I have now set the minimum tag count to 1 for the categories Image Analysis, Usage & Issues, and Development. This will hopefully get users to browse the top tags and add one that makes sense.
However, I do not think mandatory tagging is sufficient to fully address the issue of poor tagging quality. The only way I see to fully address it is for members of the forum team to edit posts improving the tags as their time and interest allows.
There are some technical things we could do, though, to make things smoother. For example, in the future, we will probably be able to auto-tag posts based on their content. So e.g. if someone writes “ImageJ” in the title or body of the post, the imagej
tag could be automatically added. And similar for the other software partners. This auto-tagging feature has already been requested by other Discourse users, and it seems likely to be added in the future; see this topic for details:
@anon96376101 One option for now is to simply monitor tags like imagej
that interest you, and rely on others to add tags to topics when they are missing but appropriate.
Alternately, one way to catch “software agnostic” questions is to bookmark a search that excludes posts tagging specific software packages. For example, it appears to work to search for #image-analysis status:noreplies -tags:cellprofiler-analyst,cellprofiler to list all Image Analysis topics with no replies yet that are not tagged with cellprofiler
or cellprofiler-analyst
.