Talk:Bracket

From Wikipedia, the free encyclopedia

[edit] Terminology

Dear LC, This article is mainly about punctuation in English text.

Outside the programming context, "curly braces", or even "braces", is much more common than "curly brackets", but I do think the distinction is useful in programming contexts. I have attempted to deal with the diffrerences in usage in the associated article. I was concerned that readers accustomed to calling them "braces" would be confused by "brackets".

"Brackets" is basically a little-used synonym for "braces", and is given as such deep in the definitions in both the Webster's Unabridged and the Webster's Collegiate. The Random House Unabridged gives "brace" as a synonym for "bracket" only in the architectural sense. The OED does not give "bracket" as a synonym for "brace", nor do three college English handbooks I have lying around. My ancient Roget's Thesaurus includes both in a category called Vinculum, but since "hyphen" and "ox-yoke" are also in there, that isn't really data. The more modern Random House Word Menu carefully distinguishes "brackets" from "braces". Of course, none of these sources is particularly sophisticated about computer usage.

It is certainly true that the "class of brackets" includes braces, as both this article and the punctuation article clearly state, but the class also includes parentheses, and even though I turned up a reference in the Oxford English Dictionary to "curved brackets", again, hardly anyone actually calls parentheses "curved brackets".

See the Talk page under Punctuation for the sharply contrasted etymologies of "braces" and "bracket".Ortolan88

I'm likely wrong on this, but it's been my understanding that "brackets" in reference to the family of symbols is British English, and that in the US there's a distinction made between "parentheses" () "brackets" [] and "braces" {}. As far as I know, most Americans would never refer to parentheses as "brackets." 161.11.130.249 17:30, 7 June 2006 (UTC)

Ortolan88, I saw your etymology, and thought it was great. :-) My intent wasn't to emphasize one term over another. I just wanted to fix two mistakes.

One mistake was that the Punctuation page claimed the symbols were given with their Unicode preferred names, where appropriate, but then it didn't give the Unicode preferred name. So I fixed the name to match the claim. It would have been just as good to fix the claim instead. I was just trying to fix the contradiction.

The other mistake is the use of the redundant phrase curly brace. The standard terms are curly bracket (to distinguish from other types of brackets) and brace (of which there is only one type, so no need to distinguish). According to the dictionary used by the Associated Press, there is a brace but not a curly brace. The same is true in the Random House Websters Dictionary. And the same in the Merriam-Webster's Dictionary. And the same in the American Heritage Dictionary. And the same in the Unicode standard index (the one that lists multiple names for each symbol). I see from Google th at some people use the redundant term curly brace, but since it isn't in the standard references to which I have access, it appears to be slang rather than a standard term. I could have replaced the redundant curly brace with either curly bracket or brace, but I used both so I wouldn't offend anyone. Apparently unsuccessfully. :-) --LC

I'm not offended. I've called them curly braces for 40 years, but what the heck? Braces is definitely more elegant. Funny that I gathered all that information without noticing my own term was out of line with the dictionaries! I think I'll stick with curly braces in my own life, but what you've done for the 'pedia is fine. Ortolan88

If brace and bracket don't actually share an etymology, the article really shouldn't say this:

Presumably due to the similarity of the words brace and bracket (they share an etymology), many people casually treat brace as a synonym for bracket.

Falcotron 07:08, 15 June 2006 (UTC)

I didn't want to change this without verifying that Ortolan88 was right. I found a variety of references that support the contrasting etymologies (e.g., http://www.etymonline.com/index.php?term=bracket and http://www.etymonline.com/index.php?search=brace), and none that disagree. (If these words are related at all, it's via an as-yet-unattested PIE word for both arms and legs.) So, I changed this to be correct (s/they share an etymology/although they do not share an etymology/). Falcotron 22:16, 17 June 2006 (UTC)

[edit] Crotchets?

I'd like to see a source for the claim Square brackets are called crochets in Great Britain.

I've lived mostly in Great Britain for 32 years, and I've never heard them called anything but "square brackets"

I put that in a couple of years ago and I just took it out. Turns out I had misread a definition. Crochet is an obsolete term for bracket, but only for the architectural bracket, not the typographical bracket. Sorry about that. Ortolan88 03:22, 20 Aug 2004 (UTC)

[edit] Unicode angle brackets 2004

0x2329 and 0x232A are "LEFT-" and "RIGHT-POINTING ANGLE BRACKET". 0x3008-300B are "LEFT" and "RIGHT ANGLE BRACKET", and the corresponding thing with double, in the "CJK Symbols and Punctuation" Unicode block. There's also the "presentation form for vertical" versions (which include the other kinds of brackets, among other things), which are in the "CJK Compatibility Forms" Unicode block.

They are, I think, used to enclose names of things in Chinese (but I'd have no idea, since I can't actually read Chinese). I'm not sure what the "lenticular brackets" are for, either. They're also unsuitable for normal brackets, since they're very wide.

I'm just surprised that I don't have a font with angle brackets (OS X has many unicode symbols, anyway). --Elektron 05:29, 2004 Jun 14 (UTC)

U+3008–U+300B, according to the Unicode standard, are unambiguously wide. So, they're probably just "fullwidth form" variants of the more familiar angle brackets, for use in text consisting mostly of ideographs. The standard doesn't mention the lenticular brackets at all, outside of the code charts. They, and the wide angle brackets, are probably included in Unicode for compatibility with earlier asian character sets. - — G↭a⇅a | Talk 05:07, 18 Jun 2004 (UTC)

Ancient Chinese, I'm told, didn't have : or ! or , or ? (that was borrowed). Lenticular brackets are used in (Hong Kong) Chinese newspapers, when I checked a couple days ago, so it definitely has nothing to do with earlier character sets. There's probably some CJK reference somewhere that says what it means. --Elektron 07:36, 2004 Jun 18 (UTC)

I'd like to know why the Unicode Angle Brakets appear in grey color in the article.

