The orange/not orange experiment

44 points by mechanical_fish ↗ HN
I didn't think that this new Orange Name feature, which is currently the talk of HN, was a big deal until I read this comment in a completely unrelated thread:

http://news.ycombinator.com/item?id=467657

Without trying to be harsh, and given your 'orange' status I realize I'm walking on thin ice here ;)

[Note how, unprompted, the responder bows and scrapes!]

... and suddenly I realized that I've seen this Frontline documentary before. It's the legendary blue-eyed/brown-eyed experiment:

http://en.wikipedia.org/wiki/Jane_Elliott

"On that day, a Friday, she decided to make the brown-eyed children the superior first, giving them extra privileges like second helpings at lunch, access to the new jungle gym and five minutes extra at recess. She would not allow blue-eyed and brown-eyed children to drink from the same water fountain. She would offer them praise for being hard-working and intelligent. The “blueys” on the other hand, would be disparaged. She even made the blue-eyed children wear crepe paper armbands.

"At first, there was resistance to the idea that blue-eyed children were not the equals of brown-eyed children. To counter this, she used a pseudo-scientific explanation for her actions by stating that the melanin responsible for making brown-eyed children… also was linked to intelligence and ability, therefore the “blueys” lack of pigmentation would result in lack of these qualities. Shortly thereafter, this initial resistance fell away. Those who were deemed “superior” became arrogant, bossy and otherwise unpleasant to their “inferior” classmates. Their grades also improved, doing mathematical and reading tasks that seemed outside their ability before. The “inferior” classmates also transformed – into timid and subservient children, including those who had previously been dominant in the class. These children’s academic performance suffered, even with tasks that had been simple before.

"The following Monday, Elliott reversed the exercise, making the blue-eyed children superior. While the blue-eyed children did taunt the brown-eyed in ways similar to what had occurred the previous Friday, Elliott reports it was much less intense. At 2:30 on that Monday, Elliott told the brown-eyed children to take off their armbands and the children cried and hugged each other."

I'm not sure what to conclude about this. Except that I want to issue a desperate plea: Please don't go on to reinvent the Stanford Prison Experiment!

28 comments

[ 1.5 ms ] story [ 76.1 ms ] thread
The flaw in your argument here is that in both of your examples there is nothing that the people could do to change their status via behavior.

I think the hope here is that posters who have proven themselves to be valuable to the community will have a tiny bit more perceived authority and thus help preserve the tone of discourse we have all come to enjoy. If anyone feels the desire to see there user name in orange, it seems fairly easy to accomplish that goal whereas in your examples there is no such opportunity.

I think the majority of regulars here aren't going to care too much one way or the other about the color of their name and will continue to post as they always have, and newcomers will have a goal to reach for that will encourage behavior that is in line with what we have had up to this point.

I still agree with this, as I was initially inclined to agree with it. It's a few upvotes, people! Surely nobody takes it that seriously!

And yet the sight of someone, out of the blue, acting out some elaborate deference to an orange-named person was quite jarring. Yeah, it was a joke, but the disarming smile is always part of these things, even when it's serious. Which I don't think it is, here, and at any rate it's too early to tell. But I remain... uncomfortable.

It's much like the disturbing feeling I get whenever someone suggests that the solution to the site's woes (whatever they are this week) is to give those of us with really high karma an array of superpowers that we can use to smite evil wherever it occurs. (Fortunately, that really has been treated as a joke, so far. The awesome powers granted to high-karma people -- you can change the color of your own menu bar -- are cute and completely appropriate.)

It's not a big deal, but if I had the power to turn off my orange name I probably would.

I think maybe that deference you noted is one of the intended consequences though. What I have heard overwhelmingly lately is A: HN as it stands now is a great community and B: People wish to preserve the current tone but are worried about the influx of new people who may not 'get' what HN is about.

I see the new 'orange' feature as a way to give a tiny bit more authority to users in good standing with the community, and as such it increases their ability to help maintain the tone as new users join.

the influx of new people and the change (loss?) of culture are at such odds with me desire for inclusiveness and internal (personal) regulation, or free will.
The point I would ask you to consider then is that a focused topic, even if it is as broad as "Hacker News" will necessarily exclude certain other topics and as a corollary it will exclude or discourage certain types of behaviors.

