Talk:curses (programming library)

Ncurses

I created a subsection that links to ncurses. I'm moving the section that talks about the date to the ncurses page. — Preceding unsigned comment added by Aronzak (talk • contribs) 2008-11-04

The pccurs14 archives found on various FTP servers (or pccurs14.arc on the SIMTEL CDs) clearly indicate that ncurses already used the name ncurses before 1990, if the pccurs author got that right in his README.NOW. –Be..anyone (talk) 08:43, 10 February 2014 (UTC)

The original sources do refer to the name of the library as "ncurses", but the package overall was referred to as "pcurses". That's not an inconsistency, although perhaps a little obscure. TEDickey (talk) 09:07, 10 February 2014 (UTC)

I added this to the intro, from the ncurses man page:

ncurses is the approved replacement for 4.4BSD classic curses.

because I felt it was important to realize that it has effectively replaced curses. BMJ-pdx (talk) 19:55, 12 June 2024 (UTC)

Termcap

The article says:

curses was originally implemented using the termcap library.[citation needed]

The document ncurses-intro.doc, distributed with ncurses says:

Historically, the first ancestor of curses was the routines written to
provide screen-handling for the game rogue; these used the
already-existing termcap database facility for describing terminal
capabilities. These routines were abstracted into a documented library
and first released with the early BSD UNIX versions.

Is this valid as a citation? Do we need something more close to the origins of curses? 87.222.26.18 (talk) 19:35, 19 June 2009 (UTC)

No - it's written by an uninformed source, whom I'd quote seldom Tedickey (talk) 20:43, 19 June 2009 (UTC)

I would quote it. It is a primary reference, and something is better than nothing. An tag can be added if people is unconvinced of the validity. 189.235.173.179 (talk) 15:55, 24 January 2010 (UTC)

I'd tag it right after "rogue", since there's no reliable source that I recall for it. Tedickey (talk) 16:25, 24 January 2010 (UTC)

BSD Curses - Ken Arnold relationship

I've seen a revert against NetBSD BSD Curses being derived from Arnold's without a "reliable source", not that I care, but the link is very obvious:

• The manual page of the library lists him in the AUTHORS section (see [[1]])
• The source itself (with CVS history available) retain the original 1981 and later Berkeley copyright (see [[2]], even though NetBSD started importing the code from 386BSD in the early 90s (that import was from cgd)).

Thanks, 66.11.179.30 (talk) 21:46, 6 February 2010 (UTC)

Look at the CVS history - virtually none of the code base could be attributed to Arnold. The mention in the documentation is honorary rather than factual Tedickey (talk) 21:48, 6 February 2010 (UTC)

Relation to vi

Sometimes it is incorrectly stated that curses was used by the vi editor.

vs.

However, not all Curses-based software employs a text user interface which resembles a graphical user interface. One counterexample would be the popular vi text editor, which while not being CLI-based,

This is simply a contradiction. I won't update the entry because I am not quite the expert here.85.127.139.217 (talk) 08:49, 22 February 2011 (UTC)

no contradiction; there are two distinct statements: (1) the original vi did not use curses, rather the curses library was derived from vi, and (2) the appearance of vi differs from some applications which are written using curses to imitate GUI applications. TEDickey (talk) 09:30, 22 February 2011 (UTC)

Yes, but the second sentence containing the vi reference is under the heading 'Curses-based software'. What has (2) to do with this article except that it is a (misleading) general remark? Why not mention a curses-based modal text editor? (I know there aren't any but I hope you get the point.) (btw. thanks for cleaning up my entry to the talk page, I was a little hasty) 85.127.139.217 (talk) 17:50, 22 February 2011 (UTC)

There are curses-based text-editors which use menus (which would be an example, rather than a counter-example). vi is a counter-example since its interface is unlike a graphical interface. TEDickey (talk) 01:44, 23 February 2011 (UTC)

History

"Curses was originally implemented using the termcap library.[citation needed] A few years later, Mark Horton, who..."

