Template talk:IPA consonant chart

Links should not be underlined

IPA links should not be underlined. Interferes with legibility. kwami (talk) 07:14, 1 November 2008 (UTC)

I've made an error when using {{IPA}} template. It has been fixed. -✉ Hello World! 07:57, 1 November 2008 (UTC)

Merger proposal

N Not done and not likely to be done (See #Closing: Will not be merged) -DePiep (talk) 21:50, 13 July 2010 (UTC)
Propose Template:Consonants(edit talk links history) into Template:CSS IPA consonant chart(edit talk links history).

• Reason: Duplication. Both templates have the same subject, the IPA consonant chart.
• Direction of merge: the target title is more descriptive, so we want to use that one. Number of usages (transclusions), that would use a Redirect: not looked at. -DePiep (talk) 10:21, 21 June 2010 (UTC)

Related topics, but not essential to this proposal:

- Parallel situation, proposed merging: Template:Vowels(edit talk links history) and Template:CSS IPA vowel chart(edit talk links history).

- CSS could be dropped from the title, since we should not be concerned with technique here. Btw, there are 2 charts here, so plural is more correct: charts. -DePiep (talk) 10:21, 21 June 2010 (UTC)

This is the actual chart. {{Consonants}} is just a shell that transcludes it. Same with the vowel templates. There is no duplication. — kwami (talk) 10:37, 21 June 2010 (UTC)

Actually, the two Vowel-templates differ from the pair of Consonatns-templates. The Vowels do not transclude, the Consonant pair does.

The topic and title-meaning and content is the same: IPA Consonants chart(s). Could you explain what this transclusion adds? -DePiep (talk) 10:55, 21 June 2010 (UTC)

I imagine it was done for ease of editing: one template for content, the other for format.

You're right, the vowels do not transclude. Perhaps they should. The have diff functions: 'vowels' is used as a navigation template for vowel articles, whereas CSS vowels is used within a text, as at IPA. — kwami (talk) 20:21, 21 June 2010 (UTC)

We are in the process of planning a significant change to the {{Consonants}} navigational template. (Of course it's taking a long time; volunteer labor, you know.) As is already the case with the {{Vowels}} navigational template, which has non-IPA symbols on it that therefore do not belong on {{CSS IPA vowel chart}}, it seems likely that in future we may have non-IPA symbols on {{Consonants}} that, being non-IPA, do not belong here. I think it's distinctly preferable to not throw artificial barriers in the way of separate evolution of the two templates. --Pi zero (talk) 00:03, 30 June 2010 (UTC)

1. Isn't it perfectly OK to merge and then mark these non-IPA symbols, as is done in the IPA charts (2005)? btw, they are already in now, these links show as bullets.

2. Here at consonants we have only one table set: this one (which is different at Vowels). Introducing separate evolutions means splitting (by copying probably) this one. Keeping the wrapper {{tl:Consonants}} would mean split that one two - that's four to maintain and use. To me that's too complicated to be useful.

3. btw, who is "we"? -DePiep (talk) 06:34, 2 July 2010 (UTC)

I remarked in the parallel vowels thread that I think the templates are logically separate, which suggests they should be separate templates, but they could just as well have a single copy of the markup with a parameter to generate the two variants. In this case, for the moment at least, the parameter would actually not have to do anything; it'd just be there providing for possible future use.

As best I recall, possibilities for changing the constants navigation template are something Ƶ§œš¹ and I were discussing some time back (at one or another somewhat out-of-the-way consonant page), and Ƶ§œš¹ raised the subject again comparatively recently at the Consonants template. I don't clearly recall whether anyone else was involved in the earlier discussion, nor whether there was any evidence whether Ƶ§œš¹ had also had other conversations on the subject. (I think Kwamikagami added to that earlier thread quite recently, though not in a way that addressed changes to the template; if I set out to track down the thread, though, I might start with that clue.) --Pi zero (talk) 10:16, 2 July 2010 (UTC)

1. On merging IPA and non-IPA symbols. If I understand your main point well, that would be good. I support: ultimately we have IPA and non-IPA consonants merged into one template, with an option by parameter to show the non-IPA's. Usage like: {{Consonants|ShowNonIPA=yes}} would show the non-IPA symbols. Same setup for Vowels: e.g. {{Vowels|ShowNonIPA=no}}. Would imply TODO for Consonants: the now visible non-IPA's (and the related footnote, the bullet) would be behind the switch. TODO for Vowels: introduce the switch into the (already more general) Vowels-tamplate, with non-IPA's like ɪ̈ and ʊ̈ and the footnote behind the switch. The CSS IPA Vowel template would become a redirect.

2. On bare tables and navigation template. Another, different merge-topic here is the 'logical' difference currently at {tl|Consonants}}: Consonants has a bare tables-template ({{CSS IPA consonant chart}}) and a navigation-wrapper ({{Consonants}}) that includes (transcludes) these bare-tables (Again, this is not the situation with the the vowels-templates ({{Vowels}}). Merging question: keep the wrapper separate? I'd say no. I suggest merging the {{Consonants}} and {{CSS IPA consonant chart}} into one navigational one. Everyplace where you'd need the bare-table, the navigational one is allowed & correct too (say end of section). I may point out, today the bare-table one is used nowhere, all 177 usages are the navigational wrapper. That would mean TODO: Consonants: substitute the bare-table template once into the now wrapper. bare-tabel template unused (redirect). TODO in Vowels: merge the CSS IPA Vowel template into the Vowels template (which has the navigation wrapper already). Anyway, I oppose any outcome that would introduce more templates, for reasons of -- well, it's called 'Template' for a reason. I find the current ones doing a very good job in reproducing the IPA charts, which is exactly the base for describing and navigating the charts in one.

