I like the idea, though not that format you suggested.

Why not just something like quote with a node parameter?

<quote node='673567'>
What we need:
<OL>
<LI> we need to attribute the original quote
<LI> we need something short to type
<LI> we need a "reply, including quote" button which includes the quot
+e and attributes it well, perhaps even linking back to it.
</OL>
</quote>
[download]

Then the rendering code can generate the user name, date of original post, links to author and referenced node, etc.

I'm not sure what the best way is to actually input the reference, but if we want the semantic meaning to be understandable for the next generation of parsers, it should be rendered as the attribute 'cite' with a URI to the quoted material.

See http://www.w3.org/TR/html401/struct/text.html#h-9.2.2.

That is a question of what we use to mark it up (i.e. quote like code) versus what is generated (pre or blockquote) to then be rendered by the browser. So the HTML generated by perlmonks could (and possibly should) generate the cite attribute when generating the output.


___________
Eric Hodges

I applaud your approach of stating more than one reason for a change, for considering and listing what the change would include, and for coming up with a possible solution.

The solution I use now is generally to just use the node id tag for the node I'm quoting ([id://673567]) and to blockquote anything I feel I need to restate verbatim in order to address.

The advantages of what I currently do are:

1. it requires no PM changes
2. it connects the reply back to the node
3. I only want to quote as much as necessary anyway, so copying and pasting is better for that than quoting the whole node and deleting large sections of it.
4. it doesn't force the PM code to track a potentially very large separation of potentially deeply nested punctuation that might be used for many other reasons inside the quoted section for every node as it's posted (this is an advantage over your specified possible alternative, but not over every possible solution)

It has disadvantages, but I don't think they're that bad:

1. I might have to type the name of the monk to whom I'm replying if I'm afraid the thread structure and the node id tag aren't clear enough (that's a pretty big if)
2. it does take a little bit of manual interaction, but then so does trimming the quote rather than rudely making a one-line response to a 200-line quoted section

I have to agree. Personally, I would just be happy if I could use <bq> instead of <blockquote>!!

YeS!!!!

<bq argel>
  Personally, I would just be happy if I could 
  use <bq> instead of <blockquote>!!
</bq>
[download]

That idea is more smoking hot than barbeque! (bbq)

I like the idea. A very frequently-used thing gets a short easy-to-type name.

Users would only have to function as selection criteria for the bots.

I don't think it would be too difficult to whip up a Perl-script using Algorithm::NaiveBayes to make the selection for you.

Hey, we could have a fully automated PerlMonks site! Anybody up for writing a troll-script?

CountZero

A program should be light and agile, its subroutines connected like a string of pearls. The spirit and intent of the program should be retained throughout. There should be neither too little or too much, neither needless loops nor useless variables, neither lack of structure nor overwhelming rigidity." - The Tao of Programming, 4.1 - Geoffrey James

Anybody up for writing a troll-script?

I thought

I think what would really be best would be if you had a bunch of bots that worked like the monkey in the proverbial shakespeare cage, constantly preping possible responses.

had been produced by one... 8-)