Template talk:IPA-en/Archive 1

Archive 1

Link target tweak

"[…] {{IPAEng}} is appropriate, as it links to Help:IPA English pronunciation key […]"

should now be tweaked to:

"[…] {{IPAEng}} is appropriate, as it links to Help:Pronunciation […]"

Would do, but page is protected. — ¾-10 00:51, 1 December 2007 (UTC)

Done. Thanks for catching that. kwami 01:09, 1 December 2007 (UTC)

BTW, thanks to Kwami

I am but an armchair linguist, so when other people on WP were screaming that Kwami didn't know what he was doing re transcription and pronunciation help, I didn't have 2¢ to add and wasn't prepared to argue either way. But I have come to believe that they are mostly wrong and he is mostly right. He is the one who is effectively sorting out the different needs and addressing each one (e.g., "For a broad, non-regional transcription of an English word, as when giving the pronunciation of a key word in an article, {{IPAEng}} is appropriate […] For foreign words that are not assimilated into English, regional pronunciations of English words, and non-standard English dialects, {{IPA2}} would be more appropriate […] For phonemic transcriptions that follow Australian pronunciation, use the templates {{IPAAusE}} and {{pronAusE}}.") The relationship of phonetics to phonemics is complex, and every part of it gets highlighted differently somewhere on WP. Sometimes the blind people groping the trunk and the blind people groping the tail don't realize that they're both touching the same elephant. I currently believe that Kwami sees more of the elephant than they do. Sometimes an IPA character is used to represent a phoneme, not a phone. (A lot of people don't realize that.) Which may be to say, it is being used as a variable whose value may be its phone or also several nearby phones that may regionally express the same phoneme. Probably we ought to have a separate set of characters for this variable usage, but we don't. We could invent one, but most people will avoid using it. And there would have to be a slightly different (or sometimes very different) one for each language. Which may be why people resist it—they instinctively grasp that IPA allows for interlingual (phonetic) comparison, and when you lead them toward the phonemic, which also has its proper applications, they say "hey! Where'd the phonetic go?!". They don't want phonemic systems that don't also maximize phonetic comparison. That is, they do want both at the same time. But they fail to understand that you can't have both in the same transcription. I'm digressing; let me get back to the basics. Sometimes laypeople need phonemic transcriptions for words in their own language. Sometimes they need phonetic ones for foreign words. (A corollary of the previous two sentences is that a native English-speaking reader of EN-WP and an EFL reader of the same EN-WP could in many cases best be served by different transcriptions of the same English word.) Sometimes a language has both naturalized and imported pronunciations that coexist. In each of those cases, laypeople need a different help chart to explain the different transcription style. There doesn't have to be one transcription style, nor one help page/chart, that addresses all of those cases. In fact, the point is that there can't be. You can't transcribe broadly and narrowly at the same time. Which one you do should really depend on the context—on the target audience—not on your personality. They both have applications. With EN-WP having multiple target audiences simultaneously, I could envision multiple transcriptions for a single word (although I know that such a system is currently a pipe dream, and maybe most people would never understand its purpose, so maybe it's never to be.) Anyway, I'm digressing again. Let me end with thanks to Kwami—I think his help has been a positive force. — ¾-10 00:51, 1 December 2007 (UTC)

Thanks for your support. Actually, there are several articles with phonetic transcriptions in English, such as when giving a person's own pronunciation of their name. I think that's fine, as long as we clarify what we're trying to communicate. (There's one with a historical pronunciation, where I get reverted each time I try to note that that is what it is.) There are also some phonemic transcriptions of foreign words, but in such cases I think we should always link to an article on the phonology of the language in question, so our readers can properly interpret what they're seeing. The IPA is designed to be used for both (basically main symbols for phonemics, diacritics for phonetics), which is where [brackets] vs. /slashes/ come in. kwami 01:06, 1 December 2007 (UTC)

Simple English interwikis

Hi, please add an interwiki to "simple:Template:IPA-en". Thanks. — Cheers, Truth's Out There ^–talk– 16:26, 4 August 2008 (UTC)

