Requested revision
Standard: | IEEE Std 802.1Qcc-2018 | Clause: | various |
Clause title: | various |
Rationale for revision
The text "boundary between the network media and PHY" is problematic due to terminology/abbreviation usage in the context of 802.1Q. This text is found at various locations throughout the standard (16 matches in D0.2 of the next Q-Rev). Detailed explanation: a) PHY is not in the list of abbreviations in clause 4. b) PHY is (unfortunately) not defined as a physical layer device/entity, although this was likely intended when the text was crafted. Instead, it is defined as an abbreviation for "Physical Layer" in the title of IEEE Std 802.11 and ISO/IEC 8802-11. Both aforementioned standards (with titles) are listed in the normative references in clause 2 of 802.1Q (i.e., in the same document). [I also had a quick look into 802.11-2016: PHY is also listed in clause 3 as abbreviation for physical layer]. With this definition of PHY in the 802.1Q document context, network media itself is currently part of the physical layer(PHY) in the context of 802.1Q - network media is included in PHY. Thus, there is currently no proper/unambiguous definition of a boundary by the text in question (technical problem), although an unambiguous definition was the intent when this text was crafted. c) "network media" is a problematic term itself. In the overall context, it is often used as a plural form of "medium". The boundary referred to in 802.1Q seems to be of a single one (i.e., single Port) most of the time (-> "medium"). To give an example of usage usage, "network media" is used in 8.5.1 of IEEE Std 802.1AS-2020 as the plural form of medium. It is further listed in 3.26 of IEEE Std 802.1AS-2020 in singular form with addition of the term "physical": "network physical medium". I guess "network" was added in IEEE Std 802.1AS-2020 to avoid some ambiguity within this standard. However, it seems preferable to just use "physical medium" at the affected places in 802.1Q - at least I am not aware of any ambiguity in 802.1Q that would require the extra word "network", and 6.22 of IEEE Std 802.1Q-2018 uses "physical medium" without the extra word "network" as well.
Proposed text
Options for discussion: 1) Avoid using "PHY" - one set of replacement phrases to start discussion could be the following: "boundary between the physical medium and the station [Port]", "boundary between the physical medium and the end station [Port]", "boundary between the physical medium and the Bridge [Port]". This would be quite similar to the definitions of the "reference plane" in 802.1AS (3.26 and 2nd sentence in 8.4.3 of IEEE Std 802.1AS-2020). By this, it may also be possible to avoid talking about a "reference plane" in 802.1Q at all ("reference plane" is always found before the text in question in 802.1Q, but it is likewise undefined in 802.1Q). 2) Add a definition of PHY such as "A device that connects a Port with a physical medium" or similar, and keep PHY in the text in question. 3) Add two definitions such as "PHY" and "PHY device"/"PHY entity", where "PHY" is "Physical Layer", and the other is as defined in 2). If 1) is accurate enough (e.g., if physical medium has clearly determinable "start" and "end" points), this may be the better choice instead of 2), because 2) results in two different definitions of the abbreviation "PHY" in the same document (802.1Q). Option 3) avoids two different definitions of the abbreviation "PHY" in the same document, however, it may add more text than needed to resolve the ambiguity.
Impact on existing networks
Likely none
Originator
Name: | Johannes Specht | Email: | johannes.specht.standards@gmail.com |
Affiliation: | PNO e.V. | ||
Submitted: | 2021-03-04 |