tea.mathoverflow.net - Discussion Feed (anonymity in ratings)

Todd Trimble comments on "anonymity in ratings" (10494)

Todd Trimble — Thu, 11 Nov 2010 07:36:32 -0800

Right, the correct definition is minimum path length (as Henry slyly hinted). The minimum length of paths from Erdos to himself is 0. There are various formalizations of this which can be left as an exercise to the reader.

Qiaochu Yuan comments on "anonymity in ratings" (10485)

Qiaochu Yuan — Thu, 11 Nov 2010 03:11:48 -0800

@powerpuff: Erdos has the empty joint paper with himself! By your definition, an isolated vertex wouldn't constitute a connected component of a graph.

Will Jagy comments on "anonymity in ratings" (10477)

Will Jagy — Wed, 10 Nov 2010 14:39:50 -0800

Random note, I deleted my comment to Gerald's MO answer and made it an MO question.

HJRW comments on "anonymity in ratings" (10476)

HJRW — Wed, 10 Nov 2010 13:37:59 -0800

powerpuff, I'm glad I'm not taking your course on metric spaces.

powerpuff comments on "anonymity in ratings" (10475)

powerpuff — Wed, 10 Nov 2010 13:00:59 -0800

@Cam, I disagree. In fact, I think that Erdos should be considered to have Erdos number 2 --- after all, Erdos has no joint papers with himself, although he has many joint papers with people who have joint papers with Erdos.

EDIT: Apologies to those who missed the (possibly not very funny) joke and who seemed instead to think that I was trying to reformulate the foundations of graph theory.

Cam McLeman comments on "anonymity in ratings" (10474)

Cam McLeman — Wed, 10 Nov 2010 12:12:29 -0800

Off-topic: I've always wondered if summing one's Bacon number and one's Erdos number was really the most appropriate way to define the Erdos-Bacon number. Maybe the 2-norm of the [Erdos-number, Bacon-number] vector?

Edit: Presumably we could axiomative enough properties of the number to uniquely define it. For example, you should clearly have an Erdos-Bacon number of 0 if and only if you are simultaneously Erdos and Kevin Bacon.

Will Jagy comments on "anonymity in ratings" (10473)

Will Jagy — Wed, 10 Nov 2010 11:04:53 -0800

Gerald, I asked you a question in a comment to your answer at

http://mathoverflow.net/questions/45477/closed-form-functions-with-half-exponential-growth/45479#45479

about local solutions only to f(f(x)) = sin x around x = 0. Essentially the question is whether there is an (odd) analytic solution, i.e. nonzero radius of convergence for the formal power series solution. I screwed up, I was assuming my series would have a similar graph to one that has already been posted. That cannot happen for very long, as within the radius of convergence, should that be nonzero, the formal power series gives a genuine solution to f(f(x)) = sin x. There is a trichotomy here, given Ax, there is a functional square root given by x sqrt(A), where the case A=1 is special, so for the local analysis A >, <, = 1 matters.

I wonder if Danica McKellar looks at MO? Her Erdos number is 4, her Erdos-Bacon number is 6.
http://en.wikipedia.org/wiki/Danica_McKellar

geraldedgar comments on "anonymity in ratings" (10471)

geraldedgar — Wed, 10 Nov 2010 08:21:30 -0800

Interesting. I wonder if someone searching Google for a swimsuit model will end up in MathOverflow? Our newest recruiting tool...

Storkle comments on "anonymity in ratings" (10442)

Storkle — Tue, 09 Nov 2010 07:37:27 -0800

Wait, you mean Mariano is a he???? I always thought that Jarah Mariano was posting here: http://sportsillustrated.cnn.com/2009_swimsuit/models/jarah-mariano/09_jarah-mariano_15.html

Gerry Myerson comments on "anonymity in ratings" (10423)

Gerry Myerson — Mon, 08 Nov 2010 21:06:34 -0800

Mariano, I had always taken it for granted that you are that reliever for the New York Yankees, http://www.baseball-reference.com/players/r/riverma01.shtml

Harry Gindi comments on "anonymity in ratings" (10416)

Harry Gindi — Mon, 08 Nov 2010 14:03:44 -0800

Meanwhile, this is another reason why you should aspire to become a moderator. Then you can see who downvoted you and exact revenge.

