This is an archive of past discussions about Template:IPAc-en. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
There's now a proposal at Talk:Pron-en for a template like this that would have a hover-over cue for each letter or digraph, which we haven't been able to implement satisfactorily with other IPA templates. If we do this, I think @l, @n, etc. should be given dedicated calls. kwami (talk) 21:03, 25 November 2009 (UTC)
This template produces IPA, but input can be given in IPA, SAMPA and a selection of multicharacter forms. However, there are some conflicts between the two coding schemes. In SAMPA, all valid IPA symbols have their IPA meaning. If other variations are needed, capitals or diacritics are used. For example the simple vowels "a, e, i, o, u, y" are all used in their IPA meaning [a,e,i,o,u]. SAMPA symbols "A, E, I, O, Q, U, V" are used to represent [ɑ,ɛ,ɪ,ɔ,ɒ,ʊ,ʌ]. For single symbols, the current translation table neither supports true IPA, nor true SAMPA. This should be remedied. Either by accepting the SAMPA way of not interfering with IPA, or by splitting the template into separate ones for IPA and SAMPA. Perhaps the later is preferable, so editors or a script can easily go through and convert all SAMPA to IPA. In my opinion, with modern computer technology, SAMPA is rather outdated and should be used only transitionally. −Woodstone (talk) 04:56, 6 January 2010 (UTC)
I note that the removal of the SAMPA-like translation tabel has been reversed. With that, my fundamental objection revives. In my view a prime requirement is that any (lower case) valid IPA string should pass unmodified. As stated above, there is no conflict between IPA and SAMPA, so let's remove the conflicting translations. −Woodstone (talk) 06:43, 1 March 2010 (UTC)
I have removed the offending instances for single symbols (including the ones followed by a colon). There are still many strings that exhibit the same conflict. My preference would be to remove them as well, but the risk for mishaps with them is much lower (e.g. ar, er, ah, aa, oo etc). −Woodstone (talk) 07:45, 2 March 2010 (UTC)
underscores
still the problem with the IPA being underlined, which makes it difficult to read. I've reverted the few instances of this template because of that. kwami (talk) 10:04, 28 February 2010 (UTC)
In my browser (Chrome) the dotted underline used is a little lower than the normal solid underlines and does not actually cross any parts of the IPA symbols. Anyway the great advantages of the hover popups easily outweighs a minor elegance issue. −Woodstone (talk) 05:07, 1 March 2010 (UTC)
It seems that Woodstone & I do not see the underlining that Aeusoes1 & kwami do. I have tested on both Firefox & IE, could the people who see underlining test this on the browsers they have available and post their results? --deflective (talk) 13:32, 19 March 2010 (UTC)
The template has been edited so that skins that showed the IPA underlined no longer do so. Unfortunately this results in a loss of functionality for other skins (they used to underline phoneme groups when they were mouseover) but this seems unavoidable. -- deflective (talk) 05:47, 20 March 2010 (UTC)
So is this show on the road now? Should I be using it instead of {{pron-en}} and {{IPA-en}} from now on? Lfh (talk) 14:33, 20 March 2010 (UTC)
I'd appreciate it if you do, let me know if you have any problems with it. There is a Greasemonkeyscript that can help you convert all existing English IPA templates on an edit page into phoneme separated {{IPAc-en}}. -- deflective (talk) 21:15, 20 March 2010 (UTC)
Message received and understood. One thing - when there's no sound file, it leaves a gap where the speaker sign would be. Can that be fixed? Lfh (talk) 16:44, 21 March 2010 (UTC)
Superscript
Why is there an "i" superscript next to the audio sign, like here:
Some editors insist that for every sound file there is a legal requirement to have access to an attribution statement. In earlier versions this looked like (info), which is rather obtrusive, especially when several pronunciations are near each other. So it was reduced to the legally acceptable minimum of just a small superscript i, linking to the sound file description. −Woodstone (talk) 05:03, 20 March 2010 (UTC)
Thanks. A better solution would be that when one clicks on the "microphone" picture the attribution page is showed, while when one merely hovers over the microphone the sound is played. But this could be hard to implement. Oleg Alexandrov (talk) 06:03, 20 March 2010 (UTC)
If we're going to be taken to a new page every time anyway, we might as well have the attribution there. kwami (talk) 07:41, 20 March 2010 (UTC)
Activating a player on hovering would almost certainly be blocked by security suites. Now a click on the speaker plays the sound right away, but unfortunately requires a browse "back" to see the page again. A click on the i bring you to the attribution page. I fear that is as close as we can get. −Woodstone (talk) 08:53, 20 March 2010 (UTC)
functional - now what?
It looks good to me. From this point out, I think we can safely use this, with any changes at the template level.
A couple things to think about, assuming they're feasible:
For those of us who have link preview turned on, it would be nice if that were overridden. That might be a fix for the link preview rather than this template, though.
When we click on a specific letter, can we be directed to the appropriate line of the IPA key? Something using anchors, maybe?
I still wonder about having a pop-up for obvious letters such as /b/, but that's something we'll get a better feel for as we get used to it. (Maybe only have anchor links for the non-obvious letters, with /b/ just directing to the top of the key?)
I have been manually changing (with a Greasemonkey script) some high profile pages to raise awareness of the template and work out final kinks. Generally I try to leave the page as unchanged as possible but recently I've always put the pronunciation at the top of the page in brackets, eg England, since it's a common convention.
This does mean that some unconventional IPA is being exposed (eg, {{IPAc-en|n|ɪ|u}} in New York) but I make no effort to try to correct these pronunciations since I don't have the experience.
People who have the knowledge could add pronunciations to popular pages that need them (top 100 lists are good too).
Those other pronunciations are local and so shouldn't use this template. The English IPA key is incorrect for them. I partially reverted your New York edit. Only English IPA templates should be switched over. kwami (talk) 19:36, 20 March 2010 (UTC)
That makes sense. Is there a reason why the IPA templates are so fractured? Why isn't there one IPA template with everyone using the appropriate subset of characters? -- deflective (talk) 21:06, 20 March 2010 (UTC)
We could add another parameter for language, I suppose. Didn't seem necessary. Gets to be a pain to type if you need to enter six parameters for a minimal output. kwami (talk) 22:19, 20 March 2010 (UTC)
At the moment the parameter is just included in the template name ({{IPA-en}}, {{IPA-de}}, etc) so that any improvement to one template needs to be propagated over all the rest. And, as we've just seen, phonemes aren't necessarily cleanly left out of a language so there are situations where French sounds shows up in a local English pronunciation. Also you wouldn't include the language parameter in the vast majority of circumstances.
