Aristotle++. Thankyou for putting into words, something I have been wanting to say ever since seeing some of the replies to 202482, but couldn't word appropriately.

If I post a node, and it stays stubbornly at 0. Fine. Whatever I said was not very useful, possible only had meaning in the context of the thread etc. Not wrong nor bad or most important of all, not technically invalid. Just nothing worthy of anyones vote. That's fine.

As you said, if it gathers some ++'s great. The effort expended was worth it. A few someone's found it interesting or useful.

But when a post, especially one that has taken some effort, and contains (as 202437 does), a working solution, and embedded proof that it does work (at least under some circumstances) and it starts gathering --'s, I go back and check:

1. Does it contain factual errors?
2. Does it contain something blatently rude, condesending or facecious?
3. Did I misread the OP's needs or answer the wrong question?

If so, I usually try and correct my errors or admit to my mistakes by striking (in deference to the unwritten "don't change history" rule, despite the fact that I disagree vehemently with both the sentiment and the reasoning) whatever was wrong and correcting or apologising.

The reason why I chose to withdraw a post entirely, was if working code is somehow wrong, then it must be because it is dangerous or misleading, in which case, allowing it to perpetuate is a 'bad thing', and so I didn't. Instead, I chose to attempt to come up with a better solution.

That I lost a few XP, Bah, Humbug! Simply makes up for some of my other posts that have gathered a positive rep dis-proportionate to the effort I put in like this where all I did was give a pointer something I had recently read or had bookmarked. Or this where I was simply the first to notice something glaringly obvious. Or this that represents the simple 10 second cut&paste of readily available information. It rankles slightly that these get ++'d when other nodes containing (what I believe to be) good Perl, which I've put considerable effort into, go by hardly noticed, but you quickly get used to the fact that such are the fickle mannerisms of the Monastery.

I know what you mean about unexpectedly disproportionate amount of votes - my second highest rep node is one of the quickest replies I wrote to date. I was surprised to log in next day and see I had gotten a boatload of XP - and even more surprised when I found out where they came from.

The "stubbornly remaining at 0" phenomenon is one I've occasionally grumbled about too.

But I disagree about the reaction to downvotes to what seems to me to be a good reply. In that case, I update the node with a remark to the effect of "I've gotten a couple of downvotes here. Would anyone please care to explain to me what about my writeup was out of line or incorrect?". If noone speaks up, I leave it at that and consider myself confirmed that I just stepped on a few cargo cultist toes by going against the party line. If someone does reply and gives a good explanation, all the better - I've learned something, and others who read my node can reach the same conclusions about why they shouldn't do it the way I did. If there really is a very compelling reason, I will <strike> my post and point people at the reply, so that a bad meme doesn't get perpetuated. That's what I understand as the sentiment behind the "don't rewrite history" principle: if you make mistakes, you and others can learn from it.

Granted, updating with an explanation request only really works so long as the thread is alive.

Makeshifts last the longest.

But, I really love your second highest rep node. As the person who asked the question said in response, it was a very clear explication of a technique that, for many, is mysterious.

Simple explanations that makes things clear is among the highest ideals we can aspire to in the Monastery, in my opinion.

Now, you can argue that it may be more or less deserving than some other nodes, sure, and I think this is your point. Individual reputations of nodes is not a ranking.

Yes, you've read the answer. Go read again. That's what node rep is: feedback. To the author, to be precise.

You have already stated that you care a great deal about your nodes' reps; it gives you pleasure. I do too, and I would imagine that many other monks around here do as well. The rep is your reward, and you want to get this reward as much as possible. This means that node rep encourages good behavior by positive reinforcement. In other words, you are doing something to recieve a reward, whether that something is write more nodes or write better nodes.

Positive reinforcement can be very powerful, and is often the best way to encourage a certain behavior. However, not everyone is alike, so the system doesn't always work as we'd like. Some don't care about the rep, or don't want to have a high reputation. In such cases, or any case where the frequency of the unwanted behavior increases, the node rep acts as a punishment.

