78

After thinking about

Should the weight of downvotes be increased?

which I declined for the reasons specified in my answer, and

Please charge rep for questions after threshold

I have a proposal.

How about if question upvotes were only worth +5 rep instead of +10? And this was applied retroactively? (answer upvotes would be unchanged from the current behavior, so this would be question specific.)

This seems to be a better solution to the particular "shore, there is always gold" problem without all the rather serious negative repercussions of increasing the punitive value and/or cost of downvotes across the board.

22
  • 2
    I don't think that it's the solution but it's at least a solution.
    – perbert
    Commented Mar 19, 2010 at 4:02
  • 7
    Or, better yet, just don't give any rep for questions at all. Completely disincent users from repeatedly asking questions that have been asked a dozen times before, and let the answers to real problems be the question's reward. Commented Mar 19, 2010 at 7:54
  • 7
    @software questions definitely have value; placing them at +0 is not correct IMO. We're cutting the value in half which is a pretty big change. Commented Mar 19, 2010 at 8:05
  • 16
    @Jeff, what would be really really cool would be to see a before/after graph of number of users per reputation, once you've finished the recalc.
    – Benjol
    Commented Mar 19, 2010 at 8:24
  • 7
    @Jeff, how about putting Alert bar on everyone on SO when it starts (like when you did on moderator voting), saying that "There's going to be Reputation Changes and Recalc anytime soon, and You could lost up to 50%", to prevent extra claims about their reputation lost?
    – YOU
    Commented Mar 19, 2010 at 10:13
  • 32
    Maybe there could be a notification that says "thank you for being a valuable member of the stackoverflow community for over a year. -500."
    – user27414
    Commented Mar 19, 2010 at 13:11
  • 4
    Ouch, I just reached 10k rep but I feel like I'm going to loose a lot of rep and access to moderation tools :'(
    – marcgg
    Commented Mar 19, 2010 at 14:17
  • 2
    @Jeff, I was a few days away from 10k on Meta, now I'm hosed. Commented Mar 19, 2010 at 16:44
  • 58
    It's a little disingenuous to ask on Meta "Should the weight of question upvotes be reduced?" then 5 hours later you announce on the blog that you are doing this. 5 hours that were in the middle of the night for most North American and European users of the site, no less. Why even bother with the pretense of having a discussion when you don't allow time for people to make their counterpoints?
    – Kip
    Commented Mar 19, 2010 at 17:59
  • 16
    And especially, what was the point of asking us what we thought about change #2 if we said we didn't like it and then you implement it anyway? Hell, it's your site and you can do whatever you want but I suggest in the future you just do it without asking if you're not really interested whether we agree or not.. At least it's more upfront Commented Mar 19, 2010 at 18:20
  • 5
    This hurts normal users (like me) who ask good well thought out questions. I have felt a part of the SO community for a while now, but I feel I am getting smacked in the face on this one. Good Questions are the bread and butter of SO. Find a way to punish bad questions, not questions in general!
    – Vaccano
    Commented Mar 19, 2010 at 19:24
  • 2
    It's much harder to get 30 votes on a question than it is to get 15 on an answer [edit for numbers: SO currently has Good Answer x 8k and Nice Question x 10.8k]; why should the easy one be worth more? This change may make sense for questions that get one upvote, or two upvotes and two downvotes, but it doesn't seem to work as well with actually good questions.
    – mmyers
    Commented Mar 19, 2010 at 20:10
  • 3
    @mmyers: I think the point is that it takes very little effort to ask several hundred middling questions and accumulate +1 - +3 on each one (even though it might look like 2 downvotes followed by 3 sympathy upvotes). For a net gain of hundreds, even thousands of rep. Commented Mar 20, 2010 at 1:32
  • 11
    Looks like this decision just got reversed.
    – CaldeiraG
    Commented Nov 13, 2019 at 19:56
  • 5
    @CaldeiraG - at least they didn't bother with "the pretense of having a discussion". make up your mind SE. Commented Nov 14, 2019 at 3:53

18 Answers 18

76

I'm fine with that. Something that's bugged me for a while is that while I don't mind helping out by editing poorly-asked questions, it seems wrong that some users come to rely on this, coasting through asking one lousy question after another, occasionally garnering significant amounts of reputation post-cleanup.

