Contributors metainfo: reputation and more

As asked by Vieniava in "How to make a list of user's reputation?", some of us came up with interesting ideas on how to fill an updated list with the reputation scores of the contributors to Answers.

I took the initiative to compile a public list of users with meta info:

• position (desc ordering by reputation)
• id
• nickname (truncated to fit the page)
• reputation
• # of comments
• # of questions asked
• % accept rate
• # of posts answered
• # of accepted answers

The code used to compile the list is available at the bottom.

EDIT

TMW team implemented a page with the metascores: http://www.mathworks.com/matlabcentral/answers/contributors

Please refer to it and congrats to the team!

Use this function to retrieve info from the link above:

function [metainfo, elapsedTime] = metainfo(type,order)
% METAINFO - Retrieve metainfo on contributors to www.matworks.com/.../answers
%
%   METAINFO Retrieve data sorted by reputation in descending order 
%
%   METAINFO(TYPE,ORDER) Specify TYPE and sorting ORDER as
%       type :  'reputation'
%               'questions'
%               'answered'
%               'accepted'
%
%       order:  'asc'
%               'desc'
%   
% Examples:
%
%     % Standard call (rep, disc)
%     info = metainfo;
%       
%     % Sort by question answered in descending order
%     info = metainfo('an','d');
%
% See also: URLREAD, REGEXP

% Author: Oleg Komarov ([e-mail]) 
% Tested on R14SP3 (7.1) and on R2009b. In-between compatibility is assumed. 
% 28 feb 2011 - Created

tic

% Check # inputs
error(nargchk(0,2,nargin))

% Retrieve inputs
if nargin == 0
    type  = 'reputation';
    order = 'desc';
end
if ~exist(type,'var')
    sortTypes = {'reputation','questions','answered','accepted'};
    type  = sortTypes{strncmp(type,sortTypes,numel(type))};
    order = 'desc';
end        
if ~exist(order,'var')
    orderTypes = {'asc','desc'};
    type  = orderTypes{strncmp(order,orderTypes,numel(order))};
end        

% Build url string
url = ['http://www.mathworks.com/matlabcentral/answers/contributors?'...
       'dir=' order '&sort=' type '&page='];

% First read
[page, ok] = urlread([url '1']);

% Catch number of pages to read
if ok
    totcon = regexp(page,'>1 - 50 of (\d+)','tokens');
    totcon = dataread('string',totcon{1}{1},'%d');
    nPages = ceil(totcon/50);
else
    error('Cannot read ".../contributors?page=1"')
end

% Loop over contributors pages
metainfo = cell(totcon,7);
metainfo(1:end,1) = num2cell(1:size(metainfo,1));
for p = 1:nPages
    if ok
        endpos = 50*p;
        % Id, Rep
        expr = '><a href="\/matlabcentral\/answers\/contributors\/(\d+)';
        data = regexp(page, expr,'tokens');
        if 50*p > totcon; endpos = 50*(p-1)+numel(data); end
        metainfo(1+(p-1)*50:endpos,2) = [data{:}];
        % Nickname
        expr = ['"Reputation: (\d+)">([\w\ ' reshape([repmat(92,1,137);33:59,61:64,91:97,123:126,161:255],1,[]) ']+)</a></h2>'];
        data = regexp(page, expr,'tokens');
        metainfo(1+(p-1)*50:endpos,[4,3]) =  cat(1,data{:});
        % Qcount, Ans, Acc
        data = regexp(page, '<span >(\d+)</span>[A-z<>"-\s\/]+','tokens');
        metainfo(1+(p-1)*50:endpos,5:7) = reshape(cat(1,data{:}),3,[]).';
    else
        error('Metainfo import stopped. \nCannot read ".../contributors?page=%d"',p)
    end
    [page, ok] = urlread([url sprintf('%d',p+1)]);
end

% Convert to doubles
metainfo(:,[2,4:7]) = cellfun(@str2double,metainfo(:,[2,4:7]),'un',false);

elapsedTime = toc;

Now if it can be charted day by day...

