I partially agree. But not for the reasons you think.</conversatron> The main reason I agree is because people need something to do when they perceive a problem. For people who see nodes that they consider too off-topic, it isn't very satisfying to respond by not approving, not replying, and not up-voting. And down-voting or sending a private /msg isn't very satisfying because it is usually under most people's radar.
The main reason I disagree is because, from experience, "off topic" is a slippery slope and officially trying to have such a category leads to a lot of mislabeling and even more (mostly useless) haggling.
A practice that I haven't noticed much of recently (probably mostly due to my lack of looking) that was all the rage not too long ago, was that of considering nodes for "edit: mark OT". That seems to meet the requirements you stated above near perfectly, IMHO.
Of course, my impression of "mark OT" behavior is that it doesn't work very well. But it doesn't work very well for reasons that would apply to any alternate system for separating out "off topic" nodes. People just don't agree what is "on topic". I see plenty of nodes asking questions about Perl that the author has marked "OT", usually, it appears, because the question is about how to do something in Perl rather than directly and purely about Perl itself.
So "mark OT" is the best we have. I don't see us getting anything much better. We might one day have an official "OT" marking that prevents a node from being approved. But you don't surf via nodes that are affected by approval status so this shouldn't matter to you. And I think making it official in that way would be a mixed blessing. And we have a lot more important changes to the approval system that have been written and tested for years that have still not been applied. So, I'm not holding my breath.
Having tried to form a consensus (and nearly succeeding but being thwarted by Petruchio jumping into the conversation w/o reading the context and refusing to drop his "point" that was only tangentially related) on improvements for dealing with extremely off-topic nodes (like requests for money), I'm not holding my breath there either. That is, I don't think it will be easy to find agreement on technical solutions for "dealing with" off-topic nodes.
At this point, I'd suggest better documenting what is on-topic and what is off-topic, being sure to stress that PerlMonks is for questions primarily of interest to many Perl programmers, not just for "pure Perl" questions (note that I consider even "pure HTML" questions to be a grey area, not something that is clearly off-topic). I think this thread gives more evidence that keeping the focus of PerlMonks wide is not an unpopular idea. And I suggest documenting "mark OT" as the official "way to deal with" nodes that are merely off-topic (but more than just marginally so).
I don't think "mark OT" will work very well. And so I consider it mostly a waste of time. But it also has a rather small down-side -- that is, it will be fairly easy for me (and some others) to mostly ignore. But some will probably find the categorization useful. It will likely be useful as an outlet. And it may be useful in progressing the definition of what the focus of PerlMonks is or should be, which would probably be the biggest win -- having a clear, official statement, based on input from many users, to point people to when there is conflict over whether something is on- or off-topic.
