# “Is there a system behind the magic of algorithm analysis?”

This question came up recently and was quickly closed as a duplicate of our reference question referenced in the title of this post. I'm not complaining about the reference question: it does a fine job of describing a lot of the problems and solutions one might have in this area.

My (ongoing) problem is that it seems to me that we're often too quick to pull the "duplicate of 'Is there a system...' " trigger. For example, in the question I referenced, the OP's intent was abundantly clear: "How do I analyze a loop that (1) iterates over every other index and (2) the upper limit decreases on every instance of the loop?"

That's not (IMO) answered by the reference question. Rather it was a very specific elementary problem that could be quickly answered with no mention of the reference question. While some people might disagree, I'd also argue that this falls squarely into the cs.SE ballpark of on-topic questions.

I've made this rant before; it's as if a student in my algos class came to my office with that question and I said "Here's a text on algorithm analysis; read the first two chapters and don't let the door hit you on your butt on the way out."

My point here is to again request that we stop and think a bit before hitting the close button. Before I enter duck-and-cover mode I suggest that we say to ourselves, "Does this reference question really address the poster's question, or would a hint or two be more appropriate?"

• Quick answer: With only the rule for translating for-loops, the question could be answered in minutes. No insight was needed, but it could be derived from the solution. Therefore I thought pointing to the reference (and the other examples!) was sufficient. You'll note that I also gave a hint that would help apply the method layed out at the reference question. – Raphael Sep 23 '16 at 16:58
• In general, I don't think "the answer is quick and easy" is a good reason not to apply policy. – Raphael Sep 23 '16 at 16:58
• By the way, bad title; it doesn't say anything about your question. – Raphael Sep 23 '16 at 16:59
• The comparison with your algorithms class doesn't work. If you're teaching a class, it's your job to assist the students. Further, unless there's a really creepy CCTV system in your university, there's no way of saying "Look up the answer I already gave in office hours last week." In contrast, if you're contributing to a site with an archive in your spare time, it makes sense not to duplicate effort and to focus effort where it will achieve the most. – David Richerby Sep 23 '16 at 17:02
• @DavidRicherby. Thanks. You might recall that we had this discussion almost a year ago. I guess we'll just have to agree to disagree. My point then and now is that the user experience in reality and in my analogy are the same. While I agree that building an archive is an admirable goal, that's certainly not our only reason for being on this site. – Rick Decker Sep 23 '16 at 17:38
• @Raphael. Too bad a two-character emoticon response to your title quibble would be too short for a comment. I'll have to settle for "heh". – Rick Decker Sep 23 '16 at 17:42
• @RickDecker I don't recall but thanks for reminding me so I don't need to say any more. :-) – David Richerby Sep 23 '16 at 18:05