A few years later than... what??? When did curses come about? There is not a clue in this article. —Preceding unsigned comment added by 12.19.148.2 (talk) 00:40, 8 March 2011 (UTC)

The previous paragraphs and reference give some context - looks like 1977 TEDickey (talk) 00:53, 8 March 2011 (UTC)

Ed Carp

The (nonreliable source) [[3]] says only

- January 8 Ed Carp has announced that he is working on a pre-ALPHA

release of the Linux curses library with support for function and arrow keys (hooray!) and ACS support (???). Send mail to Ed at

erc@apple.com if you would like a source copy ( ~100k compressed).

There is nothing in that, or any source currently available which supports the anon-IP's assumption that this material is sourced (or notable for that matter). At a minimum, a supporting source must support each part of the given statement, and anonymous comments don't really suffice either. TEDickey (talk) 08:54, 19 October 2012 (UTC)

It was widely known in the user community at the time of the curses work done by me, and lack of attribution is in no way "vandalism" or any less valid. There are many assertions in the curses and ncurses wiki pages that are not attributed, but are allowed to exist. 12.176.206.100 (talk) 16:10, 27 October 2012 (UTC) Ed Carp, erc@pobox.com

So provide a WP:RS - I already looked, found none to support the disputed points. Without those, the edit is unsourced and apparently self-promotional. TEDickey (talk) 16:46, 27 October 2012 (UTC)

A recent edit added a link to a manufactured source. That's not suitable as a reliable source TEDickey (talk) 22:10, 15 October 2021 (UTC)

PC Curses

The files in the http://purl.net/xyzzy/dos/pccurses.zip link are dated October 23, 2000, rather than 1993 as the edit indicates. The inclusion in the "Different lines of development" comment, is problematic because the wording makes it appear that PC Curses was not derived from pcurses TEDickey (talk) 09:27, 10 February 2014 (UTC)

The DOC/CURSES.NEW file in the archive has the 1993 date, the archive itself has the 2010 date reflecting the newest file. There is also a smaller cursesrc.zip (174 instead of 589 KB) with the same DOC/CURSES.NEW, and without binaries, but I guess the name of the bigger pccurses.zip is better for the purpose here (= reference for the Bjorn Larsson libraries.) Looking for version 1.4 I also saw something claiming to be version 1.5, but I didn't look into it, you wanted a reference for the date, not newer stuff. –Be..anyone (talk) 12:26, 10 February 2014 (UTC)

I overlooked that (there's no dates in the file; 1993 appears in a comment in two of the files copied from ncurses around 1995). However, that source doesn't appear to be useful (since it has been modified without detailing the dates associated with the changes, and there appears to be no way to obtain that information). If you refer to http://sourceforge.net/projects/pdcurses/files/pdcurses/, you can see that there are versions of PDCurses and PC-Curses. The 1.4 there matches (aside from uppercase filenames) the other source. Piecing the details together on Wikipedia would be original research. I don't do that. Lacking an external/knowledgeable source (I found none) which does that work, there's not much to go on. Reviewing what I did find, I'm left with the impression that there was little or no continuing development on PC-Curses past 1990, and that the later version you cited was a one-shot by some unidentified developer. For practical purposes, that line of descent was subsumed by PDCurses in 1992. TEDickey (talk) 23:13, 10 February 2014 (UTC)

PC curses had nothing to do with PDcurses, tip: compare the code. The sources are still published, SIMTEL was mirrored worldwide, and FWIW I have the two SIMTEL CDs (rather expensive about 20 years ago.) The 1993 reference in a 2000 ZIP is also wayback. –Be..anyone (talk) 02:10, 11 February 2014 (UTC)

NTcurses based on PCcurses. MINIX used PCcurses (google cache, direct access forbidden).[4][5]. –Be..anyone (talk) 04:08, 11 February 2014 (UTC)

Another (commercial?) library I vaguely recall is "AEwindows curses", find "AEwindows" on this horrible page, allegedly by "Aspen Scientific". –Be..anyone (talk) 02:38, 11 February 2014 (UTC)

