systems that don’t care about guilt”—which isn’t really a clear contrast, and it felt obvious that I was missing something.
The next paragraph helped a little, though again its connection to the previous paragraph felt unclear—“So, if you...” sounded like it would be a straight follow-up to what had been just said, but it was referring to animal training, which had only been discussed for one sentence a while back, with the (seemingly unrelated, since it didn’t seem to talk about rewards or positive reinforcement) meditation discussion in between.
Then there came the example summary, which made me go, “oh, okay”, but it still felt confusing since the summary of the example didn’t seem to be talking about the same thing as the actual text of the example.
Then we got to example two. It was still unclear to me what the most important gist of the first example had been, and now we seemed to jump to an entirely different subject. It was also structured differently from the first example—unlike the first example, it didn’t start by giving a single paragraph’s worth of description of a system that failed when a single variable was changed. Instead, the first four paragraphs were basically an introduction, then it suggested changing a single variable in the fifth, and aside for a brief “that would make a lot people comfortable”, it didn’t even say anything about the negative effect of only having a single-variable change until we were already in the seventh paragraph. And unlike in the first example, where we contrasted two different techniques (one guilt-based vs. one reward-based), this example contrasted a partial and full technique (only switch to “yes means yes” vs. switch to “yes means yes” and also make people more assertive), again making the structure of the example more dissimilar to the previous one.
If you have two examples from very different domains, it helps a lot to have them follow a similar structure—that way, the reader can at least pick up on the structure and let that guide their interpretation of what the relevant similarities and dissimilarities in the examples are. The more different the structure, the larger the risk that the reader gets confused about what they’re expected to pay attention to—like happened to me.
I don’t know how common my reaction was, but on my part the problem was also exacerbated by the fact that “let’s switch to yes means yes” didn’t seem like a particularly obvious example of something where you’d need to change several variables in order to make it work well. I first ran into the idea that we should rather be promoting a norm of “yes means yes” rather than “no means no” some months back, and then it seemed to me like a clear and obvious improvement. It didn’t occur to me before reading your post that such a change would also require any major adjustments to people’s assertiveness—and in fact this still doesn’t feel particularly clear to me, and feels more like the kind of minor change that would occur automatically and which would be just a minor hassle at most. So that also made it harder to intuitively grasp what the example was trying to say.
If you want to start with two examples, by the time the reader has read them, that person should already have at least a vague idea of what those examples were about. For me that didn’t happen, and the examples actually felt disconnected with the overall thesis even after the thesis had been presented in the “overall summary” paragraph.
Thanks for taking the time and energy to explain all of this. Agreed that my writing could have been much more cohesive and better structured, and I appreciate the explanation of why, where, and how. I will take this into account fo the next thing I write.
If you’re mostly only interacting with the type of women who are part of the LW community, you probably have an unusual view on female assertiveness among (American) women—most of us are much more assertive than average when it comes to sexuality.
It’s probably a cultural difference. I’ve also frequently heard people on this site mention that “men are expected to make the first move”, but that doesn’t seem to be true in Finland either, at least based on my experience.
But the examples are long, and with no introduction whatsoever, I can’t tell what they’re examples of, so I don’t know what to do as I’m reading the examples. (I confess I didn’t read this post, and this is the other reason why aside from the formatting.)
Lol, so if I take the time to reformat, you’ll read it? :) The title is actually the summary. Its long for a title and short for a summary, but was the best I was able to come up with as a non-professional writer having already spent a fair bit of time editing and not wanting to spend more on that particular nuance.
I actually did this based on writing advice from Eliezer—he talks about putting examples before summary.
I do agree that a short intro would be an improvement—I had one written that my proofreader didn’t like so I ended up just posting without since my title is already a bit of a summary.
In general, “start with examples” is often good advice, and I sometimes do it myself. But I think that it’s somewhat problematic here, since the examples are somewhat long and, furthermore, the way they’re structured makes it somewhat unclear what the reader should be focusing on.
Actually, “these examples could be clarified” is probably more useful advice for me to give than “put a summary first”, given that they’d feel a little confusing even if you did have the summary first. I’ll elaborate.
(Obvious caveat: the follow is just my impression, other people may have found this clearer.)
The first example is clear enough until the end of the first paragraph. Then the second paragraph starts with
We start out with “to contrast”, but for me at least, it wasn’t clear what the two things that were being contrasted were. Now that I know the right interpretation, I think the intended contrast was something like “eliminating guilt when you’re running a guilt-based system hurts, versus, eliminating guilt while also providing an alternative positive motivator helps”.