Mariano comments on "anonymity in ratings" (10415)

Mariano — Mon, 08 Nov 2010 13:56:03 -0800

Micheal: it has lead to surprisingly few flame wars around here, at least. Most of the little heat I' ve seen on MO has actually happened in comments and answers, where there is no anonymity (apart from the inevitable amount of anonymity provided by the fact that you really have no idea of who, really, "Mariano" is, of course...)

As I said earlier, I don't see the point of this thread. Are you aware that "we" cannot change the fact that votes are anonymous?

Michael Hardy comments on "anonymity in ratings" (10407)

Michael Hardy — Mon, 08 Nov 2010 13:10:34 -0800

I think anonymity leads to flame wars because it is impolite, whereas arguing is polite.

Harry Gindi comments on "anonymity in ratings" (10406)

Harry Gindi — Mon, 08 Nov 2010 13:09:25 -0800

You need two people to flame war as well.

Michael Hardy comments on "anonymity in ratings" (10404)

Michael Hardy — Mon, 08 Nov 2010 13:08:44 -0800

But arguing is civilized, whereas flame wars are not.

Harry Gindi comments on "anonymity in ratings" (10403)

Harry Gindi — Mon, 08 Nov 2010 13:08:15 -0800

Anyway, the value of anonymity is that you need two people to argue.

Michael Hardy comments on "anonymity in ratings" (10399)

Michael Hardy — Mon, 08 Nov 2010 13:06:40 -0800

@Harry: Clearly our experiences differ.

(I remember a time when "speedy" deletions (i.e. an administrator deleted without discussion) often resulted in flame wars, but now the disciplines for speedy deletion have become civilized.)

Harry Gindi comments on "anonymity in ratings" (10398)

Harry Gindi — Mon, 08 Nov 2010 13:03:14 -0800

Maybe things are different now, but when I was a wikipedian, every deletion thread I involved myself in became a flame war.

"The good old days".

Michael Hardy comments on "anonymity in ratings" (10397)

Michael Hardy — Mon, 08 Nov 2010 13:03:00 -0800

"I don't follow. Could you clarify the question?"

Participants are identified by Wikipedia user names (which in many cases are actual names). It was proposed in this thread that anonymity prevents "flame wars" that "would never stop". Did those flame wars happen because of identification of participants?

Michael Hardy comments on "anonymity in ratings" (10395)

Michael Hardy — Mon, 08 Nov 2010 13:00:49 -0800

I'm not saying there are NEVER flame wars in Wikipedia deletion discussions. I can cite some. But they're rare. The words "would never stop" don't come to mind.

Harry Gindi comments on "anonymity in ratings" (10394)

Harry Gindi — Mon, 08 Nov 2010 13:00:42 -0800

I don't follow. Could you clarify the question?

Michael Hardy comments on "anonymity in ratings" (10393)

Michael Hardy — Mon, 08 Nov 2010 12:57:45 -0800

@Harry: How many of them do you think would have been prevented by anonymity?

Harry Gindi comments on "anonymity in ratings" (10392)

Harry Gindi — Mon, 08 Nov 2010 12:54:27 -0800

@Michael: I have participated in many flame wars on wikipedia.

Michael Hardy comments on "anonymity in ratings" (10390)

Michael Hardy — Mon, 08 Nov 2010 12:52:27 -0800

@ Andrew Stacey: Unlike most of what's been posted in defense of anonymity, your comments seem like something a sensible person could think.

Ryan Budney comments on "anonymity in ratings" (10387)

Ryan Budney — Mon, 08 Nov 2010 12:42:49 -0800

Michael, I'm not entirely sure what the point of this conversation is -- what the goal is, I mean. The website, like much social interaction, is an experiment. It sounds like you want a different kind of experiment. That's the message I'm getting. But I'm not really sure what to do with this, other than mark down that you have an idea for a different kind of forum.

Andrew Stacey comments on "anonymity in ratings" (10386)

Andrew Stacey — Mon, 08 Nov 2010 12:33:02 -0800

I'm sic of this discussion ... (sorry, couldn't resist).