I don't really have the energy to make a case for fully integrating the IPA templates, just wondering what advantage there is in separating them. -- deflective (talk) 22:37, 20 March 2010 (UTC)
Not sure it would be a majority at all, let alone a vast one. English has a fair number of transcriptions! And wouldn't we need a placeholder for no language? But yeah, it wouldn't be that much harder. kwami (talk) 22:40, 20 March 2010 (UTC)
{{IPAc-en}} hasn't needed placeholders so far, don't see why this would be any different. It would just be a matter of compiling a (big) list of all the languages that Wikipedia's IPA can handle and checking to see if any of them are flagged. -- deflective (talk) 22:57, 20 March 2010 (UTC)
I agree that there should not be a speaker icon if there is no sound file. I removed it in the template, but afterwards saw that there is (was) a switch "noicon" to control this. If we still think the option should remain available, the switch should be reversed, so the default is no icon. −Woodstone (talk) 04:31, 21 March 2010 (UTC)
I think a bot should be written to mass replace all instances of pron-en with IPAc-en . Also, the phonemes of the word/phrase should be split up with pipes "|"s so that each individual phoneme can be quickly moused over. This is probably the most user friendly way for people who are not familiar with IPA to interact it with it. Over time, those who use it the most will absorb many of the IPA symbols by repetition. If this suggestion is implemented, I believe this template will do more for the spread of IPA in IPA barren areas (read: USA) than any other previous initiative. Who's with me?! --Rajah (talk) 04:29, 7 April 2010 (UTC)
Let's just make sure that it only converts valid transcriptions. Anything that doesn't fit the key should be flagged / left at IPA-en for clean-up. There's still stuff out there with stressed schwas, stress marks where there's no stress, vowels which don't exist in English and you can't tell what they're supposed to be, etc. kwami (talk) 04:45, 7 April 2010 (UTC)
Certainly any transcription that includes s.t. not found in our key, such as y, c, x (the latter rare) should not be converted, nor should anything with plain a, e, o, i, u, nor a:, e:, o:, nor 2ary stress after primary, etc. All easy enough to code, assuming we haven't missed anything. kwami (talk) 14:14, 7 April 2010 (UTC)
Perhaps one more rewrite to create a single catch-all IPAc for all languages is in order before an automated conversion takes place. -- deflective (talk) 18:58, 14 April 2010 (UTC)
Yeah, that's cool, thanks. I don't really do much IPA stuff on Wikipedia though, just throwing it out there. I do think that another good step would be to add this to all the other IPAc-xx codes as well. (Where xx is the two letter code for the language. e.g. I was reading a Portuguese article and it had Wikipedia:IPA for Portuguese link, but the mouseover for that would be really nice. This will run into problems when the phoneme isn't in English, but I think that would be a nice step also. --Rajah (talk) 22:52, 23 April 2010 (UTC)
I'm not sure it would be a good idea to use English phonemes for Portuguese or other languages. Usually the only stuff that's truly common between them are things like [b] which don't need explanation anyway. kwami (talk) 22:57, 23 April 2010 (UTC)
I'm not suggesting to "use English phonemes for Portuguese or other languages." I am suggesting to make all other IPA(c?)-xx templates exhibit the same behavior as the IPAc-en template. On the Wikipedia:IPA for Portuguese every single IPA symbol has an English equivalent, example, or explanation. Embedding those as the tooltip text a la IPAc-en would be a good idea. --Rajah (talk) 23:12, 25 April 2010 (UTC)
The real problem is the last vowel: the script does not recognize di/trigraphs at the end of a word, so final /@r, ou, i:, O:, I@r/ etc. are all separated out into individual letters. This leaves undefined vowels such as */o/ and */a/, and is also a problem with non-rhotic place names, for now it's explicitly stating that there is a final /r/, not just a final rhotic vowel. I've hopefully corrected all of Deflective's edits up to this point, but this should be an easy enough bug to correct. kwami (talk) 02:08, 15 April 2010 (UTC)
The script has been fixed so that multi-letter phonemes at the end of a word are now recognized. If you're using the script please reload it from the website. -- deflective (talk) 02:49, 15 April 2010 (UTC)
Details for improvement
Some detail issues need improvement.
The result of the template starts with a blank. This is not clean design. It leads to examples like: "A bed /bɛd/ is a ... ", violating the manual of style for brackets.
Every symbol is linked to "WP:IPA_for_English#key", which leads to a popup obscuring the IPA string and sometimes the individual letter pop-up. Can we omit the links for individual letters? And perhaps link the leading slash to this key?
The popup of the WP:IPA contains a big letter "g" (being the first image on that page), which is distracting or even confusing. Can we make that a more neutral image?
2 & 3 are the results of a link preview gadget I believe. Conflicts between it and regular mouseovers should be resolved in the gadget's code. The average user doesn't have it enabled so most people shouldn't have these problems.
For 1, I've been using a leading icon as default behaviour: a bed /bɛd/. I'm sure there's a work around for the leading space but this works until it's implemented. -- deflective (talk) 09:04, 15 April 2010 (UTC)
I have put the icon in Wikipedia:IPA for English#key, so it will show instead of the spurious , which happened to be the first picture. It will hardly be noticed by the average user and avoids distraction for more involved editors. −Woodstone (talk) 06:12, 22 April 2010 (UTC)
The template uses ' and , for primary and secondary stress and calls it SAMPA. In fact, SAMPA uses " and % for these. +Angr07:08, 21 May 2010 (UTC)
"skins that showed the IPA underlined no longer do so"
Sorry to come late to the party, but today was the first time I have seen this template. I've stuck with MonoBook for some script functionality not available in Vector, and the underlining persists. --Old Moonraker (talk) 09:54, 25 November 2010 (UTC)
Are you talking about the dotted underline? That is kept intentionally to show that mouseovers are available. It is set lower than the standard solid underline, so it doesn't interfere with the IPA character set. --deflective (talk) 18:30, 25 November 2010 (UTC)
That is the definition I just saw for schwa.
Presumably somebody has been tampering?
"agree" is the traditional and conventional example for schwa.
Thanks, Varlaam (talk) 10:35, 16 March 2011 (UTC)
I was able to test the example sound icon ("Alabama") in Firefox but it would not work when I was using Safari. It might be helpful to have a note about possible browser issue either within the article itself or within the template itself. Thanks, Shearonink (talk) 17:11, 7 April 2011 (UTC)
Maybe this isn't the right place to put this, but at the actual Template page I was able to test the example sound icon (an .ogg file of "Alabama") in Firefox but it would not work when I was using Safari. It might be helpful to have a note about possible browser issue either within the article itself or within the template itself. Thanks, Shearonink (talk) 17:12, 7 April 2011 (UTC)
I only now noticed the addition of a host of yotised variants (2010-11-26T04:26:32 Kwamikagami). This seems to defeat the pandialetic approach taken. A recent edit adding the phrase "British pronunciation" drew my attention to it. This addition is harmful, since the whole purpose of the pop-ups is to be concise. I propose to remove the separate yotised lines (and all uses of the yotised variant in references to the template). A pop-up like "d as in dew" will be interpreted yotised by a UK oriented reader and non-yotised by a US oriented reader. −Woodstone (talk) 08:35, 8 April 2011 (UTC)
Schwa and quotation marks
1) The description of schwa (ə) being "schwa 'a' in 'Rosa’s'" is particularly confusing because "Rosa" is fairly often pronounced /ˈroʊsɑ(ː)/ (the americanization of /ˈrosa/). Why not give a clearer example, such as "'a' in 'delta'" or "'u' in 'cactus'"?