3. On other topics: these should stay separated from these two. That is: we should be able to decide on the two topics above each separately, without third-topic involvement at first. e.g. layout issues should not inhibit us from changes proposed. Another example: deprecated IPA symbols are a different beast, and require a different discussion.

-DePiep (talk) 11:49, 3 July 2010 (UTC) linkfixes -DePiep (talk) 12:10, 3 July 2010 (UTC)

Under construction: {{Show text}}. Would give the option to select in each article separately whether yes/no show non-IPA-symbols in the template. Some issues there still, but you can take a look. -DePiep (talk) 00:41, 6 July 2010 (UTC)

Better: using {{yesno}} for conditional text. See template & /doc -DePiep (talk) 10:10, 9 July 2010 (UTC)

Show non-IPA symbols: solved

The template {{Consonants}} {{CSS IPA consonant chart}} now feature the switch 'shownonIPA' yes/no. Consider this topic closed. -DePiep (talk) 10:10, 9 July 2010 (UTC)

Closing: Will not be merged

Discussion lead to a different outcome:

1. The bare tables need their own template, for in-line usage in articles, e.g. IPA.
2. The navbox Consonants can transclude these tables, and do a full navigation-task.
3. Including non-IPA letters is by option (variable "shownonIPA=no", see documentation)
4. Other changes pertain to layout, and are out of scope now.

This has already been implied. See also: {{Vowels}}, with the same construction. -DePiep (talk) 21:50, 13 July 2010 (UTC)

Layout improvement here is harmless

I changed the layout of the tables here and here, essentially using the 100% width. Then kwami reverted both, with this on my Talkpage:
(copied)

Please don't adjust gross formatting in the CSS tables for C's & V's. That's done in the template shells, {{Consonants}} and {{Vowels}}. Changing the CSS tables tends to mess up the formatting of any article that transcludes it. — kwami (talk) 10:51, 21 June 2010 (UTC)

I saw your reverts. Could you point out where the formatting gets messed up? I checked: the C-tables are transcluded into the 100%-width-template, so my edits do not influence anything outside of the wrapper template. -DePiep (talk) 10:59, 21 June 2010 (UTC)

(end of copy)

Addition: I checked all usage of this template (using AWB). All What-links-here links (177) are transclusions via the wrapping Template:Consonants. In other words: this template in itself is not used outside of the wrapper. Since the wrapper is 100% width already, changes in the tablewidth and layout here have no effect outside of the wrapper template. Nowhere. Therefor I conclude & suggest that this template can be improved as demonstrated. -DePiep (talk) 13:53, 21 June 2010 (UTC)

I can't tell you why, but it did mess up the formatting of the table within the wrapper: Instead of the charts being side-by-side, one was under the other. — kwami (talk) 20:20, 21 June 2010 (UTC)

Now I understand. But in my browser (FF), they were under each other already before my edits. Isn't that a more general presentation? (I'll try to reproduce your situation). -DePiep (talk) 22:04, 21 June 2010 (UTC)

I've used FF for years, and they were always side-by-side until your edit. I believe they were designed to be that way. — kwami (talk) 22:13, 21 June 2010 (UTC)

Probably your FF screen width&size (or what's it called) differs from mine. On that, I think we are not supposed to lay out for any specific technical setting. In general, two tables below each other, each with reasonable width, would be allowable for any user (-setting). In this template, there are about 25 serious columns together, and abbreviations all over to make it fit. I'd say, cramming these two tables in one row is not a great layout principle. -DePiep (talk) 22:34, 21 June 2010 (UTC)

Checked: if I reduce the size to where two tables appear in one row, some vertical lines disappear from the table and text is almost illegible. My tube is 16". I'd still like to use it for some more time ;-) -DePiep (talk) 22:40, 21 June 2010 (UTC)

I'm on a large screen now, but was on a tiny one last week, and didn't notice anything. Next week I may be able to check it again. But I don't see why we should force it to split on a large screen when it works just fine in a single row. — kwami (talk) 23:43, 21 June 2010 (UTC)

To force a second row is not ideal, but the current way is ugly when over two rows, and it still needs abbreviations. That's a lot of compromise. -DePiep (talk) 00:12, 22 June 2010 (UTC)

(I will be back on this. For the moment I am sandboxing with the table) -DePiep (talk) 14:55, 26 June 2010 (UTC) {od} Consider this thread cosed. Development now in the standard template-sandboxes sub the template. -DePiep (talk) 17:21, 14 July 2010 (UTC)

Differences between IPA-chart and the table here

See chart image Pulmonics, Original full IPA chart and {{Consonants}}.
I see differences, and have questions for improving the table here. -DePiep (talk) 12:50, 9 July 2010 (UTC)

Column headers

IPA has the top-row of column headers, each spanning two or more columns:
LABIAL | CORONAL | DORSAL | RADICAL | LARYNGEAL
Propose adding these to the Pulmonic consonants table here. -DePiep (talk) 12:50, 9 July 2010 (UTC)

Labiodental plosives non-IPA?

