Where in the paper do they draw a linkage between a "like" feature and self-censorship?

Is your position that somehow we're not a social networking site now, but become one by getting a "like" feature?

Self-censorship undoubtedly happens here as well, with or without a "like" feature, and the only way they draw a connection between social networking features and self-censorship has to do with audience settings -- an irrelevance here, as except for a handful of sub-forums like OMNI, the whole place is open to be read by the world at large.

OTOH, we don't have a "real names" policy like FB does (even if a few of us do use either real names or aliases long-connected to our public identities, and even if it's trivial to create a second fake account on FB), and the various boards are self-selected, so the analogy is weak there -- a point they bring up in the "related work" section that's frankly more relevant to the discussion of FT than the paper itself. Their findings would seem to indicate that by allowing anonymity -- such that people's identity here is specific to FT, or perhaps to FT and overlapping communities -- people will self-censor less.

I mean, I can come up with non-sequiturs with related keywords too:
https://developers.facebook.com/blog...button-impact/
...which has absolutely essentially equally little to do with my actual point, but it talks about like buttons and boosting engagement, no matter how unrelated to context is.

No, you needn't provide proof; you're the one who made a specific claim about documentation (that "[t]he self-censoring however is documented on other social media sites") so I asked about it; having proved and seem to be basing it on faulty premises.

As for supporting my point, I've given two specific scenarios where I think it improves the board. These are based on supposition and opinion, but they seem fairly obvious.

The cases against do not seem nearly as obvious to me, and I've tried to demonstrate up-thread why the common objection about "gaming" is pretty pointless. Similarly, while there's certainly an aesthetic objection to be had (and I'd certainly support, if it's practical to do) being able to hide the new UI for those who don't like it -- just as you can hide signatures -- I don't see the potential risk of a few sour-grapes self-censorships as being worth a feature that lets people censor others from what is essentially just a more compact way of posting "Thanks!" or "This ^^^", or that someone might be offended by having the possibility exist that someone "liked" their post.

If you can offer a specific scenario beyond 'some people might be worried about what IB might later do with the "like" data' (possible, but discussed to death already) or 'some people may post less out of sour grapes,' please do share it. Same thing I asked of folks who were claiming 'gaming' -- the only two scenarios I can come up with for that (boosting an overall like count, and partisan liking one another) are ultimately in equal parts harmless but pointless (as is the post count, especially in an "OMNI posts count" era.)

Just adding a like button won't turn this into Facebook. Indeed, just adding a full set of social-networking features wouldn't (it would be a thorough waste of time and effort, but other than making the UI more confusing for no good effect, I can't see what the actual harm of any of the likely features might be. The main privacy-concerning features -- being able to see all of someone's posts -- are already there. There's already a friends list (although it's not really useful, nor do I see a reason to try to make it "useful".)