Efficiency gain would stem from the fact that a single post will serve as a reference and it will eventually grow and stay on top as people post here. The retrieval of scores would be faster since only one urlread would be virtually needed and the number of records per author is reduced to 1 (the perfect case)

@Walter: it could be charted day by day. It would be easier if not just I had the ability to edit the initial post (in order to updated the chart)

Walter should get a handicap. ;)

I was thinking maybe log scale ;-)

Walter Roberson, do you sleep? :)

I subcontract out my sleep to villages in developing countries, through an Eco-Trade Network.

Vieniava, I was wondering the same thing.

Does the above code still worK?

I still get the following error message:

>> [metainfo, elapsedTime] = metainfo('reputation','asc')
??? Error using ==> reshape
Product of known dimensions, 3, not divisible into total number of elements, 200.

Error in ==> metainfo at 82
metainfo(1+(p-1)*50:endpos,5:7) = reshape(cat(1,data{:}),3,[]).';

Answer 1

Recent news:

... and BAM! Walter hits 20,000! Congratulations are in order again. It's impressive to see, but depressing to compare :)

Timeline:

that was 10,000 on 2012-02-05
and then 15,000 on 2012-07-21
and then 20,000 on 2013-03-02
The countdown to 100,000 begins...

Original:

Has it been noted yet that Walter has reached five figures in reputation? Wow!

Congratulations on 10000 Walter!

Cheers, Sven.

It's a feat that will not soon be repeated.

This one is impressive enough that I am up voting it. Well done Walter.

So what -- five figures isn't that m... Oh. You don't mean in binary, do you...

Holy crap. Nicely done, Walter. I join Daniel in voting this up to the top.

I don't know how I managed to sneak in another 1000 in the last 30 days -- I was mostly huddling in bed sleeping off the February Blahs.

This afternoon, July 21 2012, I reached 15000 reputation -- and I still have two lives and one power-up pill remaining!

March 3, 2013 -- apparently I'm slowing down in my old age.

You just passed 20,000 though! Almost 3 times as much as anyone else! You got 5000 points in like 7 months and it takes me maybe a year or two to accumulate 5000 points. Has anyone ever done a trendy on reputation? I think it would show Walter pulling away from the pack.

I've considered turning Oleg's code above into a trendy... I'd need to plot things on log scale however to give the rest of us some pixels on the graph.

The sampling a year ago, needed log log for useful representation.

April 1, 2013 - 21000. The end-of-year rush for assignments helps.

May 2, 2013 - over 4000 Accepted Answers.

Answer 2

EDITED: deleted content (old and replaced by Contributors page)

Average points per post might be interesting. post = (questions + answers) should be relatively easy to fetch. Points are not allocated per comment (though sometimes I think they should be, so a good comment can be rewarded), but in at least one sense should perhaps count in "posts"; that would require using "parse every Question" strategy though.

If you do decide to do points-per-post then you might want to put a minimum on the number of posts required, as otherwise someone who posts just once and gets a couple of votes could easily end up on top of the list.

I am averaging less than 2 points per post and can see several people from about John D'Errico (20 points) and above whom I would consider valid points-per-post leaders on the basis of sustained contributions. I do _not_ expect to lead the list: I'm at the top of the point totals now through volume, not through quality.

Walter, you are in the top cause, among others, you experienced Matlab user and smart person.
Oleg, leave the job ("reputation processing") to the Mathworks team :)

I was actually surprised at how simplistic the reputation calculation is -- no eigenvalues? no graph Laplacians? what gives?! At very least, I'd like to see credit given for an answer that remains the only answer for several days. That generally seems to indicate that it's the right answer but the asker never bothered to accept it. (I guess I should start up-voting those when I see them.)

Take a look on No 28 (reputation=2), No 35 (reputation = 0 ). Bug?

The top 20 points-per-answer amongst those who have posted at least 10 answers:

6.4 | Kenneth Eaton
3.8 | Andreas Goser
3.8 | Matt Fig
3.2 | Rob Graessle
3.1 | Martijn
3.0 | Doug Hull
3.0 | Jiro Doke
2.8 | Jan Simon
2.8 | Ned Gulley
2.8 | Seth Popinchalk
2.5 | Sanchali Purandare
2.4 | David Young
2.4 | John D'Errico
2.4 | Oleg Komarov
2.4 | the cyclist
2.3 | Siddharth Shankar
2.3 | Vieniava
2.2 | Mark Biesiada
2.1 | Andrew Newell
2.1 | Paulo Silva

And special congratulations to Edric Ellis, who has earned his reputation of 66 with just 5 answers.