The labiodental plosives p̪  • b̪
Question: The chart says: they are non-IPA (grey). Should we mark them as such? -DePiep (talk) 12:50, 9 July 2010 (UTC)

Labiodental plosives: different symbols

The labiodental plosives p̪  • b̪ are shown as ȹ (qp ligature) and ȸ (db ligature) in the chart.
Question: To be noted somewhere? Strangely, voiceless labiodental plosive only shows the ligature, not p̪. -DePiep (talk) 12:50, 9 July 2010 (UTC)

Missing bottom line Lateral flaps

The chart shows a ninth row, named Lateral flap. The symbols in there are in the second table, "non-pulmonic and other symbols": ɺ • * (non-IPA).
Question: should the row be added or why deviate from the IPA-chart? -DePiep (talk) 12:50, 9 July 2010 (UTC)

non-IPA symbols

There are five non-IPA markings (including the 9th row "Lateral flap") on the chart image. Also, in [table in IPA] there are three more.

• я* epiglottal trill
• ɢ̆ * Uvular flap
• ɬ̢ * voiceless retroflex lateral fricative
• ʎ̥˔ * voiceless palatal lateral fricative
• ʟ̝̊ * voiceless velar lateral fricative
• ɺ̢ * retroflex lateral flap
• ʎ̯* palatal lateral flap
• ʟ̆* velar lateral flap
  

Question: can I put these symbols in the table here? -DePiep (talk) 12:50, 9 July 2010 (UTC)

The lateral flap and epiglottals are frequently dropped from IPA charts to save space, and that is a consideration for our template. This has the unfortunate effect of leading non-linguists into thinking they are somehow not normal pulmonic consonants, but it's a bit difficult to justify an extra row for the lateral flaps. (Not to say it shouldn't be done, it's just not obvious it should be done.)

The template is largely a navigation aid. As such, IMO it is nice to link all available articles to cells in the template, even if, like the dentals, they do not have separate IPA letters. Most of your "non-IPA symbols" here are perfectly good IPA. ȹ and ȸ are not IPA, p̪ and b̪ are. IMO there really is no reason for the non-IPA symbols in a template; they're in the chart, greyed out, just because they are sometimes seen in the literature. Likewise, you might want to stick to official IPA, like ɢ̆ ʟ̝̊ ʎ̯, for the other cells in the chart. — kwami (talk) 01:24, 10 July 2010 (UTC)

I'm not an phonetics-scolar. I try to improve the wiki-charts involved. So any IPA-knowledge is welcome.

re Space versus dropping information: Space should not be a concern here. There really is no reason to compromise an IPA-chart for this reason. e.g. the German Wiki uses full width for the Consonants: de:Konsonant. Only trying to cram two tables in one row creates this problem (and the right one is only useful for someone who already knows what to look for -- little navigational help). If space is a problem, we'll use our freedom of layout to solve that.

re IPA/non-IPA: Kwami explained here that it's about the (basic) symbol is not defined in/by IPA, and not the character does not appear on the IPA chart. Usage of the switch "shownonIPA=yes/no" should be corresponding to this.

re row for the lateral flaps: Space should not be a concern (an extra row, ie vertical, the least). And the unnecessary general "other laterals" can be to the point. Shall I propose that extra row in a sandbox, or would it be cancelled anyway for phonetic/IPA reasons? -DePiep (talk) 08:34, 13 July 2010 (UTC)

New column headers in the sandbox

See: Template:IPA_consonant_chart/sandbox, and the corresponding navbox Template:Consonants/sandbox. Any suggestion on adding the top columnheaders? -DePiep (talk) 12:39, 13 July 2010 (UTC)

Already moved ahead. -DePiep (talk) 17:22, 14 July 2010 (UTC)

Template layout improvements

Continued: previous discussion here at Kwami's talkpage. Will continue here from now. -DePiep (talk) 23:24, 14 July 2010 (UTC)

RE Kwami, if the bow on the affricates makes them not clear or legible, that's a font-issue, and also a IPA-definition-issue. We should not 'solve' that by not showing them. What we can do at Wikipedia is: show them as good as we can. Also, because bowed and unbowed glyphs are both used in IPA-world, we'd do good by showing both options. So we can have two rows for the affricates like here, where affricates are visible in two variants. todo: add row/columnheader, show them big enough, arrange the table nice, etc. Can be done. -DePiep (talk) 00:23, 15 July 2010 (UTC)

The tables forced on one row or more

This is what I propose. I'd say the three small tables can be in one row. Of course the spacing etc. needs attention. But if a two-row-split is not wanted/allowed/reasonable, let's not spend time on the details. To me, spreading the small tables over three rows like here, I don't see anyones wish come throu. -DePiep (talk) 23:58, 14 July 2010 (UTC)

Kwami, RE your latest version . This was my suggestion/proposal. Of course the small tables in one row #2. But this is already an elaborated solution. But before any time spend on a layout and table-shuffling, I'd like to know if you are willing to test & accept a two-row solution. If yes, let's meet in a sandbox later on. If not, we'll have to stop trying.-DePiep (talk) 00:31, 15 July 2010 (UTC)

{{od} In the template:IPA consonant chart/sandbox:

• Made the three smaller tables same width, 1.5em*10 columns (Affiricates) = 15em, * font-size 120% = 18em total. That's for the other two too then
• All IPA-letters are (still) in font-size 120%, which is good for the eye
• The placement is done by "float" (CSS), which (per browser situation) drops the tables, tetris-style, to or near their preferred position.
• Made the tables rectangular (for the eye too) by introducing a colorset (following wiki-guidelines on color-set, readability/accessibility and systematic usage). Took yellow (hue 60) as the base color. Now the tables are more alike (except for the height and cellsize).
• Tried some minor extra's.
• See also: template:Consonants/sandbox and, recommended, template:Consonants/testcases
  -DePiep (talk) 23:30, 15 July 2010 (UTC)

btw, the version I overwrote at the sandbox is here. -DePiep (talk) 23:37, 15 July 2010 (UTC)

In this version, Kwami has changed the floating tables into a enveloping, fixed table. I think this is not an improvement, because:

- RE: Stand alone usage tables: The bare tables can be used (4 together) stand alone in an article (i.e. not in a navbox-wrapper, footed, and so not showing in the Printable version). If stand alone usage were not needed, the table could be in the navbox directly, Example: IPA, or it's sandbox-copy {{Consonants/testcases}} (this testcase shows the stacked sandboxes, very useful).

- RE: Rigid table vs floating: Putting tables in a supertable is very rigid, while using the CSS floating style lets the user's browser organize the placement, according to each user's screen & browser setting. It allows for zooming, textzooming, different screensizes (widths!) etc. Kwami, is this not your first and foremost preferred option: having all in one row when possible (as in your screen & settings, I understand), and also nicely when rearranging is needed (as on my screen, incidentally). Here you wrote: I can't tell you why, but it did mess up the formatting of the table within the wrapper: Instead of the charts being side-by-side, one was under the other. — kwami (talk) 20:20, 21 June 2010 (UTC). Could you describe which general solution you'd like? Floating (yes, needs some improving, but just write what the grand idea should be in this, pls), or rigid (either vertical or horizontal - you know my opinion by now), or a third way.

-DePiep (talk) 08:56, 16 July 2010 (UTC)

What you're calling one row, I see as two, and what you call three, I see as one.

I think the three small tables need to be linked somehow. Else they scatter almost randomly across the page, and look like a mess. Any way they could be stacked when all tables are allowed in a single row, but lie side by side when forced below the main table? That would probably be ideal.

That is,

.................. .................. ..................	.....
	.....
	.....

if the screen is wide enough, and

.................. .................. ..................
.....	.....	.....

if it's not. — kwami (talk) 09:07, 16 July 2010 (UTC)

(intermediate: I reverted yr edits in the regular {{IPA consonant chart}}, the sandbox version you copied into production was not suitable. It messed up beyond our preferrences, i.e. not a matter of style etc.). -DePiep (talk) 09:20, 16 July 2010 (UTC)

Okay, I'll revert to before we started all this. The way you have it is an absolute mess. — kwami (talk) 09:20, 16 July 2010 (UTC)

this version (with the three tables) a mess? It was good for days! I'd like you to diff between "not looking nice" (I could agree or not) and "disrupting a page" (which is revertable). -DePiep (talk) 09:46, 16 July 2010 (UTC) (signed afterwards)

(To be clear, Kwami: for the moment the only topic is layout and floating. Other changes I will not discuss for now, but I might revisit then afterwards. Anyway, the Consonants testcases are working (using Consonants, not IPA). Now I can read yr reaction above) -DePiep (talk) 09:46, 16 July 2010 (UTC)

RE your examples: I agree, that is the ideal situation. Float can do that by nature (needs tweaking for sure). My first try was: make the 3 tables equal width, which produces a nice column when stacked (yr example 1.) Is what I did. (15em*120%=18em @ 100% font). Second try, solve problem 2: The bad outcome I saw in some situations is like:

.................. .................. ..................
.....	.....
	.....

(with borders more out-of-line then shown); right-below is the disturbance of course.

Suggested solve A: make the three together width smaller than the big Pulmonic table. That way, they will always fit under the big table, not forcing (CSS-floating) to put one on a "third" row. But then we'd have to drop the equal-width feature, which would make the example-1 situation less regular to the eye ("less nice"). Suggested solve B: or, alternatively, and saving the equal width nicety, we'd have to make the Pulmonics-table a bit wider (3*15em @ 120%font+ small overmargin). I'll have a look into this. -DePiep (talk) 10:08, 16 July 2010 (UTC)

Sandbox talk

Are you not watching the sandbox? I'm copying from the talk page:

We should decide on either using or not using the tie bars; it's a waste of space to have both options.

If we do use them, we need to add them to the ejectives too.

We could use an under tie bar for tɬʼ, kxʼ, kʼ etc. However, both underties and overties screw up the formatting of p̪f, b̪v, t̪θ, d̪ð, dɮ, tʃ, dʒ, tʃʼ, ɖʐ in all IPA fonts a reader is likely to be using. — kwami (talk) 00:54, 16 July 2010 (UTC)

Okay, have both here if you like. But it's no good for the actual template. — kwami (talk) 08:33, 16 July 2010 (UTC)

Here's what your last edit looked like to me:

File:Consonant sandbox temp upload.png

I'm sorry, there are many talkpages: this one, IPA-c-chart-sandbox-talk, Consonants-wrapper-sandbox, testcase-talk: I prefer (and will act so) to concentrate on this one talk page, to create a thread.

RE affricates with bow: I will not go into that now, as stated above. The issue is: layout of the tables on the page. This way of working, we can handle only one issue at a time.