Done. kwami (talk) 17:12, 4 August 2008 (UTC)

Redirect

This template links the pronunciations to Help:IPA for English which then redirects to Wikipedia:IPA for English. In comparison, Template:pron-en directly links to Wikipedia:IPA for English (so no redirect is needed). I think Template:IPA-en should act similarly, i.e. link directly to Wikipedia:IPA for English. huji—TALK 11:38, 3 January 2009 (UTC)

You're right. Other similar templates have the same issue. I'll fix the ones that aren't protected. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 18:53, 3 January 2009 (UTC)

proposal

What about a short popup crib sheet if you hover over the transcription? Maybe something like:

{{pron-en/test|ˈhaʊdi}}?

(The colors can be adjusted to s.t. less gaudy, of course) kwami (talk) 08:51, 11 February 2009 (UTC)

That would be good if it is limited to the symbols actually occurring. If not, it would be too big to be of much use. Perhaps with a JAVA script that could be arranged. Someone actually implemented something similar already quite a while ago. See at User:Nohat/IPA.js. −Woodstone (talk) 14:25, 11 February 2009 (UTC)

It's not too long. Did you hover over the key? I haven't seen Nohat's script in action, but some people turn off javascript. kwami (talk) 21:18, 11 February 2009 (UTC)

switches

I'm slowly converting the tranclusions of this template to a null or IPA switch, so that the default can eventually be changed to "English pronunciation: ", in order to move it in line with the other IPA-xx templates. kwami (talk) 14:57, 16 May 2009 (UTC)

help link

Was the "help" link (as found in {{Audio-IPA}} deliberately not used in this template, or is it missing through oversight? In my experience, most Microsoft users cannot play the .ogg sound files without downloading additional items, and most casual users will not know this. Could a technical "help" link be included following the "listen" text? --EncycloPetey (talk) 03:55, 26 November 2009 (UTC)

This issue, which affects all the IPA-xx templates, has now been raised at Template talk:Usage of IPA templates as well. Lfh (talk) 13:37, 9 December 2009 (UTC)

IPA Amalgamation

I've talked to people who have said that IPA is so difficult to understand that they don't even consider Wikipedia when trying to figure out how to pronounce a word. With simplicity in mind I've written a template with mouseovers that show how to pronounce each phoneme when pointed at. This way even a complete layman can puzzle out the pronunciation with (relatively) little effort.

Example:
{{IPAc-en|ˈ|b|æ|t|ə|l|ʃ|ɪ|p}} → /ˈbætəlʃɪp/

It is currently implemented under {{IPAc-en}} and combines the behaviours of {{IPA-en}}, {{pron-en}} and the original {{IPAc-en}}. Details in current {{IPAc-en}} documentation.

I propose that, after testing, the code in {{IPA-en}} is replaced with the current {{IPAc-en}} and all incidents of all three templates be changed to {{IPA-en}}. Ideally with the template calls edited so that each phoneme is a separate variable allowing mouseovers to be shown.

This will require some editing to separate all phonemes by '|' but I believe that the increase in usability for a sizable audience is worth it. I am also developing a Greasemonkey script that will automatically located all incidents of the three templates on an edit page and convert them to {{IPA-en}} calls with individual phonemes. --Deflective (talk) 12:31, 6 December 2009 (UTC)

This is really good. The problem I see is that even though the template allows for putting /əl/ of battleship (that is, {{IPAc-en|ˈ|b|æ|t|əl|ʃ|ɪ|p}}) with the mouseover saying "the 'le' in 'people'", there doesn't seem to be a way to keep editors from avoiding digraphs. People will get the wrong impression about the pronunciation with editors transcribing fire as {{IPAc-en|f|a|ɪ|ə|r}} instead of {{IPAc-en|f|aɪər}}. or {{IPAc-en|f|aɪ|ər}}.

Similarly, {{IPAc-en|dʒ|iː}} creates /dʒiː/, but that seems like a small problem to fix.

Another minor problem I see is that the narrower characters might be more difficult to hover over but I don't really see a way around that.

