WikiProject Computing / Networking | (Rated C-class, Mid-importance) |
---|---|
WikiProject Telecommunications | (Rated C-class, High-importance) |
---|---|
The media itself
It is not clear that for instance a cable itself is considered physical layer or that only the signals are physical layer. When doing a bottom-up aproach based on the OSI model the first step is to check the wiring. (According tot Cisco CCNA 4.0 module 1) So, is a cable itself considered physical layer or not? —Preceding unsigned comment added by 85.145.222.194 (talk) 22:23, 26 May 2010 (UTC)
rewriting of article
Listen, I'm considering rewriting this article in the following way:
- removing the discussion of 2B1Q (perhaps mentioning it as an example) and the "sublayer"
- addressing the fact that the medium can be electrical, RF, or optical
- addressing the complexities of physical layer signalling by linking to articles on: (assuming these exist)
- adding a short discussion of physical layer interfaces in a typical PC
In general, I feel that the existing article somewhat confuses the role of the interface and overly simplifies physical layer processing as mere electrical pulses on a wire. This may be true for old standards, but certainly does not hold for any physical layer standard developed since the mid-1980's.
Any comments? RobertYu 21:41, 15 Sep 2004 (UTC)
- Good points. I have tried to address them. Mange01 00:16, 9 November 2006 (UTC)
physical mail system
Would not a better analogy for the physical mail system be the vehicles that transport the mail, rather than "a specification for various kinds of paper and ink"? BevanFindlay 22:16, 18 July 2006 (UTC)
- Agree. Or perhaps the entrance to roads that transport the vehicles that transport mail? Mange01 00:16, 9 November 2006 (UTC)
- Agree, changed to roads. Conrad.Irwin 15:48, 14 December 2006 (UTC)
missing embedded system physical layers
I'm missing embedded systems in this context. If I'm looking for ISO/OSI there is only information about TCP/IP and PC's and so on. What if I use a simple microcontroller and want to comunicate with other hardware on a PCB?
Yes, we need to mention I²S, I²C, Microwire, Serial Peripheral Interface Bus, System Management Bus, 1-Wire, X10 (industry standard) ... please add them to the article. At least the article does mention EIA-485, which is one of the more popular embedded system interfaces.
- Done ~ Kvng ( talk) 17:38, 25 January 2016 (UTC)
names for the various parts
Is there a standard pinout for putting EIA-485 on a RJ-11 6P4C or RJ-45 8P8C connector? (Does 10BASE-T specify this, or are the signals on a 10BASE-T cable *not* EIA-485 signals?)
I would like to fill in the holes in this table:
- "RS232D", aka "EIA/TIA-561", is the standard for "RS-232 on a
RJ-458P8C connector" - EIA-530 is (?) the standard for "RS485 on a DB-25 connector"
- ???? is the standard for "EIA-485 on a DE-9 connector"
- ???? is the standard for "EIA-485 on a
RJ-458P8C connector" - RJ50 is the standard for "???? on a ???? connector"
- 10BASE-T is the standard using ???? on a
RJ-458P8C connector.
This table distinguishes "the electrical voltages" from "the mechanical plug shape and how the wires are twisted" from "some standard that defines which voltage goes on which physical conductor". Is there a name for these 3 things? Yes, OSI model calls all 3 things taken together the "Physical Layer", but is there 3 different names for these 3 part?
--65.70.89.241 17:35, 2 August 2006 (UTC)
- First, you'll want to correct the naming of "RJ-11" and "RJ-45" connectors. You mean modular connectors. RJ11 and RJ45 are telephone wiring standards. (Also, they don't have hyphens). I have recently updated some of the wikipedia articles on these to make that more clear. The RJ50 article is apparently incorrect in this same way, but I haven't had a chance to look up what an RJ50 registered jack really is.
- Bryan Henderson 17:50, 9 September 2006 (UTC)
- I've seen some pinouts that could be used for EIA-485 on a 6P4C or 4P4C connector [1], giving 1 pair for data and 1 pair for power, apparently defined by the CANopen standard.
- Alas, when I try to fill in the small gaps in my knowledge, I start to realize that what I don't know is vastly larger than I thought it was.
- Thank you for pointing out those things are called modular connectors.
