I'm not going to get into a revert war over this, nor would I care to try to undo all the single item categories you have created. However, I totally fail to see the advantage. Please explain what you think you are accomplishing and how this improves user experience. Dankarl (talk) 01:26, 23 August 2012 (UTC)
- My apologies for creating a problem in your world. I assume you disagree with the state categories being created to distribute U.S. items.
- My first big exposure to Commons was to participate in categorization of the Lewis Hine TVA photographs. I approached the categorization process believing the existing categories were 'etched in stone' resulting from logical decisions by someone in authority. Perhaps not a bad approach to begin with. By the time I had spent a few weeks on the project, I was throughly frustrated because there seemed to be little, if any, reason as to why a given subcat would exist for Tennessee or for a specific year in a decade or not. I left hundreds of files categorized at the U.S. or decade level because a Tennessee or year subcat didn't exist. Many times, it wasn't because because there weren't already files for Tennessee or for the specific year. The relevant subcat just hadn't been created and as far as I could see nobody cared. I reexamined some categories hoping that 'someone in authority' would add subcats when it was appropriate. If it happened, I seldom saw it. After I finished the Hine project, I went back to a number of the categories with partial diffusion and began adding subcats and diffusing for some categories.
- My addition of subcats is based on trying to improve consistency and reduce the number of times an editor needs to reexamine a file before it reaches an appropriate subcat. I wouldn't expect to initiate diffusion of a category unless there were several dozen files there. However when a category is already partially diffused for a fixed finite set of entries (e.g. years or states), I believe consistency and categorization efficiency are improved by creating subcats whenever a file exists. I dislike what I see as the alternative which is to repeatedly reexamine every file in the category and count the number of files for each possible subcat until the count reaches a 'magic' number. For a finite set of subcats, it's likely that the next file added to the category can be directly categorized into the correct subcat without numerous periodic reviews.
- However for an indefinite set of subcats (e.g. Buildings in Chicago), I wouldn't create a subcat unless multiple files already existed because the probability of the next addition being for an existing subcat is much less than for a fixed finite set. However if a category reaches more than a couple of hundred files, I would want to create subcats for the larger groups of multiple files to improve access to the groups of files, even if each subcat is small.
- When a sub-sub-cat exists directly in a given category (e.g. a city cat in a U.S. category already partially diffused with states--but without the state for that city), I would add the missing sub-cat and move the sub-sub-cat to the appropriate sub-cat.
- Regardless of the rationale above, I've clearly caused harm to and distraction from the editorial process. Again I apologize for the problem.
- What do you consider a minimum number of entries to create a diffusion category?
- When do you believe an intermediate level category should be created to address the example above where city categories are directly included in the U.S. category?
You may have a better rationale for when and how to subcat. Also missing seems to be any guidelines on the definition of categories on the category page. E.g. Courthouses in the U.S. has a number of questions. Should former courthouses be included or should they always be a separate categorization? Should Federal courthouses include buildings repurposed for special courts (e.g. Bankruptcy)? :I'd like to hear your thoughts or suggestions about how to approach these situations.
- Category:Alumni by university or college in the United States
- Category:Censuses in the United States
- Category:Christian theology
Practice of spiritual disciplines
- Category:study of scripture
American Civil War
- On the battlefield at Stone River from LoC
- Federal Camp at Johnsonville, Tenn. from LoC
- Pittsburg landing from LoC