I'm trying to find a more efficient way to collapse all the stats...and the fcn below should also be updated...bugs...there will be for now. Testing requires more than 20 minutes now (especially because of my connection...). Btw is there any way to avoid loading the whole page but just to a certain point??

I wonder if Mathworks could add a page sub-structure, such as

answers/1058-contributors-metainfo-reputation-and-more/serialnum

where the serialnum was incremented each time some content of the page changed -- especially a new answer, a new comment or an answer or comment deleted, but a upon an edit of some sort would probably be beneficial as well for other purposes. The same reasons you would want to know about text edits might want to know about voting changes a well; perhaps two or more fields could be used.

For stats purposes, you would look at the serial number of the structure changes; if it was the same as what you had recorded, then there are no posts from new people. For people wanting to implement "feeds" of their own, the serial number of the content itself would be the one to test against. For people wanting to do "mirrors", a change in any of the three fields (including voting info) indicate the page needs to be recopied.

Matt Fig continues to spoil my plans for utter domination -- curse you, Snoopy!!

Looks like I'm slowing down -- barely 300 more reputation points earned in the last two weeks. Matt is accelerating, though!

Walter, sleeping too much lately :D!

Answer 3

EDITED: deleted content (obsolete code, replaced in first post)

Nice function, but the charData output runs into trouble if the user name is longer than 29 characters - as is true for user #1695138.

Yes, I know. I was working on a semidefinite version of a single function...but really don't have time (especially to test it) to update now.

Answer 4

Various congratulations are in order.

1. Walter's 1500th answer
2. Walter breaking 2500 rep
3. Matt Fig's 500th answer
4. Paulo breaking 1000 rep (OK, jumping the gun, but it's imminent)
5. Jan's 100th accepted answer
6. Jan breaking 800 rep
7. Andrew stealing 5th place
8. Sean's 300th answer
9. My 300th answer, but most important of all...
10. My rep reaching the appropriate value of 666. Mwahahahaha!!!!

So by voting for this answer I've made you less evil!

Gah! How evil of you.

Congrats to Paulo for (factually) reaching 1001!

yay! thanks Walter although the number still means nothing to me besides that I'm spending too much time here :(

7. Wait - stealing? I paid the admins good money!

Answer 5

Congratulations to Jan Simon for reaching 1517 reputation, and thus becoming our newest Community Editor in Waiting!

Thanks, Walter. Now I'm not waiting just seconds for the characters to appear, but also hours, days, weeks and month until this problem is fixed. Now I'm waiting for something more?! Nevertheless, I like MATLAB Answers.
I'm not sure if a fixed limit is really helpful. I think the TMW team should decide this dynamically.

Answer 6

I am not sure but I don't think Andrei, Kaustubha, or IA got their props for making it to editor.

Congratulations to Sean de for passing Matt and moving into 4th.

I cannot tell if Jan's jump was earned or magical, but well done getting past 5000.

Alas, Jan magically jumped backwards again :(

But Andrew wandered over the 1500 mark sometime today! Editor Andrew!

I hear angels singing ...

I don't know what props are unless you're talking about regionprops(). I also don't know that I'm an editor - I've never gotten notice of that and don't see any special links on any Answers page for "us editors." I'm #10 on the list - how high do you need to get to get special privileges?

Props: 'Slang term for "accolades", "proper respect", or "just dues".'

Editor: last time I checked, they had to enable each person individually. Not just by setting a flag either: it seemed to require a downtime.

There is no special sign or sigal marking editor status: the main way to know that you have it is that suddenly there is a Delete control for _all_ content. Oh yes, and also a big black button with glowing black lettering that says, "Do not click this!" (Urrr, sorry about that last system crash!)

Answer 7

Azzi Abdelmalek is the first to cross the old editor boundary (1500 rep) and now is in excess of 1600 rep. Well done.

Answer 8

Dear Walter,

10'000 answers are a large and great piece of work!

Thanks.

Thanks!

