A couple of other comments can be found
here.
I just want to respond to some of the questions posed at that link.
One of the stated rationales for this proposal "is that readers can use the subgroup categories to focus on articles in their particular discipline". Is there any evidence to suggest that readers currently use categories to indirectly search for articles, rather than searching directly for the article itself? Certainly I have never thought to do so. At least based on my own experience, the sole rationale for subgroups would seem to be the benefit it might confer to editors and possibly authors, not to readers.
I agree this is the primary reason although I would say it might benefit authors more than editors.
As currently framed, the other rationales need to be elaborated as they do not make clear the benefit of the proposal to these editors or authors. A clear argument for the advantage of subgroups is especially important given that these subgroups could also complicate, not simplify, the work of authors (e.g. in categorizing their articles) and editors (e.g. in deciding who has responsibility for approving a given article).
In the current proposal subgroups do not have a role in deciding which articles should be approved. They do have a role as a nucleus of writers who want to work together on a topic. In this role there is no doubt they will move articles towards approval but workgroups take over when the approval process begins. As to complicating or simplifying the work of writers, I suspect that any group that starts a subgroup is likely to want to be categorizing their articles.
Both the term and concept of "subgroups" are not ideal if these groups of people are suppose to span multiple workgroups, rather than be subsets of individual workgroups. In the same way that "labels" could become a more preferred organizational tool than hierarchical "folders" (e.g. with GMail), we may want to apply a similar concept here. That is, editors and authors and articles could be identified by a more generous and overlapping set of category tags than is currently available. Rather than requiring editorial consent to create these category tags, perhaps these tags could be automatically approved once they have been tentatively applied to some minimum number of articles, say 100. Such a requirement would also limit creation of entirely spurious or redundant category tags (which is a possible problem for the subgroups as proposed).
Well they could be "groups" instead of "subgroups"? We have had this discussion a few times and no one has really come up with a good alternative. Maybe now the proposal is close to being voted on users here will rack their brains for a preferred title. I'm fine with changing the name to groups.
As to editorial consent, there is none as such. As proposed, subgroups can exist without editorial consent. Editorial consent is only required to be affiliated with a particular workgroup. The latter is an advantage if there is competition for tagging an article. A subgroup affiliated with a workgroup will get priority. This is relevant as there are only three possible subgroups per article. As an aside, should we allow more?
The three limit was to address your point with regard to "entirely spurious or redundant category tags". If only a few subcategories can be added to an article then the subgroups need to be well thought out and not too narrow. Also, only subgroups affiliated with a workgroup would be able to tag a popular article (given the priority rules). In this way there is some editorial control when needed.
This needs an amendment about minimum membership. Perhaps 5 editors? Otherwise it will become to much subject to small clique formation. I'm not sure how to propose it at this stage, but perhaps those submitting it will accept it as a friendly change.
Given the critical mass issue right now this would not work. In the future, when there are enough authors to invoke such a rule, don't you think that such groups would just not succeed? Possibly the better approach would be to add another factor that legitimizes subgroups as being membership number. I am a little reluctant to add such a line as it seems to weigh against niche topics which might well be legitimate.
