Fogbugz-on-demand, 37 signals and PB-Wiki use sub-domains with custom URLs for each user group to access their login page. Thus, it is not even possible for a user using these services to log in to the system (which can log in to any user of the service).
At first I thought it was a terrible design flaw, but now I think it is a really good idea because of: 1) Additional security, it may not be that much, but now you need 3 pieces of information: 1 subdomain URL, 2 logins , 3 passwords
2) You can leave the same logon to different accounts at the URL of the subdomain.
3) This provides greater isolation, load balancing, and ease of deployment, since you can have subdomains running under different servers that are more isolated from each other.
4) You can gradually deploy changes to a specific set of subdomains that are marked as BETA users before deploying them to all users.
Are subdomains for entering web applications in the future? Are there any pitfalls with this design approach that I don't see? Is it too incomprehensible for users who are not programmers to understand the concept of a subdomain?
source share