Possibly because of bad browser/OS support. They don't appear at all on OSX (they're replaced by some funny character, which upon enlarging is a square with a funny arrow in it, and "MISC. 2300 23FF TECHNICAL" written on it, hm). Elektron 19:36, 2004 Aug 30 (UTC)

[edit] Crystallography

Crystallography uses these things to denote whether a plane or direction is being discussed, and whether it's generic or specific. I have put up a basic account of their use in Crystallography#Notation, but you may want to wait a few days for it to stabilize before you copy it into this article. This may also convince someone to finally flesh out Miller index.--Joel 06:14, 16 May 2005 (UTC)

I've noticed brackets used in legal writing and legal transcripts. What is their purpose in the context?

[edit] Vinculum

I'm removing vinculum from the main part of the article, because, well, it is a single object that does not bracket text. It does not require the sort of pairing that brackets do. It just happens to "bracket" expressions or groups of symbols in mathematical usage. A google search for "bar-bracket vinculum" yeilds few results, most wikipedia clones. I've made a note of the vinculum later in Bracket#In mathematics —siro χ o 05:26, September 7, 2005 (UTC)

[edit] Unicode

The "angle brackets" in the article appear to be the CJK ones which are fullwidth. I tried changing them to the angle brackets in the "misc technical" code block but they were changed back by the software, so presumably they are the NC form? Anyway, the fullwidth ones look silly in the article :/ porges 05:07, 10 October 2005 (UTC)

[edit] Angle brackets

The current article looks good, but I think the current symbols used for angle brackets should be changed to simple < and > . My computer just shows them as blank squares. It doesn't recognize the code used. The computer I'm using is quite common, thus a probable majority of people trying to read this article will just see blank squares instead of < and > . I'm tentatively going to change them.--Firsfron 17:09, 8 April 2006 (UTC)

The symbol you added is <. That's a greater- or lesser-than sign. An angle bracket is 〈.--Primetime 17:44, 8 April 2006 (UTC)

... which show up as blank squares on my computer, and many others. Wikipedia can't really illustrate an angle bracket when many (most?) people won't be able to see it.--Firsfron 01:46, 10 April 2006 (UTC)

With my browser, it doesn't show up in the table of contents. But when one scrolls down, it displays. Is it the same with you? You are correct that this is a problem with Internet Explorer. Other browsers do not have the same problem, but using the {{unicode|symbol goes here}} template causes most symbols to display in IE outside the TOC. Best wishes, Primetime 15:07, 10 April 2006 (UTC)

I'm running Firefox on my Mac (OS X) and all I see are question marks for the angled brackets.. Scktwrnch 14:56, 30 September 2006 (UTC)

I'm running Firefox 2.0 on Windows XP and I also see question marks. On exactly what browser are they working? And why hasn't this been fixed by now? It's been over a year... RobertM525 11:41, 11 November 2007 (UTC)

Wouldn't it be possible to make the letters that don't show everywhere link to an image? That way everyone could see which letter was meant (with some inconvenience though). I'm against replacing them with images, what do we have unicode for? 128.208.3.75 03:57, 29 December 2006 (UTC)

I removed the sentence "In some (slightly older) Spanish books, double chevrons may enclose speech." because I am convinced the author of that sencence meant guillemets. If not I request a citation. 128.208.3.75 04:07, 29 December 2006 (UTC)

[edit] Parentheses and appositives

I removed this:

For example, "George Washington (a rich slaver) was not the wooden figure with wooden teeth that many think him". Indeed, such an interjection is called a parenthesis, and may also be set off with dashes or commas.

Because I've never seen a style guide make a claim that an appositive phrase would be set off by parentheses. I'll try work back in the use of parenthesis. SchmuckyTheCat 23:30, 6 May 2006 (UTC)

[edit] Computer languages

The discussion on use of brackets in computer languages would be highly misleading to someone who didn't already know about the subject. It should be either (a) more complete, or (b) more obviously nothing more than a random survey of examples. Wording like "... are used to..." makes it sound like these are universal, and especially common, uses of brackets.

In the case of function arguments, this is true. But using parens to bound lists is pretty much restricted to the Lisp family, while many other languages use brackets for the same purpose (and braces for associative arrays). Square brackets are no more ubiquitous in regexps than parens or braces (and similarly for shell patterns). Brackets may not be as common as braces for blocks (the Smalltalk family vs. the C family), but they're not negligible.

And the most common use of parentheses in computer languages is completely ignored--grouping subexpressions: "(2 + 3) * 4", "ASSERT((foo<bar, baz>(qux)))", "(1, 2), 3", the parens around test expressions in C control statements, etc.

Also, some languages use other symbols essentially as brackets, with the same character serving as "left" and "right" (see "| i |" in Smalltalk, similar to "|2|" in math).

And it's a little strange to not even mention how braces are used in CSS, and then use CSS braces as an example a few lines later.

Finally, it's strange that a handful of these uses appear in the general discussion above. For example, under "Types of brackets": "In computer programming, curly brackets sometimes denote the beginning and ending of a sequence of statements or define a scope." Falcotron 07:03, 15 June 2006 (UTC)

[edit] Brackets in SM nicknames

In the internet age, brackets [ ] and { } are used in nicknames by SM people to show so-called "ownership". I am tempted to add something about it at the end of the article, but SM is always a controversial thing (and surprising in an article of this kind, I suppose), and I am no expert on this, so I am still hesitant to mention it. Would it be appropriate to mention this use of brackets in the article, along with their use in mathematics, sport and law? --Blue Elf 23:08, 28 June 2006 (UTC)

If it's actually a widespread practice (and the different kinds of brackets have different well-understood meanings), I don't think it is inappropriate as such. However, as with internet culture in general, I would be wary of "canonicalizing" a convention that may only be in use in a few BDSM forums/chatrooms and not in others, however much the particpants that use it may consider it standard. I therefore think we should insist on explicit sourcing for such information. Henning Makholm 14:53, 30 June 2006 (UTC)

[edit] American & Canadian English

I speculate that in the United States, the unqualified word "brackets" normally refers to [] and "parentheses" refers t (), but in other countries, "brackets" means (). This article talks nothing about this. Anything to add?? Georgia guy 00:03, 16 July 2006 (UTC)

The article does talk about this. Look at the anonymous addition to the Parentheses section on 14 July.

