Still more on WSE and WS-Security
The wse 2.0 samples have helped tremendously. The ws-security is beginning to make some sense. I am still having trouble with some of the samples, especially those involving X509. I think that is user error. I don't really need to encrypt the messages but I want to know - so I will make the WSE samples work.
I was confused about how the password digest was created. There was a comment in one of the articles I read about it being the hash of a combination of the password, nonce and creation date. I finally found a spec at http://www.oasis-open.org/committees/wss/documents/WSS-Username-11.pdf
A tool that I've found very useful is SoapScope at http://www.mindreef.com. I purchased a copy to help me look at messages and it's been worth it.
I have implemented username signing in my DosEquis “learning” project. I already had an authentication scheme in place using a custom soap header, ala Yasser Shohoud' book.
One thing that I don't understand though is how a person perusing my web service site or wsdl would know about ws-security. I was looking at the microsoft.com webservice stuff and was wondering the same thing. If my client app had obtained a reference to a webservice via UDDI how would it know or be able to find out. More things that I don't understand about “real-world” webservices.
The next big thing is to start working with DIME. My DosEquis project will be collecting, processing and returning large amounts of data.
I also plan to start using the WS-I tools on my webservices. I have listened to Yasser Shohoud's web cast http://msdn.microsoft.com/msdntv/episode.aspx?xml=episodes/en/20030724WEBSVCSYS/manifest.xml and just need to “code“.