So does node rep mean something more? Definitely, though not in every case. It is a way of making good behavior increase and bad behavior decrease, which is important in any community. The reputation of a particular node, then, is important not only to the author, but to the entire community. Then again, I am by no means a psychologist, so I may have it all wrong.

elusion : http://matt.diephouse.com

Yes. Maybe it wasn't clear enough, but I wasn't saying that the concept of node reputation is useless - far from it. My point is that the individual reputation of a specific node is a useless number, even if the concept is not. As I said in my concluding sentence, "it's a form of social sanction, nothing more".

Makeshifts last the longest.

1. You don't know if the other people reading it will in fact understand or get your post in the context it was intended, especially since this is a global community. For example the word fag in England vs. fag in America makes a big difference or did at one point anyway.
2. It exposes a side of you that could be perceived as less then professional and could have the adverse effect of people skipping your posts in the future because they assume that you have simply posted another "gag" node that has no real meat.

I submit that node reputation is pretty much useless to anyone but the auhtor of a node.

This is why I liked the Node Tension idea so much -- at least for node authors to see

"rep: 10" ... is that 10 good votes, or 60 bad votes and 10 sympathy votes? was my post well recieved or not? oh well... i guess I'll never know.

Opportunity cost would be nice to know too. I'd like to know how many people read my post and then voted for somebody else. ;) Seriously, a post that gets 10 reads and 10 votes may be much better than a post with 1,000,000 reads and 100 votes. Knowing this could help solve Warnock's dilemma.

That said, perlmonks works perfectly for the community. Conversations are on-topic. Flames and attacks are rare. Wrong answers are quickly corrected. Changes at this point will only help authors seeking meta-enlightenment.

Yes, that is an interesting proposition. After some thinking, I do think others shouldn't be able to see any more than just the node rep as such - for a reason I can't quite put into words yet. But letting the author know the precise stats would provide them with even better feedback, which is all the concept is about.

Makeshifts last the longest.

kelan

Yak it up with Fullscreen ChatterBox

• Whether the thread has been front paged.
• Whether you're one of the first replies.
• How deep in a thread you reply.
• The popularity of the subject of the root node.
• The actual content of your post.

If only you could see all the reputations of all the nodes in a thread, then you could compare. A reputation of 15 where all the other posts don't get above 10 is likely to be better than a reputation of 25 where all the other posts score 40 or above.

Abigail

I think your list is spot on. I've written nodes where I took a guess at the answer (I marked it so) and the questioner replied with "no, that's wrong". But the thread got front-paged, and the node's rep (it was the second below the root) rose to 25 or so.

On the other hand, I wrote an answer to a different question after there had been some discussion and clarification in the thread, and the questioner replied with "yes, that was correct, thank you." The node was very far from the root, wasn't front-paged, and languished around 2. So I agree that content ranks far below placement.

In my own mind, though, the rep-2 node that got the "thank you" ranks much higher. Replies trump votes every time.

Now, given that this thread's slipped off the front page, and my reply is eight or ten nodes down, I predict it will stay at zero!

Your factors are too specific. They both overlap and fail to account for some large forces. Timing, for instance. What day of the week was it posted and was it right before a holiday? Presence on daily/weekly best nodelets is another example. There are bound to be many such factors. I don't think trying to enumerate them is the right approach to understanding the node rep system. I take a simpler view.

Node reputation is a function of:

• The quality of the node's content and presentation.
• How much attention or visibility the node gets.
• An X-factor for things like personality voting.
The function is quality * visibility + x. It's quite simple really and just what would be expected. There are some things about this community that need to be understood before it all really makes sense though.

One important one is the fact that the "quality" of the node is relative to the observers and is a measure of respect, agreement and interest. Many people may upvote a mediocre node if it is simply better than what they themselves would have written. On the other hand, sometimes a particularly good node won't get many upvotes because judging its quality requires knowledge or skill that most don't possess.

Another important one is that this community tends to make its points through positive reinforcement rather than negative reinforcement. People tend to upvote more than they downvote.

