Thanks for this.
However, allow me to disagree.
Usability issues are NOT a specific requirement for my journal. If the process "assigning a new reviewer" requires three different steps (create a new reviewer, create a new userid, go to the reviewer database to find the reviewer), which should in theory only require a single step (enter the new reviewer information, all other steps are done by the system automatically in the background), then this is NOT specific to my journal.
I am in the permanent business of having to give tech support to authors and reviewers etc because of the many, many usability issues which have not been taken into account during development. Most of these are not rocket science but really simple to address and to avoid. IF YOU HAVE SOME FUNDING LEFT, PLEASE PLEASE HIRE A USABILITY ENGINEER TO DO A SIMPLE HEURISTIC USABILITY ANALYSIS.
Alternatively I invite you do work as a managing editor for 1 week in my editorial office to get a feel for the issues higher-volume journals face.
OJS should be scalable, but at the moment it is still really only suitable for microjournals with few submissions.
In the last 24 hours alone, I had already authors who accidentally did not upload their supplementary files because they press the "continue" button before they press the upload button, reviewers forget to click the "submit" button because it is counterintuitive having to press yet another button after already uploading the comments (why isn't the "recommendation" drop-down list simply in the window with the comments textboxes?), etc etc etc. The list is longer than I can document.
What is "journal-specific" is merely the fact that I am dealing with an extraordinary number of manuscripts, and even more reviewers (we assign 4 reviewers to each manuscript. If one step - such as assigning a new reviewer - takes 5 minutes instead of 1 minute (would be possible with a better algorithm), and I have to do this 10 times a day, I have spent 50 minutes for a task that could have been done in 10 minutes.
Where other editors may get only 1-2 emails from authors or readers who have problems, I am getting 3-5 per day.
That's a LOT of wasted time.
Low-volume journals with one submission per month may not be bothered by these problems, but OJS is (still) not really recommendable for high-turnover journals (and my journal is still small compared to larger journals with >1000 submissions per year).
Don't get me wrong, I have the utmost respect for the OJS development team, and OJS2 seems very well engineered. However, I still wish you had a usability engineer (or at least somebody with an eye for usability) on your team.
And as I said elsewhere, the issue of not being able to load email templates into the decision emails to the author, as well as the issue of providing wrong status information to the author, ... i don't know what to say about that, but it almost makes me want to change back to OJS1.
Some points on your suggestions:
Some of the issues you point out are fairly specific to your situation, i.e. short turn-around time, frequent creation of new user accounts in a large database, etc., and they are probably best fine-tuned via customizations to OJS. We constantly struggle to find a balance between flexibility and simplicity -- if we incorporate options for every variant, the user interface will become unwieldy and the codebase will become unmanageable. These forums also serve as a developer exchange, and I'm happy to help with customizations.
Open Journal Systems Team
Don't miss the First International PKP Scholarly Publishing Conference
July 11 - 13, 2007, Vancouver, BC, Canadahttp://ocs.sfu.ca/pkp2007/