Both are due to tradition among linguists. (1) Rosa's is the standard example of a [ə] that contrasts with [ɨ] in roses. (2) Linguists have been indicating glosses with single quotation marks since time immemorial. For (1), I have no problem with changing it. For (2), I think field-specific practice should take precedence over Wikipedia's very general style manual. —Angr (talk) 20:10, 6 May 2011 (UTC)
Well, the Journal of the Linguistic Society of America Language Style Sheet says that single quotes should be used for quotations (which opposes Wikipedia policy), but that "any cited linguistic examples" should be in italics, never quotes.[1] Anyway, best practices are not always the best practices, and Wikipedia gives specific reasons why single quotes should not be used, such as "double quotation marks are harder to confuse with apostrophes" (a point made particularly apparent in the example "schwa 'a' in 'Rosa’s'", where Wikipedians have tried to solve this problem by changing the straight apostrophe to a curly apostrophe, which is also in violation of the Wikipedia Manual of Style). Also, as {{IPAc-en}} is used mostly in articles that are not about linguistics or directed toward linguists, and the entire purpose of the template's creation was to be more accessible to the public than {{IPA-en}}, it seems it should be written in the style of the public. — the Man in Question(in question)20:45, 6 May 2011 (UTC)
Is it even possible to have italics in mouseover text? If so, that would be better to quotation marks of any variety. —Angr (talk) 22:03, 6 May 2011 (UTC)
I supported single quotes during development simply because they're clearer in the confined space of small mouseover popups. There are several constrictions in order to make mousovers browser compatible (no IPA symbols, no font styles) so quotes are required and, in this context, we are often using four quotes in ten characters (eg, "g" in "guy"). When you're skimming a half dozen of these they're easier to read with single quotes.
It is my opinion that, with single quotes, an example other than Rosa's should be used but I've deferred all these decisions to people with pronunciation experience. -- deflective (talk) 22:34, 6 May 2011 (UTC)
In other places there have been discussions about schwa and some agreement seems to exist to use about as the example. This is not part of a minimal pair, but has practically no chance being wrongly interpreted. Rosa's being a proper name is not as unambiguous. −Woodstone (talk) 12:02, 7 May 2011 (UTC)
Since the examples aren't glosses but, well, examples, I agree that double quotes would be in order here. The nice thing about Rosa as an example is that it indicates a minimal pair with roses. Using another example is probably fine, but I reiterate that I seriously doubt that /ˈroʊsɑ/ or /ˈroʊzɑ/ is at all common. — Ƶ§œš¹[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]15:56, 7 May 2011 (UTC)
Using "about" seems perfect to me. I would still like to see the single quotes changed to double quotes to conform with Wikipedia policy. That said, I have an American bias toward double quotes, where single quotes just look like a bunch of ill-placed apostrophes to me. — the Man in Question(in question)04:48, 8 May 2011 (UTC)
So how do we get the ball rolling (either by taking action or by involving either people in the discussion)? At least on the use of "about" if not also the quotation marks. Also, another suggestion: since "roses" and "Rosa's" is/was used as a minimal pair, and removing "Rosa's" would negate the necessity of using the example "roses", I think that "basket" is a better example than "roses". — the Man in Question(in question)07:53, 17 May 2011 (UTC)
I've gone ahead and changed "Rosa's" to "about", but I don't see the advantage to "basket" over "roses". —Angr (talk) 09:38, 17 May 2011 (UTC)
"Listen" icon without an audio file is inappropriate
Unresolved
– Existing code still leads to violations of MOS:ICON.
|icon adds a "listen" icon even though there is no accompanying audio file. This is confusing and breaks our guidelines on superfluous icon use. In the first instance, support for this parameter should simply be dropped; following that (which I'll do in a few days), there needs to be a bigger discussion of the use of an icon to replace a textual description here, especially due to the inline nature of the template. Chris Cunningham (user:thumperward: not at work) - talk14:51, 8 May 2011 (UTC)
Useless and confusing parameters should be removed rather than simply avoided. This is just the most obvious of a number of problems with the current template design (which AFAICS favours aesthetics over accessibility) which should really have been addressed before the wave of rollouts I've seen recently. Chris Cunningham (user:thumperward: not at work) - talk00:37, 10 May 2011 (UTC)
Actually, I think it's quite effective in telling people unfamiliar with phonemic representation that it's a guide to pronunciation—as a simple alternative to saying "pronounced" or (the visually unattractive) "pronunciation:". — the Man in Question(in question)07:27, 17 May 2011 (UTC)
If we take a long term view of development then there is a case for the icon flag.
The most common format for pronunciation is to include it after the word at the top of the article:
On articles that have an audio file there is no need to use the word "pronunciation" since the audio icon provides the context (the mouseovers also help). If fact, unecessary labels make an article difficult to follow since readers have to filter out the unneeded meta-information to find what they're looking for. This is why every single article doesn't start with a phrase like, "The name Alabama refers to a state." We know that Wikipedia defines words and names, starting every article by saying that it is a word or name is unnecessary and visually cluttering.
So, two or three decades from now, there will be more and more audio pronunciations and the audio icon will be included with the majority of articles. At that time, including the word "pronunciation" will become unnecessary, actually counterproductive. By providing a placeholder icon now (easily differentiated from the audio icon by color) we can standardize the appearance of the IPA accross all articles. A fractured implementation of the IPA template from one article to the next can be just as bad as taking up more space than necessary.
The question at hand is whether or not this appearance standardization is worth providing a placeholder icon, an icon that doubles as a "content needed" request for an audio upload. -- deflective (talk) 18:48, 17 May 2011 (UTC)
It'd need to be red, then, and then someone would say remove it because it's red. You call it an audio icon, but there's no audio — does there really need to be another reason to remove it? :p ¦ Reisio (talk) 07:42, 18 May 2011 (UTC)
It's called an audio icon when there's audio to link to and a placeholder icon when there isn't. It's the equivalent of putting [[ ]] brackets around a title when there's an article to link to and leaving the brackets off if there isn't. -- deflective (talk) 13:29, 18 May 2011 (UTC)
No, it's called an audio icon regardless; and when there's no audio, there's no reason to signify there's audio — it's inappropriate. ¦ Reisio (talk) 19:53, 18 May 2011 (UTC)
You tell me that I called it an audio icon when there's no audio, so I point out that I called it a placeholder icon (which is differentiated by color). You then contradict me. I'm not sure that we're capable of debating, only arguing. -- deflective (talk) 04:30, 19 May 2011 (UTC)
I thought you were defending how you thought people would see it, not your opinion of what you considered it to be. ¦ Reisio (talk) 14:11, 19 May 2011 (UTC)
It isn't a "placeholder icon" anyway. As Reisio already said, that would be a red link. I see there's a WIP reimplementation of this template in progress right now, so I'll leave off fixing this problem until that's deployed (when hopefully it should be more straightforward). Chris Cunningham (user:thumperward) - talk10:36, 6 July 2011 (UTC)
(Outdent) This really needs to get fixed. It's now mid-November, and I don't see any progress. The code for showing an icon should not actually show an icon if there is no audio file. Since it is not serving a purpose when there is no audio file, it is against MOS:ICONS. There is no guideline or policy anywhere that countenances "placeholder" images, and the entire concept was thrown out the window several years ago (kept for historical reasons, but consensus has been overwhelmingly against the idea). — SMcCandlishTalk⇒ ʕ(Õلō)ˀContribs.15:37, 19 November 2011 (UTC)
The purpose of the icon is to label the IPA.
This template was developed to serve two purposes. Primarily, it makes IPA accessible to the layman. Secondarily, it standardizes the way that Wikipedia represents English pronunciations.
There used to be many neglected English pronunciation templates in use. Most of them had small variations in way that the IPA was labeled ("IPA:" or "pronounced" or "English pronunciation", etc.). In order to make this template acceptable it was written to that it could emulate all these behaviours. But, ideally, we will move towards one standard behaviour; this template's default behaviour.
