Jump to content

Talk:RS-232

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Add the 9-pin connector here?

[edit]

(This next comment was originally a comment to the preceding section. Since this is a different discussion - "details of 9-pin connector should be added to this article" rather than "why aren't the DE-9 connector ever added to the TIA-232 standard", I have split the next comment and all that follow into a new section. Jeh (talk) 21:11, 14 March 2019 (UTC) )[reply]

I don't know why, but this article should list 25 and 9 pin in the same table, because 9 is the most common in this world. Quote from this article - "TIA-574 (standardizes the 9-pin D-subminiature connector pinout for use with EIA-232 electrical signalling, as originated on the IBM PC/AT)". • Sbmeirow • Talk • 14:50, 10 March 2019 (UTC)[reply]
Re the DE-9 connector: this has been proposed before, and the arguments for exclusion still seem valid. Articles should be about their subjects. Do you see the article title? Does it say TIA-574? No. The 9-pin connector is described in the Serial port article, to which this article links. Jeh (talk) 17:39, 10 March 2019 (UTC)[reply]
It doesn't matter how many times it's been proposed, because any topic can be rediscussed on Wikipedia. There is nothing wrong about putting a comparison column (DE-9 (TIA-574)) against another standard, unless you can quote a specific rule against it. Knock off the snarky comments about the title. • Sbmeirow • Talk • 02:15, 12 March 2019 (UTC)[reply]
Yes, any topic can be rediscussed, but it seems pointless if someone doesn't have a new argument to bring up. It's not snark to point out that articles should be about their subject. The subject here, as dictated by the article title, is very narrowly specified. Discussion of topics not covered in the RS-232 standard are off topic. I believe the term "subject creep" or "topic creep" came up the last time. It is fine to mention that other, related standards and topics exist - particularly in the "See also" section - but adding the comparison column you suggest is a bridge too far. It's too much detail about a not-topic.
Can you explain why the coverage in the Serial port article not sufficient?
Alternatively - Heck, why don't we add the corresponding pinouts for RS-422, RS-423, EIA-530, etc.? They're all related. Jeh (talk) 04:12, 12 March 2019 (UTC)[reply]
RS232 via "DB9" connector is far more popular than all of those others, including the "DB25" too. • Sbmeirow • Talk • 14:36, 13 March 2019 (UTC)[reply]
Popularity is not the point. Precision in article topics is the point. WP:COMMONNAME pertains to article titles (it's why this article is still titled "RS-232" instead of "TIA-232-F"), but not to content. The Serial port article covers the information about the DE-9 connector. Why do you think it should be duplicated here? (Note: A "DB-9" would be the size of a DB-25 but populated with only 9 pins.) Jeh (talk) 14:58, 13 March 2019 (UTC)[reply]
1) Popularity is the point, per my comments above. 2) Numerous articles have been merged on Wikipedia, making the smaller article a section within a related larger article, thus making this point moot. 3) TIA-232-F redirects to this article, making your point moot. 4) I didn't say duplicate Serial port article here, I said "add a column". Many topics are compared on Wikipedia, making this point moot. Just because something is stated in one article doesn't automatically mean it must be excluded from another article. 5) I put quotes around the common name "DB9", though you purposely overlooked the quotes. (Note: I already knew about DE-9). • Sbmeirow • Talk • 05:16, 14 March 2019 (UTC)[reply]

5. "Purposely overlooked"? No, I inadvertently ignored your quote marks. I thought your use of "DB-9" was a mistake rather than deliberate. But it's an error either way. If you were making a point, it was too subtle for me to see.

4. Adding a column describing the pinout of the 9-pin connector would be significant coverage. Yes, many things are compared on Wikipedia, in articles with titles appropriate for such coverage. For example, the Serial port article includes a table that compares the pinouts of many connectors that have been used to carry RS-232 and similar signals for serial ports. That is appropriate for an article called "Serial port". It is not appropriate for an article called "RS-232" unless the standard is modified to include significant coverage of that connector.

"Just because something is stated in one article doesn't automatically mean it must be excluded from another article." No, it doesn't. But it does mean that the material in question already has an appropriate place on Wikipedia. And if the material is not within the second article's topic area, then copying the material to the second article is not appropriate. The topic here is the material in the standard originally known as RS-232.