I noticed that using a e o in this template yields /æ ɛ ɒ/, respectively. This is a bad idea because people may not be thinking these values (e.g. I was actually imagining they'd be /ɑ eɪ oʊ/).

Also, this should only be used for English pronunciation. With other languages, English approximations with spelling break down very quickly. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 19:17, 6 December 2009 (UTC)

Definitely this template will bring more understanding, usefulness and acceptance of IPA pronunciations. Although the template accepts SAMPA style transcriptions, we should generally discourage editors to use them, because they would make verification and editing more difficult. Especially confusing are the single lower case characters that exist with a deviating value in IPA (such as a, e, o). Perhaps these should be disallowed. −Woodstone (talk) 21:53, 6 December 2009 (UTC)

I will adapt the Greasemonkey script so that it checks templates with individual phonemes and groups digraphs but ultimately, as in all things Wikipedia, it is up to the editors to check that it's correct.

Proper translations and SAMBA representations will have to be decided by people familiar with them, I'm not much help here.

And this definitely an English template, although it could easily be adapted for other languages if successful. --Deflective (talk) 22:44, 6 December 2009 (UTC)

The reason I added the basic Latin vowels was so that people could enter according to our {{respell}}ing key and still get correct results. We should remove them if we decide they're more trouble than they're worth. kwami (talk) 04:20, 7 December 2009 (UTC)

Perhaps there should be separate templates, one that takes in respellings and converts them to IPA and this one. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 04:32, 7 December 2009 (UTC)

That would make things easier to maintain. It'd be hard to find the respellings in the midst of all the IPA. We should move the AHD symbols to the respelling template too. kwami (talk) 04:55, 7 December 2009 (UTC)

If we make the template such that any single valid IPA symbol passes unmodified, we would not need to split into two templates. Only combinations would be subject to a translation table. IPA symbols that do not occur in English would have a pop-up using an international example. Do you think that would be a feasible compromise? −Woodstone (talk) 10:25, 8 December 2009 (UTC)

Excellent idea. I've edited it so that it use “curly quotes” rather that 'straight apostrophes', I think it is more legible this way. Also, I think we should provide several examples at least for phonemes with significantly perceptually different allophones (e.g. voiceless stops "“t” as in “tie”, “sty”, “water”", and maybe vowels before /l/ and /ŋ/); and, maybe, saying more explicitly that /ɪ/, /ɵ/, /ʊ/ and /i/ mean "either /ɪ/ or /ə/", "either /ə/ or /oʊ/", "either /ʊ/ or /ə/", and "either /ɪ/ or /iː/": my proposals are "either short I or schwa, as “e” in “roses”"; "either schwa or long O, as “o” in “kilogram”", etc. --___A. di M. 16:04, 8 December 2009 (UTC)

I agree about the reduced vowels. But I really don't think we need multiple examples to explain which sound /t/ is. This is just a quick-n-dirty reminder of what's in the full IPA for English key. Shorter is better: people shouldn't need to spend any significant time reading these. We're primarily catering to Usonians, who won't need to be told that the /t/ in "told" is the same sound as the /t/ in "stole"; non-native speakers will mostly be familiar with the IPA anyway. kwami (talk) 07:45, 9 December 2009 (UTC)

We have an excellent body of examples at WP:IPA for English. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 17:57, 8 December 2009 (UTC)

We still need to eliminate the underscoring, and {{nowrap}} may be a good idea. kwami (talk) 08:22, 9 December 2009 (UTC)

The newly introduced "curly quotes" are not an improvement. They are double quotes creating distracting spacing and at the size displayed they are indistinguishable from straight quotes anyway. The simple straight single quotes give a clearer picture. −Woodstone (talk) 09:33, 9 December 2009 (UTC)

Actually, do we need to explain consonants like /b/ and /t/ at all? For those with slow internet connections, wouldn't it be better to only link the IPA symbols that aren't obvious? It might also make it easier to access the cues for narrow vowels if the adjacent consonants aren't constantly popping up. The only consonants we'd need are the non-roman ones plus /j/.

