Content deleted Content added
m →Layer names and number of layers in the literature: improper nouns should't be possessive |
|||
Line 198:
Different authors have interpreted the RFCs differently regarding the question whether the link layer (and the TCP/IP model) covers physical layer issues, or whether the hardware layer is outside the scope of TCP/IP. Some authors have tried to use other names for the link layer, such as ''network interface layer'', in an effort to avoid confusion with the [[data link layer]] of the seven-layer OSI model. Others have attempted to map the TCP/IP model onto the OSI Model. The mapping sometimes results in a hybrid model with five layers where the link layer is split into a data link layer on top of a physical layer. In literature with a bottom-up approach to Internet communication,<ref name=Forouzan/><ref name=Comer/><ref name=Stallings/> in which hardware issues are emphasized, those are often discussed in terms of the OSI model.
The internet layer is usually directly mapped into the OSI model network layer, a more general concept of network functionality. The transport layer of the TCP/IP model, which may be described as a host-to-host layer, is mapped to OSI layer 4 (transport layer), sometimes also including aspects of OSI layer 5 (session layer) functionality. The OSI
However, the TCP/IP model has never been altered by the Internet Engineering Task Force from the four layers defined in RFC 1122. The IETF makes no effort to follow the OSI model although RFCs sometimes refer to it and often use the OSI layer numbers. The IETF has repeatedly stated{{Citation needed|date=April 2009}} that Internet protocol and architecture development is not intended to be OSI-compliant. RFC 3439, addressing Internet architecture, contains a section entitled: "Layering Considered Harmful".<ref name=R3439>{{citation
| |||