'Adds nothing new' reject category

[a_reader2003]

I started responding to Kelley's post point-by-point, but thought it 
might be clearer if I re-arranged it a bit, put the issues under 
headings, and brought in some points that other people have made so 
far. Hope this helps focus the discussion; we must sort this out soon.

HOW REJECT CATEGORIES WORK

Carolyn commented:
> BTW, the major reject category at present is OT, accounting for
> nearly 60% of all rejects. But as I have pointed out, the 'adds
> nothing new' might become more important in the future, so it is
> important to agree on what goes in it.

Kelley responded:
Certainly the percentage of OT posts should decline soon, yes?
Carolyn, I know I already asked this (didn't I?), but the rejected
posts -- they're not categorized in any way aside from 'adds nothing
new', is that correct? There's not a "Reject -- OT" or "Reject --
'me too'" and so on, right?

Carolyn again:
Yes, the percentage of OT rejects should decline when the separate OT 
list comes into being. Also the movie posts should reduce.

Your second assumption is closer - rejected posts get coded reject-
OT, reject-movie, reject-adds nothing new, whichever is most 
appropriate (there are 11 reject categories). Rejected posts don't 
get any other coding except sometimes 'Just for a laugh', if a post 
has to be regretfully rejected as not making any kind of canon point 
but is nevertheless amusing. 

As I said in my paper, posts which have been coded to any of the 11 
reject categories would not be offered to members for searching, but 
they are obviously still there for Admin or whoever to have a look 
through if they want to.  [Hopefully we will make 'Just for a laugh' 
available though - which is not a reject category, but an Admin 
category in section 4].

HOW SHOULD THE  'ADDS NOTHING NEW' REJECT CATEGORY BE INTERPRETED?

Kelly (Corinthum) in post 434 described the reject dilemma very well 
(paraphrasing):
'Try to look at it from the future user's perspective - if a post is 
coherent and elaborates even a little, then keep it. Try to think if 
the post would be helpful if you were looking for opinions on 
something.'

Anne (posts 385/425) asked pertinently 'how picky are we?' and 
suggested that this reject category should be used for:
- v.short, obviously minor posts
- longer posts which still added little to the discussion
- mainly personal opinion without much reference to canon

Carolyn suggested it should also be used for:
- initial questions which are subsequently repeated in later answers
- borderline FAQs
- very simple freqently-stated ideas with no new viewpoint ['I think 
Snape is a vampire..']

David, however, queried whether we should use a tighter definition:

In effect, I have been treating the posts as if they were in the 
Pending Messages queue, rejecting those that would not have made it 
to the list if that system were in place, and categorising the 
others. The criterion for pendings, when I was doing it, was 'does 
the post make a valid canon point?' Is that sensible? 

Kelley (Elf) responded:
David, can you give me a few post numbers as examples (examples of 
what you're rejecting under this criterion, I mean)? Are they OT, are 
they glorified fluff one-liners, etc.?

Carolyn also commented:
 I showed this to Kelley the other night and she agreed with me that 
the stuff there added zilch to any debate. 

Kelley agreed & explained:
Yep, those were definitely pointless. I think, in my mind, the 
phrase "adds nothing new" is rather more narrow -- it makes me think 
of a legitimate post, but one that's already been said, if that makes 
sense. Similar to just after GoF came out and we kept getting all 
the "Did anyone notice that gleam in Dumbledore's eye?" question -- 
it had come up a hundred (or thousand) times of course, but someone 
new who'd not seen all the discussions would think they'd caught some 
subtle clue that the rest of the world missed. Legitimate topic, but 
rarely was anything 'new' ever added (same as the above re number of 
students, wave 1, etc., too, really).

Carolyn: 
So, to summarise, there will be posts which make a valid canon point 
within the HPfGU rules, but which can nevertheless still be rejected 
because it is such a slight point, and/or one which has been repeated 
so often that it adds absolutely nothing to include it again.

PRACTICAL ISSUES IN USING THE CATEGORY

Barry said (422):
'It is easy to use - too easy sometimes - need to stop and think 
whether it is a new take.
Boyd said (423):
Hard to remember what went in before when in the middle of coding
Anne said (425):
Tends to be fairly liberal in giving people benefit of the doubt
Kelly said (offlist):
Has tended to be more generous as get further in to posts and points 
become more complex.

Carolyn:
There isn't an easy answer to this. Either stop and check the posts 
previously coded to 'adds nothing new' (not practical when there are 
thousands of them), or, err on the side of doubt and code a post up 
to the main catalogue rather than reject. 

There is also the relationship between this reject category and 
the 'FAQ' reject category. Anne defined this very accurately 
recently  - only click 'FAQ' if the post is not only nothing new, but 
also features on the HB list of FAQs.

SECOND EDIT & FINAL PRESENTATION TO MEMBERS
 
Dave commented:
I'd be interested to know, too, if something of the sort underlies 
Kelley's reservations about the 'adds nothing new' category as there 
is not usually a reason to reject these for posting to the list.

Kelley responded:
To explain, my reservation on that is more when it comes to a given 
topic that keeps getting rehashed (number of students is the example 
that springs to mind); it's a legitimate topic, one that keeps 
resurfacing, and little, if anything, added is ever 'new'. I'd had
the thought that each time the topic is brought back up it would be 
identified as such (wave 1, wave 2, etc.), though.

Carolyn:
Kelley is referring here to something that was in my original idea 
for the catalogue presentation to the members. I had envisaged 
repeated debates being presented as 'wave 1', 'wave 2' under the 
various category headings (eg 'school population').

This concept is still there in the current version of the proposal 
(MEG paper). Although using the 'adds nothing new' reject code will 
remove a certain layer of fluff, repeated one-liners etc, it is 
unlikely to remove anything substantive that has been posted. I know 
for sure that I have already coded up at least two major 'waves' of 
argument about the Hogwart's school population, for instance, maybe 
more. These waves will simply consist of more on-focus posts than 
otherwise.

In the MEG paper, I suggested one way of presenting the posts eg:
Snape
posts xx-xx
posts xx-xx

where the xx's could be dates. This idea corresponds to the 
original 'wave' idea, but there are probably other more interesting 
ways of doing it. Presentation is  the part of the project I am keen 
to get on to asap, but feel we can't start to tackle it until the 
legal issues are resolved.

So, says C, limping exhausted to the end of this... 

- are there any further views on how we should use this reject 
category?
- Kelley, is this explanation suitable for MEG, or do you want some 
additional words or for me to re-draft part of the paper??

Carolyn