As for diphthongs & affricates being broken up, s.o. could go through every once in a while with AWB and make replacements. Not too hard for maintenance, even with 10,000 transclusions. kwami (talk) 10:01, 9 December 2009 (UTC)

Interesting thought. I'm afraid however that people will think there is a malfunction. Reading off all symbols from the queues gives a nice flow. I personally am not troubled by the narrowness of the symbols. −Woodstone (talk) 10:14, 9 December 2009 (UTC)

Agreed, the frustration caused by the inconsistency about whether or not a symbol has a mouseover would far outway any advantage it might have. I also think that the single apostrophe ' is less cluttered looking than quotes " inside the small mouseover. --18:31, 9 December 2009 (UTC)

{{nowrap}} now added. --Deflective (talk) 18:51, 9 December 2009 (UTC)

An idea could be using custom tooltips for groups of phonemes: (pronounced /kwɔrk/ or /kwɑrk/). (Of course, you don't need this template do do that.) --___A. di M. 08:40, 11 December 2009 (UTC)

This template will work for standard digraph phonemes

• {{IPAc-en|tʃ|ɔər}} → /tʃɔːr/

which standardizes the mouseovers and saves people from typing them out all the time (the point of templates in general) but still allows custom mouserovers

• {{IPAc-en|{{H:title|'qu' in 'quick'|kw}}|{{H:title|'ark' in 'park'|ɑrk}}}} → [deleted so as to avoid triggering category for ill-formed templates. — kwami (talk) 22:57, 31 August 2011 (UTC)]

Unfortuneatly the IPA tag doesn't work inside mouseovers so we can't put the groupings (eg, /ɔrk/) in them. -- deflective (talk) 18:54, 11 December 2009 (UTC)

Manual IPA Amalgamation

There doesn't seem to be any progress in changing the {{IPA-en}} template to the new implementation. I still think this is an improvement and consider it worthwhile to manually convert existing {{IPA-en}} calls to {{IPAc-en}} if that's what it takes.

To help with this I've updated the Greasemonkey script so it will convert all instances of {{pron-en}}, {{IPA-en}} and {{IPAc-en}} on an edit page to a properly grouped phoneme separated version of {{IPAc-en}}. The script is available to anyone interested, it isn't exhaustively tested so please let me know if there's problems with it.

I will begin making changes by the end of the week unless there's a reason not to. -- deflective (talk) 07:22, 4 January 2010 (UTC)

Hi Deflective. I notice you are picking up on the laudable prospect of enhancing WP's use of IPA. However one issue is not settled. I rather object to the distortion of valid single IPA symbols in the translation table, as happens for a, e, o, u, y (ɒ, ɛ, ɔ, ʌ, j). We should find a way out of this, since it makes the template not generally usable for loanwords and will prevent extension to full IPA. How about allowing only upper case for these few letters in the translation table? So A, E, O, U, Y become ɒ, ɛ, ɔ, ʌ, j, but a, e, o, u, y remain their untouched a, e, o, u, y. −Woodstone (talk) 03:47, 6 January 2010 (UTC)

I am definitely against a, e, o, u, y or A, E, O, U, Y translating to anything. That conversion masks mistakes people make in transcribing IPA. Generally, when people put those letters, their transcription should be checked anyway since such people are most likely to make transcription mistakes. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 06:51, 6 January 2010 (UTC)

Woodstone, I don't think we should expand this to the full IPA. Anybody who can handle a foreign phonology can handle the IPA. These conventions are aimed at poorly educated monolingual Americans, who only understand their own dictionaries. Also, we don't use those vowels in English transcription, so there's no conflict.