Good news: solution B seems to work (make Pulminics-table wider, so the three small ones will not exceecd its width). What's needed: tables NP+A+CAC together are 3*15em=45em+margin

The Pulmonics table was 42em, so by adding only 3em it wins over the small tables! Of course I put them in the left column (6em -> 9em) using width:* and setting overall width in the topline ({| ... width:46em); rm the wrapping table to allow floating.

I created a second, separate stack of sandboxes:

{{IPA consonant chart/sandbox2}}, ({{Consonants/sandbox2}}, {{Consonants/testcases2}}.)

Please take a look. Now I'll take a read&look at your works. -DePiep (talk) 11:37, 16 July 2010 (UTC)

-DePiep (talk) 11:37, 16 July 2010 (UTC)

RE the printscreen: you're right, that's fault beyond just ugly. Thank you for showing me. Wished I'd seen it before, must be controllable.

So we'll have to deliver a big improvement in from the sandbox, all edits right in one go innit? That's do the tests then. -DePiep (talk) 11:49, 16 July 2010 (UTC)

Kwami, you realise one sandbox was sub this talkpage Template Talk:IPA consonant chart/sandbox? I'd suggest I'll leave you alone on the regular {{IPA consonant chart/sandbox}}. Then I'll use the /sandbox2-stack mentioned above, trying not to mix up... btw1, the ugly printscreen is caused by '{| align="right"'- styles in these tables. That overrules any floating. btw2, Maybe I a logout for now. Enjoy your sandboxing, and we'll meet here again. -DePiep (talk) 12:13, 16 July 2010 (UTC)

So, how does this bring home your tables examples 1 and 2 from above: {{IPA consonant chart/sandbox2}} ? -DePiep (talk) 05:05, 17 July 2010 (UTC)

This is what I see currently (16" tube; zoomed + until small col/rowheaders are a bit readable. Less zooming-> next to each other, but illegible regular text):

Noting: the right-hand block is placed irregularly. This is not good for a final situation, wiki should accomodate both our situations. Also, Wiki MoS on images layout might be more decisive (like: side them, left or right; take care of text flow) -DePiep (talk) 18:42, 23 July 2010 (UTC)

That (sandbox2) looks really good. At full screen, all four tables are stretched out in a single row. At less than full screen, the three small ones stack up atop each other, to the right of the main table, which is the same height as the three of them. At an intermediate browser size, there is main table, non-pulmonic, then a stack of affric + dbl artic, which looks okay.

I agree that what you see on your screen is not acceptable. My only criticism with the new version is that, IMO, the tables crowd too closely together. Perhaps that extra 3 em of space that you need, or part of it, could be an extra, invisible, column on the right side of the main table? That should force a gap between it an the small tables, improving legibility IMO, without otherwise changing the layout.

I'd say, let's go ahead and edit the actual template to match. If you can improve the table spacing (I don't know where you added in the extra 3 em), even better, — kwami (talk) 19:50, 23 July 2010 (UTC)

(edit conflict)

- (ec: saving my edits first, will read after) Thank you for looking into this, and your reaction here. Good that the layout is OK with those screenwidths you checked. (If you want to try from Sandbox2 ahead, I suggest you copy Sandbox2 to "your" sandbox so called #1; I could do that for you, re naming & links etc, if asked here.).

- On the +3em in left column: re the margin you'd like could be elsewhere, see below. The construction "width:*" works great here, is my experience. It also solves sub-1em rounding effects etc. And, as an after effect, I would like to keep the lefthand column wide, because if there's space, we could enlarge the 67%-font there (later on).

- About inter-table space, this is what I used & discovered (at index CSS by C3W). CSS uses these words style, float, align:right, span, div. I don't wanna be teaching, but you might appreciate it. Or skip it.

- In CSS, the straight cell-text is in an inner box. From there, walking to the outside, you pass: 1. padding (between text and cell-border), 2. cell-border, 3. margin. Each can be 0 width or more. What I set to 2px (now 0.25em) was these outer margins, right and bottom only, of each table of the 4. That is why they keep equal (white) distance, whatever arrangement.

- If you want more intertable whitespace, experiment with these margins I'd say (They are in the top row of each table (=4x), and -right and -bottom are mentioned (=2x)). At the moment, eacht table has the same space (figures) here.

- I changed (just now) the this margin-widths from "2px" to "0.25em", because our width-calculations are totally em-based, which is crucial. Also, this way the inter tablespace grows with the font-size (to 120%).

The margin cannot be colored (by me, tried), it would be illustrative.

Please keep commenting ideas and experiences etc here. {{IPA consonant chart/sandbox2}} -DePiep (talk) 20:47, 23 July 2010 (UTC)

Technical next:

-propose I copy /Sandbox2 into /Sandbox. Its "yours" for experimenting & checking.

re Trying more inter-table whitespace: see above. It's the "margin-right:0.25em; margin-bottom:0.25em" code now, for each table (=4).

If we are satisfied we can agree here: copy from a tested & agreed sandbox into live. (To me this is prefered to retype the edits).

No hurry, we still have some days left before Wiki is due to be finished. -DePiep (talk) 20:58, 23 July 2010 (UTC)

I didn't think CSS was supported by WP. (?)

I'll take a jab at it.

What's "due to be finished"?

— kwami (talk) 21:09, 23 July 2010 (UTC)

How does sandbox1 look?