Sure - read the notes in pdcurses 2.0 - its "rewritten". Makes for some interesting comparisons. Wikipedia is not the place to introduce that, lacking an external review which provides the analysis. ntcurses is derived from pdcurses, according to its readme (I have a copy, for reference). Again, analysis (requiring a WP:RS is lacking). AEwindows is perhaps something to investigate. (you omitted mini in the text - similar comments would apply) TEDickey (talk) 09:14, 11 February 2014 (UTC)

By the way, Pavel Curtis's posting to usenet of the sources named the package "pcurses". Likely, the author of the code knew better what his program was named than someone making a derived work from the code, and that's how the majority of sources read TEDickey (talk) 09:22, 11 February 2014 (UTC)

The given NTcurses link says PCcurses, I don't know the package. The pcurses vs. ncurses business might need a better explanation, the "ncurses in 1993 vs. uses of this name in 1990" confused me. For the PC curses case just not talking about the name will do, after all we're sure that it's the same library by Pavel Curtis. One difficulty here might be that you have a UseNet background (BSD, *NIX, Linux, etc.), while I come from a FidoNet background (DOS, OS/2, SVR3.2), knowing GiGo, UUCP, or Emily Postnews only from a huge distance determined by modem speeds in the early 90s. –Be..anyone (talk) 22:47, 13 February 2014 (UTC)

hmm - I've done work on OS/2 (and MS-DOS), starting around 1990. But the Usenet aspect is a reasonable assessment. The last edit to the topic looks okay. Offhand, I've not come across anything substantial to use as sources for the interval between say 1986 and 1993, perhaps because I overlooked the PCCurses thread until this discussion. The earliest I touched that are - referring to changelogs, was using PDCurses 2.1 in October 1993. The sources that I did find recently are all of the sort which require analysis to use - takes a lot of time to develop that TEDickey (talk) 23:38, 13 February 2014 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on Curses (programming library). Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20140223040754/http://read.pudn.com/downloads88/sourcecode/windows/339526/PCCUR/README.NOW__.htm to http://read.pudn.com/downloads88/sourcecode/windows/339526/PCCUR/README.NOW__.htm

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 16:37, 15 August 2017 (UTC)

Crediting Mary Ann Horton

The reliable sources all state "Mark Horton", because that was factual at the time Horton did the work in question. The existing redirect provided the necessary information TEDickey (talk) 08:08, 17 July 2019 (UTC)

The reliable sources all state that "Mary Ann Horton" is Horton's name. The existing redirect provided an incorrect, older name for her. Using a no-longer-used name for someone is incorrect style, and is especially so when it is a trans person's deadname. If it is important to include that name to conform to the documentation from the time, the style expects it to be relegated via a use-mention distinction, in a subsequent clause or footnote. That would be an acceptable resolution. But the blunt naming of the person using their deadname is widely considered rude and offensive, as noted in the style guide cited. Kcrca (talk) 16:01, 17 July 2019 (UTC)

To be clear, I am willing to redo the edit so it references her birth name rather than uses it as her name, such as via a footnote. Kcrca (talk) 17:13, 17 July 2019 (UTC)

Actually (looking around for sources), all of the Usenet sources, the RFC, etc., say Mark. On the other hand, there's the personal website of Horton. By the way, some of the "Mark" sources are later than "1987". There's some after-the-fact revision reflected in the Horton topic. Keep in mind that Wikipedia is not a reliable source of information. TEDickey (talk) 01:06, 18 July 2019 (UTC)

When someone changes their name, their currently used name is used in referencing them. There is a Wikipedia page for Mary Ann. There is absolutely no dispute that this is her name, and thus should be used here. See: https://wiki.riteme.site/wiki/Wikipedia:Naming_conventions_(people)#Self-published_name_changes

Thalia42 (talk)

That's what's done in the Wikipedia topic, of course. It has no effect on the reliable sources which are being discussed here, since those precede the name-change by several years, and actually are the basis for the topic's notability TEDickey (talk) 07:57, 18 July 2019 (UTC)

The RS of the time are not the only relevant RS. When someone changes their name, they have a new name. Calling someone by their deadname is frankly offensive. I can be convinced that linking the name used at the time with their name now has value, making it worth a mention in a footnote, as in the style I keep citing and you have not yet responded to. People are generaly a RS about their own name, especially when that name goes into common usage as Thalia42 cites, for which some notice from you would also be appreciated. What is your suggestion? That I find a RS and cite it when I use her name? That seems both irrelevant (why is there a cite about a person's name in this article about curses?) and problematic (calling out her name change as somehow controversial, requiring a RS). There is no RS cite where The Matrix uses the current attribution (in the first sentence) and mentions the original attribution in a footnote. It is just there.