A fully implemented IPA pronunciation with an audio link doesn't need other labels because the audio link's icon acts as a label. In order to standardize pronunciations to this appearance the icon flag was introduced. The purpose the icon serves is a label, something that the placeholder images mentioned in your links do not do.
I'm perfectly willing to concede my point if a consensus decides against my reasoning but this shouldn't be decided without one.
I do request that if you want to vote against the icon flag that you give some thought to what Wikipedia's default IPA behaviour should be. Personally, I don't think any text label should be used even if the icon is removed. -- deflective (talk) 03:54, 20 November 2011 (UTC)
It’s academic since the template is protected and no one who can edit it can be arsed to do so, but let me ask you...
...why do we need a consensus to undo a change that was made without a consensus? That was opposed from the start? That continues to be opposed?
We don’t, of course, but I wonder if you have considered those questions. ¦ Reisio (talk) 21:35, 26 January 2012 (UTC)
The consensus process is required exactly because it is protected. Consensus was reached before through the editing process but once something is protected this method no longer possible.
Note that reaching a consensus does not mean that the decision is unopposed.
Have you considered my question? What is your preferred method to display pronunciations? Do you think they're common enough that we don't need to label them anymore? -- deflective (talk) 19:47, 29 January 2012 (UTC)
Remove the stupid icon. I have clicked those stupid things more than once and got pissed off that there was no audio to hear. Now, I don't try anymore. Rendered useless. Nicknicknickandnick (talk) 08:45, 15 February 2012 (UTC)
Long 'a' in 'bade'
I like the hover-over popups that we now have on (many) IPA spellings to provide real-world examples of how those glyphs are supposed to be pronounced. (I previously complained here about it being too hard for the average person to find a simple key for IPA pronunciation.)
But I notice that the popup for /eɪ/ says, "long 'a' in 'bade'." As far as I know, most of the world -- including myself, an American -- pronounces "bade" with a short a, exactly like "bad"; the long-vowel pronunciation is considered nonstandard. Dictionary.com shows the IPA as /bæd/. Can somebody who knows where the relevant sourcetext for these popups is located please correct this? I'd suggest "long 'a' in 'face'," in keeping with the first example given for the glyph on this project page.
Could someone add the voiced t, [t̬], as in American/ˈbʌt̬ər/ (butter) and also the symbol for u in influenza (/ɪnfluˈenzə/)?
If you want to transcribe the details of various dialects, you shouldn't use this template. {{IPA-endia}} is provisionally set up for this purpose, though it just links to the generic IPA key for now.
You mean a reduced /u/, parallel to the city vowel /i/? How would we distinguish this from /uː/? The OED has the latter, and MW doesn't distinguish it. (There are times when I've felt the need for this too, but I didn't know how to be consistent about it.) — kwami (talk) 17:42, 4 June 2011 (UTC)
No, only ɑː does. But plain ɑ now produces an error message and is flagged, so it can be fixed to match the key it attaches to. — kwami (talk) 19:47, 2 September 2011 (UTC)
Spaces In The Template
This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.
There may be a mistake in this template. A space is following right after the opening parenthesis. — Preceding unsigned comment added by Curb Chain (talk • contribs)
Sorry for the delay. I think the error is quite easy to fix. But while examining the code I discovered such horrors of inefficiency that I determined to rewrite the whole thing; hence the delay. — Martin (MSGJ · talk) 19:44, 29 June 2011 (UTC)
Rewrite
You've said that you are rewriting {{IPAc-en}}. As the original author, I'd like to request that you let me see the changes before you implement them.
The template is hopelessly inefficient, but I got sidetracked while rewriting it! If you want to see my progress, have a look at Template:IPAc-en/sandbox. I'll certainly post here before deploying so you can look over the code. — Martin (MSGJ · talk) 16:18, 12 July 2011 (UTC)
I've made the edit you suggested. If those two templates are now identical, perhaps you'd like to redirect that one here? — Martin (MSGJ · talk) 16:18, 12 July 2011 (UTC)
I came across the use of IPAc-en for the first time when seeing it used mid-sentence like this: /ˌstiːvənˈmɒfət/). I was fairly horrified at the amount of clutter this introduced into the leading sentence of the relevant BLP. When pron is removed, precisely the same amount of information is conveyed - /ˌstiːvənˈmɒfət/ - in half the space, without disrupting the flow of text. Surely it goes without saying that any expression written in IPA describes pronunciation! Alternatively, I can see the value of US: /ˌæləˈbæmə/, since it specifies US pronunciation as opposed to any other, although it's unclear which regional pronunciation this means.
I suggest that the pron variable should be rethought with a view to making IPAc-en as compact as possible. I'm not suggesting that any less information should be provided to the reader, but that redundant uses of pron which don't add value should be discarded. After all, IPAc-en is being used in-line in article text, and in the leading sentences of articles. Rubywine . talk18:41, 30 August 2011 (UTC)
I agree: the ledes of our biography articles tend to be cluttered to a rather ridiculous degree. IMO such things (pronunciation, full dates, foreign forms, transliterations, etc.) belong in the infobox anyway. I've sporadically removed the 'pron' parameter from instances of this, but not consistently. — kwami (talk) 19:44, 2 September 2011 (UTC)
This is a legacy format left over from templates like {{pron-en}}. I agree completely that this style is needlessly cluttering and I always strip the 'pron' tag when converting older templates to {{IPAc-en}}.