Actually, your change from pxs to ems was an improvement. You might prefer different spacing that what I have, depending on what looks good on your screen. — kwami (talk) 21:17, 23 July 2010 (UTC)

(edit conflict) Lost my text. Minor. ("due to be finished" is a joke. CSS is in WP, but with little documentation). Will take a look. -DePiep (talk) 21:30, 23 July 2010 (UTC)

Just looking at Sandbox1. So you changed the Pulm-table -2em to 44em (which indeed is where it happens). Now at my screen, with the three below the Pul-table, they stick out (because they're calculated to be under 46em, not under 44em.) I recall that was a problem in some screensettings (cannot reproduce this problem now with sandbox1, so no argument).
And my question: do you want that space (right side off Pul-table) to be so much extra? Can't we give all margins some extra space? Must say, this is for the eye mostly (but the space for a more 100% rowheader font I'm losing...). A demo at {{Template:IPA consonant chart/Sandbox2}}, in a minute. Say 1em width every margin, from 0.25em. -DePiep (talk) 21:55, 23 July 2010 (UTC)

No, it doesn't need to be that wide. 1 em is fine.

You're right. I thought the margin would protect the small tables from getting jumbled up, but it doesn't. They can't be any wider than the main table.

I'd like less vertical space between the small tables. The three of them together used to be the same height as the main table. Now they're higher. — kwami (talk) 22:17, 23 July 2010 (UTC)

Horizontally: OK, so we talk 1em horizontally, for each table for now. Must keep in mind & eye that any rightmost table also takes this space, possibly forcing a newline for a tabel for this 1em. (I accept). I like & prefer that the small tables have this horizontal space too (rest of eye). Calculations (think 100%font=OK, all is in em's) Widths: 3 small tables each = "width:15em;", 3x15=45 em's. Inter table whitespace: 3x1em=3em. Total: 48em (OK with me, so Pulm-table must be 48em or more MINUS 1em margin-right=47em). Let's try this.

Vertically: later more, too much now (should not interfere with horizontal. Can be separate). In /Sandbox2 in a minute. -22:57, 23 July 2010 (UTC)

Vertically: in /Sandbox2, changed margin-bottom from 1.0 to 0.5em. Reduces the three-table-stack (but to me not relevant). Never thought of height, if we need space, we take it. signed -DePiep (talk) 23:10, 23 July 2010 (UTC)

The main table is now wider than the three small tables together, so I'll take out a little extra space. The height looks good. — kwami (talk) 23:38, 23 July 2010 (UTC)

Shoulda been 47em. But I narrowed the small tables to 1/2 em, which means 46em for the main table; since the small tables form a group of their own regardless, it shouldn't matter that they don't have the same margin as the main table. — kwami (talk) 23:43, 23 July 2010 (UTC)

(edit conflict)

Concluding

Kwami, clearly we agree about the way of table-arranging (i.e. by floating through CSS). The tests & talks do converge in this. Other questions are unrelated can be dealt with later. I'd like you to take any of the two sandbox, as is, and put it into production-wiki. Talk we keep doing here, por favor. These are our trying sites:
SANDBOX

• {{IPA consonant chart/sandbox}}
• {{Consonants/sandbox}}
• {{Consonants/testcases}}

SANDBOX2

• {{IPA consonant chart/sandbox2}}
• {{Consonants/sandbox2}}
• {{Consonants/testcases2}}

-DePiep (talk) 23:57, 23 July 2010 (UTC)

eh, right now I do like /sandbox2 very much. You're right. Copy from sandbox into live wiki please! Right now! -DePiep (talk) 00:02, 24 July 2010 (UTC)

Beautiful, Kwami. Width & height both look great here! (that is: three aside in one column, and three below in one row). And lining nice. Now will you let me sleep please ;-) ? -DePiep (talk) 00:09, 24 July 2010 (UTC)

(ahum, maybe not every pix, I'm disconnecting already. But for me: into wiki article space please) -DePiep (talk) 00:12, 24 July 2010 (UTC)

All done. Good work, and good night! — kwami (talk) 00:54, 24 July 2010 (UTC)

 Done

Nice job we did. Here's a Written Barnstar, Written Barnstar for each of us. -DePiep (talk) 12:33, 24 July 2010 (UTC)

Further changes

Separeted into two sections from main entry by — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 14:16, 27 July 2010 (UTC).
Separation by: -DePiep (talk) 18:49, 27 July 2010 (UTC)

Further changes (sidenote)

aeusoes1

:

On a sidenote, the three boxes are below the chart on my browser (IE) no matter the window's width. This isn't a problem, but for some reason the width of {{Consonants}} doesn't adjust to this so that it's absurdly wider than the consonant chart. I'm not sure if there's a way to fix that (either here or there). — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 14:16, 27 July 2010 (UTC)

Indeed, behavior described is not as expected and indeed, it's not destroying the page, but not nice.

Browsers Safari and FF behave correctly, so I suspect IE, probably in rendering the "float" and "align" settings of the tables. I'll take a look around on this. -DePiep (talk) 12:58, 30 July 2010 (UTC)

See the sandboxes {{IPA consonant chart/sandbox}} (this version) and {{Consonants/sandbox}} (this version).

The IPA chart sandbox has one change: Pulmonics table from align="left" to float="left".