I held off replying here for a few hours, waiting for another milestone to be reached as well. I am now past 10000 answers, 3200 accepted answers, and 18000 reputation. Which makes me a Community Editor "a dozen times over" :-)

Jan, so is 1000 answers. Congratulations on exceeding that today!

Jan reached 1000 Accepted Answers. I remember that was a difficult milestone to reach, representing a lot of work.

Yeah. I've written 2460 not accepted answers. Some of them belong to threads I've posted multiple answers in, some of them are simply wrong (and Walter found the problems frequently - thanks!), some of them have been off topic because I always try to see the deeper sense behind a question, some of them are ignored, but in most cases other answers have been better.

Did somebody found out how to inject SQL code in the interface to solve questions like: Number of my unaccepted and unvoted answers, or ratio of comments to questions to comments to answers ?

Don't they do that kind of thing in the Trendy forum? (BTW I meant Accepted answers.)

Answer 9

Congratulations to Matt Fig for making it to 500 before I made it to 1000! He increased from 388 to 501 in about a day and a half, during which time I only increased from 901 to 966.

Numerical projections last night based upon answer acceptance rates were that it would take another 1310 questions answered each before he caught up, so clearly his solutions have been found vote-worthy.

Yee haw! I made it to 1024 and beyond! I am now a Level 2 Matlab User, with a gain of +1 Wisdom and +1 Charisma!

So, next level 1048576...
Btw, can you give a look at the function in the first post...it stops whenever finds 'strange' chars (I don't understand much about character encoding).
Thanks

Answer 10

Has anyone done a statistical analysis of the distribution of reputations? In my trials, exp() of a polynomial looked to fit a bit better than a plain polynomial. I sampled over the first 22 reputations; degree 5 was the least unsatisfactory. Degree 6 did fit the sample data better, but of course even degrees have to go back up, leading to a projection that further people would have high reputations. Degree 7 complained that the polynomial was badly conditioned.

I didn't sample far enough to get in to the bulk of the values. At the moment, 298 people have non-zero reputations, with 66 of those being reputation 2 and 91 being reputation 1.

I'm curious as to whether it matches some common distribution such as pareto.

Answer 11

Congratulations to Paulo for reaching 501! One good lap and he could overtake Matt!

Answer 12

Congratulations to Jan for leaping past 500!

Thanks, Walter. I have to decide between leaping and sleeping.

Answer 13

Congratulations to Paulo for reaching 100 accepted Answers!

Answer 14

Congratulations to Matt Tearle for reaching 500!

Thank you. Amusing thing #1: you noticed this before I did. In fact, I noticed only because of this post! Amusing thing #2: forget the elegant one-liners, forget the cunning MATLAB guru tricks, the answer that takes me past the 500 milestone is... http://www.mathworks.com/matlabcentral/answers/3791-how-to-store-5-or-6-files-output-into-same-excel-file

(Using my personal definition of "amusing", obviously)

Answer 15

Congratulations to Matt Fig for reaching 100 accepted Answers!

Thanks Walter! BTW, we need to talk Oleg into removing that first 4 page answer to ease the scrolling here.

It has been a long day, goodnight...

Answer 16

Congrats Walter for answering 1000 Questions!

Thanks!

Answer 17

Congratulations to Matt Fig for reaching reputation 1001!

Let's see... that's 500 in 27 days... now carry the 1... Ah, about 130 a week.

I knew there was a reason you were keeping a record! I bet you have even plotted some of this stuff, no? ;)
Thanks Walter.

It was only one tinny little line plot! Oh yah, and a few polyfit line plots of course. No more than a handful of spline plots. But no log scale plots, I swear!

Congratulations, too, on your amazing 38% acceptance rate. I'm going to have to start studying your methods!

Answer 18

Congratulations to Andrew Newell, who joined the 500 club, and is now presently at 525!

Andrew, you will be sent an autographed picture of Miss Jelin -- as soon as we can convince Oleg to give it up.

I still haven't got that picture ...

Answer 19

For once I get to scoop Walter:

Congratulations to Walter R for successfully avoiding any Y2K issues in rolling over to the big "double M".

Miss Jelin will be kicking herself for missing out.

I'm a Unix person: the important roll-over is 2038!

Nearly there. Be glad Helen Chen isn't Mayan, or your rep would have just reset.