...Although to be fair, I have almost nothing to lose if this suggestion is implemented...

8
  • 20
    I have always agreed that there is a substance to this concern, particularly observing the type of users who tend to benefit from it. Commented Mar 19, 2010 at 2:32
  • 1
    You could counteract this by giving points to the editor. The details of this though.... I don't know. Commented Mar 19, 2010 at 3:03
  • 4
    @George: yeah, that's been suggested before... IMHO, it's a can of worms best left unopened.
    – Shog9
    Commented Mar 19, 2010 at 3:15
  • 4
    @George: Significantly outweighed by potential abuse and difficulty of implementation to avoid it. I don't mind editing to improve the question/SO without getting rep.
    – Gnome
    Commented Mar 19, 2010 at 3:16
  • @Shog9, I have a "suggested improvement" feature functioning just fine on my CT, if anything I wish more people would suggest more improvements, I even created a Kaizen badge for people with tons of approved improvements. The only slight annoyance is that it can take a while for your change to go live.
    – waffles
    Commented Mar 19, 2010 at 5:55
  • 2
    This change nails the good questions and the bad ones equally. SO is hurting what gives it life. For questions (mine at least) upvotes are the reward for taking the time to make it usable not only by me, but by someone who hits it from Google. If we are going to hurt all questions instead of finding a way to just get the bad ones, then all questions will suffer.
    – Vaccano
    Commented Mar 19, 2010 at 19:40
  • 1
    What will be the point (rep based at least) to taking the time to make a great question when a mediocre/poor one will get you your answer. There are enough people answering questions out there that one of them will likely say the info you need. No need to worry about those that come after, there is no rep in it for you. Get your answer and get back to life. (I do not like this attitude, but that is what this change is encouraging.)
    – Vaccano
    Commented Mar 19, 2010 at 19:40
  • 3
    @Vaccano: well, you still get rep, just not as much. And unfortunately, there are already a lot of people with that same attitude - frankly, I'm not all that confident in the motivating ability of rep when it comes to asking good questions, since as both you and Jeff note, the real rewards for asking questions are the answers themselves. IMHO, the best part of asking a good question is the good answers that tend to result from it - whether that's sufficient motivation remains to be seen...
    – Shog9
    Commented Mar 19, 2010 at 20:25
34

If you think changing the weight of downvotes would cause a stir, this one would cause a storm.

In particular lets look at 3 lists:

People most affected by the change with 3k rep or more

User               Reputation QUpvotes AUpvotes QDownvotes ADownvotes Penalty
-----------------------------------------------------------------------------
George2            3233       275      0        9          0          100.00 
Blankman           4850       476      0        98         3          100.00 
mrblah             3545       343      0        21         0          100.00 
Shore              3184       302      2        209        0          99.34  
Joan Venge         6362       608      7        162        6          98.86  
Thanks             4916       438      12       73         1          97.33  
Edward Tanguay     11743      1019     34       21         1          96.77  
Masi               6052       613      31       170        1          95.19  
dotnetdev          5236       400      23       23         16         94.56  
jasondavis         3442       263      16       19         2          94.27  
acidzombie24       6054       568      39       71         7          93.57  
oo                 4314       786      56       43         8          93.35  
LazyBoy            3474       195      17       29         6          91.98  
kunjaan            3565       310      29       32         3          91.45  
KingNestor         3881       348      33       7          3          91.34  
ProfK              3119       275      33       14         1          89.29  
B. Tyndall         4929       405      49       9          0          89.21  
Jeff V             4251       364      47       4          1          88.56  
Readonly           6644       563      73       11         2          88.52  
Brian G            3654       301      40       19         6          88.27  
raldi              3005       248      36       27         10         87.32  
Simon              3023       225      41       4          5          84.59  
Jader Dias         5254       461      87       15         4          84.12  
Jeremy Rudd        6920       492      96       52         11         83.67  
Guy                5808       446      98       19         7          81.99  
Alex               5219       359      93       3          9          79.42  
Mat                3107       229      60       9          1          79.24  
minty              3115       225      63       4          0          78.13  
Teifion            8021       522      149      43         37         77.79  
JL                 3431       219      63       8          10         77.66  
stalepretzel       3128       229      66       14         12         77.63  
boris callens      4834       313      91       3          10         77.48  
ripper234          5397       334      100      23         8          76.96  
Lukasz Lew         3370       238      72       12         1          76.77  
yesraaj            6454       425      129      52         5          76.71  
Wilco              3425       220      69       2          1          76.12  
Hanno Fietz        5281       408      128      4          4          76.12  
flybywire          7250       537      171      37         13         75.85  
Ramesh Soni        3116       209      68       12         22         75.45  
Daniel Magliola    3498       224      73       6          1          75.42  
Ray Vega           5953       414      138      3          3          75.00  
J. Pablo Fernández 5048       348      117      6          7          74.84  
Daniel Silveira    3063       195      70       6          10         73.58  
Charles Faiga      3333       219      84       5          8          72.28  
Wayne M            3194       195      75       3          10         72.22  
Dan                3896       248      96       15         9          72.09  
Chris              3753       305      119      31         17         71.93  
Luke               3256       201      80       2          2          71.53  
Pure.Krome         5664       366      146      16         14         71.48  
Ethan              3743       234      95       3          3          71.12  

