48

When I was going through my suggested edit history, I found the following stat oddity:

enter image description here

As you can see, I have 1,707 suggested edits in total (which is correct), but according to my reviewer stats, I only have 933!


The odder thing is that when I look at a review audit, that number shows up correctly!

enter image description here

What's going on?


EDIT: This appears to not be a client-caching issue; this is the response from the server for this suggested edit (the field of interest being "moreInstructions":

{
    "isUnavailable":true,
    "instructions":"<div class=\"review-status\">\r\n                    <strong>\r\n<span style=\"color: #0A0;\">Approved</span> <span title=\"2015-02-27 00:28:45Z\" class=\"relativetime\">15 hours ago</span>:                    </strong>\r\n    </div>    \r\n        <div class=\"review-results\">\r\n            <a href=\"/users/899126/chrisforrence\" class=\"comment-user\">ChrisForrence</a> reviewed this <span title=\"2015-02-27 00:28:45Z\" class=\"relativetime\">15 hours ago</span>: <b>Approve</b>\r\n        </div>\r\n        <div class=\"review-results\">\r\n            <a href=\"/users/3208640/farhang-amary-ferh%c4%99g\" class=\"comment-user\">Farhang Amary &#39;Ferhęg&#39;</a> reviewed this <span title=\"2015-02-27 00:23:25Z\" class=\"relativetime\">15 hours ago</span>: <b>Reject</b>\r\n        </div>\r\n                    <div class=\"rejection-reason cool\">This edit does not make the post even a little bit easier to read, easier to find, more accurate or more accessible. Changes are either completely superfluous or actively harm readability.</div>\r\n        <div class=\"review-results\">\r\n            <a href=\"/users/609087/keety\" class=\"comment-user\">keety</a> reviewed this <span title=\"2015-02-27 00:15:52Z\" class=\"relativetime\">15 hours ago</span>: <b>Approve</b>\r\n        </div>\r\n        <div class=\"review-results\">\r\n            <a href=\"/users/1907093/jonathan-crowe\" class=\"comment-user\">Jonathan Crowe</a> reviewed this <span title=\"2015-02-27 00:05:09Z\" class=\"relativetime\">15 hours ago</span>: <b>Approve</b>\r\n        </div>",
    "moreInstructions":"<p><strong>Reviewer Stats</strong></p>\r\n    <ul>\r\n\r\n\r\n    <li><a href=\"https://stackoverflow.com/users/899126/chrisforrence\">ChrisForrence</a> has approved 960 edit suggestions and rejected 5 edit suggestions</li>\r\n\r\n    <li><a href=\"https://stackoverflow.com/users/3208640/farhang-amary-ferh%c4%99g\">Farhang Amary &#39;Ferhęg&#39;</a> has approved 7 edit suggestions and rejected 28 edit suggestions</li>\r\n\r\n    <li><a href=\"https://stackoverflow.com/users/609087/keety\">keety</a> has approved 126 edit suggestions and rejected 14 edit suggestions</li>\r\n\r\n    <li><a href=\"https://stackoverflow.com/users/1907093/jonathan-crowe\">Jonathan Crowe</a> has approved 79 edit suggestions and rejected 6 edit suggestions</li>\r\n    </ul>\r\n        <p><strong>Editor Stats</strong></p>\r\n        <ul>\r\n\r\n    <li><a href=\"https://stackoverflow.com/users/4476153/fizz-binn\">Fizz Binn</a> had 2 edit suggestions approved, and 0 edit suggestions rejected</li>\r\n        </ul>",
    "actions":[
        {
            "type":254,
            "name":"Next",
            "description":"review next item",
            "tooltip":"review next item",
            "disabled":false
        }
    ],
    "reviewTaskId":7175628,
    "content":"<snip>",
    "actionDelay":0,
    "badgeProgress":"1,764",
    "suggestedEditCount":0
}

As you can see, the server is saying that I've only rejecting 5 edit suggestions, even though it's been established earlier that I have far more than that count.