This seems to be a strange place to put it, which may be why you missed it. If this is going to be useful, it should be rewritten and moved. Possibly it belongs up top, maybe in a second paragraph before the table of contents, saying that in addition to referring to the class of all these various kinds of brackets, the word "bracket" can also mean specifically parentheses (for Americans) or square brackets (for Brits). Readers probably won't find it buried in the parentheses section, and I can't think of anywhere else it makes sense.

But I'm not sure this belongs in the article in the first place. If you read through the talk page, there's already been some discussion about American vs. British usage (and Canadian?), and I suspect that the reason nobody added anything to the article (until the anonymous edit a few days ago) is that nobody could agree on what actually is typical British usage. Also, this is an encyclopedia, not a dictionary, and I'm not sure usage information for the word "bracket" is of central importance to an article on the class of brackets. Maybe it belongs on the bracket (disambiguation) page? Falcotron 05:18, 16 July 2006 (UTC)

OK, I made both of the changes I suggested. Again, I'm not sure whether we need this information in either the disambiguation page or the main page. But at least hopefully people will be able to find it now. Falcotron 05:44, 16 July 2006 (UTC)

And, if we're going to say parentheses are "sometimes known as... just brackets" maybe we should also say the same about square brackets (typical American usage) and braces (archaic usage in both countries, I believe?). Falcotron 05:47, 16 July 2006 (UTC)

[edit] Parentheses with Numbers

Does anybody know why some people enclose a digit in parentheses after spelling the word out? e.g. There were five (5) ducks in the pond.

I know when I write up legal type documents I tend to do that to make absolutely certain there is no confusion or possibility of confusion. In proper English I was taught to only write out numbers under 100 and that was that needed to be done. For every day things it seems to be redundant. Scktwrnch 15:00, 30 September 2006 (UTC)

It's probably for those who don't know a lot of English. 67.188.172.165 05:35, 6 November 2006 (UTC)

[edit] Emoticons Redirect

I've reworded the message at the beginning as it sounded a bit awkward. It said: "For technical reasons, some emoticons redirect here, such as :), :(, etc. because : is an unsupported character. :P takes you to P."

I've changed it to "For technical reasons, some emoticons redirect here. Ones such as :) & :( will because : is an unsupported character."

I removed the last part because it just isn't needed. I think the general idea is conveyed in the explanation. --Son0rouS 11:25, 21 August 2006 (UTC)

[edit] British English

I concur with the author who wrote that in other countries the signs ( ) is called brackets. I was educated in British schools and we called ( ) brackets.

[edit] Usage in chat

On chatrooms and message boards, actions are put in brackets.

"I know she can't sing, but I like Lindsay Lohan {ducks from flying vegetables}."

There's also the "Insert Item" usage.

"I don't care if [insert nominee here] deserves to be in the hall of fame, I think that Maris should be in."

[edit] Allowing for a possible change in the number of subjects from singular to plural.

I could find no written rule that applies to whether one should use a space before the opening parenthesis in this case. For example, "If you fail to comply with this requirement, this policy shall apply in the same manner it would have applied had such policy(ies) been so maintained in force." It is my understanding that a space is not used. The Elements of Style does not address this usage however; in the index there are several possibly plural terms listed omitting the space. For example, "adjective(s)", "modifier(s)", "verb(s)", etc.

Kaos Klerik 19:39, 23 January 2007 (UTC)

In such cases, you can consider rewording the phrase. "...had such policy or policies been so maintained..." would sound more natural. Xiner (talk, email) 17:56, 11 February 2007 (UTC)

[edit] Still a Stub?

Why is this article listed as a "Stub"? It seems well beyond that to me. What else would it need to upgrade? Colincbn 16:59, 25 March 2007 (UTC)

It isn't. The "history" section is tagged as being a stub, which does not seem inappropriate given its extreme brevity. –Henning Makholm 19:22, 25 March 2007 (UTC)

Ah, I see. My mistake. Thanks for the reply. Colincbn 05:25, 26 March 2007 (UTC)

[edit] Silly Question

This might already be answered but I couldn't seem to find it in the main body of the article. What do you call the individual brackets or parenthesis? Is it "left" and "right" or "open" and "close" or something else entirely? Thanks, Colincbn 16:52, 25 March 2007 (UTC)

"A bracket" or "a parenthesis". (It is "parentheses" in plural). A "parenthesis" may also refer to the entire parenthesized expression, including bracket glyps and the content between them, or to the content alone. There seems to be no universally understood short way of overcoming this ambiguity in English, except for disambiguating explicitly (e.g. "opening parenthesis character") where one fears misunderstanding. –Henning Makholm 19:26, 25 March 2007 (UTC)

Thanks. I was wondering because of a separate article on ASCII art that needs to specify which of the two types are used while making pictures and emotions and things of that nature. Colincbn 05:21, 26 March 2007 (UTC)

[edit] Bracket (mathematics)

User:Salix alba has taken it upon himself, without any discussion, to create Bracket (mathematics) and move content to that article. Maybe that's a good idea, maybe not, but I think it would be prudent to check if there's any dissent to such a proposal after the event, if not beforehand. Do any of you good people out there in Bracketland object to having the mathematics content moved to its own article? --Aim Here 11:40, 14 May 2007 (UTC)

