Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 153 Stack Exchange communities.

What is meta?
Here's how it works:
  1. Any Stack Exchange user can ask a question
  2. The community provides support, votes on ideas, and reports bugs
  3. Your voice helps shape the way Stack Exchange operates

Surely robo-reviewers basically hammer the review pages waiting for a review to appear.

Can the number of daily page refreshes of the review pages be limited to some reasonable "humanly possible" amount - say a several hundred?

Doing so would mean that their bots would exhaust the limit quickly and go away, reducing server load and giving some satisfaction to SO coders.

The daily refresh limit could be expanded and contracted dynamically based on the size of the backlog.

When the refresh limit is reached, it could display the message "Get a life", in case a real human actually hits the limit.


There seems to be some confusion here, even though I thought it was a pretty simple idea. So looks like we add more detail for the pedants who focus on the edge cases:

  • A page refresh limit would be completely unrelated to a review limit
  • The premise is that bots trash infrequent review queue pages until they get an item to review
  • 10K page refreshes was an arbitrarily "large" number that would almost certainly require a bot, as this is roughly one refresh every 9 seconds continuously for 24 hours - hence the "Get a life" message, but maybe I'm wrong. Maybe non no-lifers have this kind of time of their hands. Let's say they get a solid 8 hours sleep and spend 2 hours on food and hygiene. If they bash away at the review pages at the rate of 1 every 5 seconds for the remaining hours of the day, they'll hit 10K. Sounds reasonable...
share|improve this question
related (not a dupe because of terribly incomplete "fix"): Drop delay for “Skip” and increase it for “action”... – gnat May 14 '13 at 5:14
Should I consider myself a robo-reviewer just because I frequent the /review page? I'm way past 1000 and I still regularly run out of reviews in the suggested edit queue and I skip often. – Jan Dvorak May 14 '13 at 5:42
@JanDvorak How many times do you refresh the page per day? 10000? 100000? See what I mean? A robot would probably be hitting the pages every few seconds. Even if you are a real human and exceed the limit, it could just display a message "Get a life". – Bohemian May 14 '13 at 8:07
Since review limits are in place users do see a Get-A-Life kind of message after 20 reviews. Isn't this a similar check to a new 'page refresh limit' feature? – Aziz Shaikh May 14 '13 at 8:31
@Aziz The 20 limit just means you've reviewed the limit, but how did you get to review them? Did it require 10K page refreshes, or only a few lucky ones? If it was 10K, then you are a bot or you don't have a life. Either way, go away. – Bohemian May 14 '13 at 9:47
Getting mixed messages here. In the question, you say a couple of hundred but in the comments, you say 10k. If the refresh limit recommendation was for 10k, then I'd agree. At a couple of hundred though (as per the question), that limit could be reached easily by eager reviewers on slow review days especially if they are working on multiple review queues. Also, I think the message would need to be something more polite than "Get a life" as these users are contributing by making the site better in most cases. – Ren May 14 '13 at 10:17
This question states that "robo reviewers" are literally bots or scripts fully automating the review tasks. Is this the case? I'm under the impression that the term meant "reviewers who blindly upvote or pass a review without examining the content". I.e., bad reviewers who are trying to grind the badges, but nonetheless human ones. Are there any demonstrable cases of bots being used for reviewing, or is this simply hyperbole/assumption on the asker's part? Is this really a problem? – Esoteric Screen Name May 14 '13 at 13:45

You must log in to answer this question.

Browse other questions tagged .