I'm not sure what "more R-specific" would mean. I could write something more complex, and I have. This particular kata, however, has a couple of specific, simple learning objectives. It was not written to "make it different" from the other kata (which I don't believe was even out of beta when I wrote this kata), but to demonstrate some fundamental concepts about the R language. The proposed duplicate does not require any demonstration of how to subset a vector, which is one of the two learning objectives here. Since I think there is utility in having an even simpler kata, an R translation there could be proposed and would add something to the available R kata as well. I invite you to do that if you'd like. Considering we still have fewer than 300 R kata available, I think that would be productive. It seems we have a disagreement about the value of having a range of kata, from the simple to the complex. I understand your point of view, but I strongly disagree with it.

This is not spamming duplicates. This kata was written in October 2017, when there were only a handful of R kata on codewars. The task is particular to the structure of objects in R, which is why I did not translate it. I brought it up in gitter before creating it, and whether this was of benefit was discussed. The consensus was that adding this to the available R kata was beneficial.

Please do not re-raise this issue. You do not like the kata. This is not an issue. You may vote on your satisfaction here, but don't raise this as an issue.

Not an issue. The reason for using the t statistic is in the description (you do not know the population standard deviation, so you need a t statistic). More details for those who are interested can be found in the linked articles. Thank you for the suggestion to use test values from different distributions. I'll consider that.

Not an issue. Using a reasonable algorithm for the problem at hand is part of the task for this kata.

not an issue. This is a simple kata by design.

It looks like there has been a change in, perhaps the library used for the test suite (a previously working function is returning an object not found error). I haven't been around in some time. If I get a free minute, I will trouble shoot this. Thank you for raising the issue.

Overlapping samples are a different problem, which can be found in the other related kata (as mentioned in the description).

Re: non-overlapping ranges, this kata is looking for a very particular case, with a very particular teaching goal. The more general case is presented in another kata (as mentioned in the description). Repeated values are allowed (in the initial sample). Drawing the same value from the initial sample more than once is not. This is what a permutation test is.

OK, I added an empty line so I could fork it. Here it is: Fixed fork. It needs approval.

Thanks for spotting this.

I tried that edit, but it's throwing a "Kumite was published with failed tests!" error, despite the run output (for both the sample tests and the full test suite) showing all tests passed. And I can't fork it, because it says it's the same as the original. Suggestions?

hi @qaz56

It's not the same function :) One is the result of a call to vector_op, which by definition (in the description), returns an object with length (a list). The other is the result of a call to iter_eq, which by definition (in the description), returns a boolean.

Ahh... colon, not semicolon. Fixed!

Hi B4B, where, exactly, would you like a semicolon?

R Translation kumited

Please carefully review and approve!

-mpx