I'm no scientist, but I was involved in a couple of projects with a research group to develop web apps that others could use to run biophysics simulations. Whe

I'm no scientist, but I was involved in a couple of projects with a research group to develop web apps that others could use to run biophysics simulations. When submitting them for peer review, there was an absolute requirement from the journals in question that the sites did not require a login to use, and not even an email address to be entered to alert the user to results/completion. Result pages and download links were to be provided at a hidden URL which was linked to from the submission page after the form was submitted. So while we did this, we also ended up maintaining emails for job alerts, but optionally so. Most users have since used their emails to run jobs as it is more convenient for them. But for the reviewers, their requirements made sense. We were submitting to journals which had entire dedicated editions for online scientific apps. Hundreds of them, all of which required peer review by scientists who were being very generous with their time. For a free service, such requirements don't seem at all unreasonable.