- Fixed the connector styles ... anyone know how to fill in the remaining gaps? -- 68.0.124.33 ( talk) 05:31, 17 July 2008 (UTC)
Lack of references.
I tagged this article for lack of references because the only reference given does not touch upon most of the subject matter in the article and computers, networking, and telecommunications is an area of expertise where it is easy to have misconceptions which, if material is unreferenced, will work their way into WikiPedia. I have 25 years of experience in this field an still find that I suffer from the occasional misconception concerning the details of some subjects. I have discovered some of these misconceptions while looking for references to articles that I wrote on WikiPedia so I am sure that finding references is a Good Thing :-) --mlewis000 00:56, 20 December 2006 (UTC)
- I changed this to a refimprove tag, as there are 3 refs now. Widefox ( talk) 17:07, 21 October 2009 (UTC)
Vote: Four and/or five layers in the TCP/IP model template and wiki articles?
Give your vote here. Should the TCP/IP model template have four or five layers? I.e. should the physical layer be a separate layer or not? And what should the the bottom layer be named in case of four layers? And is it okay to mention both the four and five layer TCP/IP models in Wikipedia articles? Mange01 (talk) 18:24, 17 July 2008 (UTC)
Requested move
- The following discussion is an archived discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section.
The result of the move request was no consensus for move. Aervanath lives in the Orphanage 04:26, 8 January 2009 (UTC)
I believe this move should be undone (the move was made along with several other {{OSIModel}}-related moves). WP:MOSCAPS issues can be tricky, but in my mind the issue boils down to whether this article is OSI-specific.
This article covers both OSI and non-OSI issues — point-to-point and bus topologies aren't OSI, but still have PHYs. While it might be possible to split the article into two parts (a general-purpose physical layer, and an OSI-specific Physical Layer (OSI) or somesuch), it seems far more sensible to leave this as a single article that covers both OSI and non-OSI issues, and name it 'physical layer' to reflect its general-purpose nature. --Underpants 15:20, 2 January 2009 (UTC)
- The main focus of this article has long been the layer of the networking model of OSI. 'Layers' don't exist by themselves, a layer is only a layer as an entity within some framework. I doubt that the term 'Physical Layer' ever causes anyone to not associate it with the OSI model. I would say that in other uses the term 'hardware level' is more common. If there are examples in this article that for some reason do not qualify as OSI-relevant technologies, they should probably be moved to the PHY or other fitting article. Kbrose ( talk) 16:37, 2 January 2009 (UTC)
- The above discussion is preserved as an archive of a requested move. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.
Sublayers
A section about the Physical Layer Signalling (PLS) sublayer was added. That is however only used in 10Base-T Ethernet, and may be moved to that article, or described in a wider context. Different IEEE protocols divide the physical layer into several sublayers. In e.g. 100 Mbps and 1000 Mbps Ethernet versions, the PLS sublayer is replaced by a Physical Coding Sublayer (PCS). See for example the IEEE 802.3a model (page 4 at http://grouper.ieee.org/groups/802/3/ae/public/mar00/booth_1_0300.pdf).
Perhaps someone would like to make a table, with one column for each protocol version, where all sublayers are presented. You may start out from the above pdf file. Mange01 (talk) 12:37, 14 April 2010 (UTC) sss —Preceding unsigned comment added by 203.126.136.142 (talk) 09:20, 5 August 2010 (UTC)
Media access control
This article incorporates public domain material from the General Services Administration document: "Federal Standard 1037C". says that media access control is handled by the physical layer. Data Link Layer says it is handled there. Data Link Layer has only one reference which does not cover this question. I'm inclined to believe the federal standard. --Kvng (talk) 13:45, 29 July 2011 (UTC)
- Don't. IEEE invented the MAC sub-layer. It's mapping to the data link layer is found in IEEE Standard 802-2001, where it is mapped beneath the LLC sublayer in the data link layer. There is nothing physical about the MAC sub-layer: it has addresses for heaven's sake. — Dgtsyb ( talk) 23:28, 29 July 2011 (UTC)
- Actually, FS1037C does not say that media access control is handled by the physical layer: it says that it performs its function using the services of the physical layer (duh). — Dgtsyb ( talk) 23:33, 29 July 2011 (UTC)
- Can you please add a ref or two to Data Link Layer? -- Kvng ( talk) 13:50, 31 July 2011 (UTC)
- Data Link Layer already refers to IEEE 802 as the source of the MAC sub-layer definitions. Also, note that IEEE 802is the ISO data link layer (ISO/IEC 8802), just add 8000 to the IEEE number to get the ISO/IEC number. — Dgtsyb ( talk) 07:24, 2 August 2011 (UTC)
Requested move, multiple
- The following discussion is an archived discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section.