That said, I also believe we should keep it for now. It eases the transition process if people who are used to {{pron-en}} can still simulate the old behaviour while we're converting. Much in the same way, I want to continue passing unrecognized symbols without an error warning until we've finished converting old templates. If we made these changes to {{IPAc-en}} now it would just make the conversion process more difficult. -- deflective (talk) 01:54, 3 September 2011 (UTC)
Creator is quite irate that I merged this fork into IPAc-en. There's one change in how the icon displays, but no discussion or documentation. Should be mentioned here in case he starts using it in mainspace again, so that we know to check it along with this template for cleanup. — kwami (talk) 20:48, 2 September 2011 (UTC)
It does rather defeat the purpose of forking to "merge" back into the original (or: unilaterally [and against policy] delete), would you truly not say?; and it wouldn't have ever existed if you people were about cleaning up. ¦ Reisio (talk) 21:04, 2 September 2011 (UTC)
No, forking rather defeats the purpose of having a template. If you want to play, use the sandbox. Don't create messes in mainspace. I'm reverting all of your edits; if you do it again, I'll delete the fork and salt it so you can't recreate it. Meanwhile I've tagged the template for deletion, since you apparently aren't actually using it for anything. — kwami (talk) 06:15, 3 September 2011 (UTC)
I don't want to play, I want to get things done, and I can't at this template. If you want to see a mess in mainspace, look around at the countless misleading pointless volume icons that we can't fix from this template code because those who have access won't and those who would don't have access. It's simply illogical for a person to add to a page a template that cannot be edited and/or isn't maintained. It's working towards the opposite of what a wiki is for. ¦ Reisio (talk) 09:24, 3 September 2011 (UTC)
If you don't want to use the icon, then don't use the icon. Since you understand enough coding to edit the template, I'd assumed that you would understand how it worked, especially considering that we document it. Don't add an "icon" parameter (which the doc page says is deprecated anyway) and the icon won't appear. — kwami (talk) 03:11, 4 September 2011 (UTC)
The edit summary in my last indicated that talk page comments shouldn't be trashed by other editors. In fairness I need to point out that the guilty party was engaged in a mass rollback of the OP's mainspace edits at the time and the deletion may have been inadvertent. --Old Moonraker (talk) 16:37, 3 September 2011 (UTC)
[outdent] Actually, it's extremely common to have multiple similar templates that perform slightly differently. WP:CFORK doesn't apply outside mainspace; articles are content subject to WP:NPOV, WP:V and other content control policies, while templates are functional pages that save editorial time by performing repetitive functions; they're completely different. Threatening to revert all of someone's edits and WP:SALT the page under discussion when you yourself are deeply involved in the editing dispute is a gross abuse of admin authority; you should certainly know better, Kwamikagami. This is not how we resolve issues around here. Absent evidence that the other user is insane or a troll, it is usually a strong indication that something at a template needs to change (often simply the addition of an option) when another experienced editor insists on creating an alternative to it. For my part, I couldn't care less what change Reisio wants to make (TLDR, and that page is a redlink again, so I can't see the code); I'm simply responding to the pissing match that assaulted my eyes when I hit this talk page. PS: Actually, I've read through this stuff now and have to agree with Reisio; having an icon that does nothing is "user-hateful" and it violates MOS:ICONS. — SMcCandlishTalk⇒ ʕ(Õلō)ˀContribs.15:25, 19 November 2011 (UTC)
Too many expensive parser function calls
Each invocation of this template (that doesn't specify a media file) adds 9 expensive parser function calls, mostly because this template seems to be trying to guess whether any of the first 3 parameters is a reference to a media file. That causes pages like Heteronym (linguistics) to be in Category:Pages with too many expensive parser function calls. Why doesn't this template use a named parameter for it instead, so no expensive parser function calls are needed at all? Something like this, perhaps, although we'd want a transitional step in there where both the old and new systems were recognized to avoid disruption while existing articles were changed over. Anomie⚔23:18, 17 September 2011 (UTC)
Yes, that's needed to be done for a while. 'Audio' might be a bit more intuitive than 'file', though?
It seems to work well enough. Shall we switch over? It will only take a couple hours to convert the transclusions, so I don't know if we need a transitional version, though you can write one if you like.
Actually, if I convert the transclusions first, they simply stop supporting the audio files, but it doesn't mess up the display. So I'll run through those, and then we can change the template itself. — kwami (talk) 02:00, 18 September 2011 (UTC)
Yeah, I merged page histories to give you blame er, credit. If the edit had my name on it, people might expect me to be able to do stuff like that! — kwami (talk) 21:05, 18 September 2011 (UTC)
The mouseover for ɔː is <'au' in 'baud'>. But this is not very helpful if you do not know how to pronounce baud. It is not a very common word, and the editors of the article baud think it is necessary to include /ˈbɔːd/, and so you get a circular definition. It is even more confusing if, like me, you pronounce baud as /bəʊd/. And I think I am not alone, because the Oxford English Dictionary gives the pronunciation as "/bəʊd/ /bɔːd/". JonH (talk) 01:12, 8 October 2011 (UTC)
I would say that Maud is no longer a common name, and may be unfamiliar to readers: note that the article Maud (given name) includes the pronunciation. But I cannot see any problem with using fraud. JonH (talk) 11:29, 8 October 2011 (UTC)
Our readers are not psychic. We should link to things they may not understand already, such as what IPA is, and articles on specific dialects of English (especially keeping in mind that a large number of users of en.wiki are not native English speakers and thus do not already know about US vs. UK English).
This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.
Replace each of the following, at all occurrences in the code:
Done. If you want other languages added, make a specific request. If anyone feels the need to revert the link addition, please edit Template:IPAc-en/pronunciation to remove the links instead of reverting the code changes to this template, as IMO the code changes are an improvement in maintainability even if the links turn out to lack consensus. Anomie⚔23:18, 19 November 2011 (UTC)
This template documentation does not explain how to indicate the separation between words. By experimentation I found that an underscore or space parameter appears to produce only a tiny (almost invisible) separation.
I'm attempting to construct the IPAc-en for Eilidh McCreadie (the intended pronunciation can be heard here, at 44:40 until 15:00 UTC on 21 December 2011). This is my attempt /iːlimˌkˈrɛdi/, but the word separation is too small.
Nor does the documentation mention the icon parameter.
That can only be /iːliməˈkrɛdi/ in standard English IPA transcription.
That might be the standard English (mis)pronunciation, but it isn't the way it was spoken in the reference that sadly is no longer available to listen to. As Eilidh McCreadie produced that programme, that must be her preferred pronunciation of her surname – no sounding of an 'a' vowel in Mc, and the primary stress on 'r' as Mc-Redy.
Added a note. However, if the space is too narrow, it should be discussed here. Inserting multiple spaces, inconsistently between articles, will only make a mess. — kwami (talk) 19:50, 12 April 2012 (UTC)
This is a marvellous template! Could the mouseover for primary stress say "primary stress follows" rather than just "primary stress" because some systems place a mark after the stressed syllable? Albedo/ælˈbiːdoʊ/ is an example. It is difficult to position over the stress mark because it is so narrow and may not stand out too well for anyone not looking for it. I don't really have a suggestion for this but maybe someone else has a bright idea. Thincat (talk) 16:16, 14 January 2012 (UTC)
I noticed my browser was not showing global file usage for some audio files linked by IPAC-en template calls. My post at Commons describes what I see. I cannot demonstrate the problem at Commons because the template is here, so here are two tests: Alabama has this code: {{IPAc-en|audio=en-us-Alabama.ogg|ˌ|æ|l|ə|ˈ|b|æ|m|ə}} which expands to /ˌæləˈbæmə/ⓘ and the global usage lists many uses; and Etobicoke has this: {{IPAc-en|audio=etobicoke.ogg|ɛ|ˈ|t|oʊ|b|ɨ|k|oʊ}} which expands to /ɛˈtoʊb[invalid input: 'ɨ']koʊ/ⓘ but (until now at least) global usage lists no uses. If I understand global file usage correctly then it should now list these expansions. I will check after this post. -84user (talk) 17:36, 11 February 2012 (UTC)
The problem is that the "file usage" counts only use via wikisyntax, either as shown to the right or as a wikilink using the "Media" namespace like Media:en-us-Alabama.ogg. This template, on the other hand, makes an external link to the audio file itself. There doesn't seem to be any way to use "Media" in combination with an image, either via the image's |link= or by piping an unlinked image as the text of a wikilink. Perhaps that should be filed as a bug. We could use the Unicode characters U+1F508–U+1F50A (like 🔈, 🔉, or 🔊), but browser/font support may not be there for it; it doesn't look too good on my computer. Anomie⚔18:39, 11 February 2012 (UTC)
Done. They may show up as a bunch of little boxes, if, like most people, you use an inadequate browser. If so, you're out of luck, unless you want to switch. I think that's why we didn't do it in the first place. (We can't format text in mouseovers.) — kwami (talk) 19:15, 11 May 2012 (UTC)
Suggestion of a different example for the CURE vowel
It's currently "moor". The problem is that it's not uncommon for speakers to
have a distinct CURE lexical set, but
pronounce "moor" with the FORCE vowel.
I myself grew up speaking such a variety of British English. "moor" and "more" were homophones, but "tour" and "tore" constituted a minimal pair for CURE vs. FORCE. My sense is that I was not uncommon in this respect.