3. I don't think you know what "moot" means. A "moot point" is one that is valid, but inapplicable to the argument at hand. For example, if you said "2 plus 2 equals 4" and I replied with "yes, but 2 plus 1 equals 3", as if that was a counter to your claim, then I would be making a moot point: 2 plus 1 does equal 3, but that fact doesn't refute your claim. As to this case, if anything is moot it's your observation that the TIA-232-F redirect exists. The existence of the redirect does not mean that WP:COMMONNAME is not being followed. WP:COMMONNAME is about titles of articles, not titles of redirects. WP:COMMONNAME even suggests that redirects be created for alternative names.

2. Yes, article merges happen. If the result is an article with a broader subject, then the article title is almost always changed to reflect that shift. Are you proposing a merge? From what article? - noting that there is no article for the TIA-574 standard. But there is no need, because the broader-topic article you're looking for already exists. It's called Serial port.

1. Popularity of the 9-pin connector is apparently your point, but it is irrelevant. The article topic here is not "serial ports that are pretty much compliant with RS-232, or at least similar in some ways". The topic here is the RS-232 standard. That standard is described in a document. That document does not mention the 9-pin connector. Therefore a serial port with a 9-pin connector is not compliant with RS-232, and this article should not give significant coverage to such ports. But! Nobody is disputing that the 9-pin connector is popular. Accordingly, it is covered in the Serial port article. Which is pointed to by a hatnote here: "For RS-232 variants, including the common 9-pin connector, see serial port." You still have not described why you consider that coverage insufficient. Jeh (talk) 19:12, 14 March 2019 (UTC)[reply]

I'll come back to this sillyness later. • Sbmeirow • Talk • 10:27, 16 March 2019 (UTC)[reply]
If you could not make any more WP:POINTy edits to the article in the meantime, that would be good. Jeh (talk) 23:19, 16 March 2019 (UTC)[reply]
Reverting my edit was hypocritical, per your comments above. Also, you don't own this article, though more proof that you act like it. • Sbmeirow • Talk • 01:52, 17 March 2019 (UTC)[reply]
Since no one else is commenting, it's a waste of my time to continue this thread. • Sbmeirow • Talk • 02:28, 17 March 2019 (UTC)[reply]
<more than a year later> I've made the annual revert of the IBM AT 9-pin connector addition to the table. Or has whoever's taken over the standard snuck it into the latest edition? Last one I saw was rev F and no 9-pin connector was standardized there. Back in the day there were many 9-pin arrangements that complicated our lives immesnely. --Wtshymanski (talk) 04:02, 6 August 2020 (UTC)[reply]
I didn't come here to add this info, but after reading this conversation I have perspective to add. All else being equal, I feel like including the DE-9 pinouts with e.g. an asterisk and a footnote is not unreasonable as a matter of pragmatism, but I see the argument that this page is purely about the standard. What I also see though is that this article appears to contain plenty of content that isn't about the standard.
The paragraph beginning "Since the standard definitions are not always correctly applied[...]" and virtually the entire section on "Cables," I think, describes elements outside of the standard, which explicitly states that pinouts are immutable. Text elsewhere in the article mentions the DE-9 repeatedly, mentions other DB-25 standards in case of confusion, mentions use of RS-232 in the PC and USB displacing it (arguably irrelevant to the standard itself,) lists other standards somewhat related to RS-232, and so on.
I am not speaking hyperbolically when I ask if that should all be removed on this justification. I would be happy to do the work of carefully excising it all and relocating it as needed - I was already looking at editing this page because I felt it was somewhat lacking in clarity. If removing everything that isn't core to the spec and replacing it (if appropriate) on the serial port page itself, treating this one as the theoretical and that one as the practical, would resolve this dispute, it seems worthwhile to me. Otherwise I think this argument (explicit or implicit) will continue until the focus of the article is made clear.Gravislizard (talk) 21:19, 14 August 2020 (UTC)[reply]
Go for it. Articles should be about their subjects. We've got articles for serial cable, the IBM serial port, game controllers, and each kind of Pokemon. --Wtshymanski (talk) 01:53, 15 August 2020 (UTC)[reply]
I have moved this material to Serial port#Pinouts ~Kvng (talk) 17:27, 7 May 2023 (UTC)[reply]

