Just a few things I noticed while forcing a test article through the software:
1. It would be great if there was a way to force the software to skip a production step, in such a way that you don't have to send endless fake e-mails to yourself to force the step through. Admittedly I was testing. But being able to bypass the author when desperately needed might be really useful. Like, say the author is doing fieldwork in the wilds of Tunisia and calls you on her satellite phone to say, "It's fine! Whatever! Just go!"--we wouldn't be able to do that. Similarly, it's unclear to me whether the author could just approve something during, say, copyedit, without uploading a new file.
2. As I noted in another forum, the Comment feature needs tweaking. Some kind of default title should be provided automatically if the commenter fails to enter a title, so that the comment hotlinks from the article page. Currently, if a title isn't input by the commenter, then the entire thing elides and you have to view all the comments to see it. In addition, the target audience for my journal is big on commenting and likes threaded comments. It is apparently impossible to "Add a New Comment" from the Comment page, only comment to existing threads, which makes no sense. This is definitely something that is going to be perceived as a problem.
3. This is just a typo: if you click the Information > For Librarians link, the breadcrumb on the associated page is incorrect. It should read Home > For Librarians, but it reads Home > For Authors. Still, it would be cool to fix this in future versions, because just finding the page to fix this is such a hassle that I still haven't bothered to do it.
FWIW! No need to respond, really. (The journal I'm testing is here: http://journal.transformativeworks.org/index.php/twc
and the site is not yet live.)