Q: Is behaviour in IE now better? But, funnily, now in my FF and Safari I see the effect described ... That is, the three small tables always stay under the big one, however wide the page by zooming. (Stroked, based on edit typo; see next entry) -DePiep (talk) 13:28, 30 July 2010 (UTC)

Improved version. float:left was edited wrong, is now in style=""-construct. Behaviour in my FF and Safari now correct.

Question: how do the two sandboxes look & behave in IE? -DePiep (talk) 13:39, 30 July 2010 (UTC)

• {{IPA consonant chart/sandbox}}
• {{Consonants/sandbox}} (navbox)

- DePiep (talk) 21:48, 30 July 2010 (UTC)

Further changes (grand suggestion)

aeusoes1

I'm in the mood to make some suggestions to this chart that will move it further away from the official chart and enhance the presentation:

1. Create an alveolopalatal column. Put it next to the palatal column so that the symbols used for both palatal and alveo-palatal sounds (namely, [ɲ c ɟ ʎ]) can be present in both columns.
2. Remove the qualification of "pulmonic" from the chart. This way we can add nonpulmonics.
3. Add a click row.
4. Add an ejective row.
   1. This may only be feasable for stop ejectives, in which case we can add further ejectives in a sidebox.
5. Add an implosive row.
6. Create an affricate row. They have a place of articulation and are a distinct manner of articulation. We still won't have a place for [tɬ dɮ], but they can remain on the side as "other affricates" or join [ɫ ǁ tɬʼ] in an "other lateral" category. — [aeusoes1 aka] Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 14:16, 27 July 2010 (UTC)

Looks good and comprehensable.

I am not an IPA-person yet, I more about columns and tables and codes. So the IPA-details I cannot understand and will not discuss too deep. But this I understand: if you go away from IPA-chart, there is a long and insolvable discussion between IPA-knowers. So please discuss & conclude here on talk first, before anything is changed. Especially since the current table is sensitive to any (technical) change. There are a lot of compromise in there already!

Also, I want to add that user aeusoes1 here, already has altered the table away from the IPA-one: [1]: putting three fricatives halfway into approximants. Which is not IPA-chart, and undiscussed.

So please, discuss & conclude here first, IPA-friends. -DePiep (talk) 19:41, 27 July 2010 (UTC)

After these changes it would not be an IPA-chart any more. Still, the proper IPA-chart might be needed on wiki. I suggest: continue this development in new template:

• Template:Integrated consonant chart. Then this IPA-chart can stay IPA, and development there is more free re other sources etc. -DePiep (talk) 12:35, 30 July 2010 (UTC)

It would still be an IPA chart. However, IMO the clicks won't work: they don't correspond well to the points of articulation of non-clicks.

I'm not too crazy about the idea of the other mergers either. Maybe the affricates, but the non-pulmonics are independent of the lenition cline which the chart currently partially follows. They're also so obvious symbolically that it seems almost silly to include them in the main chart. The problem IMO with including ɕ, ʑ is that we conflate another dimension, obscuring the character of those Cs. Why not add a column for labiovelars if we're going to do that? — kwami (talk) 13:58, 30 July 2010 (UTC)

zh.wikipedia has a column for alveolo-palatals, including ɕ, ʑ, and the unofficial symbols ȵ, ȶ, ȡ and ȴ. It doesn't have a separate column for the labio-velars. Double sharp (talk) 10:26, 13 May 2012 (UTC)

zh: uses the two-year old template we had here. The ones we use now on en: here are very well evolved to show the beter thing. And, btw, are you tailing me, Double sharp? -DePiep (talk) 23:08, 13 May 2012 (UTC)

No, I'm not. :-) I just happen to have a secondary interest in these things. Double sharp (talk) 07:36, 14 May 2012 (UTC)

"a column for alveolo-palatals, including ɕ, ʑ", you write. To me it looks, at zh; with our 2-yr old template copy, they are in separate columns (ɕ unlinked, ʑ linked to [2]). But on the same row. Which column you refer to? -DePiep (talk) 23:50, 14 May 2012 (UTC)

I'm referrring to the column with "齦顎" (which is Chinese for alveolo-palatal) at the top. Double sharp (talk) 05:00, 15 May 2012 (UTC)

If we start adding 2ary articulations, like [ɕ, ʑ], then we should add all 2ary articulations, which would be a huge chart. If we add non-pulmonics, we should add all of those, which would also be a huge chart. But when the inventory of a language is given in the literature, the pulmonic, glottalic, lingual, linguo-pulmonic, and linguo-glottalic consonants are presented in separate charts, unless they're conflated because there are only a couple of them.

If we do this with the consonants, why not with the vowels? We could add nasal vowels to the vowel chart, glottalized vowels, strident vowels, etc., just as people do when giving the inventory for a language. — kwami (talk) 05:55, 15 May 2012 (UTC)

cʼ palatal ejective

I discovered there is an article on cʼ palatal ejective. The symbol is not in this IPA-chart. It may be expected as a non-pulmonic ejective, of course. Question: shall we put it in here? Possible:

• No, leave it out. They're just examples anyhow.
• OK, but on the second row (with tɬʼ tʃʼ)
• OK, but it the proper top row, whatever the visual consequences (the table would become wider, disturbing the layout. Would need sandboxing for sure)

I have no preference but for a non-disruptive sequence. Any suggestions? -DePiep (talk) 18:34, 5 August 2010 (UTC) linkfix -DePiep (talk) 18:35, 5 August 2010 (UTC)

Okay if it doesn't mess up the layout, but it's not an important sound. — kwami (talk) 20:03, 21 August 2010 (UTC)

Redlinks

There's really no reason to have the redlinks on this template. Per WP:NAV:

"Red links should be avoided unless they are very likely to be developed into articles, and even if they do, editors are encouraged to write the article first."