I think the biggest "problem" (although I'm hard-pressed to call it that) is the fact that visibility is a multiplier. This, more than anything else, makes it difficult to compare node reputations. Very good replies deep in a thread often don't get half the reputation as poor replies early in a thread. Replies to front-paged nodes invariably have higher reps. Etc., etc., etc. This is the problem everyone is really talking about when they claim that node reputation doesn't mean anything or is as useful as random numbers.

In any case, I think this is a good model of the node reputation system but I'd be interested to hear any criticisms.

-sauoq
"My two cents aren't worth a dime.";

If only you could see all the reputations of all the nodes in a thread, then you could compare.

Although it isn't a complete solution, you can choose to order the notes in "Best First" order in your user settings. That will allow you to tell whether one note's rep is equal to or greater than another one but it won't give you any idea by how much. It will also only help you to compare the immediate children of a particular parent note; it won't help with notes in separate subthreads.

-sauoq
"My two cents aren't worth a dime.";

    * How deep in a thread you reply.
... and how late.

  p

The requirement to add a comment was discussed before. The problem with it is that it has to be anonymous for the voting system to work as it does; and then someone can evade it by simply providing "you suck", "." or other worthless "reasons". Bottom line, we're no wiser than before.

I do agree with your sentiment that downvotes should be explained, and while I'm not going to pretend I always do, I make an effort to adhere to that. If I'm downvoting an obvious troll or similarly worthless node, I rarely care to post an explanation, but if I'm downvoting for technical reasons or because I think someone was out of line, I remind myself to be explicit about it - that will both help the original poster understand why they're getting admonished as well as let me know if my own judgement on the matter is approved in the form of votes on my explanation post.

Makeshifts last the longest.

I enjoy the voting system in the above way, because though
I can't see who is applauding through the footlights, I
respect the audience in general.

Aristotle, your node explains this idea perfectly! I hope we all gain benefits from it!

-- JaWi

"A chicken is an egg's way of producing more eggs."

To me personally, node reputation is much more significant number in the database than the XP rating. And I have learned this lesson during my stay here at perlmonks.

When I first came here, I somehow thought that a person with higher XP is also automagically more knowledgable in Perl, therfore I was tending to think that their advice is better (not even listening to what others are saying). Sort of "I mean, hey, he's got >1000XP, he surely knows more than the next guy with just about 250XP (I'm not even reading Anonymous postings)" feeling. Wrong. It took me a few months to understand that XP only means how active the user was on perlmonks.

It was only then when I started to take judgement which reply to my question is better (I mean if they were offering different approaches) on the node reputation number. Of course, a totally wrong answer would be flagged with an answer and explanation why it is wrong, but sometimes there is also a choice between right answers. I have found that reputation number is usually a decent way to choose. Just my personal experience.

I know that I have more than 1000XP myself, and a lot of new people *MAY* think that I am quite knowledgable in Perl ("Hey, this guy is an abbot -- he *MUST* be good at Perl") while I am not. Well, I know the basics, but you definately shouldn't judge my knowledge on my XP.

Just my $0.02

So what are my recent voting/posting habits?

• I rarely use all my votes. Not because I can't find enough deserving nodes, but because I have limited time to spend here.
• I scan the nodes and reply/vote only on the ones that interest me. So if everyone else casts their votes for the same reasons I do, I deduce that they don't share my immediate interests.

--Jim

A fair number of people seem to agree that a node's rep is in no way indicitive of it's quality (and don't hesitate to point it out, again and again and again...). What I'm wondering is how to improve the voting system so that you can get an idea of which posts to bother reading based on a brief rating. Is it possible? Is something like slash closer? What other features would help?

Thank you for your opinions :).

So a flawed system is better than no system at all? I would argue the opposite, but I don't care, and I have to go. There was really no point to this post. Move along now.

I would argue that all systems of social sanction are flawed, yet I'd rather live in a world with one than without.

Makeshifts last the longest.