The term "flame war" is emotive so let's avoid it.

Voting is anonymous because this site (and its siblings) are places where discussion is deliberately discouraged. Oftentimes, I get into a discussion with someone about an answer or a comment, and it quickly swamps more time that I would like to devote to this site. Due to the nature of this site, I usually - with a clear conscience - simply stop participating when I feel that a "discussion" has started. If I had to leave my name every time I voted, I would not vote because although I feel that I could defend every vote, I really don't want to have to because I haven't the time. I want to be able to come to MO in a spare 5 minutes, have a quick look at what's going on, see if I can help in any way, then leave.

Part of that "seeing if I can help" is to comment on others' work. The purpose of MO is to get quality answers to quality questions. As "quality" is in the eye of the beholder, the more beholders, the better the judgement of the quality (to a first degree of approximation - but that's a different discussion). Since I want to do this quickly, I want to be able to just say, "Yup, this is good" or "Nope, this isn't" without getting in to a huge discussion about it. I do try to leave a comment if I think it will help, particularly if I am voting against something. But, as I said, on CW questions I'm less inclined to do so - partly because I don't think that anyone can justifiably feel put out if I vote down on an answer to a CW question and partly because I don't really like those questions on the MO site so rarely look at them and even more rarely vote on them.

So voting on MO is an integral part of the feedback mechanism. It's like the wikipedia talk pages, only faster. But as with wikipedia talk pages, the voting is not the thing, the whole thing, and nothing but the thing. It's merely the eminence grease.

Michael Hardy comments on "anonymity in ratings" (10385)

Michael Hardy — Mon, 08 Nov 2010 12:20:04 -0800

"Michael, if you 'look about you', you will see evidence that some people take down-votes personally enough to start long threads on meta complaining about them."

But ONLY when such votes are anonymous has this happened!

What "would" happen when politeness accompanies them is untested.

Michael Hardy comments on "anonymity in ratings" (10384)

Michael Hardy — Mon, 08 Nov 2010 12:18:21 -0800

So voting has to be anonymous because otherwise flame wars "would" (sic) result (even though they don't result from non-anonymous votes to close and even though they don't result from votes for deletion on Wikipedia, and even though no one can cite experience to support this and therefore they have to say "would") and down-votes are alleged to be the principal means of maintaining standards even though their rationales are secret, so they cannot communicate what the standards are, and even though actual verbal communication of standards takes place (verbal communication must be considered a far inferior means of communicating) and it is actually held that the site "would" (sic!) be "ruin[ed]" (sic) if voting continued to exist but were not anonymous, and that last point is too self-evident to require any support beyond its mere assertion.

But it is conceded that people are less polite when they are anonymous.

HJRW comments on "anonymity in ratings" (10381)

HJRW — Mon, 08 Nov 2010 12:09:30 -0800

Michael, if you 'look about you', you will see evidence that some people take down-votes personally enough to start long threads on meta complaining about them.

+1 Mariano & Ryan.

Harry Gindi comments on "anonymity in ratings" (10379)

Harry Gindi — Mon, 08 Nov 2010 11:57:22 -0800

I think what Harry is getting at is the format of the site has the voting as an integral component.

Yes, but I was too lazy to type out the explanation. Thank you, Ryan.

Ryan Budney comments on "anonymity in ratings" (10378)

Ryan Budney — Mon, 08 Nov 2010 11:53:38 -0800

Michael, I think what Harry is getting at is the format of the site has the voting as an integral component. Traditional forums have a very top-down governance structure which causes problems. By having votes you're empowering a larger proprotion of the community to make types of administrative decisions. Questions with lots of upvotes are more visible so they tend to get more time on the front page. There are other aspects of the site that make it more community driven as well -- by accumulating votes you develop the perhaps poorly-termed "reputation" and that gives you abilities to vote to close and such.

Michael Hardy comments on "anonymity in ratings" (10377)

Michael Hardy — Mon, 08 Nov 2010 11:36:14 -0800

"Isn't that the whole point or something?"

You've lost me with that one.