People most affected by the change with 10k rep or more

User             Reputation QUpvotes AUpvotes QDownvotes ADownvotes Penalty
-----------------------------------------------------------------------------
Edward Tanguay   11743      1019     34       21         1          96.77   
Claudiu          10447      520      427      45         33         54.91   
Kip              10647      404      471      6          36         46.17   
Allain Lalonde   10869      413      501      7          39         45.19   
DR               10106      387      503      8          12         43.48   
Jason Baker      24839      963      1261     19         55         43.30   
Thomas Owens     22704      874      1277     64         96         40.63   
Robert Gould     15586      508      852      29         59         37.35   
Simucal          17628      512      1017     4          14         33.49   
Cheeso           11969      301      724      6          50         29.37   
Jeff Atwood      16648      334      822      7          35         28.89   
e-satis          12295      339      838      16         44         28.80   
Mark Harrison    15932      415      1028     21         62         28.76   
Vilx-            10589      278      723      4          53         27.77   
Andreas Grech    12313      267      699      9          23         27.64   
Jason S          12789      302      808      2          38         27.21   
dbr              15992      348      1017     5          55         25.49   
lomaxx           11745      222      656      4          53         25.28   
Michael Stum     25669      543      1610     5          98         25.22   
Mason Wheeler    11973      266      817      13         16         24.56   
Matthew Scharley 12702      263      896      4          44         22.69   
Roddy            10264      198      684      2          28         22.45   
Daok             17405      330      1195     11         123        21.64   
Sam Saffron      13794      246      941      6          23         20.72   
Soviut           10254      195      769      6          54         20.23   
John Nolan       10333      164      650      0          25         20.15   
RSolberg         10781      166      664      6          24         20.00   
Justin Standard  11289      100      405      3          9          19.80   
nickf            32868      606      2462     13         141        19.75   
splattne         18670      294      1215     8          34         19.48   
Ash              10692      183      775      2          23         19.10   
rlbond           10810      200      848      3          50         19.08   
Naveen           12801      211      918      2          46         18.69   
Andy White       12077      201      881      3          36         18.58   
Adam Davis       28375      505      2228     23         80         18.48   
FlySwat          23068      467      2074     32         281        18.38   
Dan Herbert      12125      179      814      3          24         18.03   
oxbow_lakes      14562      232      1096     14         54         17.47   
Yuval A          14984      241      1142     22         102        17.43   
Mike Stone       10427      146      696      3          52         17.34   
Robert S.        11168      130      651      6          43         16.65   
Roger Lipscombe  10289      142      715      0          19         16.57   
Kibbee           17928      254      1281     12         109        16.55   
Juliet           13873      188      975      12         29         16.17   
Jim McKeeth      12614      166      861      5          13         16.16   
AraK             13047      193      1026     4          37         15.83   
Rob Wells        12405      171      915      19         55         15.75   
Brian R. Bondy   36633      602      3296     19         125        15.44   
Unknown          11385      161      882      12         58         15.44   
Keith            15159      214      1204     6          65         15.09   

