You'd be better off just sharing benchmarking code if it's not too big, plus that way people can set the optimizations to maximum for their machine and have more representative results. Otherwise all you're going to be benchmarking is the optimization/CPU tuning settings the binary was built with (which might not be necessarily bad - if you're going to be releasing whatever product you're making as a windows-only executable, there might be no problem with this approach, it's up to you to pinpoint your target audience).
Oh, and of course, the malware thing. Yes, fewer people will download your benchmark to give it a try if it's an executable. I don't think it's forbidden by the rules here since I see it all the time in the beginners section, but I'd recommend not posting too many, and making the most of the few people that are going to be downloading it (and, yes, this goes without saying, but don't abuse this by actually posting malware, or you'll be swiftly banned by a moderator).
The slowsort algorithm is a perfect illustration of the multiply and surrender paradigm, which is perhaps the single most important paradigm in the development of reluctant algorithms. The basic multiply and surrender strategy consists in replacing the problem at hand by two or more subproblems, each slightly simpler than the original, and continue multiplying subproblems and subsubproblems recursively in this fashion as long as possible. At some point the subproblems will all become so simple that their solution can no longer be postponed, and we will have to surrender. Experience shows that, in most cases, by the time this point is reached the total work will be substantially higher than what could have been wasted by a more direct approach.
- Pessimal Algorithms and Simplexity Analysis