After talking to Elizabeth, and based on new information I’ve learned since this was posted, I’ve updated the OP to reflect what we’ve learned and avoid giving anyone the impression this change had a bigger impact than I believe that it did. The central points remain unchanged.
There were enough different changes that I’d like mods to reimport to ensure both copies reflect the same changes.
I greatly appreciate the time Zvi put into talking to me and that he updated the post. For posterity, I would like to note that this covers ~30% of the change I wanted. In particular, I’ve gone from feeling confident Zvi’s estimate of the impact of the change was that it was sufficient to clear the problem, to not knowing what he thinks the estimate is, aside from >0 (which I medium-confidence disagree with, but that’s a different issue). This may or may not be because the information isn’t there: it could also be that it’s there but I didn’t read carefully enough to derive it (and then we can debate how much effort is reasonable to expect from readers).
I don’t expect further edits based on this, but it seemed like useful information to share.
Flagging here my having-slept-on-it update policy going forward for this particular post:
If there’s a factual error anywhere, or a grammar/typo problem, I will continue to fix this particular post. If I say X happened and it didn’t, I should fix it, and I should fix technical mistakes.
If there’s an impression problem, where it’s felt this gives the impression that I’m claiming the change did more to solve the logjam than it did, or that we should be a certain level of excited, or something similar, nope, not going to change this post further for that, this is as far as I’m willing to go.
If there’s additional new information that seems relevant, beyond what I added this time, that’s a thing to be fixed with additional posts, and if they get written I will modify this particular post to link to those new posts, but not to directly reflect new information, again unless it causes a factual error as per #1.
Flagging the update policy I have in general as well:
If the post is timeless—as in, being used permanently as a link by myself or others on a regular basis, and continuing to get readers, or about to go into the yearly review, or something—and there’s a factual error, I’ll fix it.
If the post isn’t timeless—it’s a weekly Covid post, or didn’t become much of a thing, or whatever—then I’ll fix errors within about a week, and otherwise consider myself free to not worry about it and move on, to avoid creating an undue burden.
Anything beyond fixing factual errors in posts that don’t get special attention/hits is entirely optional and requires exceptional circumstances for that to change. Typos and grammar errors and such will be fixed when it seems worth it, otherwise not.
After talking to Elizabeth, and based on new information I’ve learned since this was posted, I’ve updated the OP to reflect what we’ve learned and avoid giving anyone the impression this change had a bigger impact than I believe that it did. The central points remain unchanged.
There were enough different changes that I’d like mods to reimport to ensure both copies reflect the same changes.
I greatly appreciate the time Zvi put into talking to me and that he updated the post. For posterity, I would like to note that this covers ~30% of the change I wanted. In particular, I’ve gone from feeling confident Zvi’s estimate of the impact of the change was that it was sufficient to clear the problem, to not knowing what he thinks the estimate is, aside from >0 (which I medium-confidence disagree with, but that’s a different issue). This may or may not be because the information isn’t there: it could also be that it’s there but I didn’t read carefully enough to derive it (and then we can debate how much effort is reasonable to expect from readers).
I don’t expect further edits based on this, but it seemed like useful information to share.
Flagging here my having-slept-on-it update policy going forward for this particular post:
If there’s a factual error anywhere, or a grammar/typo problem, I will continue to fix this particular post. If I say X happened and it didn’t, I should fix it, and I should fix technical mistakes.
If there’s an impression problem, where it’s felt this gives the impression that I’m claiming the change did more to solve the logjam than it did, or that we should be a certain level of excited, or something similar, nope, not going to change this post further for that, this is as far as I’m willing to go.
If there’s additional new information that seems relevant, beyond what I added this time, that’s a thing to be fixed with additional posts, and if they get written I will modify this particular post to link to those new posts, but not to directly reflect new information, again unless it causes a factual error as per #1.
Flagging the update policy I have in general as well:
If the post is timeless—as in, being used permanently as a link by myself or others on a regular basis, and continuing to get readers, or about to go into the yearly review, or something—and there’s a factual error, I’ll fix it.
If the post isn’t timeless—it’s a weekly Covid post, or didn’t become much of a thing, or whatever—then I’ll fix errors within about a week, and otherwise consider myself free to not worry about it and move on, to avoid creating an undue burden.
Anything beyond fixing factual errors in posts that don’t get special attention/hits is entirely optional and requires exceptional circumstances for that to change. Typos and grammar errors and such will be fixed when it seems worth it, otherwise not.