[1] We will remove material of the following types:
…
To a very limited degree, material that seriously threatens LessWrong’s long-term values, mission and culture.
Could you say more about what is meant by this?
A possible argument for being deliberately vague (!) in this sort of situation is that telling people in advance exactly what bad things you'll punish helps hypothetical adversaries find bad things that you don't know how to detect.
Doing justice to this is probably going to be a full blogpost at some point, although we figured it was better to state it explicitly here to at least start the communication about it. Some initial thoughts:
1) Although I'm not confident about where exactly to draw the line, the team has chatted about classes of posts that we might want to take down, apart from the other ones listed above (i.e. obvious spam and obvious criminal activity, etc).
2) I expect it to be extremely rare for us to even consider that – I'd consider it a very big deal, and I think it's extremely important that LW users don't need to worry about that, even when writing controversial posts.
3) But, I think it's really important the LW team acts with integrity, both individually and as a team. For integrity to be credible (both to ourselves, and to others), we need to think in advance how we might handle extreme edge cases, and then communicate honestly about that, at least at the meta level.
We could generate non-zero cases where we would regret having committed to a policy of never deleting content, so it seemed better to transparently not commit to that.
I consider us to have something of a philosophical-debt and backlog of writing up our thoughts in more detail (both in terms of how we think about the site overall, as well as how we think about the responsibilities we have to our users). In the past couple months we've started making progress on this, which includes posts by Ruby, as well as habryka starting to write up thoughts on his shortform feed. (With this particular comment on integrity being most relevant to the current discussion, and this post by Ruby establishing the general frame in which we'll be writing things up)
All of this does, of course, mean that there's a limit to whether and how much people should trust us.
Our goal over time is be transparent enough about our policies that people can make informed decisions about whether to trust us to make calls like this, and meanwhile to be clear about which commitments we are making and which we are not.
I think it would be useful to distinguish between "deleting" content, and only allowing the OP to see it. While reasons for not being transparent about what you'd delete make sense, having to back stuff up* in case it gets deleted (as opposed to "taken down") would be a pain.
*Particularly posts (etc.) which require time and effort to polish into a good, publishable ("post worthy") form.
Oh, yes. To be clear. Whenever we delete anything, we still allow the author to access the content. We've never deleted anything in the sense of making the content inaccessible to its author and don't plan to ever do so.
Also, as we discussed this a bit more today, we realized that in most cases we can probably address our concerns by changing the post to "unlisted", rather than removing it from the site entirely. So, people can continue to have discussions there, but they'll need to get a link to it from somewhere else.
(It might also make sense to add some kind of flag on those posts to let people know that the mods don't consider the content representative of LessWrong, since people who get linked to random articles from reddit don't always have the right context to understand how LessWrong relates to the post, a la some neoreaction culture war stuff)
Quick note: Occasionally the LessWrong team makes announcements, or posts like this one that are intended as a longterm reference for how LessWrong works. Those posts might sometimes be set to "frontpage", which is a bit of an exception to the "no meta on frontpage" rule. Just wanted to flag that this an explicit exception sometimes made for key site information.
For clarity, I'll say that this post does not announce anything new. It merely clarifies that which was already the case in the minds of the LessWrong team and in the site design, but we had possibly not communicated nearly clearly enough.
Posts on practically any topic are welcomed on LessWrong [1]. I (and others on the team) feel it is important that members are able to “bring their entire selves” to LessWrong and are able to share all their thoughts, ideas, and experiences without fearing whether they are “on topic” for LessWrong. Rationality is not restricted to only specific domains of one’s life and neither should LessWrong be.
However, to maintain its overall focus while still allowing posts on any topic, LessWrong classifies posts as either Personal blogposts or as Frontpage posts.
The two classifications
Personal blogposts
Frontpage posts
This system allows LessWrong members to write about whatever is of interest to them while ensuring that only members who wish to see “off topic” content see that content.
*The LessWrong team will make some exceptions in the case of announcement posts we believe should have maximally broad visibility, e.g. the FAQ, and give these Frontpage status.
How to view personal blogposts
Personal blogposts, with their laxer restrictions, are not shown by default on LessWrong’s homepage. To view Personal blogposts, you can:
Personal blogposts and the Recent Discussion Feed
In some cases, the moderation team will hide comments on Personal blogposts from the Recent Discussion feed on the homepage. This is done if the moderation teams feels a discussion is veering in directions which are particularly controversial, political, or unproductive.
I (and I believe the rest of team) are not opposed to such discussions per se, but believe that we shouldn’t be drawing marginal attention to these discussions. In particular, the team does not think it's ideal for newcomers to encounter these discussions when first exploring LessWrong.
What does this mean for me?
Our classification system means that anyone can decide to use the LessWrong platform for their own personal blog and write about whichever topics take their interest. All of your posts and comments are visible under your user page which you can treat as your own personal blog hosted on LessWrong [2]. Other users can subscribe to your account and be notified whenever you post.
[1] We will remove material of the following types:
[2] In the future we might add various user-page/personal blog customization features like custom backgrounds, curating which posts and comments are shown first, etc.