The reasons for splitting are many. This article seems primarily concernerd with typesetting so a full treatment of the many mathematical uses, is beyond the scope of this article. It also allow the new article to be placed in various mathematical categories, and serves as a more appropriate link target for mathematical articles which need to use the term. --Salix alba (talk) 12:06, 14 May 2007 (UTC)

That sounds reasonable. —Ben FrantzDale 12:51, 14 May 2007 (UTC)

[edit] Curly bracket example

"Select your animal {goat, sheep, cow, horse} and follow me" sounds to me rather artificial, like something an NPC in a video game would say. Can we get a more natural-sounding example, preferably one from a moderately well-known work of literature? Neon Merlin 04:41, 17 July 2007 (UTC)

[edit] Hoinkies

Does anyone have the ref cited to see if this is for real? I've used Unix for 25 years and never heard this terminology. It isn't in the Jargon File and doesn't google well. I have to say it sounds like utter bs to me. Quale 01:20, 19 October 2007 (UTC)

I've removed it because it's so rare -- certainly not used "often" as the writer had suggested -- and IMO not remotely notable enough for this article. 86.160.226.159 (talk) 12:16, 16 May 2008 (UTC)

[edit] Angle brackets 2007

I've removed the Unicode characters placed by an editor because they render in my browser (Firefox, recent version) as "? ?", which is much worse than any typographical discrepancy between the ordinary "< >" characters and the "correct" ones. (Same result with Internet Exploiter, meaning that 99% of the readers here would see the wrong thing in any case). Please do not reinsert the Unicode characters. +ILike2BeAnonymous 06:43, 25 October 2007 (UTC)

Works fine for everyone else. SchmuckyTheCat

Wikipedia never uses kludges to get around the fact that not everyone can see all Unicode characters correctly. We went through this several years ago when deciding whether to use IPA or XSAMPA to represent pronunciations. Not everyone can see all IPA characters, but everyone can see all XSAMPA characters. Yet we went with IPA because it's standard and correct, and XSAMPA is just a kludge. Incidentally, the angled brackets show up fine for my recent version of Firefox on both my home and work computers. More likely, you just don't have a font installed that includes the characters. The {{Unicode}} template currently selects the following fonts in order of preference: "Code2000", "Code2001", "Free Serif", "TITUS Cyberbit Basic", "Doulos SIL", "Chrysanthi Unicode", "Bitstream Cyberbit", "Bitstream CyberBase", Thryomanes, Gentium, GentiumAlt, "Lucida Grande", "Free Sans", "Arial Unicode MS", "Microsoft Sans Serif", "Lucida Sans Unicode". Of these, at least Arial Unicode MS, Code2000, and FreeSerif include the angle brackets, and maybe others do as well. —An gr 15:18, 25 October 2007 (UTC)

Hmm... maybe the problem is with the MediaWiki software. I just noticed that the symbols used on this page are not Unicode's normal angle bracket symbols, U+2329 and U+232A, but rather the CJK angle brackets symbols, U+3008 and U+3009. But when I tried fixing them, nothing happened -- it just switched back to the CJK symbols. Still, Arial Unicode MS and Code2000 both have those symbols, so if you have either of those fonts installed, they should display correctly. —An gr 17:15, 25 October 2007 (UTC)

I think editors should resolve this issue by discussion rather than reverts. I agree with IL2BA that the two most used browsers (Firefox and IE) can't display characters properly which post a problem for 99% of the editors. So please just use the "< >". Chris! _ct 05:24, 27 October 2007 (UTC)

As I said above, Wikipedia never uses kludges to get around the fact that many users can't see Unicode characters properly. Frankly, that's their problem, not ours. And it isn't true that Firefox doesn't display these characters correctly: I use Firefox both at work and at home and they display just fine for me. If someone's using Firefox and the characters don't display, it's not because of Firefox. More likely, it's because they don't have a font installed that includes the character. —An gr 08:52, 27 October 2007 (UTC)

P.S. It works on IE for me as well, so these brackets can be displayed on that browser too. Whatever the reason is why IL2BA can't see the angle brackets, it isn't the browsers. —An gr 11:49, 27 October 2007 (UTC)

Excuse me. It is our problem. No, since Wikipedia is not exclusively for those who can see unicode and nobody WP:OWN this article, I see no reasons for us to use unicode here. I mean, why can't we just use "< >" instead? There is no problem with using these symbols. Everybody can see them. Chris! _ct 23:25, 27 October 2007 (UTC)

We can't just use < > instead because they are the wrong characters. They are not angle brackets. We don't write "ae" instead of "æ" when writing in IPA or Old English. We don't write "p" instead of "ρ" when writing in Greek. And so on. If we used < > instead, we would be lying to our readers. —An gr 06:00, 28 October 2007 (UTC)

I'm sorry; I just can't hold my tongue any longer. This is just plain stupid. Take, for instance, this statement in a recent edit summary by someone restoring the Unicode characters here:

If your browser doesn't display Wikipedia pages properly, get a better browser.

The reason this is stupid is that this statement is implicitly being thrown in the face of every reader of this article, not just me as an editor. The fact is that in my browser, which is Firefox, one of the world's most widely-used programs for this purpose, after Internet Explorer, the Unicode characters do not display at all. All I see is "? ?". Which means that somewhere around 90% of all readers who read this article will see the same thing.

The problem is not a "better browser". We cannot admonish our readers to "get a better browser" if we use a problematic character set in our pages here and they end up seeing "? ?" instead of the intended characters.

Just to be clear about this, when I say that Firefox is not the problem, I mean Firefox and the "normal" selection of fonts found on home computers. You may be correct that I'm missing a font needed to display these Unicode characters. But that also means that very many of the other readers of this page will also be missing those fonts. I don't remember seeing any warning messages on the front page of Wikipedia alerting me that I need to have certain fonts installed on my computer in order to render the contents of pages therein correctly. To do so would be an unreasonable requirement; it basically means forgoing certain features of fonts that may or may not be installed on the typical reader's computer, and instead using a more "vanilla" font that is more likely to at least render something in the majority of cases.