That is the pattern I will post in a day or two unless someone has a better idea. Tedickey, if that is not acceptable to you, we need to take it to a noticeboard to get other input. I firmly believe you are misreading the rules, partly because you seem not to be reading the rules I am citing, just repeating the RS rule as if it was the only one that applied. So please either (a) give me a better way to do this that follows the style (and basic courtesy), (b) engage in this discussion with the other rules we have cited, or (c) leave my upcoming change alone. Kcrca (talk) 08:28, 18 July 2019 (UTC)

Following your line of reasoning, you'll have to support your statement by getting the reliable sources changed as well. Have a nice day TEDickey (talk) 08:33, 18 July 2019 (UTC)

Are there any other rules that apply? Is it possible you are misreading the rule set by focusing on one rule? Can you point me to some clause of the RS rule that clearly (or even by a reasonable interpretation) overrides the rules we are citing? Help me out here. I don't want to think you misunderstand basic English usage and courtesy, but that needs a specific rule-based logic that shows how the rules we are citing end up with the conclusion you prefer rather than the one we do. Kcrca (talk) 15:58, 18 July 2019 (UTC)

This discussion would also be helped if you said what specific kind of data you were looking for in a source. Are you wanting a source that says her current name is changed from her birth name? A modern source that says the work was done by Mary Ann Horton? A source from the time that says the work was done by Mary Ann Horton? Which specific asserted fact is missing a needed reliable source? I have been assuming you wanted a source on the name change, but you haven't actually said which fact needs an RS. Kcrca (talk) 16:22, 18 July 2019 (UTC)

Any original source for Horton related to this topic says "Mark", because as I pointed out (and you're obviously aware), the work was done and completed years before the change of name (see this for example, making it 5 years). Several of the existing references for this topic support that comment. That said, the name-change is irrelevant to this topic. The redirect served the purpose of relating the two (and dwelling upon it in this topic would be undue). In other topics (e.g., Princeton College), redirects are used similarly, to help the reader see the discussion in context. TEDickey (talk) 08:29, 19 July 2019 (UTC)

Reading the Horton topic on the other hand, it's vague about 1987, and mentions that for some time both names were used. The reliable sources presenting technical work said "Mark", into the 1990s. That's indicating a deficiency in the Horton topic — but irrelevant to this topic. TEDickey (talk) 09:19, 19 July 2019 (UTC)

Show me a source for your argument that someone who is KNOWN to have changed their name is appropriately referred to by their prior name, because the reference uses a source from prior to their name change. That is incompatible with the Wikipedia rules on how people should be named. There is no argument about what her name currently is. (If you need sources for it, I would point you to the Wikipedia page for her.) So what's your argument here? That if you reference Marie Curie's research based on a paper from before her marriage you would use Marie Skłodowska because that was the author's name on the paper? That makes no logical sense. Thalia42 (talk) 21:53, 19 July 2019 (UTC)

So let me lay this out clearly. I want the page to use her real (and current) name, not her deadname. Your position is that, to do that, I have to find a RS that shows that she used that name at the time the work was done. Can you tell me how you see that in relation to MOS:GENDERID? Because I see that as clearly saying that, in almost all cases, you use the current name. It gives examples of where this is done, such as on The Matrix page or with Switched On Bach. According to this, and Thalia42's other reference, you seem to be asking for something that is not only not required, but is anti-required. Can you please comment on MOS:GENDERID? It is the specific rule I am citing, and you haven't once mentioned it. That feels like you are not taking my arguments in good faith, though I hope this is not true. [And thanks, Thalia42, for being here with me on this.] Kcrca (talk) 00:52, 20 July 2019 (UTC)