There are some people who have no distinct CURE lexical set at all, or have it only after /j/. For them, no example will be useful.
However, my suggestion is that there are a not insubstantial number of people who do have a distinct CURE lexical set, but don't have it in the specific word "moor", but do have it in "tour". Since
I'm not aware of any speakers who are the opposite way around (i.e. who do have a "more"-"moor" split but not a "tore"-tour" split), and
In support of my anecdotal impressions, I refer you to Hannisdal's dissertation, page 150, where "tour" is much more likely to have a distinctive vowel than is "moor". (It also shows that "tourist" would be an even better example). Hannisdal's dissertation covers RP and RP-like newsreaders in Britain. Grover cleveland (talk) 16:16, 12 May 2012 (UTC)
That may be a better choice. I think the reason we picked 'moor' was orthographic: the ⟨oo⟩ paralleled 'food' for the equivalent non-rhotic vowel. — kwami (talk) 21:23, 12 May 2012 (UTC)
Why does the pronunciation displayed vary from the Wikitext?
In the Eutheria article, the pronunciation {{IPAc-en|juː|ˈ|θ|iːr|i|ə}} is displayed as /juːˈθɪəriə/. Why not /juːˈθiːriə/? The latter is how I have heard the word pronounced. Peter M. Brown (talk) 00:49, 25 July 2012 (UTC)
See WP:IPA for English, there is no contrast between /iːr/ and /ɪər/ in English; both are different ways to transcribe the vowel of NEAR. The latter is how we transcribe it in our diaphonemic transcription system and our desire to be consistent is compelling enough that the IPAc-en template doesn't allow for /iːr/. — Ƶ§œš¹[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]01:10, 25 July 2012 (UTC)
Speaker icon on the Pleiades (Greek mythology) article
There's a speaker icon before the IPA stuff in Pleiades (Greek mythology), but there doesn't seem to be any audio associated with it. Seems to me the speaker icon is misleading. Could anyone here comment on why we have the speaker icon? It seems to be associated with IPAc-en. thanks --Tagishsimon(talk)17:33, 26 September 2012 (UTC)
Not only is it misleading, not only is it the wrong color, not only do the unlinked and linked images differ in appearance, not only is it a dense spot of color in a sea of thin characters, not only has the person who introduced it ceased to use it, but these concerns have already been addressed here, and ignored. Welcome to Wikipedia, "the free encyclopedia that anyone can edit". ¦ Reisio (talk) 01:07, 13 October 2012 (UTC)
There are thousands of those. We should probably have a bot remove them. Or we could change the coding so that it displays "pron." or some such. — kwami (talk) 17:43, 26 September 2012 (UTC)
This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.
The current icon doesn't match Wikipedia's link color, and does not align to its 11×11 pixel grid. I've made a new version , can somebody replace the old icon with it? (I couldn't replace the old icon on commons—it's not updating)—Kelvinsong (talk) 03:02, 6 January 2013 (UTC)
{{IPAc-en|UK|ɜː}} currently renders as {{nowrap|{{IPAc-en/pronunciation|UK}} /ɜr/. It is a rubbish to me, but I'm (almost) not a vocal speaker and may misunderstand the reason to display a rhotic phonetics where the input suggests a non-rhotic pronunciation. In any case, I doubt that the /ʃɑːˈtrɜrz/ pronunciation (from {{IPAc-en|UK|ʃ|ɑː|ˈ|t|r|ɜː|z}}) actually exists. Incnis Mrsi (talk) 11:52, 6 February 2013 (UTC)
This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.
Could we perhaps make it so that users browsing with Navigation Popups enabled don't get previews of Help:IPA for English#Key over and over again? I believe that all this would require is throwing in a <span class="nopopups"> in front of {{#switch:{{lc:{{{1}}}}} (and, of course, a </span> after {{#if:{{{50|}}}|{{H:IPA|{{{50}}}}}}}). This should not effect template behavior for users without Popups, and I don't think it would be controversial, as the same has been done for {{topicon}} and {{pp-meta}} (the former per my request, but no complaints so far). — PinkAmpers&(Je vous invite à me parler)18:01, 29 March 2013 (UTC)
Hi Pink. :) Could you put together a sandbox version and set up a couple of test cases? It would be useful to see how this would actually work in practice. Best — Mr. Stradivarius♪ talk ♪18:19, 29 March 2013 (UTC)
I've also noticed this template being transcluded on some user talk pages, which categorizes them in Category:Articles including recorded pronunciations along with mainspace pages. Is there a way to add something like a 'nocat' parameter or some sort of namespace recognition coding so that only mainspace pages get categorized? -- Ϫ02:56, 4 September 2013 (UTC)
The different feature sets of the IPAc-en and IPAc-fr templates pose a consistency problem for articles that require both templates. In the cogito ergo sum article, I want to use IPAc-en for the Latin cogito ergo sum and IPAc-fr for the parallel French je pense, donc je suis.
IIUC, the -en template requires the "/" delimiter. Neither template allows inclusion of punctuation other than spaces (using different marks). For parallel display, the best seems to be:
But I would much rather remove the slashes as they are particularly awkward for phrases (or at least have them only at the start and end of the phrase). I imagine others would prefer if the French version adopted the "pronunciation:" or "IPA" prefix. Thanks, humanengr (talk) 08:03, 19 June 2013 (UTC)
a:Orthographic commas don't belong in transcriptions. It doesn't look like IPAc-fr is maintained; we use {{IPA-fr}}. — Lfdder (talk) 09:03, 19 June 2013 (UTC)
Thanks -- based on your response, I was about to change to /ˈkoʊ.ɡɪ.ˌtoʊˈɛr.ɡoʊˈsʊm/ and /ʒəpɑ̃sdɔ̃kʒəsɥi/. I now see you edited the cogito IPA-la from oʊ to oː. In exploring that earlier, I saw that Help:IPA for English fn 17 says the latter is a common transcription. But the former shows up many times more frequently in Google searches. In any case, I see that IPAc-en converts oː to oʊ. Shall I go ahead with the changes I indicate above? humanengr (talk) 19:01, 19 June 2013 (UTC)
{{IPA-la}} is for transcriptions of Latin words as they'd have been pronounced in classical Latin. There's no diphthong [oʊ] in Latin. — Lfdder (talk) 19:09, 19 June 2013 (UTC)
One last issue, if I may -- I had included secondary stress for the last syllable, so it would be ˈkoː.gɪ.ˌto. "Secondary stress … falls on the final syllable (ultima) in a trisyllabic word with a short penult." (per About.com citing Richmond) Reference.com also shows the secondary stress. Is that not correct? Thanks, humanengr (talk) 20:40, 19 June 2013 (UTC)