You mean this voting is the whole point of the site? No it's not; the point is to ask and answer questions about mathematics. Or so I thought. If voting things up or down is the point of the site, then the site is worthless.

Harry Gindi comments on "anonymity in ratings" (10376)

Harry Gindi — Mon, 08 Nov 2010 11:31:32 -0800

@Michael:

Isn't that the whole point or something?

Michael Hardy comments on "anonymity in ratings" (10375)

Michael Hardy — Mon, 08 Nov 2010 11:13:15 -0800

"ruin the site"! Amazing. These votes are so crucial to the site, then? I'd never have guessed they were so regarded. Is that something I was expected to know? Does anyone else regard them that way?

Mariano comments on "anonymity in ratings" (10374)

Mariano — Mon, 08 Nov 2010 11:04:26 -0800

Well, here the votes for closure are not anonymous...

I don't understand the point of this thread.

The word "would" appears here. Assertions about what would happen, if not based on experience, should not be qualified with the word "clear".

To clearly convince you we'd have to ruin the site, then. Cool.

Michael Hardy comments on "anonymity in ratings" (10373)

Michael Hardy — Mon, 08 Nov 2010 10:54:26 -0800

"look about you"

My only experience of forums with such voting practices in which such votes are not anonymous, is Wikipedia, where votes for deletion of articles fail to lead to flame wars.

That's what I see when I look about.

Henry Wilton: What specifically is _your_ experience of flame wars in forums with such non-anonymous down-voting?

And can anyone else answer that?

HJRW comments on "anonymity in ratings" (10372)

HJRW — Mon, 08 Nov 2010 10:44:33 -0800

Michael,

If you seek evidence for Qiaochu's assertion that

If everyone knew who was downvoting them the flame wars would never stop

look about you.

Michael Hardy comments on "anonymity in ratings" (10371)

Michael Hardy — Mon, 08 Nov 2010 10:39:33 -0800

I'd certainly be prepared to state reasons for any of my downvotes.

Verbal expression rather than downvotes seems obviously to be the means of supporting standards, since downvotes alone are cryptic.

Harry Gindi comments on "anonymity in ratings" (10368)

Harry Gindi — Mon, 08 Nov 2010 10:23:29 -0800

I, for one, am prepared to defend any individual downvote I've made, but shiver at the thought of being open to defend all of them.

I'm not prepared to defend any individual downvote I've made. I used to make some pretty cynical downvotes..

I would be willing to defend any and all downvotes I've made in the past six months, however.

Cam McLeman comments on "anonymity in ratings" (10367)

Cam McLeman — Mon, 08 Nov 2010 10:20:09 -0800

It is very clear (from experience no less) that anonymity on the internet alters individual behavior, typically in the direction of decreasing attention being paid to conventional social norms. Typically this is a burden that a site must overcome to maintain civil discourse, but here it serves a valuable purpose in the guise of downvotes (given, of course, that the moderators have done such a good job of keeping textual discourse so civil). I, for one, am prepared to defend any individual downvote I've made, but shiver at the thought of being open to defend all of them.

Ben Webster comments on "anonymity in ratings" (10365)

Ben Webster — Mon, 08 Nov 2010 10:18:21 -0800

There is a "suspicious voting patterns" area that moderators can access (I believe 10k+ users can't). At the moment, it's not showing any suspicious downvote patterns, though, of course, no algorithm is perfect. Andrew is right that the most effective course of action in these cases is to email moderators@mathoverflow.net.

Also, on the merits of your proposal, we've already had this discussion: http://tea.mathoverflow.net/discussion/265/signed-voting/ and my recollection was that not many people took a similar view of the matter to the one you've suggested.

EDIT: This seems very out of context now; I wrote it when Andrew's was the last post, and then forgot to press return...

nng comments on "anonymity in ratings" (10364)

nng — Mon, 08 Nov 2010 10:18:02 -0800

I guess we're not in Wikipedia anymore.

Michael Hardy comments on "anonymity in ratings" (10363)

Michael Hardy — Mon, 08 Nov 2010 10:15:13 -0800

"downvotes are the principal means by which the standards and conventions of the site are adhered to"

I don't see how they can do that when rationales for downvotes are kept secret. Verbal statements of objections are obviously efficient means of communicating them. But I _still_ don't know the reasons for the down-votes in question in this case; I've seen only speculation (and speculation is what it is) from people not claiming any responsibility for them.