So far as your insistence that these are the "wrong" characters, I also challenge that. They may not be exactly typographically correct, technically speaking. But consider all the various texts published in the world that use these characters, each set in a different typeface which renders these angle brackets slightly differently. There is wide variation in how these characters are rendered exactly from font to font: the angle of the opening, proportion of width to height, thickness of line. So what? What matters is that the reader is being presented with a form of the character that is likely to display at all on his or her screen.

Angr, would you mind telling us why and how these are the "wrong" characters?

As an extra-credit exercise, I wonder if someone here would be willing to do us a favor and put in a comparison chart, here on the discussion page, which shows all these various character encodings, so that we might be able to evaluate them and report which ones do and don't display with our own browsers. I would if I knew how, but I'm not familiar enough with the character sets to do so. +ILike2BeAnonymous 07:49, 28 October 2007 (UTC)

I'm sorry, these 90% etc figures that you keep throwing around are complete and utter nonsense. I use Firefox (newest version) at home and at uni, and the unicode characters display perfectly fine. As has been stated several times, the problem cannot therefore be in the browser. There is a line that Wikipedia needs to draw between making it accessible for everybody and maintaining integrity in our articles. The greater-than, less-than symbols are typographically and visually distinct from angle brackets. Using < > instead of proper angle brackets is like using (a) instead of @. It is simply incorrect. Wikipedia cannot possibly be expected to cater to all users in all situations all the time if it comes at the expense of article accuracy. Rather than detrimenting Wikipedia's integrity, maybe you could try this. Maelin (Talk | Contribs) 08:41, 28 October 2007 (UTC)

Just a thought, Unicode U+2329 and U+232A are represention of angle brackets, not the definition of an angle bracket. I've seen many a maths paper using plane old <,> for angle brackets well before the invention of unicode. --Salix alba (talk) 08:59, 28 October 2007 (UTC)

Sure, and I've seen many linguistics papers that use < > for orthographic representation. It's a kludge for typographic convenience, but it isn't correct. —An gr 11:38, 28 October 2007 (UTC)

I'm copying what I said about the bracket issue in another place: The candidates for the left bracket are U+003c (<), U+2329 (〈), U+3008 (〈), U+27E8 (⟨) and U+2039 (‹). The first of these is the "less than sign" so that is probably out for wikipedia purposes, the second of these looks promising but Unicode has made it canonically equivalent with the thrid and their notes have a remark that this canonical equivalence has made this character roughly useless since the third comes from the CJK range and it's size is therefore too large too fit with latin or mathematical text. So I guess I would vote for using the fourth (U+27E8, ⟨) with mathematics (it is defined as "MATHEMATICAL LEFT ANGLE BRACKET") even though it may be poorly available in fonts since it has only recently been introduced and the fifth (U+2039, ‹) "SINGLE LEFT-POINTING ANGLE QUOTATION MARK" for denoting dialog and orthography in linguistics. Comments? Stefán 20:30, 28 October 2007 (UTC)

I do not think a web browser can function without reading the "<" and ">" characters, so I think any browser can display them, which is optimal. Anyway, Mediawiki automatically replaces < with < (<). --Henry W. Schmitt (talk) 18:41, 15 January 2008 (UTC)

(The following posted as a reply to user Henning Makholm, and as a general comment)

I think your insistence on using only the "official", "approved" versions of the angle-bracket characters is counterproductive to this article.

1. Consider the context: this is supposed to be a general article for a lay readership, not a specialist article in an academic journal. If this article were in a journal of mathematics, for instance, then it would be entirely appropriate to insist on the typographically-correct characters, and even to warn readers that they must have current software installed on their computers in order to correctly render these characters. But there is no such requirement with Wikipedia. So far as I know, there are no warnings or alerts to readers that they must have certain versions of software installed in order to see the content of articles here, so the assumption would be that anyone with a functioning web browser should be able to read articles here and have their contents rendered correctly. (This leaves aside the matter of text-only browsers, but that's not relevant to this discussion anyhow.)

2. Your insistence on the "correct" characters leaves readers like me with missing information in the article. This is a bad thing. Instead of recognizable characters, I see "? ?" throughout the article. This is not a tolerable situation.

3. Again, considering the context, what is the big deal with insisting on the "correct" characters in any case? Let's look at what we're talking about here: a couple of characters consisting of two lines joined in a vertex, pointing left and right. I still fail to see how the article would not be well-served by simply using the ordinary keyboard charcters "<" and ">" in most places (not all, as explained below) in order to actually show those characters to the reader, without requiring them to have any particular level of software installed on their computer? The "correctness" that you so doggedly insist on is a barely-perceptible difference between glyphs, perhaps in the thickness or length of the lines, or the angle of the vertex. So what difference does this really make? What compelling case can you make that accuracy is being ill-served here by using the ordinary, garden-variety characters on everyone's keyboard? Characters that most people (i.e., those who aren't academic specialists) call "angle brackets"?

I reject the argument, made here, that what I'm proposing is tantamount to allowing "(a)" to represent the "at sign (@)"; it is not even close. The previous is an obvious contrivance cobbled up to look (faintly) like something else. The difference here, between the keyboard "angle brackets" and the typographically-correct glyphs, is tiny by comparison, and nobody will mistake the ordinary less-than and greater-than symbols for anything else.

I'm not saying that each and every instance of "< >" should be replaced here; it is appropriate to discuss the subtleties of differences and the use of different glyphs, as well as coding systems such as Unicode, etc. But the article should show instances of these characters to all readers, regardless of the state of their browser or operating system (within reason, of course; nobody's saying that we should expect correct results on a machine with, say, only three fonts installed, or something equally ridiculous).