"scheduled maintenance" my backside! It was you, wasn't it, Walter -- your 2038 rep rollover crashed MATLAB Answers!

How will you cope with the guilt? All those desperate people needing urgent help on their secant method homework. All on your conscience!

What happened to the logo next to your name, Matt? I hope this is a change in MA rather than your employment situation....

or we are all working for mathworks now ahahah the symbol is irrelevant :P

While they might be fully justified in firing me for spending too much time here, I think they chose leniency because, after all, I do spend my time telling everyone how great logical indexing is :)

Paulo's almost right, but don't expect a paycheck anytime soon. Apparently the logos have been removed so that we're all, like, one, man. [sing kumbayah here] Or something like that.

Answer 20

Please welcome Oleg to the 500 club!

Welcome Oleg to the 500 club, also Matt Fig for his new Editor status !!!!

Hey, Matt, format this. ;)

Thank you guys :)

I think Andrew meant,

HEY MATT CAN U FORMAT THIS PLZZZ?! ITS URGENT!!!!!!

Seriously, though, congratulations to both of you. I will follow your future careers with interest.

Answer 21

Sean de took a break from his studies long enough to reach 500 reputation! Will he be granted his Master Of Fine Answers??

In less than two years, Sean has increased his reputation 10-fold, to over 5000! Congratulations!

Answer 22

subsref({'Matt Tearle'},substruct('{}',{Accepted>=100}))

And Andrew too! Congrats!

Thank you. Passing 100 is my favorite milestone so far.

Answer 23

Now that the first 6 on the list have all reached 100 or more Accepted answers, is it time to start a new round? :)

Or do we wait for the first 10 to all reach 100+ Accepted Answers?

Party now!

Now you've got me trying to imagine what a MATLAB party would be like... We could play Find and Fseek, and Bobbing For Logical Indices, and Hula-Loops... and we could have Pi à la ModeManager...

Pin the Meaning on the Question ...

Perhaps, just as needed: Pin the Answer to the Comment!

Answer 24

Fittingly, time to congratulate the author of this thread. 600 rep for Oleg!

Thanks! Although for the next week I will have to reduce/stop my answering frequency (exams)

Answer 25

Thanks to Oleg, who took time from his exams to reach 100 accepted answers!

Thank you Walter! My exams ended today with a mediocre performance in econometrics...was thinking to change completely direction and study computer science or go for maths, but afraid to start over after 3+1 years of uni...

Answer 26

Please welcome Kaustubha Govind to the 500 Club!!

And congratulations to Kaustubha for having reached 100 Accepted answers!

Answer 27

Congratulations to Arnaud Miege for reaching 500!

Arnaud also has 97 Accepted and so will likely be at 100 within days.

We now have 10 Charter Members of the 500 Club :) Dues statements will be sent out shortly ;-)

Congratulations to Arnaud for reaching 100 Accepted Answers!

Answer 28

A big congratulations to Matt Fig for reaching 2004 reputation!

Speech! Speech! Speech!

Thanks, Walter!
No comment at this time... Haha.

And your acceptance rate is still going up! I am beginning to suspect an inside job.

44% right now! Since you first pointed this out above I have been trying to get to 45%. That is about as good as I could hope for with the amount of questions I answer...

There are others who have near 50% or more: Great job to them!

And congratulations for reaching 2500!

Answer 29

Congrats to Walter for reaching 3,000 points - a double editor!

O.k., so it is a little early (2999 points as of now), but I may miss the actual rollover tonight. It's a preemptive congratulations.

I don't open my xmas presents early ;-)

Okay, now I can celebrate!

I'd like to thank my team of substitute-sleepers, without whom this would never have been possible!

I knew it!!!

So now we know what all of Santa's elves are doing during the off months ...

Answer 30

Congratulations to Jan Simon, who just become the fourth person to reach 1000 reputation!

Congratulations to Jan also, may he continue to provide good answers for a very long time :)

Thanks! I hope to provide answers *in* a very short time, if the preview can be disabled in the near future ;-)

Answer 31

Congratulations to Sean de for promotion to Elite 1000 status! Sean is now entitled to complimentary pretzels when his Answers are delayed for 4 hours or more!

Answer 32

Congratulations to Andrew Newell, who has reached 1000 reputation!