People least affected by the change with 10k or more

User             Reputation QUpvotes AUpvotes QDownvotes ADownvotes Penalty
-----------------------------------------------------------------------------
R. Bemrose       11858      0        1024     0          70         0.00    
Lou Franco       13064      0        1267     0          53         0.00    
bobince          34163      0        3147     0          99         0.00    
Charles Bailey   22539      0        2181     0          20         0.00    
Chuck            20818      0        1815     0          44         0.00    
Guffa            31067      0        2829     0          97         0.00    
Eric             13203      0        1166     0          34         0.00    
Vinay Sajip      14838      0        978      0          32         0.00    
gimel            12090      0        1082     0          59         0.00    
Brian Agnew      29391      0        2763     0          69         0.00    
mquander         11771      0        1170     0          42         0.00    
Richard          11056      0        994      0          40         0.00    
Developer Art    16304      0        1721     0          67         0.00    
HLGEM            12414      0        1092     0          77         0.00    
Brian            10111      0        920      0          7          0.00    
Stephan202       12783      0        1339     0          29         0.00    
JP               18107      0        1609     0          77         0.00    
jitter           11701      0        941      0          41         0.00    
Brian            22019      0        1897     0          82         0.00    
ShuggyCoUk       12030      0        926      0          12         0.00    
CMS              50297      0        4820     0          129        0.00    
Jörg W Mittag    10832      0        954      0          37         0.00    
Leon Timmermans  12303      0        1195     0          33         0.00    
Bombe            11681      0        1086     0          103        0.00    
Welbog           15006      0        1381     0          57         0.00    
Michael          17262      0        1572     0          28         0.00    
Eric Lippert     15599      0        1717     0          18         0.00    
mjv              10850      0        943      0          26         0.00    
Pascal Thivent   16417      1        1355     0          29         0.07    
Matthew Flaschen 10703      1        1258     0          69         0.08    
John Feminella   17190      2        2077     0          47         0.10    
Rob              10955      1        1011     0          43         0.10    
JaredPar         67345      10       9261     1          251        0.11    
caf              10417      1        888      0          12         0.11    
Michael Burr     36206      4        3541     0          52         0.11    
Freddy Rios      10982      1        829      0          37         0.12    
Joachim Sauer    15552      2        1576     0          41         0.13    
David Dorward    16555      2        1573     0          124        0.13    
dirkgently       19512      3        2210     1          90         0.14    
Mehrdad Afshari  60582      12       8281     1          198        0.14    
Dav              11852      2        1365     0          38         0.15    
Shog9            23235      3        1967     1          89         0.15    
MSalters         14991      2        1261     0          60         0.16    
Sinan Ünür       22494      5        2662     1          79         0.19    
John Saunders    28844      6        2953     3          221        0.20    
unwind           28276      6        2841     0          101        0.21    
Daniel Roseman   10868      2        836      0          21         0.24    
SQLMenace        11960      3        1181     1          90         0.25    
sth              16989      4        1547     0          32         0.26    
Fredrik Mörk     22548      6        2169     0          32         0.28    

As you can see, it does punish some problem users quite effectively.

But ... it also punishes some non problem users quite harshly, dbr, michael stum, you and me would be hit with a 12% reputation penalty. Which is odd, cause overall I think the questions I ask on SO are pretty good and if anything we need better questions on SO.

