I'd be curious to see how things breakdown using JSPerf...

lol I know what you mean.

Nice Kata! Denying those functions would have made it interesting :)

Hey thanks, that one looks pretty good. I changed the tests in order to use that function and I also added some of them to the initial tests.

I unpublished while I make changes to the test structures.

Working on adding functionality to the images returned so you can click and get a textbox with the JSON array of image data. @JHoffner, can you give me some clientside javascript that would create a light box or similar that would be in keeping with how you like to display stuffs on the screen? I want the test images to be able to be cliked and provide the user with the image's data.

Added the tests to the new space for initial test fixtures.

Still need to identify why my results are different than MrKishi's solution (which looks righ to me.

The JS sandbox times out after 3 seconds. This was implemented to naively try to prevent runaway scripts. Its a pretty small execution window for some of the harder challenges on this site so we may increase that to 4 or 5 seconds. Are there any optimizations you could implement within your solution?

I reviewed my code and found a bug left over from the first version where I had implemented blur and was averaging the results instead of normalizing the matrix due to an incorrect understanding of the algorithm. This bug only shows up for the kernels that are not normalized (7th and 8th tests). Try submitting your solution again.

I simply feel that its problem statement doesn't provide enough to go on without (in my opinion) undue side research

I'm at a loss on what you want me to write exactly. @nklein, is there anything I've omitted you think I should write?

Well, I guess if I denied people from using Array.prototype.concat, Array.prototype.filter, Array.prototype.push, there would be a lot more solutions where people actually move values around.

I, on the other hand, enjoyed this problem as a break from doing the Fibonaccinumbers again with a slightly different interface. Programming skill, for me, can't be measured in a vacuum without at least occasional reference to "what would I do with this?"