As is seen many times in the Monastery, monks will oftentimes ask a question without using Super Search to see if the question has been answered before. I suggest adding a link to Super Search next to Add your question or the preview button on SOPW, as a friendly reminder. I would hope this would cut down on repeat question, though people would still have to click on it, and ask now what my fellow monks think.

"Cogito cogito ergo cogito sum - I think that I think, therefore I think that I am." Ambrose Bierce

Replies are listed 'Best First'.
Re: Super Seach Link on SOPW
by Ao (Friar) on Feb 02, 2004 at 23:21 UTC
    Kutsu++, maybe even auto-magically perform the Super Search for them based on what they use as the Title when they click on the preview button for root-nodes.
Re: Super Seach Link on SOPW
by Theo (Priest) on Feb 02, 2004 at 18:38 UTC
    Wouldn't do any harm, but I don't think it would get worn out from being hit a lot, either. Go for it though!

    (so many nodes and so little time ... )

Re: Super Seach Link on SOPW
by graff (Chancellor) on Feb 03, 2004 at 03:25 UTC
    I don't know the internals about how Super Search handles the indexing of nodes or massaging of query terms. 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.

    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:

      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$/