Answer 33

Congratulations to Paulo for achieving 1500 and editorship!

(Apologies for the delay in noticing; I have been traveling.)

Thanks Walter, this spy still doesn't care about Reputation points, just here to learn and help others at the same time :)

Answer 34

Congratulations to our newest editor-in-waiting, Sean de!

Sean was not available for comment at press time, but rumors are that he and his cousin R. Goldberg have something in the works!

Answer 35

August 31, 2011: yesterday evening I passed 5000 reputation, and this morning I passed 3000 questions Answered.

Time for a nap, I think.

Walter congratulations, you are also testing the system (MATLAB Answers) while helping others :) thanks for sharing your huge knowledge with us.

Walter you should receive MathWorks T-shirts or any other TMW goodies for your help ;)

Answer 36

September 4, 2011:

Congratulations to Oleg for reaching 1500 reputation, making him eligible to join the motley crew of Community Editors!

Thanks, happy to join the evil editors!! :)

Oleg welcome to the editors and thanks for sharing your knowledge and for all the help provided, the same goes for all regular contributors and editors, good work!

Answer 37

Oct 9, 2011 22:22 UTC-5:

It was a steep climb, but I made it! I crept over 6000 reputation yesterday, and reached 1000 Accepted Answers late tonight, having also reaching 3700 Answered questions.

The 1000 Accepted Answers point was a hard one to reach! It's getting tougher and tougher to edge out the other great answers ;-)

Dec 16, 2011

Sometime yesterday or today, I passed 8500 reputation and 5000 Answers. Oh! Oh! Jiminy Crickets!

Jan 1, 2012

This evening, I squeaked through to 1500 Accepted Answers. Now that was tough work!

March 28, 2012:

While I wasn't looking today, I passed 2000 Accepted Answers. My glasses were broken for a week recently so I didn't think I'd reach that this month.

June 27, 2012:

Today I passed 8000 Answers and 2500 Accepted Answers! Porridge for all my tomte !

Answer 38

Congratulations to Jan and Fangjun, both of whom have exceeded 2500 reputation!

Kaustubha Govind reached the 1000.
And Matt Fig concentrates on other fields of work. Hi Matt!

Kaustubha has so many quality answers that I am startled she is not higher. I should take more time to vote for her answers.

And Matt -- come back! We need your humpday puzzlers!

Answer 39

If someone feels like playing with curve fitting, here are the figures as of last night for Accepted Answers for the 770 people who have had any Answers accepted:

