Wow! But...

I like this, because I've always had two major gripes about the old NN list:

(a) For any top-level node listed as new, I don't get to see how many replies it has -- I have to leave NN and go to the particular "wing" (SoPW, etc) or to the node itself to see if there are replies. (I've learned to deal with this well enough, using tabs on my browser, but knowing the quantity of replies, without having to leave NN, would really be helpful.)

(b) For any top-level node that is no longer "new" for me, I won't see any new replies in that thread unless I sift through the "Notes" portion of NN (which, as sintadel points out, is pretty unwieldy).

Your new Threaded version solves both problems, so it's definitely cool. But it strikes me as overkill. While it is much nicer/easier to scan than the variable-length "15 Nodes per Page" format of the various "wings" (SoPW, etc), it has lost the economy and compactness of the original NN's "one line per thread" format.

I'd like to suggest an alternative design, which is (I hope) an easy simplication of the threaded approach you've shown us. It would keep the "one line per thread" format of the old NN list, and instead of showing the thread's internal structure and depth, there would simply be a number in the right-hand column, saying how many replies have appeared in the thread (regardless of depth) since the viewer's "newness" threshold.

I'm assuming it's fairly easy and not too costly to run a query that gets all nodes where "age < threshold_age" (or "date > threshold_date"). To the extent that titles on replies are consistently following the new format of /Re(?:\^\d+): (.*)/ it should be pretty easy to locate and keep track of the root nodes (title eq $1) for all the new replies.

With that, your idea of using the font to indicate new things in an old thread will work fine to differentiate between just two cases: new root nodes (with 0 or more replies, all of which must be new), vs. old root nodes with 1 or more new replies (not counting replies that aren't new).

With that sort of setup, I could easily imagine not needing or having a "Notes" section at all. But I expect a lot of people like the idea of having direct links to specific new replies. Well, now that you already have a means for showing the tree structure of a whole thread, why not apply this on an "as-needed" basis: consider a one-line-per-thread layout on the "new improved NN" list like this:

 [Thread title (font shows new vs. old)]  [author]  [# of new replies]

  - Thread title links to the whole thread, just like old NN
  - author links to OP's home node, just like old NN
  - "# of new replies" links to a full thread tree,
      where font highlighting shows which replies are new
[download]

This way, you only pay the price of laying out a full tree for one thread at a time, when a viewer asks to see a particular tree -- and considering how big some of these trees get, it makes sense to have a resource that just puts one whole tree on one page. I like it. Do you think you'd try it?

If you do, someone (e.g. me) would probably ask that this tree-display resource be used throughout all the "wing" sections on PM: wherever there is a link labeled "# of replies", it would be really sweet for that to take you to a tree display, instead of going directly to "the whole magilla" (especially since the title link already goes there).

---

---

• Are you posting in the right place? Check out Where do I post X? to know for sure.
• Posts may use any of the Perl Monks Approved HTML tags. Currently these include the following:

  <code> <a> <b> <big> <blockquote> <br /> <dd> <dl> <dt> <em> <font> <h1> <h2> <h3> <h4> <h5> <h6> <hr /> <i> <li> <nbsp> <ol> <p> <small> <strike> <strong> <sub> <sup> <table> <td> <th> <tr> <tt> <u> <ul>

• Snippets of code should be wrapped in <code> tags not <pre> tags. In fact, <pre> tags should generally be avoided. If they must be used, extreme care should be taken to ensure that their contents do not have long lines (<70 chars), in order to prevent horizontal scrolling (and possible janitor intervention).
• Want more info? How to link or How to display code and escape characters are good places to start.