Please, let's not become so anal here that we only create confusion, rather than imparting information, which after all is (supposedly) the whole point of this exercise. +ILike2BeAnonymous (talk) 19:58, 16 January 2008 (UTC)

A better example would be saying that the ℃ character is preferred over °C. --Henry W. Schmitt (talk) 21:00, 16 January 2008 (UTC)

Of course, ℃ is not only not preferred, it's deprecated. —Random832 21:12, 16 January 2008 (UTC)

(Replying to next-to-last comment) Yes, that's an excellent example. (Reply to last comment) Huh? What's your point? To which glyph are you referring? ILike2BeAnonymous (talk) 21:28, 16 January 2008 (UTC)

This has already been discussed at length. The facts still stand: < and > are not angle brackets, and it would be a lie to claim that they are. This is not just a feel-good website; it is an encyclopedia. An encyclopedia must not lie. This is not negotiable. It holds even if some readers who lack the fonts to show the truth would be able to view the lie. The article does show instances of angle brackets to all readers who are able to view the image in the sidebar. People with pure ASCII-based text browsers cannot see angle brackets no matter how we do, and they would not be helped by our lying to them. –Henning Makholm 12:51, 17 January 2008 (UTC)

Sorry, but this question has still never been definitively answered here. Please tell us: in just what way do "<" and ">" (the standard keyboard characters) not represent angle brackets? What typographical tests do they fail? What would be the harm in using them to represent these characters? Who would even notice, much less object? Please be specific and thorough in your answer. +ILike2BeAnonymous (talk) 18:22, 17 January 2008 (UTC)

You have been told the difference in form and function several times, still visible on this talk page. Your continued and unexplained refusal to read those explanations strains my ability to assume good faith. –Henning Makholm 17:59, 19 January 2008 (UTC)

They are not angle brackets because that's not what they are definitively called. The Unicode names are "003C LESS-THAN SIGN" and "003E GREATER-THAN SIGN". The ISO/IEC 8859-1:1998 standard also calls them "Less-than sign" and "Greater-than sign". They have the Unicode property "Sm" (symbool, math), not any of the punctuation properties Po, Pe, Pi, Ps that other types of brackets have. And since there are specific characters which are called angle brackets, then the less-than and greater-than signs most definitely are not such. This would be similar to saying why not use "l" instead of "1", since in some fonts they certainly look the same, and if you don't have modern equipment (e.g., a really old typewriter), you don't have this newfangled one digit, so you have to use lower-case L...who will notice anyway? As far as the appearance, in general the less/greater signs are significantly different from angle brackets in most typefaces. The signs are generally vertically centered, and usually of x-height, are much more squat and wide (probably about an en-space), and of also of a similar weight and color of other mathematical symbols and letters. Angle brackets on the other hand are usually full height (they extend into the descender/ascender regions), are much more narrow (with a less acute angle), and are often of a lighter weight than letters. They look different. Now, it is very true though that the less-than and greater-than signs are often used as adequate substitutes for actual angle-brackets, and as such can be effectively angle brackets in some cases. And the computer-use section I think fairly describes this usage. But this is just like ??( and ??) can also be effectively square-brackets in some cases too. That doesn't actually make them real square brackets though. -- Dmeranda (talk) 17:20, 18 January 2008 (UTC)

Strange, my browser (Mozilla FireFox) is no longer showing the angle brackets either.... just question marks. --Henry W. Schmitt (talk) 02:24, 23 January 2008 (UTC)

The characters shown were changed on 2008-01-18 by User:Henning_Makholm from U+3008 to U+27E8 (and the corresponding closing bracket), and this is probably what broke it for you. See the table in the next discussion section. I don't think it is just a browser issue, but more of a platform issue. When using FF under Linux I can see both sets of characters, but FF under Windows XP can not display U+27E8. Although I haven't tried, IE, Mac, or other environments are certain to have still different results. That being said, I'm not sure that the recent change was a correct one; as U+27E8 is specifically only for mathematical display whereas U+3008 is a general punctuation bracket. I don't necessarily agree with the CJK rationale for avoiding U+27E8, especially why that is somehow more important than avoiding math symbols in place of general punctuation. (U+27E8 would be appropriate in the discussion about mathematics usage, but U+3008 seems more appropriate to me for general text use). This has been talked to death, but I still don't think there's any consensus except that it most certainly should not be the less-than sign (and even that is contested by some). The Unicode standard is just too ambiguous on the matter, and it becomes even more confusing because different browser/OS combinations have different often-poor rendering capabilities. -- Dmeranda (talk) 18:21, 23 January 2008 (UTC)

U+300x are explicitly defined as being for use with Chinese/Japanese/Korean script[1]; they are not appropriate in English prose. Similarly, the normative Unicode database defines U+232x as canonically equivalent to U+300x, which means that for all Unicode-conforming purposes they should be considered the same character. Therefore U+232x are not appropriate in English prose either. In particular, they cannot be considered "general punctuation" when they are the same as CJK brackets. Browsers will often display them using CJK-width glyphs, which make them look quite wrong. –Henning Makholm 04:04, 24 January 2008 (UTC)