Michael Hardy comments on "anonymity in ratings" (10362)

Michael Hardy — Mon, 08 Nov 2010 10:11:46 -0800

"As downvotes are the principal means by which the standards and conventions of the site are adhered to"

I find that surprising and counterintuitive. Can it be supported by arguments? It seems to conflict with the experience that is Wikipedia. Articles there do get deleted, and those who proposed and argue for deletion are identified by name, but that is hardly the "principal means by which the standards and conventions of the site are adhered to".

Michael Hardy comments on "anonymity in ratings" (10360)

Michael Hardy — Mon, 08 Nov 2010 10:09:42 -0800

"It's clear that there would be far fewer downvotes if one's names were attached to them."

The word "would" appears here. Assertions about what _would_ happen, if not based on experience, should not be qualified with the word "clear". Here's what _would_ happen if someone were to try to create an encyclopedia by using a wiki, so that contributions would come from anyone who chose to contributed. It would consist only of grafitti and would be ignored and unknown. That was "clear" to some quite sober and experienced people before they had experience with it.

I am comfortable with others disagreeing with me, and with expressing my disagreements with them, so I would down-vote postings when the poster knows who's doing it, and state my reasons if asked. I doubt that there's anyone here who hasn't done a lot of that sort of thing.

Cam McLeman comments on "anonymity in ratings" (10356)

Cam McLeman — Mon, 08 Nov 2010 10:04:58 -0800

As to the actual question "What is the advantage of such anonymity?":

It's clear that there would be far fewer downvotes if one's names were attached to them. As downvotes are the principal means by which the standards and conventions of the site are adhered to, this would be a very bad thing.

Harry Gindi comments on "anonymity in ratings" (10355)

Harry Gindi — Mon, 08 Nov 2010 10:04:47 -0800

Dear Robin, are you constantly writing poetry, or do you have a compulsive habit of hitting the return key?

=D! (No disrespect intended).

Michael Hardy comments on "anonymity in ratings" (10354)

Michael Hardy — Mon, 08 Nov 2010 10:03:13 -0800

@jbl: Are you saying no one should raise questions about politeness here?

I am comfortable with people disagreeing with me. I am not always comfortable with their expressing it in offensive ways.

Robin Chapman's way of raising a new issue here in this thread is the right way to do it: He used words. ("polemic" and "spam"). If he hadn't done that, I wouldn't have suspected based on down-votes that anyone thought my posting was spam or that I had put it there as pretext for a polemic.

(I remain amazed by Robin Chapman's seeming inability to understand that the question with which the "memorable titles" thread began was "what makes the title of a paper memorable for them", and the request for titles was tacked on as an afterthought.

Robin Chapman comments on "anonymity in ratings" (10352)

Robin Chapman — Mon, 08 Nov 2010 09:58:05 -0800

Mr Hardy, you are becoming vexatious. To deal with
your rhetorical questions, please read the question
and your reply.

Michael Hardy comments on "anonymity in ratings" (10349)

Michael Hardy — Mon, 08 Nov 2010 09:53:35 -0800

"pretext for a polemic"? Did I write such a polemic? Where is it?

Michael Hardy comments on "anonymity in ratings" (10348)

Michael Hardy — Mon, 08 Nov 2010 09:50:52 -0800

@Robin Chapman: You still haven't finished reading the original question in the thread. It did not ask only, nor even primarily, for such examples.

My answer was on-topic because it dealt with what the question ACTUALLY said. It was in no sense spam.

Robin Chapman comments on "anonymity in ratings" (10347)

Robin Chapman — Mon, 08 Nov 2010 09:48:43 -0800

The question asked for interesting examples of titles of papers.
Hardy's "Calculus Made Honest" reply provided no examples,
but instead gave a title of a book the poster was vaguely
thinking of writing. It did not attempt to answer the question,
but served merely as a pretext for a polemic. I am only surprised
that this reply didn't recieve many spam flags.