I'm confused. Per your edit and last comment, I thought IPA-la didn't allow oʊ -- and as of now we're using IPA-la. But that re-raises the larger point I should have started this discussion with -- given that cogito is commonly mis-pronounced with a stress on the second syllable and sometimes, à la Italian with a soft g, I thought it more important to upgrade to IPAc-en to provide the mouseover instructions (maybe prefaced with "Classical Latin pronunciation:", "Classical:", or "pronounced:") than to stick with IPA-la for "transcriptions of Latin words as they'd have been pronounced in classical Latin." I'd rather be helpful than correct. So, if we go that route, are you saying that IPAc-en would be ˈ|k|oː|.|g|ɪ|.|,|t|oː|_|ˈ|ɛr|.|g|oː|_|ˈ|s|ʊ|m (which I see outputs as /ˈkoʊ.ɡɪ.ˌtoʊ ˈɛr.ɡoʊ ˈsʊm/) but IPA-la would be ˈkoʊ.gɪ.ˌtoʊ ˈɛr.goʊ ˈsʊm? Also, do you have a source for the "ˈkɒ-"? Lastly, do you have a pointer to the list of IPA-la valid characters? humanengr (talk) 22:15, 19 June 2013 (UTC)
IPA-la allows just about anything; it just wraps what you type in it inside square brackets and class IPA and links it to Help:IPA for Latin
That's just how it's 'properly' pronounced in English; you don't need to preface it with "Classical Latin pronunciation" (because it's not, actually)
A follow-on: The length of the entry turns out to be awkward on small screens (phones), which leads to a suggestion to allow line breaks at spaces. If that's not possible, does splitting this case into UK and US as per Alma Mater make sense here?humanengr (talk) 17:52, 20 June 2013 (UTC)
Everything's wrapped inside class 'nowrap'. Nothing we can do without changing the code of this template I think. — Lfdder (talk) 09:25, 22 June 2013 (UTC)
I don't think so, the OED will say when a pronunciation is distinctly British or American, e.g. [4]. The first one definitely shouldn't be marked US either way. Collins follows BrE pronunciation. Something like "/ˈkoʊ.ɡɪ.ˌtoʊˈɜːr.ɡoʊˈsʊm/, also /ˈkɒ.ɡɪ.ˌtoʊ/" would be fine. No one says you absolutely must keep the comma inside the slashes. — Lfdder (talk) 19:20, 22 June 2013 (UTC)
I don't have OED access at the moment -- what did that entry show? I now see that Collins also offers ˈkɑdʒəˌtoʊ ˈɜrgoʊ ˈsʌm if one selects the "American" radio button as opposed to ˈkɒɡɪˌtəʊ ˈɜːɡəʊ ˈsʊm for the British. AFAICS, Collins is the only use of the "soft g" which, from what I had seen elsewhere, is not recommended. How much authority should we give to Collins? So, it's either your last suggestion or including only the ˈ|k|oʊ|.|g|ɪ|.|ˌ|t|oʊ| version. humanengr (talk) 20:56, 22 June 2013 (UTC)
OED has "/ˈkəʊɡɪtəʊ/ /ˈkɒ-/" for cogito. Soft g also in NODE. [5] I imagine it's fairly rare, esp. btn people somewhat versed in Latin. I think we should turn a blind eye on this one. :-) — Lfdder (talk) 21:08, 22 June 2013 (UTC)
I invariably hear a faint /a/ in "comma", not a true schwa. How about "ago" or "again". A clearly unstressed initial syllable is clearest. −Woodstone (talk) 16:56, 13 March 2014 (UTC)
Hmm, it's not easy to find something universally unambiguous. I suspect the people who pronounce "about" with an 'æ' would pronounce "ago" or "again" with an 'æ' as well. Are there any examples that aren't spelled with the letter 'a'? -- Dr Greg talk 20:05, 13 March 2014 (UTC)
I think the least confusing option, even though it's stressed, is "the". By the way, this all arose from the confusion that is evident from the edit comments of this and this edit. -- Dr Greg talk 17:15, 24 April 2014 (UTC)
+1 for "the", but it's only /ðə/ before a consonant — before a vowel, it's /ðiː/. So I'd suggest something like "'e' in 'the car'" as a replacement tooltip, or something else with a consonant immediately following. DavidPKendal (talk) 18:01, 24 April 2014 (UTC)
I guess that would work, although it's unusual to give a two-word hint. I suppose another option would be "'o' as in 'button'", even though this template treats /ən/ as a separate case from /ə/. (For what it's worth, schwa lists some more examples.) -- Dr Greg talk 16:08, 25 April 2014 (UTC)
Could the confusion be that, with small print, the tool tip looks like it's saying "/a/ as in about"?
"Button" is no good for people who pronounce it with a syllabic en. (That's why we list it separately in the first place.) Two-word tips might be confusing.
What about words like aware, award, awake, etc. Are they are pronounced with an /æ/ in these dialects?
If all the following are pronounced with /æ/, maybe we need to find something at the end of a word:
Ideally I'd like to avoid the letter "a" because of recent edit-warring at Clitheroe, where one editor refuses to accept |ˈ|k|l|ɪ|ð|ə|r|oʊ| as the correct pronunciation on the grounds that "it's not pronounced with 'a'". (At the time of writing the page is locked to prevent edit-warring.) I've asked that editor to suggest some alternatives.
How about the 'e' in 'boomerang', 'geranium', 'operation', or any other word containing "ər" followed by a vowel, and spelled with only a single "e"? (I.e. where the "r" isn't rhotic, so, if I understand correctly, should be encoded in this template by |ə|r| rather than |ər|.) -- Dr Greg talk 15:31, 26 April 2014 (UTC)
That editor seems to carry some prejudices about the IPA. The 'e' in geranium is an /ɪ/ in RP. Boomerang or operation would work, I suppose. Let's avoid any schwarrs. — lfdder15:42, 26 April 2014 (UTC)
Not heresy, pronounced with an /ɪ/ in trad-RP; not hegemony, stressed on the 2nd syllable in BrE. — lfdder16:02, 26 April 2014 (UTC)
/ər/ is out because of the effect of the /r/; it's often just a syllabic ar. Most reduced vowels spelled 'e' or 'i' are actually schwi, not schwa. — kwami (talk) 16:56, 26 April 2014 (UTC)
As some of you know I’m new to IPA, but want to try to contribute. The e in therapy might be a potentially useful suggestion. I note that Theropoda seems to contain the same issue in the pronunciation of “theropod” but “Theropoda” seems to work?--Trappedinburnley (talk) 18:13, 26 April 2014 (UTC)
I also think it wouldn't hurt to be less timid about using a stressed syllable as an example. Any transcription using /ə/ should have stress marks anyway. DavidPKendal (talk) 19:09, 26 April 2014 (UTC)
Kwami, could you clarify your comment? Are you saying that "boomerang" and "operation" are no good? I think neither of these are /ɪ/, nor can they be omitted, i.e. they'd not be pronounced as "boomrang" and "opration". -- Dr Greg talk 20:42, 26 April 2014 (UTC)
For GA speakers, they're not boom-ə-rang, op-ə-ration, but boom-r-ang, op-r-ation. Using them as examples would suggest that sofa is to be pronounced as rhotic sofer – which it is in Boston, due to hypercorrection, but not in GA. (Similar thing with hurry, which has the NURSE vowel in GA.) — kwami (talk) 00:37, 27 April 2014 (UTC)
Is it worth considering that wherever possible, the keywords of the lexical set all end in a voiceless alveolar or dental consonant? Or does this just make muddy water more muddy... I am not sure I like "syrup" (though I am very fond of its maple variety) but cannot easily explain why-- maybe something to do with the fact that although it is the unstressed syllable that is being used as the example, the "syr" has multiple possible pronunciations. Could we identify a word that has no ambiguity in either the example syllable or others around it? Or is English just not cut out for that maybe? Just sayin'... KDS4444Talk03:00, 28 April 2014 (UTC)
What about hocus-pocus or famous? Carrot or human?
