Many of our writers have not done structured authoring, or they are such experienced writers that their first reaction to DITA is that it’s too restrictive and they’ll find a way to do things the way they’ve always done them. To identify the need for remedial training, or even to identify the need for a specialization or different output rendering, I think we’ll keep tags like i and b. Automated searches for presentation tags in DITA can make them serve as a “honeypot” either for tag abuse or to identify an unidentified need.

Thanks for your comment.

I like your idea of rendering the text in orange to call out tags that need attention. And, I agree that no standard is going to fix the world’s problems. I just don’t want a standard to perpetuate problems, and I think leaving tags that are solely representational does that.

There is a trade-off between allowing, then “correcting,” this kind of markup versus not allowing it, but possibly getting other kinds of tag abuse. Given that trade-off, I prefer the second choice, at least with professional writers, though that may put me in the minority.

Thanks for your note (I edited your first comment to escape the italic and bold tags).

I see the point, though it concerns me that authors can so easily fall back on those tags. I guess if I were doing a specialization, the first thing I’d do would be to remove the representational tags. I’ll check out the search you recommend.