Few comments about "autodiscover" to Exchange Architects and bright minds that did think that:
- autodiscover has a clear higher TCO than previous Exchange versions to publish Exchange Services to the Internet
- autodiscover architecture make Outlook 2003 a best-choice over Outlook 2007 for Outlook Anywhere (and Outlook 2007 is no more in Exchange CAL)
- it's simply unmanageable for a medium company hosting many internet mail domains inside the organization (every domain added requires certificate re-submission & web service reconfiguration)
- it's unbelievable that not implementing autodiscovery method (complex, costly, difficult to manage) make simple, robust & consolidated features of Outlook/Exchange like OOF, Calendaring & Meeting scheduling, ... no more usable from outside the company (eg: Outlook anywhere)
- it's unbelievable that an "autoconfiguration" method like autodiscover in my opinion is and should be, it's required even if Outlook client is already configured. The result is that without autodiscover in place outlook (anywhere) is working at 50%.
- it's unbelievable that "out-of-the-box" e2k7/o2k7 do LESS for the users (outside company) than e2k3/o2k3
Information Technology is always getting more complex, and thinking in a complex way doesn't help to get things working. I doesn't understand why my outlook (anywhere) is able to send and receive mail, but not to synchronize GAL, place a meeting, setting Out-of-office, configuring Voice Mail...
If I'm able to talk to a CAS to access Mailbox, AND Active Directory (for authentication & c.) why shouldn't I get FROM THAT WAY even configuration info for which Autodiscover is required?
Why CAS doesn't do for me the initial simple autodiscovery query getting the XML (in other words proxying autodiscover like it proxies other access request)? It seems so simple to me...someone could answer me?
Thanks!
Alessandro
PS: It seems to me that you (Microsoft people) design a product with no respects for your users, and without historical memory of your products, often reinventing the wheel...am I wrong?