from what I understand, R does just that without any macroscopic issue

> a <- 1
> b <- c(1,2,3)
> class(a)
[1] "numeric"
> class(b)
[1] "numeric"
> a+b
[1] 2 3 4
> sum(a)
[1] 1
> sum(b)
[1] 6
> for(e in a) print(e)
[1] 1
> for(e in b) print(e)
[1] 1
[1] 2
[1] 3

Adding more tags means more to consider when tagging a submission. For example, there’s an OpenBSD and a Unix tag; should every OpenBSD article get both tags? A Rails tag could be added for those that just want to see Ruby but not Rails, but then every Rails article now needs two tags or they won’t show up if someone’s browsing /t/ruby. I already do a lot of cleanup of submissions to add additional tags because a lot of people only pick one tag and submit it.

I’d like to avoid adding tags for very specific things that will rarely get used and rarely get filtered. “compsci” is a decent generic tag that can be used for programming that doesn’t have a specific tag (although “philosophy” is getting over-used to apply to articles that don’t really have anything to do with philosophy) until enough submissions show up about a certain topic that a tag can be added.

As for voting on tags, if you want to write all of the code to do voting on tag recommendations and submit a pull request, I’ll consider it, but otherwise just create a meta post or send me a message to suggest a tag.

As for the specific examples cited here, I’d have a hard time saying that anything about machine learning, functional programming, immutability, and Haskell are not compsci or not appropriate for being posted here.

the video of the presentation is also available http://www.youtube.com/watch?v=lWIP4nsKIMU