But notice that as written, the second paragraph doesn’t actually talk about systems that also eliminate guilt, but rather about systems which don’t increase guilt (ie. systems that do not include a guilt-increasing component). And on top of that, there is the mention of “[m]editation does not use reward during the meditative process”, which makes it sound like meditation also didn’t have any particular positive motivators either. You do mention that people often do add extra positive rituals to meditation, but it currently sounds like a sidenote that admits the existence of an exception to the general rule, rather than the general rule itself.
So, at least to me, by the second paragraph my idea of the contrast you were trying to make was something like “eliminating guilt when you’re running a guilt-based system hurts, versus, there exist systems that don’t care about guilt”—which isn’t really a clear contrast, and it felt obvious that I was missing something.
The next paragraph helped a little, though again its connection to the previous paragraph felt unclear—“So, if you...” sounded like it would be a straight follow-up to what had been just said, but it was referring to animal training, which had only been discussed for one sentence a while back, with the (seemingly unrelated, since it didn’t seem to talk about rewards or positive reinforcement) meditation discussion in between.
Then there came the example summary, which made me go, “oh, okay”, but it still felt confusing since the summary of the example didn’t seem to be talking about the same thing as the actual text of the example.
Then we got to example two. It was still unclear to me what the most important gist of the first example had been, and now we seemed to jump to an entirely different subject. It was also structured differently from the first example—unlike the first example, it didn’t start by giving a single paragraph’s worth of description of a system that failed when a single variable was changed. Instead, the first four paragraphs were basically an introduction, then it suggested changing a single variable in the fifth, and aside for a brief “that would make a lot people comfortable”, it didn’t even say anything about the negative effect of only having a single-variable change until we were already in the seventh paragraph. And unlike in the first example, where we contrasted two different techniques (one guilt-based vs. one reward-based), this example contrasted a partial and full technique (only switch to “yes means yes” vs. switch to “yes means yes” and also make people more assertive), again making the structure of the example more dissimilar to the previous one.
If you have two examples from very different domains, it helps a lot to have them follow a similar structure—that way, the reader can at least pick up on the structure and let that guide their interpretation of what the relevant similarities and dissimilarities in the examples are. The more different the structure, the larger the risk that the reader gets confused about what they’re expected to pay attention to—like happened to me.
I don’t know how common my reaction was, but on my part the problem was also exacerbated by the fact that “let’s switch to yes means yes” didn’t seem like a particularly obvious example of something where you’d need to change several variables in order to make it work well. I first ran into the idea that we should rather be promoting a norm of “yes means yes” rather than “no means no” some months back, and then it seemed to me like a clear and obvious improvement. It didn’t occur to me before reading your post that such a change would also require any major adjustments to people’s assertiveness—and in fact this still doesn’t feel particularly clear to me, and feels more like the kind of minor change that would occur automatically and which would be just a minor hassle at most. So that also made it harder to intuitively grasp what the example was trying to say.
If you want to start with two examples, by the time the reader has read them, that person should already have at least a vague idea of what those examples were about. For me that didn’t happen, and the examples actually felt disconnected with the overall thesis even after the thesis had been presented in the “overall summary” paragraph.
Thanks for taking the time and energy to explain all of this. Agreed that my writing could have been much more cohesive and better structured, and I appreciate the explanation of why, where, and how. I will take this into account fo the next thing I write.
I’m glad you found it useful. :)
Someone gave an example re: the needing change in assertiveness/attiude:
http://lesswrong.com/lw/hj6/changing_systems_is_different_than_running/9547
If you’re mostly only interacting with the type of women who are part of the LW community, you probably have an unusual view on female assertiveness among (American) women—most of us are much more assertive than average when it comes to sexuality.
It’s probably a cultural difference. I’ve also frequently heard people on this site mention that “men are expected to make the first move”, but that doesn’t seem to be true in Finland either, at least based on my experience.
But the examples are long, and with no introduction whatsoever, I can’t tell what they’re examples of, so I don’t know what to do as I’m reading the examples. (I confess I didn’t read this post, and this is the other reason why aside from the formatting.)
Lol, so if I take the time to reformat, you’ll read it? :) The title is actually the summary. Its long for a title and short for a summary, but was the best I was able to come up with as a non-professional writer having already spent a fair bit of time editing and not wanting to spend more on that particular nuance.