Aeusoes, I agree that allowing simple Latin vowel letters is flirting with a real headache down the line. But we can't just not have conversion algorithms for them, because that would mean that they would come through unfiltered. We should have an algorithm to change them into a dingbat, to alert the reader to a potential error. I doubt y would be a problem (in fact, I think we'd need it), but maybe we could change a e i o u into ♦ or something. kwami (talk) 07:32, 6 January 2010 (UTC)

I wrote some comment as well on the {{IPAc-en}} page. Is that not a better place to continue the discussion? The problem is that the template uses IPA and SAMPA input, but both are partly incorrectly passed. In SAMPA itself there are no conflicts with IPA, so why introduce them here? My viewpoint is that ultimately all transcriptions use IPA. Allowing SAMPA is only to provide shortcuts for people who are not prepared to use other than the standard keyboard. Ideally an editor or script would convert the input to the template into proper IPA, so the next editor will not have to deal with the obsolescent SAMPA symbolism. −Woodstone (talk) 09:47, 6 January 2010 (UTC)

Do we need SAMPA? We can input directly in IPA. kwami (talk) 19:58, 6 January 2010 (UTC)

All objections to the template change concentrates on the SAMPA conversion. This feature seemed like a relatively harmless way to add functionality for the rare user who finds it useful. At worst, later users could hit preview on the edit page and then copy/paste the translated word.

But this is distracting from the point of the change, to make Wikipedia pronunciations usable for the average layman who finds IPA too complex. Instead of holding up the change in order to debate how to handle an editing feature lets drop it.

Without the translations, are there any objections to making this change to {{IPA-en}}? -- deflective (talk) 11:00, 8 January 2010 (UTC)

Formal Request to Change to {{IPAc-en}}'s Implementation

{{editprotected}}

As discussed in the previous two sections.

I've talked to people who have said that IPA is so difficult to understand that they don't even consider Wikipedia when trying to figure out how to pronounce a word. With simplicity in mind I've written a template with mouseovers that show how to pronounce each phoneme when pointed at. This way even a complete layman can puzzle out the pronunciation with (relatively) little effort.

Example:
{{IPAc-en|ˈ|b|æ|t|əl|ʃ|ɪ|p}} → /ˈbætəlʃɪp/

It is currently implemented under {{IPAc-en}} and combines the behaviours of {{IPA-en}} and {{pron-en}}. Details in the current {{IPAc-en}} documentation.

This proposal requests that the code in {{IPA-en}} is replaced with the current {{IPAc-en}} so that all incidents of the templates can be changed to {{IPA-en}} with the template calls edited so that each phoneme is a separate variable allowing mouseovers to be shown.

A Greasemonkey script has been developed that will automatically located all incidents of these templates on an edit page and convert them to {{IPAc-en}} calls with individual phonemes (will be updated to {{IPA-en}} once the change has been made). -- deflective (talk) 00:00, 25 January 2010 (UTC)

It's not ready yet. You still haven't taken care of the underlining, which was pointed out months ago.

Also, in the interests of maintenance, I wonder if we shouldn't keep the templates separate, with IPAc only used for those entries which are divided up into phonemes, the way we do with IPA-pl and IPAc-pl. kwami (talk) 00:19, 25 January 2010 (UTC)

You brought up underlining but refused to explain why you thought it was an issue. Other people who had the same concern were satisfied with the reasons why it was kept. If you still considered it an issue you should have brought it up over the last three weeks this change was left waiting to make sure everyone was satisfied.

Everyone universally agreed that the idea is good and this implementation was written so that it wont break current template calls. If you believed that the templates should be kept separate then you should have said so when I was going to start converting {{IPA-en}} calls to {{IPAc-en}}, which is what you seem to want.

That will be the necessity the moderator decides that {{IPA-en}} should not change. -- deflective (talk) 09:53, 25 January 2010 (UTC)

Where did I ever "refuse to explain" why there should be no underlining? I don't want to accuse you of bad faith, but that's nonsense. I've explained at least a couple times. We have longstanding consensus that the IPA should not be underlined, because the underlining grossly interferes with descenders and subdiacritics. As for consensus for underlining, I can't find any such discussion. Perhaps it's a page not on my watchlist?

