# rename tag: randomized-algorithm -> randomized-algorithms

the tag was created, deleted(?), and then re-created with a typo, and it does not let me change it. (it says something alike 'the cannot be created since already exists.. raise this issue in meta')

• Yea, seen that before. The system protects itself against too similar tags. Mods can rename tags; Shog9 has done that for us in the past, I am sure he will again. :)
– Raphael Mod
Mar 23 '12 at 7:54
• I don't think the tag is particularly suited for this question, but it is ok.
– Raphael Mod
Mar 23 '12 at 8:02

I'll take this opportunity to discuss randomness-related tags.

I don't see how applies to this question: it's about generating random numbers. A randomized algorithm is one that uses random numbers, not one that makes them: quicksort pivot, probable primes, …

This question should be tagged : although it's not about a PRNG per se, it's about generating random numbers with a particular distribution.

On top of these, we may or may not want a more general tag: (like SO) or (like CSTheory and Crypto).

• I don't agree with you on this. I read that question in the following way: "Given a random binary string, how can I construct an algorithm to do XXX" (here, XXX is output uniform numbers within range given by input, that is XXX in this case is "RNG"). This is clearly a randomized algorithm. Most of the answers there describe a las-vegas algorithm, my answer describes a monte-carlo, but all are algorithms. One can discuss their expected run time, expected amount of randomness needed, failure prob, etc. Mar 23 '12 at 23:25
• I agree with Gilles here, and I don't like the general tags. The notion of Las Vegas and Monte Carlo algorithms does not apply here as the desired algorithm does not have a correct output; we require a property of a sequence of outputs. I think your retagging edit, @RanG., was premature.
– Raphael Mod
Mar 25 '12 at 10:42
• @Raphael the field of randomized-algorithm is very broad. It contains many subfields: las-vegas algs, monte-carlo-algs, but also, analysis of required randomness for specific task. Other related subfields are extractors ("purifying" randomness), and pseudo-random generators. I think all these can be tagged as randomized-alg (the last two better be "crypto"; anyways most of the crypto algs are probabilistic). I think the question should be edited to be clearly a CS question and not MATH question, and this THIS should be the point of the discussion. Mar 25 '12 at 18:03
• PS. I didn't mean my retagging to be "one sided action" and I'm sorry if it is considered as such. Remember that all is editable, and it'll be changed as needed according to our discussion (if you prefer, it can be rollbacked now and set after the discussion...) Mar 25 '12 at 18:06
• Discussion rekindled here.
– Raphael Mod
Aug 29 '16 at 23:00