  • Do you choose "improve edit" or "reject and edit" often? – nicael Nov 20 '14 at 19:28
  • The majority of my reviews were before those options were available, but I'll do it on occasion. I should have roughly 800 edit rejections – Chris Forrence Nov 20 '14 at 19:29
  • The thing is, those options don't count neither to "approved ... " nor "rejected ... " but increments your counter. – nicael Nov 20 '14 at 19:31
  • @nicael - That's not quite what I was getting at; I've clicked "Reject" on roughly 600 suggested edits, so that 1 should be 605. I found another interesting screenshot from a review audit that does show the proper reject count, so I'll be editing my post with that. – Chris Forrence Nov 20 '14 at 19:39
  • 8
    Related on MSO: meta.stackoverflow.com/questions/272597/… – gunr2171 Dec 12 '14 at 21:39
  • 1
    I have a query here to support the weirdness – rene Dec 12 '14 at 21:53
  • @gunr2171 - Ah, good catch. – Chris Forrence Dec 12 '14 at 22:18
  • This is still an issue as of February 26, 2015. – Chris Forrence Feb 27 '15 at 0:29
  • @ChrisForrence Trying clearing your cache – QuyNguyen2013 Feb 27 '15 at 12:30
  • 1
    @QuyNguyen2013 - Unfortunately, that won't do it. I've edited my post to include what the server is returning when looking at a suggested edit, and it includes the incorrect number within the response. – Chris Forrence Feb 27 '15 at 15:44
  • Right, @ghosts_in_the_code. To clarify, I had been looking through the Suggested Edits review queue history. – Chris Forrence Mar 3 '15 at 11:31
  • I just reproduced this too. The page said I have rejected 5 edits when in fact it is roughly 600. – ɥʇǝS Apr 24 '15 at 23:52
  • Just noted the same weirdness and came here to look for this (also noted that audits show the correct result). – Leeor Aug 10 '15 at 18:18
15

This has been fixed and will be deployed in build 3746 for MSE/MSO, and build 2869 for other sites.

Many thanks to Jon Ericson for diagnosing the issue!

15

The same query is used for both displays. It's been a while since you asked the question, so current data doesn't match:

Review Result   Total 
--------------- ----- 
Approve           997   
Edit              193   
Not Sure          936   
Reject            649   
Reject and Edit    10  

A more recent example is Seth's Ask Ubuntu review stats. According to an audit:

Seth's suggested edit review stats (audit)

Meanwhile, according to a regular review:

Seth's suggested edit review stats (buggy)

Review Result   Total 
--------------- ----- 
Approve          3023  
Edit              623   
Not Sure           68    
Reject            728   
Reject and Edit     9   

Therefore, the audit result page is using just the "Reject" review result reason, while the regular result page is using just the "Reject and Edit" reason. Arguably, both displays are wrong. They should sum up both the rejections reasons.

I'll check with a developer to see if we can fix this bug.

  • 4
    At last! An explanation and some sort of roadmap for fixing this! – Nathan Tuggy Oct 6 '15 at 6:38
  • I would agree that both displays are incorrect; would 6-8 weeks be a reasonable lead time to fix this? – Chris Forrence Oct 6 '15 at 14:04
  • 1
    @ChrisForrence Lol. It certainly was done in less than 6–8 hours. – Chris Jester-Young Oct 6 '15 at 15:14
  • 2
    @ChrisJester-Young I mean, the bug's been in existence for at least 6-8 months ;) Good catch on the fix though! I'll mark your answer as accepted once those builds hit the market and I can verify that the numbers are good – Chris Forrence Oct 6 '15 at 15:23
  • @ChrisJester-Young Actually, quick question. If the review result is "Edit", should that get counted into the "Approve" count (since it's an implicit approval)? – Chris Forrence Oct 6 '15 at 18:54
  • @ChrisForrence I've deployed "meta" build 3746 a couple of minutes ago, and "prod" build 2869 is building now (about 1 minute left). Test away (soon)! – Chris Jester-Young Oct 7 '15 at 1:42
  • @ChrisForrence I've now implemented adding the "Edit" count to the "Approve" count, to mirror the "RejectAndEdit" count being added to the "Reject" count, on our internal dev system, so there's a proof of concept. (I'm waiting to hear back from Jon Ericson about whether it's desired; if so, it will appear in "meta" build 3747 and "prod" build 2870. If not, it will be rolled back. :-P) – Chris Jester-Young Oct 7 '15 at 2:16
  • @ChrisForrence After more discussions with another community manager, we felt it would be better to split improved edits into a third category (reject-and-edits are still counted as rejected). So that's what I've implemented now. – Chris Jester-Young Oct 7 '15 at 2:50
  • @ChrisForrence Builds 3747/2870 have been deployed a few minutes ago, so you should see the newly-added third category in the reviewer stats. Enjoy! – Chris Jester-Young Oct 7 '15 at 15:02

You must log in to answer this question.

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