What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 133 Stack Exchange communities.

I tried "Your math is wrong." as a rejection reason. For reference, I was rejecting an edit that said "because n+n/2+n/4+... = 2n-1", which is wrong. So I rejected the edit because the math was wrong.

This was rejected so I just went with "invalid edit."

Seems to have happened before but that is marked and is a year old, but I just got this.

Note I don't know the logic behind this. The only related questions I could find made it coming up at all sound like a bug. If my edit reason was invalid according to the system (it's not), I'm going to need some clarification so I can get around this requirement submit lovely rejection reasons in less time than it takes for someone else to reject for an uninformative standard reason.

share|improve this question
3  
Did you just created the [sorry-buster] tag for this question? Be careful, Jeff doesn't like this. –  Antony Jun 14 '13 at 15:49
1  
@Antony well I'm more than happy to go add it to the other questions. It's pretty ridiculous error text which kinda warrants a tag, IMHO. –  AAA Jun 14 '13 at 15:50
1  
I think the words "sorry buster" in the title are more than adequate. We don't make a tag on Stack Overflow for every [InvalidDoofusException] that comes along. –  Robert Harvey Jun 14 '13 at 15:57
    
@RobertHarvey was kind of making fun of the ridiculous error put in by the programmer, but we hate fun (seriously, "we hate fun" isn't documented in many memes of meta? Sigh.) –  AAA Jun 14 '13 at 16:10
    
@djechlin That's less of a meme and more of a fact... –  staticbeast Jun 14 '13 at 16:44
    
How is the math wrong? 1+2+4+8+...+n = 2n-1 –  CodesInChaos Jun 14 '13 at 17:38
    
@CodesInChaos that's a different statement. That one is true (replacing 2n with 2^n), the one in my post is false. –  AAA Jun 14 '13 at 17:43
    
@djechlin 1+2+4+8+...+n and n+n/2+n/4+...+1 are exactly the same sum for integral n and n a power of two, which are clearly valid assumptions in this context. Or if you prefer exponential notation, then 2^0+2^1+...+2^m=2*2^m-1 is also equivalent with n=2^m. –  CodesInChaos Jun 14 '13 at 17:44
    
@CodesInChaos n+n/2+n/4+...+1 is a finite sum, n+n/2+n/4+... is an infinite sum. –  AAA Jun 14 '13 at 17:54
    
@djechlin It's obvious from the context that the sum terminates at 1. It's about an iterated algorithm working on n elements where each step takes half as many comparisons as the previous one. The last step does 1 comparison and then the algorithm is finished. Doing comparisons on 1 element, then on 1/2 an element, then on 1/4 an element,... makes no sense. –  CodesInChaos Jun 14 '13 at 18:05
    
@CodesInChaos wasn't obvious to me, which is why I rejected the edit. Answers/edits should be clear. Ideally the editor would have learned that I rejected because I thought the math was wrong (more ideally I would have explained why I thought it was wrong), but now editor just knows it was mysteriously invalid. –  AAA Jun 14 '13 at 18:09

1 Answer 1

This looks like it can crop up if your edit reason needs to be pushed over the minimum character threshold and you do so by junk characters. Period. Space.

Could be the end, could be at the head. But when nudging enough characters to fill in the reason it may be time to verbosify your custom reason or just select one from the pre-drawn as the edit gets approved anyway while you were busying looking at what was being suggested.

share|improve this answer

You must log in to answer this question.

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