[1 1 4 1 2 1 2 1 3 1 1 2 1 1 1 5 1 17 1 10 1 1 1 1 1 2 6 309 1 1 1 2 58 1 4 259 1 3 3 1 2 1 1 1 1 2 1 1 1 1 1 1 1 3 12 1 1 2 1 1795 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 3 3 7 1 1 1 1 1 1 3 1 1 1 1 1 1 1 1 1 1 1 2 5 1 1 1 1 1 1 1 3 1 1 1 6 1 115 1 2 1 5 31 2 2 1 2 1 3 1 1 2 2 1 1 16 2 2 1 1 252 1 4 1 1 1 1 1 1 1 1 1 6 1 1 1 1 1 1 21 3 1 1 22 1 1 1 10 4 1 1 1 386 1 1 1 1 2 1 4 209 15 1 1 1 1 11 1 1 7 11 3 1 1 1 3 3 1 1 1 1 116 1 1 1 1 1 1 1 1 1 10 1 1 1 1 1 1 1 5 3 1 1 1 43 1 1 1 1 1 1 1 1 1 5 5 3 1 6 3 2 1 1 4 1 1 9 234 1 2 1 1 1 1 1 1 1 1 3 1 1 1 1 38 11 45 2 1 3 13 1 5 1 3 1 1 1 1 1 1 1 2 78 1 1 1 1 1 1 1 2 1 2 1 1 1 2 1 1 1 1 1 1 54 4 1 1 1 1 1 15 1 5 3 11 1 1 1 1 1 1 2 3 16 1 1 6 1 4 1 2 4 1 1 1 1 1 2 2 1 5 2 1 1 1 1 153 6 66 1 1 1 1 1 1 7 2 1 1 22 1 1 1 20 1 1 1 1 2 1 1 2 3 1 1 2 1 1 34 2 280 2 2 1 1 1 1 1 9 1 1 5 1 1 127 3 1 18 1 5 3 1 1 1 1 1 1 1 1 1 1 1 8 1 2 1 5 1 1 1 5 1 26 8 1 1 1 26 1 1 1 1 1 18 1 3 1 1 3 1 1 1 1 1 1 3 1 3 1 7 6 1 2 16 2 1 1 1 1 1 1 1 2 17 6 3 2 1 1 1 1 48 2 1 2 1 1 1 1 1 1 1 1 1 1 1 3 518 2 2 1 1 268 1 19 1 1 1 4 1 1 7 1 9 1 1 3 4 1 1 1 1 1 8 16 1 1 1 1 5 29 1 1 1 1 25 1 1 1 2 1 635 1 1 1 11 2 1 1 1 1 2 1 1 1 1 1 2 7 12 10 1 1 1 1 1 1 57 55 1 1 9 1 1 1 2 1 1 359 1 1 1 1 1 1 1 1 2 1 1 1 2 2 2 1 1 1 46 1 42 1 1 1 2 12 10 1 1 11 1 2 1 1 1 1 1 1 2 1 1 1 1 1 2 1 1 1 1 2 1 1 1 1 1 1 1 1 5 1 42 2 1 97 26 97 1 1 2 1 1 4 1 1 2 2 2 1 1 1 2 1 1 10 2 131 1 1 1 1 1 1 1 1 1 2 1 5 1 1 1 5 5 6 1 2 1 8 4 1 1 1 1 1 423 1 1 23 1 1 1 1 2 1 1 1 1 1 1 2 1 1 1 2 29 1 1 5 1 1 1 1 1 3 1 1 1 1 1 1 4 1 8 1 1 1 2 1 1 1 1 1 1 1 3 1 228 93 1 1 1 2 1 1 1 2 1 16 3 3 44 12 2 1 1 1 1 1 1 33 1 1 2 7]

I cannot seem to make much of it, even if I remove the obvious outlier. The tail of 1's is so long that it distorts everything else. Though if I remove the 1's, then log-log tends to a straight-ish line.

Answer 40

Congratulations to Jan for reaching 4000 reputation!

Jan reached 5000 reputation!

Jan, do you prefer Import or Domestic?

I like French and Swiss cheese, but prefer German milk because it needs less resources for the transport. But I'm not sure, which object is meant by "Import or Domestic" and googled this term. The first hit tells me: "Foreign car or domestic? Reputation isn't the only factor".

While Walter could be refering th cheese (I like that idea) or cars, typically "import or domestic" is a question asked in the US about ones preference for "beer" (which encompasses ale, lager, stout, etc).

@Daniel, Thanks, I've known that this forum is fine to improve social interactions and intercultural communication.

@Walter: I'll take a Guiness after sunset and think of you and the other forum members. I have no idea if Guiness is "import or domestic" in Germany, because I drink one beer per year only. I'm not sure if you reached the 15'000 before I got 5'000. Anyhow, you've earned 3 bottles of beer.

According to the US norms, it is the country in which the beer originates that determines if it is an import, regardless of where the beer is brewed. In the UK, it is the location of the brewery that determines import/domestic status. By the US standards, Guiness is Irish and is therefore an import everywhere expect Ireland.

Amusingly, I myself would much prefer the milk (or the cheese.) Just a preference. But Black Forest Cake (Schwarzwälder Kirschtorte) doesn't taste right if it doesn't have Kirsch!

Answer 41

Congratulations to Andrew, Daniel, and the cyclist, all of whom have exceeded 1500!

For the historical record, the people who have currently reached or exceeded 1500 include,

andrei bobrov
Andrew Newell
Daniel
Fangjun Jiang
Image Analyst
Jan Simon
Kaustubha Govind
Matt Fig
Oleg Komarov
Paulo Silva
Sean de Wolski
the cyclist
Walter Roberson
Wayne King

... and today I also snuck over 12000 !

I passed 7000 questions answered sometime in the last couple of days.

I guess I should be happy that Answers hasn't had any down time in the past few weeks, but it would be nice if I could get the keys.