16
  • 3
    What does the penalty column represent? Percent of votes affected? Percent of reputation lost? Would some people actually get crushed entirely? Yikes! Commented Mar 19, 2010 at 5:43
  • 8
    thanks for putting this together -- it actually looks fine to me. At least, the effect is what I had in mind. Commented Mar 19, 2010 at 5:43
  • 3
    @dmckee that's percent of votes that are question votes. If it's 100% you would lose half your rep, roughly. Commented Mar 19, 2010 at 5:44
  • Thanks for putting numbers to the issue. As someone affected by it, does it really feel like punishment or is there some "questions were slightly overvalued, and this corrects that"? Either way, also as someone significantly affected, do you think it would be an improvement to SO?
    – Gnome
    Commented Mar 19, 2010 at 5:49
  • 1
    So now I'm a "problem user" ? I have no problem with the change, I will probablly still use SO as I do today. For me, it's an invaluable tool for finding answers, and I rarely have the time to post answers (very rarely there's a question that my skills alone can help. Usually there are dozens of people ready to write answers, so I don't feel this is hurting/leeching the site).
    – ripper234
    Commented Mar 19, 2010 at 8:04
  • 1
    -1 for incorrect figures - your calculation ignores bounties/downvotes. The later are negligible for most users, but the former are not.
    – Casebash
    Commented Mar 19, 2010 at 8:13
  • Glad I'm not in that top list! I think I'm going to take quite a hit for this as well, though I don't think I've often asked questions that weren't for a real problem I needed to solve.
    – Benjol
    Commented Mar 19, 2010 at 8:21
  • 1
    Wow - I'm on a list higher than Eric Lippert, it must be wrong! ;)
    – ShuggyCoUk
    Commented Mar 19, 2010 at 10:10
  • Considering I'm facing a back-of-the-envelope 60% penalty (silly bounties), I wouldn't really call 12% that harsh
    – Casebash
    Commented Mar 19, 2010 at 10:14
  • 2
    The data is about 6 months old, isn't it? Commented Mar 19, 2010 at 11:07
  • @Daniel not 6 months, maybe 3 or so. have not gotten around to loading latest, but I suspect the trends and affected users still hold
    – waffles
    Commented Mar 19, 2010 at 11:42
  • 2
    It totally ignores that upvotes != reputation. If you get many votes on a day that doesn't translate into equally much reputation. Also community wiki posts (which are often highly upvoted), are probably not accounted for correctly.
    – sth
    Commented Mar 19, 2010 at 12:51
  • Wow, I'm number three among 10k+ users? Even still, it looks like I only stand to lose 1010 rep, which keeps me over 10k. Now, on Meta, I stand to lose a lot more...
    – Kip
    Commented Mar 19, 2010 at 13:49
  • OUCH! Taking nearly a 20% hit!
    – RSolberg
    Commented Mar 19, 2010 at 15:29
  • How did you calculate this?
    – unlobito
    Commented Mar 22, 2010 at 2:18
32

This makes a lot of sense; I've seen a user with over 7000 rep, with less then 30 upvoted answers. The majority of this rep (6k) is from questions. Many of them showing a complete lack of effort on the users part.

1
  • 9
    If you do lots of self-answers and answering of old questions then you tend to get a lot of answers without upvotes
    – Casebash
    Commented Mar 20, 2010 at 0:20
15

Could this not apply to meta?

As questions have different meaning here, I don't think it should apply.
A good bug report or feature request deserves the +10 per vote

For the rest of the sites, I agree.

7
  • 5
    i agree. on SO, questions will come naturally, but answers need to be encouraged. on Meta, it's kind of the opposite: good feature request ideas or good bug finds should be rewarded, while the "answers" are just discussion that will happen naturally
    – Kip
    Commented Mar 19, 2010 at 13:54
  • 1
    (i'm really miffed that i'm probably going to go below 10k on meta)
    – Kip
    Commented Mar 19, 2010 at 13:55
  • I'm bummed out that I'm days away from 10k on Meta. It was in sight. Commented Mar 19, 2010 at 16:52
  • 1
    +1 On meta, upvotes on answers ("Yes, I agree") worth less than questions.
    – mmx
    Commented Mar 19, 2010 at 17:28
  • 2
    I disagree; good discussions and feature requests on meta tend to be rewarded with a lot of upvotes, so applying the same change here should not have a calamatous effect on reputation. (Keep in mind there are still lots of n00b questions here too, which do deserve to be affected the same as on SO.)
    – Ether
    Commented Mar 19, 2010 at 17:58
  • 5
    @lance we will keep +10 / +10 on meta Commented Mar 20, 2010 at 9:30
  • @Jeff, thanks, now I'll finally get to see those 10k tools. I don't mind the 5pts/question, just didn't like the retro. Commented Mar 20, 2010 at 20:52
11

By my count, I'm about to lose 710 rep on SO here, which knocks me down a level in terms of what I can do on the site (bye-bye, editing!). I, too, laboured under the impression that asking good questions -- providing the impetus to other users to provide answers -- was a desirable thing. I can't say that I'm against the change as a whole, but I'm not a fan of the recalculation that's coming with it.

I've asked a few questions that have tons of views and that people seem to find a real use for, over and over again. How is it that those are suddenly less valuable?

For what it's worth -- and given that there's been a blog posting about it already I'd say that it's not worth a lot -- I don't like this.

