Plausible yet pseudonymously provided & unverifiable info should not be promoted to Main, IMO. If he was willing to own his claims and at least his bio of working for a corporation that did something similar to what he indicates, that would be one thing, but he didn’t.
(I’m not going to criticize him for not being willing to risk the NDAs, but that doesn’t mean we should endorse the post and try to spread this post as far and wide as possible.)
While I read this post, and before I got to the grandparent comment, my reaction was “Wow, this is enough of a surprise to my current model (and most of the evidence is the personal statement of an unverified person on the Internet) that my probability of it being a fabrication is significant”. I don’t usually get that from a LW post.
Furthermore, the effect of promoting a fabrication is way worse for accusations of major malfeasance and racism than for most minor personal anecdotes.
(This is not saying that I think the post is probably a fabrication! I give it about a 2⁄3 probability of being true.)
We don’t generally (for instance) discredit posters for pseudonymity. Doing so for a particular post, but not others, invites the question — Why this one?
Most posts make arguments that readers can assess using public sources: if the argument and citations are good, they can stand separately from the poster. This post is personal testimony.
Second what gwern said, but more politely. This is an excellent post, and if it substantiated its claims (in general, not in NDA-violating specifics) it would definitely belong in Main, but as it stands it’s unverified (and perhaps unverifiable for good reasons), and thus I think we should leave it on Discussion (but I’m upvoting the post).
Promotion implies some significant level of endorsement (of the quality of evidence if not necessarily of the conclusions), and LW should be careful not to abuse the trust reposed in it by many readers.
This should probably be promoted to Main immediately.
Plausible yet pseudonymously provided & unverifiable info should not be promoted to Main, IMO. If he was willing to own his claims and at least his bio of working for a corporation that did something similar to what he indicates, that would be one thing, but he didn’t.
(I’m not going to criticize him for not being willing to risk the NDAs, but that doesn’t mean we should endorse the post and try to spread this post as far and wide as possible.)
On reflection, I agree. Consider my earlier recommendation retracted, though I’ll leave it up for contextual reasons.
Retraction puts strikethrough, so it’s still readble but noticably retracted. You should retract.
Thanks, done—I wasn’t sure whether that would bury the comment thread or not, but it looks like it’s okay.
I don’t see this level of scrutiny being applied to most posts here.
While I read this post, and before I got to the grandparent comment, my reaction was “Wow, this is enough of a surprise to my current model (and most of the evidence is the personal statement of an unverified person on the Internet) that my probability of it being a fabrication is significant”. I don’t usually get that from a LW post.
Furthermore, the effect of promoting a fabrication is way worse for accusations of major malfeasance and racism than for most minor personal anecdotes.
(This is not saying that I think the post is probably a fabrication! I give it about a 2⁄3 probability of being true.)
I don’t either, but that isn’t a compelling argument against applying that level of scrutiny.
We don’t generally (for instance) discredit posters for pseudonymity. Doing so for a particular post, but not others, invites the question — Why this one?
Most posts make arguments that readers can assess using public sources: if the argument and citations are good, they can stand separately from the poster. This post is personal testimony.
Good point — it’s not that the poster is pseudonymous, but that they don’t offer much to distinguish a factual post from a hoax.
Second what gwern said, but more politely. This is an excellent post, and if it substantiated its claims (in general, not in NDA-violating specifics) it would definitely belong in Main, but as it stands it’s unverified (and perhaps unverifiable for good reasons), and thus I think we should leave it on Discussion (but I’m upvoting the post).
Promotion implies some significant level of endorsement (of the quality of evidence if not necessarily of the conclusions), and LW should be careful not to abuse the trust reposed in it by many readers.
I think you missed something.
Ah, you replied to gwern while I was reading the comments and writing my reply to you.