11

Community's auto-protection of questions is implemented such that quick upvotes to answers from new users make those answers cease to trigger the checks for automatically making a question protected.

Recent case when I saw this happen in a popular question, when just a funny picture got 2 quick upvotes made me recall The Trouble With Popularity and wonder how often does this happen:

it's too addictive and too easy, and in the absence of any moderation, the community would do nothing but add and upvote the easy, fun stuff...

Can we have stats on how many questions escape (or delay getting) protection because of upvotes gained in the checked question - compared to questions that get protected? Preferably per site, to account for differences in protection criteria and in voting patterns at different sites.

Improve this question
16
  • FWIW I was thinking about making sort of rough estimate using data explorer but dropped that - since it's about troublesome posts, such results would be substantially skewed because of deletions (after all, the very answer that tripped my curiosity is deleted by moderator)
    – gnat
    Commented Aug 25, 2014 at 20:40
  • 1
    I've attempted to unpack the first sentence to be clearer - it confused me at first, I thought you were talking about the reasons for its implementation, and I wasn't sure which checks or protection you were talking about. Please check my edit to make sure it still says what you intend, and doesn't contain factual errors or etc.
    – doppelgreener
    Commented Aug 25, 2014 at 23:09
  • @doppelgreener thanks! regarding reasons for implementing it that way, a while ago I gave it a thought and, as far as I can tell, current approach has a benefit that it catches very clear cut troublesome cases and avoids slippery ones (one can argue that it's slippery to count for protection answers from users who already gained rep prior to check)
    – gnat
    Commented Aug 26, 2014 at 4:21
  • even more blatant case of broken community protection: this answer from new user has got to -4 but then, received senseless upvote so that user rep has got to +11 (if anyone wonders how garbage gets upvoted, take a look at The Trouble With Popularity)
    – gnat
    Commented Oct 31, 2014 at 10:19
  • This cuts both ways... Consider this question - automatically protected at the 3rd new-user answer, though all three answers (and authors) quickly exceeded the threshold for protection.
    – Shog9 Mod
    Commented Nov 10, 2014 at 16:36
  • @Shog9 I see, interesting. Probably stats asked about here could be complemented by those related to posts and users who managed to trigger protection, like how many "triggering" answers and their authors tend to go well/bad in the longer run. Yeah I am certainly interested in learning another part of the picture... I'll probably ask another question about this kind of stats
    – gnat
    Commented Nov 10, 2014 at 16:47
  • @Shog9 now look at this example, check the vote split. 8 upvotes from hot list lemmings against 11 votes down from regulars having 125+ rep. Answerer reputation is solely from this answer and it's whopping 67, more than enough to "scare away" community protection. Is this how it is supposed to work?
    – gnat
    Commented Nov 27, 2014 at 14:36
  • 1
    That is a terrible question, @gnat.
    – Shog9 Mod
    Commented Dec 2, 2014 at 23:39
  • @Shog9 agree, thanks for closing it (my own vote has long expired). FWIW advertising it to sidebar audience is probably your fault: At smaller sites, penalize hot questions having 3-4 close votes
    – gnat
    Commented Dec 3, 2014 at 6:27
  • 1
    @Shog9 ...you might also want to take a look at TWP meta follow-up to your closure: What are “opinion-based” questions?
    – gnat
    Commented Dec 3, 2014 at 9:05
  • 1
    If it takes y'all a week to close or fix questions, @gnat, that's a deeper problem than can be fixed with superficial changes to how questions are displayed.
    – Shog9 Mod
    Commented Dec 3, 2014 at 18:20
  • 1
    For those who can't see it, gnat's example was rightly deleted by the community within six hours on a national holiday. In other words, that part of the system worked. As for the rest, there is disagreement about the suitability of the question; I sure hope we don't evaluate a question on the basis of one offensive, deleted answer.
    – Monica Cellio
    Commented Dec 3, 2014 at 18:59
  • @Shog9 advertizing bad questions to sidebar makes it harder to deal with these, but I believe that community will be capable to handle them anyway (at least that's how it went for last few years at Programmers and Workplace)
    – gnat
    Commented Dec 3, 2014 at 19:40
  • ...and who told you that goal is to close questions. Without clueless answers and upvotes brought by hot list visitors, salvaging edits would become much more viable option for many of them. It's that simple
    – gnat
    Commented Dec 4, 2014 at 11:49
  • If tricky questions don't get automatically protected, just flag and ask a mod to protect them!
    – curiousdannii
    Commented Jun 2, 2015 at 0:36

2 Answers 2

Reset to default
3

You indicated elsewhere that you'd be comfortable with some pretty loose accuracy here...

