If the indexing and query handling were such that it could support an operation like "find documents similar to this one" (a common feature on google and elsewhere), then Ao's suggestion could actually be worth the effort -- in two ways.

(1) If the person is stating their problem reasonably well and using appropriate terms, a list of "similar nodes in SoPW" that comes attached to the Preview page may well give them the answer they were looking for.

(2) If the text being previewed is so vague or misstated that the "similar nodes" list contains nothing relevant, the person might get a clue that their post will raise more questions than answers. (Well, okay, that's just wishful thinking.)

I do have a hunch that some newbies flail with any sort of search facility because they tend to use the wrong words for the concepts they have trouble with.

Yes, indeed. I've been taking note lately of nodes with questions that basically boil down to control flow (with a view toward possibly hunting down or creating some kind of control-flow tutorial to point future ones to), and take a look at the ones I've got on my list so far:

• deleting from a mysql database
• searching a file
• substuting a whole file
• making loop in perl script
• The third 'if'

Now, the title "making loop in perl script" does clearly have to do with control flow. "The third 'if'" might. The others obviously don't -- yet that was the poster's problem. Further, the title that most obviously has to do with control flow, 'making loop in perl script', corresponds to the node of which I am *least* sure that the real problem is control flow. My conclusion is that if the posters understood control flow well enough to know it was the source of their problems, they'd be able to figure out their own problems and solutions without posting their questions at all. The reason they were stymied is because they didn't understand what their problem was.

I suspect that control flow is not the only concept of which this is true. For example, I once reported a 'bug' in bugzilla that I thought was related to the handling of CSS as applied to anchor tags, but my problem turned out to have to do with the inline box model as it applied to the parent element. Not being familiar enough with the spec to understand the inline box model, I didn't realise where my problem was.

---

$;=sub{$/};@;=map{my($a,$b)=($_,$;);$;=sub{$a.$b->()}}
split//,".rekcah lreP rehtona tsuJ";$\=$ ;->();print$/
[download]