Talk:ASCII

Backspace to form diacritics.
This article should perhaps somehow mention the ability of many old-school printers to form diacritics etc. using backspace. Since ascii does not have characters like "é". E.g.:

é = e backspace '

e = e backspace e

e = e backspace _

212.178.135.35 (talk) 11:40, 2 August 2022 (UTC)
 * Can you provide a citation? Or even name a printer that could do it? Typewriters certainly could and did, but it seems that there were very few printers that could do it. It was for this reason that tilde, circumflex ("caret") etc got repurposed. For the long explanation, see tilde. --John Maynard Friedman (talk) 11:51, 2 August 2022 (UTC)
 * See also Backspace. --John Maynard Friedman (talk) 11:53, 2 August 2022 (UTC)


 * This was certainly the intention when ASCII was standardized, but I believe there was very little implementation of this. Hardware that would move the printhead backwards by one character would add mechanical complexity, and for video terminals being able to overprint two characters would double the memory needed (which was extremely expensive at that time). Hardware overprinting was mostly done by using CR and then printing the entire line of accent marks, and this was pretty rare.Spitzak (talk) 00:53, 3 August 2022 (UTC)

Cleanup
I made some bold removals as uncited opinion such as:
 * 1) Most modern character-encoding schemes When was this written? Unicode is the only modern encoding system in use today;
 * 2) ASCII has practically speaking been replaced (because of limited language support), e.g. with extended ASCII encodings, and most recently by Unicode (which supports all languages); its ASCII-compatible UTF-8 encoding (which is dominant on the web). ASCII only supports English (and few minority lanuages) and doesn't handle e.g. many loan words or given names of all American people. because
 * 3) "limited language support" is not the reason why ASCII was replaced;
 * 4) "extended ASCII" is a misnomer for ISO 8859-1. We should not feature errors in the lead, given that the true story is in the body;
 * 5) the reference to UTF-8 is way too detailed for the lead.
 * 6) replaced some instances of cite journal that I guessed used to be cite document (which atm redirects to cite journal). I guess someone is doing a cleanup in preparation for releasing cite document to do what it says on the tin. I used cite techreport, which is not ideal but cite standard has other issues.
 * 7) I severely edited Despite being an American standard, ASCII, unlike e.g. modern UTF-8 or other extended ASCII supersets, doesn't support symbols such as the cent, ¢ (or €, ©), though it does support the dollar, $..
 * 8) UTF-8 is entirely irrelevant in this context
 * 9) € is not a US native character
 * 10) ¢ is the only really serious omission but I suspect that this was another case (like ~) where the designers hoped it would be met by backspace and overtype.
 * 11) © is just one of many symbols in common use that are not supported. So we give all or none.
 * 12) Middle English: predates 1776, I think? Anyway, just makeweight.

Obviously WP:BRD applies but anyone reverting needs to reinstate the CS1/2 fixes I applied.

An observation: the lead should summarise the body but looks to me to be thin on the technical content? 𝕁𝕄𝔽 (talk) 17:07, 7 November 2022 (UTC)

Need better choice for diacritics
The word resume seems to be a poor choice as an example of the need for diacritics

https://novoresume.com/career-blog/how-to-spell-resume DGerman (talk) 20:36, 4 April 2023 (UTC)


 * I read that word entirely wrong and was confused until I spotted the word `career' in the provided link. Yeah, people ignore the spelling if the context is right...  I suppose I mostly wonder if you have a better example. Vollink (talk) 20:32, 27 September 2023 (UTC)

ANSI X3.4-1965 References
As best as I can tell, any statements about what was IN this never-published standard have come from the IBM manuals ... "used by IBM 2260 & 2265 Display Stations and IBM 2848 Display Control". From what I've read, though, those manuals do not actually claim to be following an unpublished standard, but merely refer to ASCII and go into how their use differs from the '63 standard.

That is, the table's claim that certain characters moved and then moved back are quite suspect. Did I miss a reference here, one that doesn't in-turn point back to this article (the dreaded circular fact-check that is quite the plague). I've read everything I can, but I just can't find the actual data to support these claims.

I guess what I'm saying is - it is notable and given the platform - quite useful to know that IBM deviated from the standard and how, but it does not seem worthy of claiming it is from an "approved but never published standard" without putting in a BIG caveat, you know, that cool NEEDS A REFERENCE edit-mark.

It's heartbreaking because the '67 standard makes it PAINFULLY clear that a '65 standard DID EXIST, but ... that is just lost to history I guess. I was hoping the NYPL business library archive might have something on this, but it seems that ANSI never left anything with the library (Yes, I asked). [ four paragraphs, single comment, sorry for the length ] Vollink (talk) 21:23, 17 September 2023 (UTC)

Old Latin
I'm not confident about this, but wasn't Old Latin written with a subset of the alphabet without diacritics? As opposed to Archaic or Classical Latin. DAVilla (talk) 02:11, 20 January 2024 (UTC)