So here you go. I counted all questions that got at least 1, 3 or 5 answers in the past 365 days from users who earned no reputation on the site prior to that answer's creation and are not protected and were not protected by Community during this same time period.

Then I ran this on every non-meta site except Stack Overflow:

5 Answers 3 Answers 1 Answer Site Name
--------- --------- -------- ---------
53        588       19573    Super User
10        123       9009     Server Fault
0         2         47       Stack Apps
20        151       1953     Cooking
14        103       2163     Home Improvement
4         27        1680     Game Developers
32        277       5907     Gaming
0         21        2567     GIS
3         43        1241     Photography
1         18        3040     Statistical Analysis
5         62        1828     Web Apps
2         21        1107     Webmasters
81        427       7301     Apple
0         0         168      Theoretical Computer Science
45        269       6212     English Language and Usage
11        41        978      Personal Finance and Money
4         39        983      Role-playing Games
40        442       16026    Ubuntu
7         134       6871     Unix and Linux
1         35        3383     WordPress
6         48        843      Bicycles
1         21        1970     Programmers
27        259       5203     Android Enthusiasts
1         10        524      Board and Card Games
3         36        3355     Physics
1         10        295      Homebrew
13        88        2328     IT Security
2         26        509      Writers
2         50        3600     Electronics and Robotics
9         80        2102     Graphic Design
1         38        2671     Database Administrators
24        226       3595     Science Fiction
2         19        309      Area 51 Discussions
2         31        2006     Code Review
69        176       767      Code Golf
0         9         448      Quantitative Finance
1         12        339      Project Management
0         8         509      Skeptics
1         6         2144     Drupal Answers
3         21        518      Fitness and Nutrition
4         52        933      Motor Vehicle Maintenance and Repair
13        62        668      Parenting
1         20        2567     SharePoint
7         47        1229     Musical Practice and Performance
1         38        589      Software Quality Assurance and Testing
0         3         510      Jewish Life and Learning
0         12        314      Astronomy
4         24        602      German Language and Usage
0         3         417      Japanese Language and Usage
1         4         337      Gardening and Landscaping
1         18        559      Philosophy
3         23        263      Personal Productivity
6         73        1390     Travel
1         9         454      Cryptography
0         1         408      Signal Processing
0         19        440      French Language and Usage
0         34        1045     Christianity
1         19        581      Bitcoin
2         5         239      Linguistics
0         23        550      Biblical Hermeneutics
2         13        317      Economics
5         34        533      History
2         6         171      LEGO®
4         15        403      Spanish Language and Usage
1         2         228      Computational Science
8         100       1653     Movies
2         36        512      Chinese Language and Usage
0         4         721      Biology
0         6         136      Poker
0         4         490      Mathematica
0         9         318      Cognitive Sciences
9         33        307      The Great Outdoors
4         18        176      Martial Arts
0         3         339      Sports
11        63        1327     Academia
2         20        696      Computer Science
5         63        870      The Workplace
0         26        450      Windows Phone
0         13        1133     Chemistry
1         20        369      Chess
1         56        1712     Raspberry Pi
1         21        273      Russian Language and Usage
0         31        818      Islam
1         17        1902     Salesforce
1         7         161      Patents
11        80        1446     User Experience
0         1         97       Genealogy and Family History
0         2         240      Robotics
0         0         194      ExpressionEngine
4         11        223      Politics
1         6         276      Audio-Video Production
2         43        1066     Anime and Manga
3         61        2836     Magento
7         57        1971     English Language Learners
2         6         116      Sustainable Living
0         0         50       Tridion Stack Exchange
0         1         200      Reverse Engineering Stack Exchange
3         43        1042     Arduino Stack Exchange
0         21        854      Network Engineering Stack Exchange
0         6         284      Open Data Stack Exchange
1         16        171      Freelancing Stack Exchange
0         4         924      Blender Stack Exchange
0         4         883      MathOverflow
2         8         377      Space Exploration Stack Exchange
1         8         469      Sound Design Stack Exchange
1         6         295      Tor Stack Exchange
2         19        415      Pets Stack Exchange
1         9         189      Amateur Radio Stack Exchange
0         3         145      Italian Language Stack Exchange
4         31        707      Aviation Stack Exchange
2         7         117      Ebooks Stack Exchange
3         14        110      Beer Stack Exchange
1         22        1019     Software Recommendations Stack Exchange
1         3         215      Expatriates Stack Exchange
6         19        172      Mathematics Educators Stack Exchange
1         8         181      Earth Science Stack Exchange
0         5         385      Joomla Stack Exchange
3         27        505      Data Science Stack Exchange
24        127       1002     Puzzling Stack Exchange
1         1         33       Embedded Systems Stack Exchange
1         3         183      Craft CMS Stack Exchange
1         16        327      Buddhism Stack Exchange
5         32        535      Hinduism Stack Exchange
1         2         33       Moderators Stack Exchange
5         27        374      Startups Stack Exchange
80        221       1306     Worldbuilding Stack Exchange
0         1         418      Emacs Stack Exchange
0         3         117      History of Science and Mathematics Stack Exchange
28        97        516      Lifehacks Stack Exchange
0         16        330      Engineering Stack Exchange
1         13        155      Coffee Stack Exchange
1         6         266      Vi and Vim Stack Exchange
3         11        188      Music Fans Stack Exchange
4         14        220      Woodworking Stack Exchange
0         6         280      CiviCRM Stack Exchange
1         13        362      Health Stack Exchange
3         30        636      Русский язык
0         5         101      Mythology Stack Exchange
1         8         296      Law Stack Exchange
1         18        158      Open Source Stack Exchange
2         11        452      elementary OS Stack Exchange
0         8         146      Portuguese Language Stack Exchange
0         4         96       Computer Graphics Stack Exchange
0         7         176      Hardware Recommendations Stack Exchange