5
  • 5
    Asking question is desirable and looking through some of yours, you've done a decent job (much better than me, I suck at asking), but consider this as "they were overvalued before" rather than "suddenly less valuable now". If it's any consolation, you might not miss editing others' posts, since you haven't done it in over 3 months (based on your activity tab; there's one retag, but you'll still be able to do that), and -710 puts you about 100 shy of 2k, so you could make that back up in a single day if you wanted it back.
    – Gnome
    Commented Mar 19, 2010 at 14:02
  • 1
    How are you managing this super-fast data mining? Is there a toolset I should be aware of? That's pretty cool stuff!
    – Chris R
    Commented Mar 19, 2010 at 16:30
  • As he said, Gnome is just looking at your activity tab. Are you not aware of it? stackoverflow.com/users/23309?tab=activity#tab-top
    – raven
    Commented Mar 19, 2010 at 18:07
  • There is: you can work with the public data dumps and various sites run by ordinary users (using the data dumps), but I've not bothered to get into that (yet) and just used your profile as raven mentioned.
    – Gnome
    Commented Mar 19, 2010 at 20:15
  • Wow, that was pretty quick. I did much the same thing, but doing so from the activity tab seemed painfully slow. I was hoping there was a generic toolkit for making that sort of query.
    – Chris R
    Commented Mar 22, 2010 at 14:51
11

Should the weight of question upvotes be reduced?

Yes.

10

Why is everyone so sore? It's an arbitrary number - it's as arbitrary as it has always been!

It could be in the millions on average, or in the hundreds on average.

2
  • 4
    That said, I'm losing up to 2.5k rep. Someone call the waaambulance!
    – Pollyanna
    Commented Mar 19, 2010 at 20:03
  • 3
    You know, these rep numbers mean something to people. I'm sure there is a range from mild curiosity to hourly OCD rep checking. Commented Dec 6, 2011 at 23:43
8

I'm in favour of this simply because it means we will finally see a global rep recalc. I'm often frustrated when I delete bad questions with upvoted answers and know that those users will get to keep their rep, because recalcs aren't done often enough (or ever) for most users.

PS. Thank you Jeff for the much more positive tone of conversation on this thread vs. that other one.

7

OK. I was looking forward to the increased downvote power, but I can deal with this as an alternative.

2
  • 3
    that's right, this is another (less negative IMO) way of achieving a similar result. Commented Mar 19, 2010 at 2:35
  • 2
    @Jeff: If you also take away pity upvotes (that is, upvotes on posts with net < 0 score) being worth +10, then the "problem users" should be taken care of almost entirely.
    – Ether
    Commented Mar 19, 2010 at 17:59
3

Please explain to me why I should be penalised 10% of my reputation having asked 1 question to every 8 answers?

I stupidly thought that asking questions that also happen to be useful to other people is a good thing.

This is a sledge hammer solution to a situation that needs a scalpel.