Andrew Stacey comments on "anonymity in ratings" (10344)

Andrew Stacey — Mon, 08 Nov 2010 09:35:30 -0800

Moderators can look for, and deal with, suspicious voting. If you think that someone has gone through and voted against a number of your posts, you should alert the moderators with your suspicions (via email).

(I don't know if the MO moderation tools are similar to those on SE2.0 sites, but if they are then they may already be aware of this voting spree against you.)

It certainly is impolite to vote against something without leaving a reason why on a real question, but on CW questions the answers are taken more as a poll and so comments are not generally expected (in particular, lest they descend in to an argument).

It is also worth noting (in case you aren't aware) that the '@Name' syntax doesn't work on MO in the same way that it works on other SE sites: there is no notification.

jbl comments on "anonymity in ratings" (10343)

jbl — Mon, 08 Nov 2010 09:32:56 -0800

Michael Hardy, don't you have something better to do with your time than complaining that someone hurt your feelings? It's the internet -- not everyone is going to think exactly as you do, and no one is obligated to explain themselves to you when they disagree with you. (Incidentally: you probably got down-votes because the first title is not obviously memorable and the second title is not the title of any work that exists.)

Michael Hardy comments on "anonymity in ratings" (10342)

Michael Hardy — Mon, 08 Nov 2010 09:20:40 -0800

@Qiaochu: Are you quite certain (based on particular experience, for example) of your assertion that appears to be a counterfactual conditional? It seems quite implausible to me. Since it seems implausible to me, I could assert "That would not happen". Would my bare assertion then be less well supported than yours?

Qiaochu Yuan comments on "anonymity in ratings" (10341)

Qiaochu Yuan — Mon, 08 Nov 2010 09:13:26 -0800

Voting is anonymous precisely to prevent this kind of argument from happening. If everyone knew who was downvoting them the flame wars would never stop.

Michael Hardy comments on "anonymity in ratings" (10338)

Michael Hardy — Mon, 08 Nov 2010 09:10:26 -0800

As far as I can tell, when one gives an up- or down-vote to a question or an answer, one does so anonymously. What is the advantage of such anonymity?

In the thread titled "most memorable titles", quite a large number of people answered simply by citing published papers with memorable titles, some of them employing humor to get the reader's attention or to help the reader remember them. "A Midsummer Knot's Dream" got a net 15 up votes, "The Joy of Sets" got 11 (that one's a book, not a paper), "Integrity of Ghosts" got 7, etc. Some were not particularly humorous but were memorable for other reasons, e.g. "Can one hear the shape of a drum?".

My first participation in that thread cited a paper by T. D. Parsons titled "Ancestors, Cardinals, and Representatives" with a link to its jstor URL, and like most postings in that thread, said nothing more. It immediately got one down vote, which seems irrational by itself, but that was followed almost instantly by someone going through various of my postings in various other threads and down-voting them. So I added a comment my "Ancestors" posting asking the anonymous person who did that to step forward and explain how it was so different from other replies in that thread that it would get such a different reaction. Several days have passed and that has not happened. The anonymous down-voter remains anonymous. Under the circumstances as a whole, that behavior strikes me as impolite. Would I be mistaken in thinking that?

The thread did not ask ONLY, nor even primarily, for such titles, but those are what have been posted. By I posted something else in reply to the question posed: "I was wondering if the MO-users would be willing to share their wisdom with me on what makes the title of a paper memorable for them; or perhaps just cite an example of title they find memorable?" It was a title that have considered for a book that would exist if I were to collect and extend some things I've written (only one of them an actual published paper, which I didn't cite). That title is _Calculus_Made_Honest_. It got three down-votes, and again I asked the anonymous person to explain his or her objections, and again that has not happened. But Robin Chapman posted a comment that I was off-topic BECAUSE I didn't cite a particular existing published paper (as if the question had asked simply for that). Since his comment doesn't make sense and shows only that he didn't understand the original question in the thread, at least some slight suspicion that he is that anonymous person could fall on him, and if it's not Robin Chapman, then he would be to some extent the victim of the practice of doing these things anonymously. Is that consideration outweighed by some particular advantages of anonymity?