Thanks for the mammoth list of suggestions. I think my favourite is now "quiet", mainly because it's the only one spelt with an "e". I'm trying to second-guess the editor of Clitheroe, who doesn't seem to have much of an understanding of phonetics and rejects anything that he thinks is the wrong letter, regardless of pronunciation. (See User talk:Hortimech2014.) If he gets it wrong, then I guess some other readers will, too. As a second best, any of the "-ate" suggestions would do, as I can't think anyone would seriously think they were pronounced "-æt". -- Dr Greg talk 22:34, 28 April 2014 (UTC)
We really shouldn't choose an example based on a single dispute. Also, is Clitheroe pronounced w a schwa, or is it a schwi?
Examples preceded by vowels, such as quiet, might be confusing because of the y or w sound that creeps in. (I don't know that, I'm just trying to anticipate problems.) — kwami (talk) 01:39, 29 April 2014 (UTC)
See the pronunciation at the head of upsilon. The actual OED American English pronunciation is /~ɑn/, which produces an error message in your template. The vowel is glossed as the a in palm and the o in lott, which Help:IPA for English divides between long ɑː and inverted ɒ. I presume the lack of support isn't an oversight on your part but a conscious choice: (a) for what it's worth, I disagree. We should be able to recreate exactly the IPA pronunciations of reliable sources such as the OED; in particular, we should be able to produce the sounds of General American given that we can easily label it as such. (b) Am I right in thinking that given your (wrong) options, that /ɒ/ is closer to the intended sound? — LlywelynII01:42, 6 October 2014 (UTC)
This template represents diaphonemes, abstractions of speech sounds that accommodate General American, Received Pronunciation, Canadian English, South African, Australian, and New Zealand pronunciations. For each phoneme a standard representation is chosen. Several sources use contradicting conventions, so a literal copy from those may not coincide with the choice used for this template. If you want to describe more dialectic detail, use the more general template {{IPA-all}}. −Woodstone (talk) 06:07, 6 October 2014 (UTC)
Similarly, I got an error from pasting over a ᵻ (the e in roses & business) from the OED. I'd imagine that's simply a variant of ɨ which should be supported. Does it actually have any separate use meriting its lack of support? — LlywelynII01:46, 6 October 2014 (UTC)
Automatic inclusion in Requests for audio pronunciation (English) category
There is still a problem with the UK version of ɜː. It still shows up as ɜr, which is just plain wrong. The current Earl page has an example in the intro. Can someone please fix this or point me to the code so that I can take care of it?Ordinary Person (talk) 22:56, 25 December 2014 (UTC)
One of the aims of this template is to use only one symbol for symbols that are interchangeable between all (applicable) dialects. In other words, we wanna use only one symbol to represent phonemes that are pan-dialectal. Is this not the case with <ɜː> and <ɜr>? 213.7.227.83 (talk) 00:32, 26 December 2014 (UTC)
Not done: please make your requested changes to the template's sandbox first; see WP:TESTCASES. There will need to be codes added to the template sandbox first to make sure it works as intended. As for your little table there, once the codes you put in the sandbox work, you should be able to update the documentation to add that part yourself (and an admin may request it once they carry out the other part of the request). — {{U|Technical 13}} (e • t • c)02:05, 12 March 2015 (UTC)
Done. I've used two-letter codes for Irish and Australian English. Peter238, reactivate the request if you're happy with my sandbox version. Alakzi (talk) 02:19, 12 March 2015 (UTC)
There is a request for comment taking place here on the suitability of the word "foot" as an English example of the sound and IPA symbol "ʊ". Please consider responding to the request. Much thanks! KDS4444Talk15:02, 26 May 2015 (UTC)
Lua version
I've just finished creating a Lua version of this template at Module:IPAc-en. The phoneme data is at Module:IPAc-en/phonemes, and the pronunciation data is at Module:IPAc-en/pronunciation. I've also added some more test cases, and you can test it out using the sandbox. It can handle any number of pronunciation keys and any number of phonemes, and should be faster than the existing template. One thing that it doesn't do at the moment is handle missing parameters (i.e. {{IPAc-en|,|&|l|@|'|b|10=&|11=m|12=@}}), but I highly doubt any transclusions will be using syntax like that. Let me know what you think, and if you spot any bugs. If there are no problems, I hope to make the Lua version live in a few days' time. — Mr. Stradivarius♪ talk ♪09:52, 17 June 2015 (UTC)
Also, I've just updated the module so that the underlining actually looks nice now. Before it was technically several bits of underlined text next to each other, but now it is just one, so browsers can now render it without any of the dots bunching up. (This also means that the "unsupported input" warnings are now also underlined, but I think this is definitely a price worth paying.) — Mr. Stradivarius♪ talk ♪10:18, 17 June 2015 (UTC)
Yep, I realised that after I posted that last message, and spent a busy hour or two fixing the module. So now the "unsupported input" warnings and the separators are not underlined, and the underlines look nice. What's not to like? ;) — Mr. Stradivarius♪ talk ♪13:22, 17 June 2015 (UTC)
Made the changes per module talk page and only after saving was told that the module had been superseded. Change:
"/ɒː/ 'o' in 'coffee'"
to
"/ɒː/ variable 'o' as in 'coffee'"
and
"/aː/ 'a' in 'bath'"
to
"/aː/ variable 'a' as in 'bath'"
From H module talk page (variable BATH and CLOTH vowels):
We need to say these are variable. Otherwise we will be giving people the wrong pronunciations.
An example is Denali. The name has two common pronunciations, with /æ/ and with /ɑː/. There are lots of place names like this. But while I pronounce "bath" with /æ/, I pronounce "Denali" with /ɑː/. Thus for me the claim that it's pronounced like "bath" is wrong. However, if we say "the variable 'a' in 'bath'," I will understand that people pronounced Denali two ways just as they do bath, which is what we mean.
If people object that the 'a' in Denali is not exactly the BATH vowel, then we shouldn't have these entries at all, because nearly all uses will be for foreign or place names that do not correspond to "bath" for each idiolect.
— kwami (talk) 01:44, 10 September 2015 (UTC)
Moved from Template talk:IPAc-en/doc
in Wikipedia:IPA for English#key, so it will show instead of the spurious 
, which happened to be the first picture. It will hardly be noticed by the average user and avoids distraction for more involved editors. −Woodstone (talk) 06:12, 22 April 2010 (UTC)
Unresolved
Checking... — Martin (MSGJ · talk) 14:47, 27 June 2011 (UTC)
Done. If you want other languages added, make a specific request. If anyone feels the need to revert the link addition, please edit Template:IPAc-en/pronunciation to remove the links instead of reverting the code changes to this template, as IMO the code changes are an improvement in maintainability even if the links turn out to lack consensus. Anomie⚔ 23:18, 19 November 2011 (UTC)
, can somebody replace the old icon with it? (I couldn't replace the old icon on commons—it's not updating)—Kelvinsong (talk) 03:02, 6 January 2013 (UTC)
Done, thanks. Chris Cunningham (user:thumperward) (talk) 10:54, 8 January 2013 (UTC)
Not done: please make your requested changes to the template's sandbox first; see WP:TESTCASES. There will need to be codes added to the template sandbox first to make sure it works as intended. As for your little table there, once the codes you put in the sandbox work, you should be able to update the documentation to add that part yourself (and an admin may request it once they carry out the other part of the request). — 