Note that 5 answers from new users in 24 hours is the threshold on most sites; on Programmers, ELL and The Workplace, it's 3 in 24 hours.

Since the numbers listed represent the number of questions to receive 5, 3 or 1 answer(s) in 365 days, these should be considered a worst-case estimate - some or all of the questions counted would not actually have been eligible for auto-protection even if the system did consider reputation at the time of posting.

Improve this answer
1
  • thanks! would be great to see how these numbers compare to questions that were protected
    – gnat
    Commented Feb 27, 2016 at 17:38
9

Observing for last 5-6 months how protection works (or more precisely fails to work) leaves me quite disappointed. Unless stats show that my observations somehow miss the big picture, I would say that it is probably broken, at least at the sites I am active at (Programmers and Workplace).

As an example, take a look at recent question at TWP. It has quickly got three troublesome answers (1, 2, 3) - all from new users having less than 10 on-site reputation, all with negative score (one is even already deleted).

Still, protection didn't kick in - simply because when third "answer" was posted, one of two prior ones has picked a random upvote that pushed answerer's reputation to 11 (soon it has fallen back below, because downvotes to that answer continued coming in, but it was too late). It's not the first time I see this mind you, only maybe the most clear cut one.

You know, protection criteria depends on reputation, which is driven by voting. Voting, in turn, is somewhat random... and that's fine! Voting may be weird sometimes, but in the long run, on average, in typical cases, it tends to be reasonable - that's the way how whole system works and so far it looks sufficiently reliable, despite infrequent oscillation.

If protection criteria was as reliable as the rest of the system, I would be fine with that. Thing is though, it is (to a large part) based on voting in questions intended to be protected - in popular ones, and voting in these often is anomalous... softly speaking. As far as I can see, that makes it way too fragile.

Improve this answer
5
  • 5
    Oh the irony that this question itself is now protected :)
    – Stijn de Witt
    Commented Jul 20, 2015 at 14:05
  • The question asked for stats. How does this answer the question?
    – Monica Cellio
    Commented Feb 9, 2016 at 16:20
  • @MonicaCellio note the answer was posted more than two months after the question was asked. In the absence of requested stats, all I could do was to collect observations and provide a summary of what I have seen so far
    – gnat
    Commented Feb 9, 2016 at 16:24
  • 1
    Sure, time has passed. This just strikes me as a continuation of the opinions in your question rather than an actual answer, is all. Especially on a self-answer it seems like you should either answer the question or not answer (since you can already use the question).
    – Monica Cellio
    Commented Feb 9, 2016 at 16:59
  • @MonicaCellio you have a point, originally I considered making it sort of update to the question. I drafted it in, reviewed how it would look like and decided that it would better fit to the answer. "I would say that it is probably broken... As far as I can see, that makes it way too fragile." - I did not intend to put stuff like that into question when I asked it, and I still don't intend to
    – gnat
    Commented Feb 9, 2016 at 17:03

You must log in to answer this question.

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