I missed this when it was first posted, but thanks for the congrats!

Congrats to Image Analyst, who has reached 3000, and to Sean de Wolksi who has reached 3500!

Answer 42

Congratulations to those who have reached at least 500 Accepted Answers:

andrei bobrov
Fangjun Jiang
Image Analyst
Jan Simon
Sean de Wolski
Walter Roberson

500 Accepted Answers takes a lot of time and dedication and skill (and patience!). I would say that this is the hardest of the milestones to meet!

Dear Walter, thanks and congratulations. I appreciate the respect you show by sorting the list alphabetically - just a tiny detail, but I like it, that you care.

Answer 43

Andrei Bobrov has reached Double Editor, 3000 reputation!

Answer 44

Congrats to Jan Simon, who reached 6000 reputation a moment ago!

Thanks, Walter.

And congratulations to Image Analyst, who ascended to rank 2 in the list of contributors! Due to the aim (and name) of this forum, the number of (good) answers is more important than the "reputation" number.

I'm still way behind you Jan (4666 vs. 6011) in reputation points, though I'm #2 in number of answers posted. I guess your Accepted rate being higher than mine gives you 30% more reputation points. I think I'll have a long way to go before I pass your reputation points, if ever. I doubt anyone will ever surpass Walter. Looking back over my posts I see an extraordinarily high number of questions never accepted - much more than the others. Even though they say in a comment it was solved, they never seem to mark it as accepted, or at least at a lower rate than everyone else for some reason.

@Image Analyst: I have only 43 more accepted answers. The surprising difference in the points might come from voted answers and my 36 questions. But the counting is still a little bit mysterious for me - I guess the accepted answers are counted in real time, but voted answers once per hour only, or something like this.

Last year my reputation number made a jump of 1500 points over one weekend. I asked TMW for an explanation and they found out, that this have been caused over 700 votes from one user. They asked him or her for an explanation, but did not get an sufficient answer as far as I understood. Finally I got back my old reputation number again.

Perhaps this silent follower voted again, but less striking by distributing it over several weeks instead of hours.

Anyway, my conclusion is, that the reputation number can be influences by cheating or let me calling "strange effects", but the number of given answers combined with the quality is a hard fact.

It could be a cause for users to hesitate to accept your answers, that some of your answers contain a lot of code. A beginner could be overwhelmed by this, although the answer is correct and useful. Look at the difference between the information density in the question and the answer in this Example. But this is exactly the cause, why the editors can accept answers also. I'm going to look for your answers.

But what I wanted to say is: I personally can profit from your answers and the shared knowledge, and this will concern many many of other forum users also. Thanks.

[EDITED] I've checked about 300 of your answers. It was depressing. Most of the threads without an accepted answer have a brute lack of feedback from the OPs. I've seen over 100 of your questions for clarifications without an answer. I have to stop the reviewing for today due to comming near to my frustration limit. There have been a lot of vital discussions between you and Walter, while the OPs did loose the interest.

Therefore my conclusion is: The procedure of accepting an answer does not only measure the quality of the answer, but also the interest of the OP in his/her own answer as well as the respect for the forum. Do you know understand, why I think that the number of answers is hard, while the reputation points and acceptance rates are more fluffy? But without doubt they are a useful democratic method to control the editor powers in the forum.

Jan passed 7000 reputation earlier this evening!

@Image Analyst: You have written 4 month ago:

I think I'll have a long way to go before I pass your reputation points, if ever.

I do not agree with your definition of "long", but without doubt the work is "hard".

I'm going to reduce my contributions here and care a little bit about the FEX.

I should put some effort into that too. I have several dozen short image processing demos illustrating various things and I keep thinking I should post them all as a "grab bag of image processing demos." But the number of demos keeps growing so I keep putting it off. You do have a nice collection in your FEX. By the way, had I known in advance I might have tried to meet you in person. I was walking around the University of Heidelberg in September 2011 (a Sunday sightseeing trip from Frankfurt where I was staying). I didn't know you were there.

Answer 45

A big Thank You to Image Analyst, who at this very moment is at 5000 reputation!

Answer 46

Hi walter,

Hope you are doing wel..!!! felicitations!! Crossed 18,000 reputations..Great work..