I haven't visited the page for a while, as I haven't noticed much activity there. If I missed something, then I'm sorry, but that doesn't mean my concerns are somehow irrelevant. We have tens of thousands of English pronunciations. The conversion will take a long time. Meanwhile, people will continue to add pronunciations in the old format. By redefining this template, we'll end up with a maintenance nightmare, and if s.t. goes wrong, we won't be able to easily address it. It's just as easy to switch to the new template as we go through the articles converting the IPA format. kwami (talk) 10:29, 25 January 2010 (UTC)

The only time you've ever brought up underlining was in {{pron-en}}'s discussion page and when I asked why you thought it was important you never replied (search the page for 'underlin' to find all the times it was discussed). If you can show me any time on any page that you actually explain your position then I will apologize.

My intention in changing the existing {{IPA-en}} was to consolidate the fractured number of templates that all do the same thing with small editing features or minor cosmetic differences: {{IPA-en}}, {{pron-en}} & {{IPAc-en}}. I was up front about this ultimate goal and was encouraged by suggestions from many people over the months. If you didn't think it was a good idea you could have spoken up at any time instead of waiting until I made the formal request.

It's up to the moderator at this point. If the change is not made then I will look at changing current calls to a different template. -- deflective (talk) 18:47, 25 January 2010 (UTC)

@Deflective i don't see consensus just yet. Also, if you start changing instances to a different template, some people might consider that as an attempt to bypass consensus through wikilawyring. I would advise against that. Why not discuss it a bit more, and invite people from village pumps or a few more relevant wikiprojects to review and participate in the discussion ? —TheDJ (talk • contribs) 19:38, 25 January 2010 (UTC)

Fair enough. I would welcome discussion but for the last month pretty much the only feedback comes when I try to push the change forward and then it's just to bring up a minor issue and then never respond again after it's been addressed.

If anyone is familiar with relevant wikiprojects then please point them this way, I don't know any personally.

Considering that the latest suggestion is to keep the current template and convert calls to a new template I'm not sure how resolve that with your warning of wikilawyering. -- deflective (talk) 20:19, 25 January 2010 (UTC)

I have no problem with this eventually moving over to under this name, nor do I mind consolidating the templates. I just don't think that it's a good idea to do it in one go. I'd advise changing the articles over to IPAc-en. Then, when we think there's nothing left at IPA-en, we can delete this page (template only, not talk). Then wait a few weeks for the inevitable complaints of broken links from all the articles we missed. Fix them up, wait a few more. Once we're satisfied that's taken care of, we can move IPAc-en here, and merge the page histories. We might at that point want to set up a bot to convert IPAc-en to IPA-en in all those articles.

However, there's another problem that I ran up against while doing this w Polish. AWB can only handle 25k files. If the number of articles transcluding a template exceeds that number, they can no longer be maintained. That is, we will no longer be able to search all of them for illegal characters or strings, nor will we be able to search them to ensure that they utilize the hover-over function. In fact, it will be the most recent transclusions, where new people come along and make mistakes, which will be excluded from our oversight. kwami (talk) 21:50, 25 January 2010 (UTC)

Ah, there's now a plug-in so that admins at least can review more than 25k transclusions, so that's no longer a problem. My only objection to starting now, without looking more closely, is that the IPA is still underlined. kwami (talk) 10:08, 28 February 2010 (UTC)

Well, we're here again, over a year later, and while it's possible I've gone and missed the place where any progress on this has moved to, it appears that this proposal has stagnated. At the risk of being a windbag: For as long as I can remember using Wikipedia, I have considered this commitment to IPA to be its most singularly evident flaw. It was perfectly stated by Deflective at the top of this section -- IPA may be universal, but it's so completely inaccessible to the layperson that it defeats the purpose of including a pronunciation guide at all (it would be interesting to run a poll on how many users rely on IPA vs. how many simply look elsewhere). IPA simply goes too far, almost entirely sacrificing accessibility on the altar of absolute universality. I can't think of a clearer warning bell than having the expression "it looks Greek to me" apply literally to what is often the second or third "word" in an article!