First I think we can all agree that by the strictest definition, all three pairs U+2329/A, U+27E8/9, and U+3008/9 are technically "angle brackets" (and it can be argued that perhaps less-strict definitions may be appropriate). But even with just those three, the question is not which is valid, but rather which among them is the best choice given the context in which they are being used. Yes, I agree U+3008/9 are clearly intended for use within CJK/East Asian scripts, not so much because of any unique linguistic purpose but mainly because of special kerning requirements. But don't overlook that U+27E8/9 are equally intended to only be used within mathematical contexts (it even has a different Unicode property than all other "brackets"), so I don't think it is any more appropriate for use in English prose either (although it is very appropriate within the mathematics section). I actually think that U+2329/A is the best choice for general use; and I think you may be interpreting more into the meaning of "canonical equivalence" than what I read. That only means that it is decomposable, usually for normalization purposes, not that it is to be interpreted as actually being the same character. Among the things it doesn't imply is that it is also restricted to the same script; so just because U+2329 can be decomposed to U+3008, that doesn't make it a "CJK" character. Also, just because some browsers may decide to actually perform the decomposition just so they can do a missing glyph substitution does not make it the same character either. And neither is it on the list of deprecated characters. So ignoring all font/browser-related issues, why is U+2329/A not the best choice among those three for general use in English prose? It's certainly not any worse is it? - Dmeranda (talk) 07:44, 24 January 2008 (UTC)

If I understand you correctly, only Linux computers with Firefox can see these characters? I'm all for wheelchair ramps, but only if people with legs can get up them too. --Henry W. Schmitt (talk) 23:26, 23 January 2008 (UTC)

No, visibility is not a matter of platform and browser (there have been success reports for all common combinations), only one of whether appropriate fonts are installed on the system. This may or may not be the case regardless of operating system and browser software. –Henning Makholm 04:04, 24 January 2008 (UTC)

[edit] Table for comparison

The second row to the bottom shows angle brackets in black. The bottom row shows greater-than and less-than signs in red.

Here, I've started the table I talked about earlier. I'd sure appreciate it if someone could fill in those other characters so we could compare them. I really don't even know the differences between the standard keyboard angle brackets and those other characters; if someone would be kind enough to fill them in below, we'd have a good reference here. +ILike2BeAnonymous 02:45, 29 October 2007 (UTC)

Description	Characters	Notes
Standard keyboard symbols ("less than" & "greater than")	< >	[2]
Unicode characters (left-pointing angle bracket)	U+2329 (〈) and U+232A (〉)
more Unicode characters (left angle bracket)	U+3008 (〈) and U+3009 (〉)
mathematical left angle bracket	U+27E8 (⟨) and U+27E9 (⟩)
single-left angle quotation	U+2039 (‹) and U+203A; (›)

Since I also cannot see these Unicode angle brackets, perhaps someone could describe for me how they're different from the standard keyboard angle brackets...? And could someone cite something stating that the "<" and ">" symbols aren't used for these supposedly incorrect purposes? Is this article describing them or prescribing them...? RobertM525 11:47, 11 November 2007 (UTC)

My questions exactly, and I'm afraid things are coming down more on the prescriptive side.

I wish someone would fill in that table above with those Unicode characters so we can compare notes on this (using our respective browsers). Anyone? +ILike2BeAnonymous 11:53, 11 November 2007 (UTC)

I've filled in the table and added an image everyone can see showing the difference between angle brackets (tall, with an obtuse angle) and the greater-than and less-than signs (short, with an acute angle). And we're not being "prescriptive", we're calling things by their proper names and describing their proper typographic functions. As Maelin mentioned above, < and > are no more angle brackets than "(a)" is the at sign, or "(c)" is the copyright sign, even if people sometimes use them that way for convenience. Before we compare notes, though, everyone please make sure you have a font installed that includes the angle brackets, such as Code2000, Aboriginal Serif or Aboriginal Sans Serif, Cardo, FreeSerif, or Junicode. (Arial Unicode MS has them too, but it costs money.) —An gr 12:49, 11 November 2007 (UTC)

Assuming the article isn't changed, it at the very least needs a technical warning that it contains Unicode characters not commonly found on (apparently) Windows or Macintosh machines. You shouldn't have to read the talk page to find out why the article is talking about "?" as if it were a bracket (which is what the article appears like). RobertM525 08:30, 12 November 2007 (UTC)

I would also like to know why these symbols—apparently relatively common and highly useful symbols—were left out of the Unicode for Arial, Times New Roman, Verdana, and various other common fonts. It would seem that they are being left out because they're not being used by non-typesetting-purists. In which case, it is not something like the substitution of (c) for ©, but of a shift in symbol use to something more easily accessible. (Assuming the declaration of those symbols as the right ones was made by something authoritative in the first place.) RobertM525 08:39, 12 November 2007 (UTC)

Just to supply a data point here, the only Unicode characters I can see in the table above (i.e., the only ones that don't render as "?") are U+2039 (‹) and U+203A; (›). I'm using a Wintel box w/Windows 2000, a fine example of an older computer that the owner (me) hasn't bothered to add extra fonts to, so I would imagine quite representative of its vintage. Unless one wants to specifically punish such users for their obtuseness and lack of total computer literacy, I suggest we jettison these "typographically-correct" characters. The question asked above—how these characters are different from the keyboard angle brackets—still has not been answered, and it seems to come down to one editor's (Angr) say-so; I cannot say whether they're an "authority" on the subject or not, but let's just say I have my doubts. +ILike2BeAnonymous 17:34, 12 November 2007 (UTC)

You mean you can't see the difference in the image to the right? The two don't even look similar. As for jettisoning the typographically correct characters, that's simply out of the question. We don't refuse to use IPA characters even though more people can see SAMPA, and we use characters from writing systems from all over the world, even though many people can't see them all. If this article is going to discuss angle brackets, it needs to show angle brackets, and not symbols people sometimes use instead when they can't be bothered to use angle brackets. I've given you a list of free fonts above that include these brackets; if you still can't be bothered to install at least one of them, you're clearly no longer discussing the issue in good faith. —An gr 19:03, 12 November 2007 (UTC)

I have trouble believing this is even an issue. Of course we should not use inequality signs for angle brackets; that would be completely misleading. Yes, people in ASCII-only environments will not be able to see the angle brackets, but at least they will be aware that they're not seeing the character being spoken about. Showing a different set of characters and claiming it to be angle brackets would amount to lying, plain an simple. (For the record, I don't think that U+2039/U+203A are useful approximations of angle brackets either. They are single guillemets, and not proper for any of the uses described for angle brackets. –Henning Makholm 23:27, 12 November 2007 (UTC)

Will someone please answer the $64,000 question that still hasn't been answered here:

What is the difference, typographically speaking, between "inequality signs" and angle brackets? Is there any well-defined standard for either of these things? What, is there some committee of the ISO meeting somewhere in Brussels, or Geneva, or wherever, specifying the line weight, angle, etc., of these characters?

