Use of reject codes

[boyd_smythe]

> Carolyn wrote regarding "Nothing New" category:
> 'Use this for first statement of questions, where the question is 
> repeated in subsequent replies. Also use it selectively to ignore 
> posts that repeat points that are frequently made.'
> 
> I'd be interested in your views on how easy you have found this to 
> interpret in practice. It implies that the first person to make a 
> particular point takes precedence over a later one, for instance.
> 
> Its important we are clear about this, as using this category 
> effectively is likely to become more important as we go on, 
> especially after the main reject categories (movie and OT) get their 
> own separate lists, leaving the main list for pure canon discussion.
> 
> Carolyn

boyd:
For this project to become a useful tool for group members, we'll need 
to effectively seperate the wheat from the chaff, so to speak. But as 
Barry noted, this is going to become harder as we start seeing the 
same thoughts expressed many, many times. Frankly, once we've 
categorized 100,000 posts, the next 20,000 are going to be brutal if 
we have to look back every time to see whether someone has said it 
better.

I propose we reject to "Nothing New" only if we have seen the same 
thing was said better elsewhere in our current allocation of posts. 
The result will be more posts kept, with more repeated thoughts, but 
much faster coding. Then in Phase Two (after all posts have been 
coded), we go back and sift through the posts in each category and 
either keep only the better ones or mark them with an additional 
category to cross-reference (e.g. not FP, but RP--Representative 
Post).

--boyd