Because I do have a free will, if I have the desire to behave in a certain way, and it's at odds with the goal of HN, I can choose to use any number of other outlets for my communication, or even create my own avenues.

The counter intuitive truth is that if all of the social news sites where 100% inclusive in terms of content and were 100% homogenized (that is to say, the culture's of those sites fully exhibited all forms of communication/topics) then that would actually inhibit free will because nobody would have any choices in terms of what type of content they wanted to interact with: everyone would be subject to the same exact atmosphere with no viable alternative.

i meant inclusive within the bounds of hn's goals. there are still sub communities; eg: dev hackers, startup idea folks. no one is necessarily wrong in their interpretation of hn's community and submission policy. sometimes it is not that the level of comments has gone down but that the amount of comments has gone up. this is bad for the morale of commenter, to say nothing of the time readers waste (since without collapsible threads...). maybe after 20 comments no one should be able to comment.

i'm not advocating homogeneity so much as customized views. as the community grows it becomes harder to be a part of it. there's a social quality i'm trying to explain...and a concern for how exclusion is implemented... but i've spent far too much time on hn today as it is, so i'll let it rest.

Ah, yes I do see the distinction you are making after clarifying.
by and large, this community seems to be fairly resistant to these types of things and to the general woes that plague places like reddit/digg. if an orange-namer starts getting out of line, i'm sure they'll be taken down a peg.

edited to add: about the only problem i forsee with orange names is that they become a self-fulfilling prophecy. "oh this guy is orange, he posts good comments, this is probably a good one too, i'll vote it up", leading to orange namers rarely losing orange-named status unless they really step out of line

Just want to note: that was already happening even without the orange. At a certain karmic critical mass, my comments started getting voted up more often. But I'm just as much of an asshole now as I was 9 months ago.

I agree the orange makes it worse; it's a step in the wrong direction. On the other hand, it's not introducing a new problem. And we're all way to obsessed with this silly number.

But I'm just as much of an asshole now as I was 9 months ago.

But you're an asshole we all know. You're our asshole.

I had no idea what your karma was until this comment prompted me to check... kudos. But your point is still true. I usually find myself either loving or hating what you say, and you have been prolific enough that I recognize your username and actively search for your comments on stories just to see what you have to say. As such, I'm more likely to vote on you than someone else that I haven't followed.

To me, karma is purely feedback on whether I said something of value. I could care less about the karma of others, and base my voting on the comment/submission, not who said/submitted it. I didn't realize until now that recognition is also a bias to my voting however. Thanks for being the example that revealed that to me. I'm not sure I'll change my behavior, but it's an amusing observation either way.

As the person whose comment you are quoting you completely picked up on what I was trying to signal. Of course I still posted my comment but I felt like talking back to someone 'superior', and of course that is not the intended reason for the change but one of the not-so-nice side effects.

I personally think that the real quality measurement of comments is not up to a mere algorithm and a voting system that is so susceptible to bias.

Just looking at my posting history an hour ago or so to see what if any of my '1' rated comments I should take back considering the effect on the average karma over the last of my 50 posts, and I'd have to conclude that for me that really would reduce the value of HN to < 0.

I'd quit, that simple.

Lots of the meta discussion around this subject has been a pretty good learning experience for me, I consider the time on this 'experiment' well spent in that sense, but I think the outcome to date has only been:

- more incentive to game the system

- exploding meta discussions about the subject including one open attempt at revolt

- something deep inside me that I can not put quite in to words that makes me uncomfortable about distinguishing individuals instead by 'their cover' instead of reading what they have to say.

And I'm the orange guy you were replying to. As I replied later in the thread... please ignore my oranginess. I don't like it either. I think it's a mistake. You want good comments to rise to the top, not assume that only people who've made good comments in the past are the people to listen to.

I might have only BS to say on a particular topic, yet get flagged orange because I've said interesting things about other topics.

