I know many high profile developers participate on HN, and I often see people mention projects they've worked on which I've heard of or used before. While it can be nice to read a comment completely oblivious to who the author is, often extra context can add insight. A lot of the time the HN handle is non-obvious and even different from their GitHub username (if they have one). Is there something like a public list that shows some of the more seasoned voices and what their HN username is?
Couldn't agree more, I use my own name just because I use it everywhere, but on HN it's not called out and I rarely look at the author unless I think "I bet XXX wrote this, yep they did" or "Who wrote this, I'd like to reach out to them". I think it helps, to a degree, to help prevent/lessen group-think and people taking someone's word as law.
While you're very much correct that seems rather irrelevant to the question posed to the community. At least I do not know the OP's intentions with such data and distinguishing comments may not be a concern of theirs.
I very much disagree in cases where the argument is nuanced and I don't have time to do substantial research. In this very common case we resort to heuristics, and "who's making the argument" is one of the most effective heuristics there is.
I wish that were true. Too many conversations on HN essentially drill down to 'my experience contradicts your experience/theory/proof'. It never hurts to know if the person asserting his experience actually has any experience of note.
This seems like it should be true but I don't think it actually is. It's very easy to frame up a plausible argument which is actually false, and inexperienced people do it all the time. The problem is that if the reader of the argument is also inexperienced, he will be unable to recognize the flaws in the argument.
For people like that, post author is an excellent heuristic.
> It's very easy to frame up a plausible argument which is actually false, and inexperienced people do it all the time.
The problem is that experienced people do it too, though (hopefully) less often; so, if you get in the habit of trusting arguments more because of who has made them, then you will never have a chance of catching that mistake.
> A good argument stands on its own. Who the maker of it is, is irrelevant.
Damn straight!
I'd like things to go even further in the opposite direction. Usernames should be hidden for N days, say until the article is locked for updates. You'd want to make sure the usernames remain consistent throughout an article though (so you know it's the same person you're replying to).
@dang: If you want a simple way of implementing this that doesn't require any extra storage (per user), how about HMAC(JSON.stringify([username, storyId], secret)?
I have toyed with the idea of an extension that lets you keep notes / tags / custom color by username with the idea of allowing the memory deficient to figure out the history of comments with that user.
I suppose the USENET kill file would inevitability be the results :(
Agreed, but I think there are a lot of somewhat well known figures in tech who haves accounts on here and they are trying to be anonymous, but also newer HN members may not know who they are. I don't feel like we need this mapping but I could see how some might find it useful.
We DEFINITELY don't need people's account to be de-anonymized who are trying to remain anonymous
> Agreed, but I think there are a lot of somewhat well known figures in tech who haves accounts on here and they are trying to be anonymous, but also newer HN members may not know who they are.
I think that I don't understand this post. This sentence seems to start by saying that we should respect people's desire to be anonymous, but to conclude by saying that it's a problem when … we don't know who they are?
It's probably inevitable that someone will publish a collection of predicted sockpuppet/alias accounts. It wouldn't surprise me if the HN mods had their own private collection; I think it'd be a fun project to show estimated alternate usernames next to a comments' username along with a confidence score. Naive bayes on comment text would probably do alright but there are plenty of other signals. A public list being published might suck if any of them are linked to a real identity and the real identity didn't really want them linked, but at least it's safe from novel associations, like maybe chipgap98 is _why but is also Bob from the Mozilla office down the street (in addition to Jonathan)...
I started making a spreadsheet of interesting folks but then realized it wasn't worth the effort. Some of the more famous folks (e.g. stevewoz, brianchesky) don't post any more after their initial activity years ago. And the ones who do (norvig, BrendanEich, alankay1) have names that are immediately noticeable, and I almost never read the name on a comment before reading the comment anyway.
If they want, they can use keybase.io for that. All my digital identities are linked up through that, but of course, I'm not a "well known" person in Tech.
It happened recently that I saw someone insert their email in a comment and upon researching, I determined they are my arch-nemesis. Like not in just a business sense, I have fully committed my life's work to undoing this guy's daily work. And I hate that he's gotten rich doing it.
I'm not going to call him out but it hurts low-income citizens across the country so knock it off, asshole!