RS-232C

[edit]

RS-232C redirects to RS-232#History but the term is not described there. We have a similar problem with RS-422A. What do the suffixes mean? ~Kvng (talk) 16:19, 13 December 2022 (UTC)[reply]

The suffixes are just sequential identifiers, right?. For the RS-232 article, see the end of the History section for a publishing history of the standard. Somers-all-the-time (talk) 01:32, 14 December 2022 (UTC)[reply]
Sorry, I missed that since there's an extra hyphen there (EIA RS-232-C). ~Kvng (talk) 15:30, 16 December 2022 (UTC)[reply]

Connectors - Typo?

[edit]

"The standard recommends the D-subminiature 26-pin connector up to revision C" -- is this a simple off-by-one typo? It seems worth asking, as a D-sub connector of this count exists (DA-26, a "high density" three-row style), and it doesn't seem to be referring to the "ALT A" type (which is mentioned earlier as added in revision E). This seems at odds with the DB-25 that is commonly seen and regularly mentioned in the article; and at least one revision of RS-232 I have seen, does not mention such a connector (but I don't have all revisions to exhaustively confirm this).

Also, it's not clear what citation (or more specifically, revision and page) is controlling here, as no citation is used in this section (perhaps one should be added?). So again, prudent to check before making an edit... 98.144.229.53 (talk) 14:00, 13 November 2024 (UTC)[reply]

7E1 ASCII vs 8N1 UTF-8/ISO/IEC 8859-1

[edit]

@Sbmeirow: I assume you would want to keep "historic" context here re ASCII vs UTF-8. But as I mentioned in the comment, ASCII would be wrong. Either its 7-bit ASCII encoded 7E1 (which the diagram should have called out as parity instead of b7 as it does call out start/stop bits) or its actually 8N1 encoded. As ASCII is 7-bit, it could only be some 8-bit variant (ISO/IEC 8859-* or code page). UTF-8 is just the most common one. Aeroid (talk) 18:26, 18 November 2024 (UTC)[reply]

The original "description" of the images when the photos were uploaded to Wikimedia Commons clearly describes the waveforms to be 8N1 with no parity bits.
Single byte "ASCII" is the defacto common term for RS-232 discussions since the 1970s. Since computers don't have 7bit wide memory, it means that 8bits is the common way to store it, and often transmit it too, even if it isn't the most efficient way to transmit or store it. Just because ASCII data is commonly stored or transmitted in 8bit "containers", it doesn't magically mean that it should be described as something else!
"ASCII" descriptions in hardware discussions should not be automatically converted to "UTF-8", because then readers may think that its ok to send long-bit unicode text characters to devices that assumes all text characters are one-byte ASCII characters. Just because UTF-8 is the defacto standard for webpages, it doesn't automatically mean everything else is now magically UTF-8 too. Please don't redefine these terms in hardware descriptions!
So are you going to restore the word to ASCII for this image, or do I need to do it for you?

• Sbmeirow • Talk • 22:35, 18 November 2024 (UTC)[reply]

That's what I meant with "historic". Fine with me.
I've forked the SVG for a 7E1 version, which adds parity to teh visual and is 7-bit ASCII w/o any implicit "container". The new SVG is also much cleaner and smaller. Feel free to use that.
File:RS232 Oscilloscope Trace K-7E1.svg Aeroid (talk) 19:30, 22 November 2024 (UTC)[reply]
RS-232 does not standardize anything about the transmission format other than voltage levels, slew rates, and pinout functions. Hyopthetically an RS 232 link could transmit 5-bit Baudot,TWX, FIELDDATA, EBCDIC, or Morse code. This discussion is out of place here. --Wtshymanski (talk) 00:47, 1 December 2024 (UTC)[reply]
Retrieved from "https://en.wikipedia.org/w/index.php?title=Talk:RS-232&oldid=1260483941"