except on the palms of our hands and the soles of our feet, but numerous hairs are so fine they’re virtually invisible. hair is created up of a protein referred to as keratin (the same protein in nails) produced in hair…

I agree with the part of your statement that says that LDAP shouldn’t be the only standard for access to identity data. We need a service that is much more high level. As you rightfully mentioned, SAML, Infocards and Liberty fill some of those gaps. But those standards also need to access “the raw data” somehow. And the most efficient way to do this is through the LDAP protocol.

LDAP isn’t going to go away. It is just too useful as a protocol. I think LDAP will stick around, but we’ll have some higher level services that will still use LDAP as a lower-level protocol to access the actual identity data. LDAP is more useful as a protocol than, say SQL because it’s faster, asynchronous – in short, it is highly efficient for low-level operations on identity data.

So whatever high-level identity services we’ll use, we still need a low-level identity data repository. LDAP directory servers are very useful for this purpose, and the more efficient we make them, the better it will be for the higher level services. Telcos are usually interested in massive scale, and thousands of operations per second, and they need a repository that scales to that level. Typically, they’ll deploy some identity services on top of LDAP, but for the applications that are very demanding, they’ll access LDAP directly. Efficient LDAP servers have no problem with that, whilst services based on XML are difficult to scale to thousands of operations for second – unless you have an unlimited budget