Thanks for the insight you have provided; I'll keep that in mind in case anyone in the future refers to this webpage TEDickey (talk) 13:38, 20 July 2019 (UTC)]

Thanks. I am taking you to have agreed that this is no longer in dispute. (If not, I am sorry and please mark it disputed again.) I have changed the title of this section of the talk page to be more descriptive, and will reference it in the updated article. Kcrca (talk) 02:06, 21 July 2019 (UTC)

No - you're putting words in my mouth, with all that implies. TEDickey (talk) 18:56, 21 July 2019 (UTC)

First, I stated explicitly that I might be wrong, and to inform me if I was. That was me stating my interpretation of your words, noting that it was an interpretation, and clearly stating that you might correct me. That is not "putting words in your mouth", it is "echoing back to you what I think you said". And yes, my confidence was fairly high. It turns out to have been wrong, so I will revert.

Second, what did you actually mean? Are you going to respond to my citation of Wikipedia style that says that my edit is correct and recommended? Do we need to move this to a noticeboard to get other input? Because I feel that you are completely ignoring my citation of Wikipedia rules, so that you are effectively acting in an "output only" manner. You have ignored my citation four times now, and ignored Thalia42's citation twice. You have not yet commented on this in any way. So if you are going to stick to reverting my changes without addressing my points, broadening the discussion to others is the only next step I can see. Kcrca (talk) 19:09, 21 July 2019 (UTC)

not everyone's on your personal schedule, or follows your orders. For someone talking about guidelines, you might consider reading WP:CIVIL and WP:AGF, rather than making a selective reading of guidelines which you prefer. In your preferred rendition, you have provided the reader no way to relate the references to that link. You continue to insist that any other way to do this is rude. The expressed attitudes fall into this one: WP:SOAPBOX, and ignore MOS:MULTIPLENAMES (it helps to actually read the guidelines rather than using them as a hammer). Have a nice day. TEDickey (talk) 07:47, 22 July 2019 (UTC)

This quote from the MOS relates to this discussion: "If a person is named in an article in which they are not the subject, they should be referred to by the name they were using at the time of the mention rather than a name they may have used before or after the mention." (Horton is not the subject of this topic) TEDickey (talk) 22:56, 22 July 2019 (UTC)

I had no intent to be disrespectful. I had hoped that my immediate revert after you told me I misinterpreted you would be evidence of my good faith. I'm sorry that I caused you to feel disrespected. Kcrca (talk) 04:13, 23 July 2019 (UTC)

As for offense, I offer the following sources ,,, and refer to you the page on Transphobia which has two references to back up the statement "Deliberately misgendering a transgender person is considered extremely offensive by transgender individuals." As stated previously in the article, "deadnaming" is a form of misgendering. So there is reason to view this deadnaming here as harm.

Back to the edit in question, thank you for addressing the cites. When you talk about confusion WRT the name, note that I am asking that the reference be changed to something like 'Mary Ann Horton (then known as "Mark Horton")' in a parenthetical or footnote. I think this is a clear improvement that reduces confusion. In the original version, the user was simply dropped on a page of someone with a different name. My proposal gives the reader a way to understand both modern and older references. In MOS:GENDERID, it says that references from other articles are to be driven by context. The biographical article recommendation is to always use the latest form. In this case I am arguing that possible confusion be addressed by clear explanation, but that respect is best served by using the latest form. The discussion in Gender_identity#Retroactivity gives a basic idea.

The piece you quote is general advice about name changes, only a few examples or of gender identity. And it also says "However, see MOS:IDENTITY." Which says "When there is a discrepancy between the term most commonly used by reliable sources for a person or group and the term that person or group uses for themselves, use the term that is most commonly used by reliable sources. If it isn't clear which is most used, use the term that the person or group uses." Which in this case is Mary Ann Horton, is it not? Kcrca (talk) 04:13, 23 July 2019 (UTC)

The point of the MOS is that the topic about the renamed person should be clear. But providing a parenthetical comment on each link to that topic goes beyond the advice given there (whether it is neutral or not seems to be a point of disagreement). TEDickey (talk) 20:57, 23 July 2019 (UTC)