The result of the move request was: Move. Jafeluv (talk) 17:33, 26 October 2011 (UTC)
- Physical Layer → Physical layer
- Data Link Layer → Data link layer
- Network Layer → Network layer
- Transport Layer → Transport layer
- Session Layer → Session layer
- Presentation Layer → Presentation layer
- Application Layer → Application layer
- Link Layer → Link layer
– The TCP/IP and OSI layering models are not based on proper names of entities. The layers are discussed here as generic groupings, and compared across the different protocols. Many (possibly most) books have them in lower case; some mixed (uppercase "Physical" and lowercase "layer"). In general, there is little or no evidence for the acceptance or usage of these terms as proper names, even in books that capitalize them by convention (see some preliminary discussion on this at Talk:OSI model. WP style is to capitalize only proper names, not things that others capitalize for other reasons. Dicklyon (talk) 03:44, 19 October 2011 (UTC)
- Support. Investigation of the articles confirms what the proposer of the RM suggests. These are ordinary entities referred to by generic descriptive means. There is no question of proper names – if that is taken as the best or only determinant of capitalisation, which is itself debatable. In its very first sentence WP:MOSCAPS gives plain guidance, and it can be applied directly here: "Wikipedia avoids unnecessary capitalization." An aptly restricted Googlebooks search on "physical layer" "session layer" (restrictions: "Books›Jan 1, 2000–Oct 18, 2011›Search English pages›Preview and full view") shows a marked preference for lower case in "layer", except in headings. So these capitalisations must be regarded as "unnecessary", and best avoided on Wikipedia. NoeticaTea? 04:41, 19 October 2011 (UTC)
- Support as one of the usual suspects. Noetica's contribs list is spied on daily by many editors. There is utterly no reason to capitalise the l in layer, and as he has pointed out, our long-standing house rules are to capitalise only if there's a very good reason to do so. Caps seem to be flung about all over the place in some fields and professions—perhaps it makes people feel "safe". We have more confidence here in plain English, in ordinary old common nouns that don't need dressing up. Please, let's not poke readers' eyes out with redundant capitals: it's soooo 1970s. Tony (talk) 06:53, 19 October 2011 (UTC)
- Support, obviously, as I initiated this discussion. Note that this is about lower-casing both the word 'layer' and the type of layer ('physical', 'transport' etc.). Mixed caps don't make any sense outside of the technical reasons for article names. This view is supported by the literature on the subject. --EnOreg (talk) 07:32, 19 October 2011 (UTC)
- Right, beyond the title fix, it would be all lower case in the article except at start of sentences or headings: "physical layer", as in many sources, but not in those with a more capitalizing style. Dicklyon ( talk) 12:24, 19 October 2011 (UTC)
- support, since most sources use it as a generic name, including Tannenbaum. --Enric Naval (talk) 17:42, 19 October 2011 (UTC)
- Support per usage in reliable sources. --Born2cycle (talk) 06:31, 20 October 2011 (UTC)
- Support -- SchreyP (messages) 06:25, 23 October 2011 (UTC)
- Support —Ruud 22:38, 25 October 2011 (UTC)
- The above discussion is preserved as an archive of a requested move. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.
I've worked on updating case in a few of these articles. Review and more would be appreciated. Dicklyon (talk) 18:34, 26 October 2011 (UTC)
Thanks for helping, EnOreg. In the process, I found other related bits of over-capitalization to work on, like at Talk:Internet Protocol Suite where I did an RM to downcase that one per typically usage and sources. Also the sublayers in some of these articles need work: Talk:Logical_Link_Control#Requested_move. Dicklyon (talk) 03:12, 27 October 2011 (UTC)