Yahoo Forum Archive

While I feel like we are making progress in this thread, I also feel
like it's still not completely clear what exactly has been decided
and what's still unresolved.

Dwim asked the question: "What is/are the problem/s with the current
categorization scheme that we should be trying to answer?" And Marta
and others have summarized these problems. So as a next step I
thought it might be useful run through each problem, its proposed
solution, and what consensus, if any, we seem to have arrived at.

For the sake of clarity, here is what I mean by the various terms:
Division = a group of main categories. This year we had three:
Books/Time, Genres, and Races/Places.
Main category = a group of subcategories. If a main category has
only a very small number of stories, it would have no subcategories.
Subcategories = a group of 5-15 related stories that compete against
each other. A nonviable subcategory is one with fewer than 5
stories, and those stories are called orphans until they
are "adopted" by another subcategory.

Problem: The category divisions and their presentation in this year's
form are confusing.
Solutions/consensus:
1. The new form suggested by RSF and drafted by Dwim should help
reduce the confusion. It may still need some tweaking, but we should
definitely adopt it.
2. The Books/Time division will be changed to simply "Time."

Problem: If the number of nominations is significantly lower next
year, then we may have too many categories to create viable
subcategories.
Solution/consensus: The number of nominations is too hard to predict
at this point, therefore we should avoid making drastic changes in
the number of categories for the 2006 MEFAs. In other words, we
should not eliminate any of the three existing category divisions.
However, the overall number of categories can be tweaked, if
necessary, simply by eliminating specific main categories within a
division.

Problem: "Mirror" category/subcategory combos like Romance: Rohan and
Rohan: Romance or Drama: Gapfiller and Gapfiller: Drama.
Solution/consensus: One idea is to simply make it a rule that no main
category can be used as a subcategory under another main category. I
don't think we have arrived at any consensus on a solution, but I
think we're all agreed that mirror categories should be eliminated.
It may fall to the admins and categorizers to work out the technical
details of how best to do that during the categorization process next
year.

Problem: Confusion about subcategories: how to choose them and how
stories end up in one or the other.
Solution/consensus: The new form, and Martha's category FAQ with its
list of all subcats used in the past two years, will be a great help
in dispelling the confusion.

Problem: Miscellaneous concerns about subcategories that are too
small, too large, the moving of orphans to other categories, etc.
Solution: It's good for these things to be raised in the post-
mortem. But again, I think they may be best resolved by the staff
group during next year's categorization process, as it's difficult to
discuss the more arcane points of categorization in the abstract…you
need to be actually working on the categories for it to make sense.

Well, I think that's it. But please, everyone, chime in where you
think I'm wrong, or if I missed something…

Kathy (Inkling)

Hi Kathy,

On 19 Nov 2005, at 22:58, Kathy wrote:

> While I feel like we are making progress in this thread, I also feel
> like it's still not completely clear what exactly has been decided
> and what's still unresolved.
>

I think you're right. Thanks for pushing this forward.

> For the sake of clarity, here is what I mean by the various terms:
> Division = a group of main categories.  This year we had three:
> Books/Time, Genres, and Races/Places.
> Main category = a group of subcategories.  If a main category has
> only a very small number of stories, it would have no subcategories.

Just to be 100% clear: An example of this would be "Romance".

> Subcategories = a group of 5-15 related stories that compete against
> each other.  A nonviable subcategory is one with fewer than 5
> stories, and those stories are called orphans until they
> are "adopted" by another subcategory.
>

This is where it gets a little bit confusing. A category can be
nonviable if it has fewer than 5 stories by 2 authors, and a
*mandatory* subcategory (poem, drabble, incomplete) could be inviable
for the same reason. But as for other categories, they need 5 stories
by 2 authors to be viable but if there *aren't* 5 such stories we don't
try to move stories around; we just don't have that subcategory.

So for example, if there are four stories in "Romance" about Faramir
and Eowyn, we don't try to move a fifth story from another category
into Romance so we can have a "Faramir and Eowyn" subcategory. We just
don't have that subcategory,, and those four stories go somewhere else.

> Problem: The category divisions and their presentation in this year's
> form are confusing.
> Solutions/consensus:
> 1. The new form suggested by RSF and drafted by Dwim should help
> reduce the confusion.  It may still need some tweaking, but we should
> definitely adopt it.
> 2. The Books/Time division will be changed to simply "Time."
>

Agree on all of that. Will these fix this problem? (I'm asking not just
Inkling, but anyone.)

> Problem: If the number of nominations is significantly lower next
> year, then we may have too many categories to create viable
> subcategories. 
> Solution/consensus: The number of nominations is too hard to predict
> at this point, therefore we should avoid making drastic changes in
> the number of categories for the 2006 MEFAs.  In other words, we
> should not eliminate any of the three existing category divisions. 
> However, the overall number of categories can be tweaked, if
> necessary, simply by eliminating specific main categories within a
> division. 
>

That makes sense. I was thinking about this issue earlier, and it seems
to me that this is just a problem we'll have to live with. It's
certainly not worth reducing the number of categories just for this.

> Problem: "Mirror" category/subcategory combos like Romance: Rohan and
> Rohan: Romance or Drama: Gapfiller and Gapfiller: Drama. 
> Solution/consensus: One idea is to simply make it a rule that no main
> category can be used as a subcategory under another main category.  I
> don't think we have arrived at any consensus on a solution, but I
> think we're all agreed that mirror categories should be eliminated. 
> It may fall to the admins and categorizers to work out the technical
> details of how best to do that during the categorization process next
> year.
>

I like the suggestion to just not have subcategories with the same name
as other categories. It seems simple. Thundera, I think you had another
situation, but I'm not sure I ever wrapped my head around it
completely. Would this work okay for you? How about everyone else?

> Problem: Confusion about subcategories: how to choose them and how
> stories end up in one or the other.
> Solution/consensus: The new form, and Martha's category FAQ with its
> list of all subcats used in the past two years, will be a great help
> in dispelling the confusion.
>

The way I understand Dwim's form, people will have a lot more guidance
on providing the information we need to make subcategories. I think
that will help, too. They don't *have* to know what will make a good
subcategory, just give the information.

> Problem: Miscellaneous concerns about subcategories that are too
> small, too large, the moving of orphans to other categories, etc.
> Solution: It's good for these things to be raised in the post-
> mortem.  But again, I think they may be best resolved by the staff
> group during next year's categorization process, as it's difficult to
> discuss the more arcane points of categorization in the abstract&you
> need to be actually working on the categories for it to make sense. 
>

I think it depends on the issue. Some thingss we can't anticipate,
others are really best handled case by case, but others we can look at
now. Categorisation will always be busy, and if something can be done
ahead of time, I think that's a good thing.

Marta