I don't know why the neutrality would be in question, neither piece of data (current or birth name) is at issue. Can you explain? Also, clarity is not just an issue with the topic name. Here it is an issue with the name used to refer to the topic page, which is not the same as the topic name. That seems intrinsically unclear/confusing. That this name is also very likely to cause personal pain to the referent party and offense to many others, and there is a reasonable way to make it _not_ do so, seems relevant as well. Kcrca (talk) 16:09, 25 July 2019 (UTC)

Given the effort which you have spent to promote your suggested change, it's really hard to assume you are not joking. Looking forward to constructive discussion—later TEDickey (talk) 00:25, 26 July 2019 (UTC)

I don't know what you are trying to communicate here. What is there to joke about? What have I done to "promote" this that is problematic or spamish? Why would you think I was not serious? This is a very baffling comment. Kcrca (talk) 01:43, 26 July 2019 (UTC)

Support using Mary Ann Horton's current name in article. (Here via WikiProject LGBT) There's nothing factually incorrect about referring to someone using their present name instead of their name-at-the-time. Here's a New York Times piece where a picture of a pre-Obama Michelle is captioned Michelle Obama as a college student at Princeton. As far as what Wikipedia's policies state, there is no real guidance on this issue one way or the other. MOS:GENDERID provides guidance for "main biographical articles" but for other articles, it just has the wishy-washy phrase Use context to determine which name or names to provide on a case-by-case basis. There was an RfC a couple years ago to determine how to refer to a transgender person when they are mentioned in an article that isn't their main biographical article. The RfC asked whether articles should always use their name-at-the-time, or always use their current name. There was no consensus for either approach. WanderingWanda (talk) 09:15, 30 July 2019 (UTC)

I'll add: During the Wachowski RfC I compiled a list of reliable sources that chose to use Lana and Lilly Wachowski's current names when talking about their old work, instead of their names-at-the-time. (The Wachowski sisters are transgender women.) For example, The Globe and Mail, Canada's "newspaper of record" said this: Siblings Lana and Lilly were the minds behind the groundbreaking Matrix trilogy. That article does not mention their deadnames at all, not even parenthetically.

Anyway, this conversation seems to have run its course, and currently its 2:1 in favor of using Horton's current name. If anyone else has any additional perspectives on this, please share, but meanwhile I'll go ahead and remove the "disputed" tag. WanderingWanda (talk) 04:38, 7 August 2019 (UTC)

References

FMLI example

Noticing some clutter in the picture section, I checked the more obvious one and see that the FMLI example is contrary to the stated rationale (there are plenty of free examples of curses applications). The example itself is poorly chosen, since the application is for Solaris 10 which is (ahem) in "premier support" (see this comment for clarification of that misleading term). Solaris 11 uses a different approach, which is not relevant to this topic. A good sample picture of curses would use some newer feature than FMLI, which was last documented in 1993. That is, color and line-drawing characters are assumed by the typical reader. TEDickey (talk) 00:22, 26 July 2019 (UTC)

misleading history comment

A recent change inserted an anachronistic comment about deprecation, stating (in the context of AT&T curses) that Keith Bostic deprecated it in 1995. The source given is weak because it is second-hand information (what Bostic actually said is unknown). More to the point, Bostic had no relationship with AT&T, but was one of the developers with CSRG, which was supplanted by other organizations in 1994/1995. Whether CSRG deprecated its curses library is moot; NetBSD for instance did not. Some of this is in the ncurses FAQ, e.g., non-Linux systems, comments on OpenBSD curses, and comments on NetBSD curses TEDickey (talk) 19:12, 18 May 2023 (UTC)

Every release announcement for ncurses gives the correct information regarding this issue, e.g.,

In mid-June 1995, the maintainer of 4.4BSD curses declared that he considered 4.4BSD curses obsolete, and encouraged the keepers of unix releases such as BSD/OS, FreeBSD and NetBSD to switch over to ncurses.

(going back to ncurses 4.0 in 1996 finds no difference) TEDickey (talk) 21:38, 19 May 2023 (UTC)