Happily, it appears that with my setup (vista on high contrast black theme and FF3) everybody's just blue. Not that I think I'd actually care if they were different colours; I suspect quite a few of us on here have the personality trait / social dysfunction / whatever of not giving a flying fsck about externally determined rank, and being much more interested in whether a person has earned -our- respect directly.
The beauty of the Orange name is that it only reflects your last 50 comments
good reminder. nonetheless, see noodle's comment for why this fails to console.
I'll chip in with a tiny little insight, which I think applies in this whole karmic discussion:

Complex systems encourage simple behaviour. Simple systems encourage complex behaviour.

That's a cute aphorism, but I'll mod you up if you can provide interesting examples!
Perhaps this is a bit over-the-top, but look at The Internet versus Wordpress versus Tumblr.

Given a blank page, people will create anything. You can take three HTML-only pages at random and you'll come up with 3 very different results.

Given a blogging engine, people will start restricting themselves to the form of their engine. That means you get lots of different posts, but very rarely will you see a blog using the Wordpress engine for something really standout and unique.

Given a system that divides posts into 6 types of content, the vast majority of blogs created will all start looking the same. The average tumblelog is self-referential, generates long winding tags of citations upon reblogs, and content tends to be amateur and haphazard. Very rarely is a tumblelog interesting or a source of good original content, and deviations from the norm are even rarer.

I'm in the middle of writing an essay about how medium affects content - it's taken about 5 months so far, so perhaps it's not quite even "in the middle" yet - and the gist of one of the things that I state is that the more defined your form is, the more rigidly people adhere to it. And form is good for some things. For other things, it's better to keep the form as loose as possible and let people riff on their own.

Your observation about 6 types of content is fascinating -- it makes so much sense! The medium is the message, right? So content expands to fill the container it is given.
Exactly. As you add more features, you simplify certain things, but you also limit the sorts of things people will come up with.
Constraints can also encourage creativity, because people have something to push against. Larry Niven (science fiction writer) claims that constraints make a story (although he arguably writes puzzle fiction; and puzzles are defined by constraints).

I've seen a similar idea in niche marketing, where you specialize on a segment instead of spreading yourself thin to please everyone. Does this promotes creativity, or does it just focus it? I think the focus makes it engage with reality more, and that this does promote creativity.

This is really how content constraints affect content (not how medium constraints affect content.

Typewritten essays were very constrained in their presentation (especially in the form of the manuscript that the author worked with), including academic papers, such as Einsteins.

PS: maybe I misunderstand, and by "content" you mean the presentation that people create, not the text.

Well then, here's a pretty interesting example of complex visual patterns arising from relatively simple systems: http://sprott.physics.wisc.edu/pubs/paper249/PAPER249.HTM

One physicist (Wolfram) maintains that natures rules are quite simple (I prefer "elegant"), and that beautifully complex patterns can arise from these. http://www.ur.umich.edu/0203/Oct14_02/m_wolfram.shtml

Alternatively, it seems intuitive that when given a complex system (or a multitude of parameters) that objects within the system will have less room for deviation from these given parameters.

Let's take baseball as an example. With its plethora of rules, only a few possible outcomes can occur, and the game is actually quite simple. All agents operating within the system can only act along the limitations given by the rules, and because there are so many, there is little room for deviation.

It will be interesting to see how this changes voting behavior. I know I am already voting up above average gray comments over above average orange comments. It wasn't purposeful. But easily apparent when I looked back at my voting behavior.

I haven't quite decided if I should just completely ignore the orange/gray experiment or if I should see it as an intellectual challenge. In some senses I have much more freedom to post as I please when I'm gray. On the other hand, there is seemingly more influence being orange. And there is something to be said about being gray and still having a highly voted comment.

Uhmmmm.... I think the Stanford Prison Experiment is exactly what we want.

We deliberate seek a way to enforce an exclusivity based on a shared hacker "culture" if you will.

Hey mechanical_fish, drop the meta and come back to talking with me about Drupal:

http://news.ycombinator.com/item?id=467926

(Perhaps you've already seen this & chosen not to respond, but I think it's more likely you haven't seen it since my response came ~8 days after your comment. My apologies if the former.)

(And -- regarding the orange/not orange experiment -- for some reason my erstwhile orange status seems to have goaded me into making lots of little 1 point comments which I know will take me to not orange status if the feature ever comes back -- which is not the response I would have expected in myself. Any experiment that produces a surprising result without harming anyone is an experiment worth conducting.)