13
  • 7
    asking questions is useful, but not as useful as answering. That's the dynamic we're trying to capture. And yes, it involves a bit of sacrifice from every user (including myself) except those who never asked any questions. Freaks! :) Commented Mar 19, 2010 at 12:22
  • 3
    @Jeff, being a moderator it doesn't really affect you (in regards to abilities at least)
    – juan
    Commented Mar 19, 2010 at 12:40
  • 8
    @ashh: sorry if this sounds harsh, but you care way too much about rep. You're at about 15k now; -10% would put you at 13.5k. This means that a) you would not lose any privileges whatsoever, and b) judging by your rep graph you would recoup the loss in 10-14 days. In my humble opinion, that's not a big deal. Especially considering the fact that everyone's rep would get recalculated, not just yours, so you would probably even stay on the 9th page in the Users tab rather than drop to the 10th.
    – ЯegDwight
    Commented Mar 19, 2010 at 12:41
  • 2
    It's not punishment, it's just a different way of looking at reputation. As a bonus to the answers on a question, one still gets 5 rep for each upvote!
    – Arjan
    Commented Mar 19, 2010 at 12:45
  • 2
    @Jeff you are not sacrificing anything.
    – Kip
    Commented Mar 19, 2010 at 14:01
  • 2
    @Jeff, questions are what draws googlers to SO Commented Mar 19, 2010 at 16:53
  • @RegDwight, doesn't sound harsh at all because you have misunderstood my concerns. The point is that only people who have actually asked questions are affected, and the more questions asked, the bigger the effect. It is the reduced recognition of the effort I have put into my questions that I am/was concerned about, not any loss of privileges. But I'm sure I'll get over it.
    – user135186
    Commented Mar 22, 2010 at 5:31
  • @Jeff, it's your (very useful) site so who am I to complain. But if you're looking to stop problem users such as Shore and Gold, I'm not sure why it must be applied to all users. In reality it smells like a philosophical change, so why not just say so, rather than pointing to problem users as a reason.
    – user135186
    Commented Mar 22, 2010 at 5:41
  • @ashh, I understand your point perfectly well, but you seem to miss that it's not actually your point, but rather the whole point of the rep recalc as such. You say that "only people who have actually asked questions are affected, and the more questions asked, the bigger the effect", but you miss that that's precisely the intention. The "reduced recognition of the effort [you] put into [your] question" is not an accidental byproduct of the change, it's the sole reason for the change. Your questions are valued less, thus your rep is decreased; not the other way round.
    – ЯegDwight
    Commented Mar 22, 2010 at 11:15
  • @Lance: and answers are what keep them comming back.
    – AnonJr
    Commented Mar 22, 2010 at 13:33
  • @RegDwight, Your essays on the (supposed) shortcomings in my points wonderfully demonstrate that, in fact, you care far more about reputation than I ever could or will. Your "humble opinion" indeed.
    – user135186
    Commented Mar 23, 2010 at 7:15
  • @ashh: I am not the one complaining here about losing a tiny amount of rep using pompous wording. I am merely addressing such a complaint of yours. If you don't wish your complaints to be addressed, there is no point in posting them.
    – ЯegDwight
    Commented Mar 23, 2010 at 10:46
  • @regdwight glass is half full -- your answers are worth more than they were before. Particularly since accepts and bounties -- both answer related functions -- are 100% immune to rep cap now. Commented Mar 28, 2010 at 5:50
3

I think I agree with the idea of tuning the weight of questions versus answers, but I really disagree with making it retroactive.


(Answer withdrawn)

2
2

Would the retroactive re-calc take into account the new cap?

For example, if one day you had +20 votes on questions, and then +10 on answers, would you still get 200 for the day?

7
  • I don't see why it wouldn't.
    – fretje
    Commented Mar 19, 2010 at 12:59
  • I don't really know how the recalc process is @fre
    – juan
    Commented Mar 19, 2010 at 14:42
  • 6
    It kind of "simulates" your whole history again, from the moment you registered as a user, so yes, the rep cap is taken into consideration.
    – fretje
    Commented Mar 19, 2010 at 15:02
  • We'll probably see a decrease in Mortarboard badges being awarded for a while, as some of those badges will retroactively become invalid (and so the next time a user earns one, it takes the place of the one that shouldn't have been earned previously). I can't think of any other badges that will be affected, though - no others use reputation as opposed to votes in their calculations. Then again, perhaps not, as most MB recipients are reaching the cap via answers rather than questions.
    – Ether
    Commented Mar 19, 2010 at 18:11
  • That is not true @Ether, badges are never removed when you recalc.
    – juan
    Commented Mar 19, 2010 at 18:15
  • @Downvoter: I realize they aren't removed, which is why I said we might see a decrease in new badges for a while: some MB badges will be rendered "invalid", which means that the next time you earn a real one, it will take the place of the illegitimate one you already had.
    – Ether
    Commented Mar 19, 2010 at 18:27
  • oh, I get it, i think you're right. Jeff said the process returns "how many badges you should have" and compares it with what you have, and awards them. I stand corrected. @eth
    – juan
    Commented Mar 19, 2010 at 18:31
2

The value of asking the right questions for me is as valuable than the right answers, more valuable, actually. You can't attach an answer to nutt'n. Thank goodness for the folks who took the time to ask the same question I had about something or other.

In the interest of full disclosure, the majority of my rep comes from asking questions, as well as "residual" rep increases. That puts me in a different class from the Jon Skeets of the world obviously, but the questions are at least as much what puts SO on the (search engine) map as the answers. I try to imagine what SO means to devs that have no interest whatsoever in participating in SO and just want answers via Google or Bing.

