If someone really felt motivated, they could go through all the old PMD threads looking for likely suggestions.

I have actually started working my way through one particularly fertile thread full of suggestions and feedback: PerlMonks for newbies?. I'm about half way through and so can make a few observations about the work flow involved.

The document I've been assembling lists each suggestion I found, followed by a list of comments related to that suggestion. Each comment identifies the monk, the note where the comment was found, whether the monk was for, against, expanded on the idea, or some combination of the above. Monks tend to be thoughtful, so they rarely simply say yea or nay.

I'm currently doing this in a text editor opened side by side with the browser page. There are several things here that slow me down:

• My text editor doesn't speak PM mark-up so I don't have any quick feedback as to whether I've formatted something correctly. That means I'm going to spend a lot of time just cleaning up format. What would be nice for this kind of work would to have extra private scratch pads that I could open side by side with a PM note.
• Transferring the monks name and note id to my text file is slow and error prone. If there were a way I could press a button and see the id and author of a note appear in my scratch pad that would save lots of time.
• An alternate approach (if we had the technology to support it) would be to tag and highlight selected portions of the text, but I'm not sure how useful this would really be. It turns out that a fair amount of work is involved in summarizing and categorizing suggestions. Most peoples eyes (especially mine) blur when lists don't have any narrative to help one categorize and remember the various ideas therein. I find that cutting and pasting in a text editor is the easiest way to create those categories and narrative.
• Part of the reason this is true is that there are no well defined categories for suggestions - I'm making them up as I go. I suppose we could develop a coding system. That would facilitate multiple people getting involved - provided they would be willing to learn it. On the other hand that might hinder as much as help: it ups the learning curve for volunteers. More importantly, some of the best suggestions push things in a new direction that makes one rethink categories. What would we do with such out of the box suggestions? I think we need a history of free-form experience with this task before we can decide if a coding system is a good idea, or come up with a coding system that really works. So for the short term, tools that facilitate free-form categorization would be most helpful.
• Where per-note tagging would really help would be in identifying lists of nodes that should be mined for suggestions. Tagging could also help volunteers keep track of which notes had been mined and which still needed mining. As notes progress, we simply change the tag to reflect that.
• For tagging to work as a process management tool, we would also need a page that had a links to lists displaying nodes in each stage.

Just some thoughts.

Best, beth