dave@vdelta.UUCP (Dave Arnold) (02/04/88)
Could somebody briefly describe the relationship between X.225 and X.410? When a X.400 vendor says that they use X.225 in the session layer, does this mean that X.410 is not used at all? I though X.410 was the session layer for X.400. -- Dave Arnold uunet!ccicpg!arnold!dave -or- dave@arnold.UUCP -or- dave@arnold.VOLT.COM
hrs@homxb.UUCP (H.SILBIGER) (02/06/88)
In article <30@vdelta.UUCP>, dave@vdelta.UUCP (Dave Arnold) writes: > Could somebody briefly describe the relationship between X.225 and X.410? > When a X.400 vendor says that they use X.225 in the session layer, does this > mean that X.410 is not used at all? I though X.410 was the session layer > for X.400. > -- > Dave Arnold > uunet!ccicpg!arnold!dave -or- dave@arnold.UUCP -or- dave@arnold.VOLT.COM The session layer protocol for MHS is the Basic Activity Subset (BAS) of X.225. X.410 is the Message Tranfer Agent protocol. Herman Silbiger ihnp4!homxb!hrs
cabo@tub.UUCP (Carsten Bormann) (02/14/88)
In article <30@vdelta.UUCP> dave@vdelta.UUCP (Dave Arnold) writes:
() Could somebody briefly describe the relationship between X.225 and X.410?
() When a X.400 vendor says that they use X.225 in the session layer, does this
() mean that X.410 is not used at all? I though X.410 was the session layer
() for X.400.
Since I have not yet seen a response, a late reply:
In a sense, X.410 actually IS the session layer of X.400-1984, very much
like X.409 could be called the presentation layer of X.400-1984.
Unfortunately, the OSI standardization process moved away from the
semantics of the layers as they were intended by IS 7498, essentially
removing the session layer semantics from the session layer (some
people will remember a posting last year from Andy Tanenbaum, a former
OSI guru, wondering on what happened to the session layer).
What was left in the session layer is a syntactic means to mark parts
of a communication process as relevant to a certain activity and/or as
secured by the partners in stable storage (synchronization points).
These syntactic means do not have any semantics; they were based on
T.62 (the Teletex protocols) and were defined in X.225 (I forget the
right ISO number here).
CCITT had to put back session semantics somewhere, so they invented
ROS/RTS, put them in the OSI application layer, and defined them in
X.410-1984 (again, no ISO number handy).
In 1984, nobody had an idea on how the presentation layer would look
like, so X.400-1984 could not use the presentation layer. Nonetheless,
a defined data representation was needed for X.400, so X.409 was
invented (and later introduced into the ISO groups as ASN.1).
Since ASN.1 now is used by a lot of other things than X.400, CCITT
now calls it X.208/X.209 (one is the abstract syntax, one is the
basic encoding, similar to ISO 8824/8825).
I was told that X.400-1988 makes use of the so-called presentation
layer that ISO has arrived at in the meantime; so the ``real'' session
layer RTS now sits on top of the OSI presentation layer; but the OSI
presentation layer does not give you any real presentation layer
functionality since you need that on top of (and not below) the RTS...
What a mess.
I gave up on OSI about two years ago and started working on ODA
(T.400); there is some hope that at least this part of the Open Systems
Architecture can be kept clean.
Greetings, Carsten Bormann
--
Carsten Bormann, <cabo@tub.UUCP> <cabo@db0tui6.BITNET> <cabo@tub.BITNET>
Communications and Operating Systems Research Group
Technical University of Berlin (West, of course...)
Path: ...!pyramid!tub!cabo from the world, ...!unido!tub!cabo from Europe only.