I'll grant you that there may be conventions for the use of these characters, in mathematics texts, etc., however ill-defined they may be, but I'll be really surprised if anyone can come up with an authoritative specification for these glyphs. +ILike2BeAnonymous 06:12, 13 November 2007 (UTC)

OK, here are some typographical differences:

Angle brackets have dimensions similar to those of parentheses: They typically descend below the baseline and reach at least as high above the baseline as digits and upper-case letters do. Their width is much less than their height. (Thus, the angle is markedly obtuse).
Greater-than and less-than are not nearly as tall as digits; often their height is little more than the x-height. They may float a bit above the baseline. Their width is roughly the same as their height. (Thus, the angle is markedly acute).

–Henning Makholm 22:26, 13 November 2007 (UTC)

I don't understand why you think this question hasn't been answered. Above I explained the difference in form between the two sets of glyphs (and it's also shown in the image), and the article itself explains the difference in function. As to the question whether these differences are codified by a standard or by convention, of course they're convention. The form and function of virtually all typographic characters are determined by convention; and the form, of course, varies from typeface to typeface. There are only a handful of glyphs whose form is set down by official standards, such as the estimated sign, the CE mark, and the euro sign (although most font designers ignore the specifications and draw a euro symbol that fits in with their font). But if you're hoping to undermine the argument that angle brackets and inequivalency symbols are distinct in form and function on the grounds that the distinction is pure convention, you're out of luck: the distinction in form and function between C and G is also pure convention, but you will certainly not convince anyone that the two are therefore interchangeable. —An gr 19:01, 13 November 2007 (UTC)

From the unicode character table: U+2039, U+203A are single-left angle quotation. U+2329 and U+232A are left-pointing angle bracket, U+27E8 and U+27E9 are mathematical left angle bracket and U+3008, U+2009 are left angle bracket. So <, U+2039 are both incorrect in unicode. --Salix alba (talk) 08:23, 13 November 2007 (UTC)

In HTML, XML and related, the standard marks for <a href="...">...</a> are the good old fashioned less than and greater than signs. Do these count as bracketed expressions? --Salix alba (talk) 08:36, 13 November 2007 (UTC)

Arguably yes, in that the signs occur in matched pairs, are each other's mirror images, and serve to enclose something in a more-or-less convex, partly imagined outline. Another case of < and > serving the role of brackets is in the template syntax of C++, copied for generics in Java and C#. I would consider it a mistake to typeset these syntaxes with angle brackets rather than < and >. –Henning Makholm 22:10, 13 November 2007 (UTC)

Why doesn't someone just upload a small picture of angle brackets, and then include the picture inline in the section, so the rest of us don't see question marks? bd2412 T 08:28, 8 February 2008 (UTC)

[edit] For technical reasons, :) (smiley face) and some similar combinations starting with : redirect here. See emoticon.

Anyone mind if I remove this? --Henry W. Schmitt (talk) 08:11, 11 December 2007 (UTC)

Yes, I mind. Why would you want to remove this, do you not think that people who enter ":)" into the search box and press Go would like to be presented with a link to the article on emoticons? Stefán (talk) 18:21, 11 December 2007 (UTC)

[edit] Quick Q

I've seen, in the NIV Study Bible, tiny L- shaped brackets around certain words (in a sub-script sort of style) (_└example_┘) - note that the brackets I refer to are lower down the word than the ones I've shown here), used (I believe) to indicate that a word is implied. I can't find them mentioned on the wikipedia, or as a proper symbol in word - does anyone know anything more about these? --El Pollo Diablo (Talk) 15:09, 13 December 2007 (UTC)

Hmm... doesn't sound like any bracket standardly used with the Roman alphabet. It's reminiscent of the brackets 「」 used in East Asian writing systems, but a little different. It's probably just an ad-hoc system the editors of the NIV Study Bible thought up. —An gr ^{If you've written a quality article...} 19:51, 13 December 2007 (UTC)

[edit] Warning to users

I just added a ref tag on the angle brackets section of the article, following the suggestion given by RobertM525 above. Someone please add some fonts recommendations to that note. I should clarify that I have Arial Unicode MS installed, and I still see them as ? ?. Waldir ^talk 15:02, 2 April 2008 (UTC)

[edit] Accounting use

The use of brackets to instead of minus sign for negative value seem to be an american accounting notation only. I read a couple of other european articles and no mention of that notation was made. It is also a common britsh practice ? —Preceding unsigned comment added by Ghaag (talk • contribs) 10:41, 22 April 2008 (UTC)

[edit] Three-way, anybody?

In my opinion, this article's a mess. If I didn't know what "brackets" were, I'd be no further educated after consulting this article. The worst sin I can point to is what appears to be an indescriminate switchover, throughout, between () and [] in both the examples and in the actual article narrative, making it impossible to tell, if you're looking at context, just what a bracket set is, and what it's used for. Under these conditions, it's difficult to tell the difference between the definition of a parenthesis and a bracket (did I not already know).

Maybe we can do a thing where, instead of trying to define the bracket using all possible definitions in one article, we split it and make three articles, with disambig. page. For example, 1) an American (English)-punctuation/style article, 2) a British-punctuation/style article, and 3) an article devoted to the bracket used in mathematics and tech (computer/code terminology/use, etc.). What you think? Sugarbat (talk) 16:13, 10 May 2008 (UTC)