1
  • Well, I got just one upvote, but apparently it was from the person with the juice to do something about it. Cool! Commented Nov 14, 2019 at 16:05
1

This is a great idea; could it also be changed to make a downvote on a question -5 so that only questions with a net + award reputation?

1

From my understanding of this (& the related blog post) it's partly to motivate increased voting for answers, rather than questions.
Isn't this contradictory to the aims of the electorate badge - which seems to encourage voting for questions?

1
  • 2
    The electorate badge is a benefit to the voter, while the rep is a benefit for the votee (for asking a question someone else thought worthy of upvoting). I don't see a clash.
    – Gnome
    Commented Mar 19, 2010 at 9:11
1

First off, I understand two things:

  1. Reputation is a meaningless number.
  2. This isn't necessarily directed at me (or so I hope... Jeff did say that the list that I was fifth on looked about right)

That said, it's really difficult for me to not feel penalized when I lose 20% of my reputation and probably get knocked out of the top 100. I like stackoverflow because it's a place where I really feel that learning is encouraged. And I personally love learning. I'll probably still get a big portion of my upvotes from asking questions.

I understand the decision. However, it's difficult not to feel discouraged though as my favorite part of using stack overflow will now essentially be considered to be half as important as the other part.

At least give me some kind of consolation here. Maybe a new gold badge of some kind? A top 20 question asker list? Heck, I'd even go for an increase to the bonus I get for accepting answers.

EDIT: I suppose the rep recalc didn't have as huge an effect as I thought it would. I'm still not convinced that this is the best course of action, but I can live with it I suppose.

1
  • Sure, I'd like compensation too! How about you give us an actual gold badge we can sell Jeff?
    – juan
    Commented Mar 22, 2010 at 13:06
1

Ok, I know it's a little bit late, I just wanted to say what I think : The problem is that this will discourage old users of SO to ask questions. By "old users" I mean people that had eventually red the FAQ and that are possibly interested in earning rep thous will take care of the way they write their questions. But it will not discourage any one with no reputation at all to ask a i-have-nothing-to-loose question. Thous this will possibly degrade the overall quality of the questions, and since I don't agree that much that A are twice as valuable as a Q, I don't think it is such a good idea. I guess we'll see.

0

Ya, go ahead. I'll lose a lot BUT so will Jon Skeet! Muh-ah-ah-ah!

12
  • Nope. He hasn't received that many votes on his questions.
    – waiwai933
    Commented Mar 19, 2010 at 2:55
  • George, I believe existing posts wouldn't lose any reps, not sure with reps recalc though. @Jeff, any ideas on reps recalc?
    – YOU
    Commented Mar 19, 2010 at 2:55
  • 1
    Recalc means existing stuff gets recalculated. And I would be content witnessing Jon Skeet lose even 1 point! Commented Mar 19, 2010 at 2:57
  • I'll only lose on order of 100 points. I'd has lost more the other way. Commented Mar 19, 2010 at 2:57
  • Your definition of Recalc might be correct, but I think its still depends on Jeff's implementation.
    – YOU
    Commented Mar 19, 2010 at 2:59
  • 3
    It says above that the change would be applied retroactively. Commented Mar 19, 2010 at 3:00
  • @dmckee, I see, so it effects past posts, I think I am going to see some people came here, and claim about "Hey, I lost half of my reps, Fix Please" :P
    – YOU
    Commented Mar 19, 2010 at 3:03
  • Exactly. Oh well :) Commented Mar 19, 2010 at 3:04
  • 6
    @S.Mark: The bad part is, some of those folks will be innocent. That is the big advantage that the increased downvote would have had: leaving the question heavy users who haven't accumulated a lot of ill will lightly touched. Commented Mar 19, 2010 at 3:18
  • 2
    skeet will probably be fine--if he's easily hitting the rep limit every day, any rep he loses on question votes will likely be made up for with answer votes that used to be beyond the limit, so that he'll still hit the same rep cap most days.
    – Kip
    Commented Mar 19, 2010 at 13:52
  • @S.Mark That would be affects. Effects is usually a noun. It can be a verb, but not in the way you're using it. Your construction would be valid if the recalc was creating posts. /grammarnazi
    – waiwai933
    Commented Mar 20, 2010 at 1:38
  • let the Skeet rep hatefest commence...Now! Commented Dec 6, 2011 at 23:41

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .