What is IEC 61850-8-1 MMS Compared to IEC 61850-8-3 DMS?

The paper "The Standard Message Specification for Industrial Automation Systems ISO 9506 (MMS)" has been downloaded 6,691 times from the ResearchGate ... seems to be an interesting document. 

Now, since we are working on a new protocol standard for IEC 61850-7-2 Client/Server communication (IEC 61850-8-3, DMS, Direct Message Specification), the question is: What is the difference between MMS and DMS?

Here are some answers:

DMS will be an alternative protocol for MMS ... with WebSocket connectivity, ASN.1 message schema, ASN.1 JER JSON encoding, ... very simple ... a straight forward from IEC 61850-7-2 ACSI (abstract messages) to concrete messages ... 1:1 ... no tricks ...

There are a few crucial aspects: (1) MMS as a standard that needs outdated OSI upper layers, (2) MMS is very complex and does not easily fit for IEC 61850-7-2 ACSI ... needs a tricks mapping to MMS in IEC 61850-8-1, (3) DMS communicates over WebSockets (bidirectional), (4) DMS uses JSON for message encoding (or DER binary) many tools are available, (5) the message schema of DMS is far simpler than the MMS message schema ... even both are using ASN.1 as schema notation, (6) MMS is not a good friend of many experts, (7) the Proof of Concept for RTI2 (Netherlands DSOs) is the basis for IEC 61850-8-3 ... and available at GitHub ...

From an system point of view (engineering, SCL, Models, abstract services, ... application functions, ...) there is no difference between the two protocols. From a smart implementation point of view the "stacks" could have an API that is independent of the protocols. There needs to be a tiny layer that decides if, e.g., a Report (with the payload from the application models) is wrapped with MMS or DMS ... for DMS there are two options: JER (JSON) and DER (binary) ... from a DMS stack point of view there is one line of code (!!) that is needed to encode in JSON or binary ...

Of course, the MMS message syntax and encoding are not compatible with DMS message syntax and encoding. BUT: they both carry the same payload!! ... thanks to the huge list of standardized "Signals" (models).

Both solutions run on TCP/IP ... there may be multi-protocol IEDs available some time down the road ... as we see today that many IEDs are supporting multiple protocols ... 

People asked me: Why did you not have this solution developed earlier?!? I developed something with my granddaughter when she did her Bachelor at KIT ... check my blog with some historical discussions