This idea, to include a mouse-over template that automatically translates any IPA into phonetic examples in the language of that particular Wikipedia site is brilliant, inspired and would satisfy all while harming none. It should be implemented, not merely as an optional addition (inevitably leaving most pages without it), but as a permanent fixture of all IPA pronunciation links, as has been proposed. (Indeed, I think it is the IPA that should be hidden behind a mouseover, with the pronunciation format most accessible to the majority of users taking center stage: The people who know of, and/or need IPA, can still find it, while the majority of users who have no idea what it is won't be bewildered by the linguistic equivalent of a differential equation written in Ancient Greek. But I'll be happy with accessible pronunciation guidance, however it manifests. Eunomiac (talk) 14:22, 15 March 2011 (UTC)

I do see kwami's point about underlining interfering with legibility of certain characters; that's a sound concern and it shouldn't need to be "preserved" to be considered such (though I understand the frustration at having an issue you thought resolved drudged up, however innocently). Is removing that underline difficult to implement, tech-wise?

I also appreciate TheDJ's concerns about the maintenance nightmare of needing to change any article pages that use the altered template (i.e. almost all of them). It would be much better if the IPA mouseover conversion could be implemented template-side only. What about parsing the IPA pronunciation guide into a single mouseover (instead of symbol-by-symbol), and having the mouseover operate as a simple acrostic with the symbols running down the vertical and the 'b as in buy' guidance flowing to the right of each letter? This would also make it easier to see the whole pronunciation of the word at once, without having to fiddle with perfect mouse positioning over some of the smaller characters. This may be outside the bounds of current Wiki capability, but doing it this way might absolve us of the need to alter any articles, which I do see as a formidable hurdle. Regardless, I believe this to be such an important issue that it may justify a little bit of low-level scripting if necessary to make it work, while avoiding the maintenance nightmare of updating almost every single article to compliance.

Eunomiac (talk) 14:35, 15 March 2011 (UTC)

Several of us have tried such mouseovers, such as

{{pron-en/test|here}}

and

{{Pron-en/test3|here}},

but have failed to create anything that is visible across the range of platforms we would need. — kwami (talk) 21:41, 15 March 2011 (UTC)

The project continues slowly. I am converting older pronunciation templates (this now includes {{IPA-audio}}, {{Audio-IPA}}, {{pron-en}}, {{Pron-en}}, {{pronEng}}, {{IPA-en}}, and {{pronounced}}) to {{IPAc-en}} with the Greasemonkey script. The conversions need to be checked manually so it takes a while.

Mouseovers are setup the way they are because of technical limitations. For example, some browsers can't handle IPA symbols in mouseovers and template code can't effectively parse IPA words for individual characters.

Possibly the most effective way to speed changes would be to to get lists of pages that use the old templates. Anybody know if there's a way to google wiki edit pages? -- deflective (talk) 21:48, 15 March 2011 (UTC)

I don't understand what you're asking. "What links here" will give you those lists. — kwami (talk) 22:16, 15 March 2011 (UTC)

Ah, there it is. For some reason I never thought to look at the toolbox. Thanks. -- deflective (talk) 23:15, 15 March 2011 (UTC)

If you want to use the list for s.t. else, you can generate it w AWB and then export it as a txt file. — kwami (talk) 00:29, 16 March 2011 (UTC)

regex code for AWB maintenance

With a little regex, AWB makes it pretty easy to ferret out funky IPA transcriptions. (Though of course not simply incorrect ones.) If anyone's interested, I've posted some of the code I've come up with at Template talk:Pron-en#regex. — kwami (talk) 01:10, 29 April 2010 (UTC)

use [brackets] instead of /slashes/

-Stevertigo (w | t | e) 14:57, 12 June 2010 (UTC)

That seems like a bad idea for English, since we're using not just a phonemic but even a diaphonemic transcription. +Angr 17:43, 12 June 2010 (UTC)

I've seen exclamation points used for a diaphon(em)ic transcription: !sɪt!. Not sure if that would be a good choice. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 19:12, 12 June 2010 (UTC)

I've seen vertical bars used for it, e.g. |sɪt|. +Angr 20:28, 12 June 2010 (UTC)

Ooh! Who did you see do that? — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 21:09, 12 June 2010 (UTC)

I don't remember. +Angr 16:06, 13 June 2010 (UTC)