I tried removing them, but apparantly my coding skills aren't up to snuff. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 18:50, 21 August 2010 (UTC)

Yeah, they can be taken out. (I just created stubs for a couple of them.) The problem was that the remaining sounds got misaligned, messing up the voiced-voiceless pair format. — kwami (talk) 20:01, 21 August 2010 (UTC)

Oh I can believe it. This wouldn't be the first time I've botched up this table, so no hard feelings. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 21:32, 21 August 2010 (UTC)

Bilabial approximant and Dental approximant

I think that Bilabial approximant and Dental approximant shall be linked. --84.61.155.9 (talk) 18:03, 8 July 2011 (UTC)

How or in what sense? -DePiep (talk) 22:22, 8 July 2011 (UTC)

ɥ and ɫ not a pair?

See the Co-articulated consonants (Approximants). Is the vertical line between ɥ and ɫ appropriate? Or are they a regular voiceless--voiced pair, as most other cells have? -DePiep (talk) 16:05, 17 July 2011 (UTC)

From chart 2005.svg the IPA chart I learn that ɥ is voiced (so should be on the right side of a pair then).-DePiep (talk) 16:53, 17 July 2011 (UTC)

Both are voiced. They are not a pair. I guess that's the reason for the seperate cells. -- machᵗᵃˡᵏ🙈🙉🙊 18:54, 17 July 2011 (UTC)

Answered then. Thank you. -DePiep (talk) 23:48, 17 July 2011 (UTC)

table fix

Can somebody fix the table, the borders are not right. Please? --Tyranitar Man (talk) 06:13, 10 December 2011 (UTC)

Done -DePiep (talk) 10:33, 10 December 2011 (UTC)

Also fix the positions of the retroflex/palatal/velar/uvular nasals. Already done. --Tyranitar Man (talk) 11:39, 10 December 2011 (UTC)

OK in Template:IPA_consonant_chart but not in International_phonetic_alphabet#Consonants, can't see why -- 86.70.86.147 (talk) 23:15, 11 December 2011 (UTC)

Edit request on 11 June 2012

I would, for the sake of completeness, like to see the symbol for the voiceless palatal approximant j̊ be inserted next to its voiced counterpart on the  IPA chart since it is a sound that exists in a few languages and has its own article on Wikipedia. Skwuim (talk)

Skwuim (talk) 17:32, 11 June 2012 (UTC)

Without comment about the value of the request, I'm declining to do the request until there is consensus. Any other auto-confirmed logged in user will be able to make the change. --RA (talk) 20:46, 11 June 2012 (UTC)

Palatal and alveolopalatal nasals

...are not the same. There are languages that do distinguish them. See Irish phonology for examples, where it uses [ɲ] for palatal and [n̠ʲ] for alveolopalatal. Is it even accurate to describe alveolopalatal consonants as dorsal instead of coronal? - Gilgamesh (talk) 03:48, 18 June 2012 (UTC)

Yes, they are not the same, just as dentals and alveolars are not the same. Yet ⟨n⟩ is used for both of those. ⟨ ɲ⟩ is typically used for the alveolopalatal, just as ⟨n⟩ is typically used for the dental. Yes, alveolopalatals are dorsal, because they do not involve the front of the tongue. ⟨n̠ʲ⟩ would be a palatalized postalveolar coronal, and would go in the postalveolar cell; I doubt any language contrasts that with alveolopalatal, just as I doubt (m)any langs contrast [s̠ʲ] and [ɕ]. — kwami (talk) 06:41, 18 June 2012 (UTC)

Adding interwiki link to Template in Hebrew - [he:תבנית:טבלת עיצורי אפ"ב]

Could someone please add it? The template in Hebrew is still being remodeled but it exists--EsB (talk) 18:00, 14 July 2012 (UTC)

Pharyngeal stops

Why are the pharyngeal stops in the gray area?. I am aware they don't exist in any spoken language and it's very rare to hear them, but they are possible to pronounce.--Adamsa123 (talk) 10:03, 29 July 2012 (UTC)

I notice that there was a discussion to that effect at talk:pharyngeal consonant but don't see an agreement that they can. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 14:12, 29 July 2012 (UTC)

Ladefoged writes, "No language makes stops consistently in the upper part of the pharynx". Starting in 1993, The IPA charts show voiced pharyngeals as being impossible. But voiced epiglottals occur, so that's quite odd. Esling seems to be the authority here, so I'd say we go with whatever we can find there. — kwami (talk) 19:01, 29 July 2012 (UTC)

Yes, that's odd. However, contrary to what we present at IPA, the chart currently presented by the IPA site presents voiceless pharyngeal stops as possible (File:IPA chart (C)2005.pdf). What does Esling say about this? --JorisvS (talk) 20:07, 20 September 2012 (UTC)

Affricates

With all those affricates to the left of the main chart, wouldn't it make more sense to move them to two rows in the main chart (one for the central ones and one for the lateral ones)? --JorisvS (talk) 08:48, 5 June 2015 (UTC)

• Support. That's what's done on the Alemannic Wikipedia, but they only have three or four articles about affricates. Peter238 (talk) 09:01, 5 June 2015 (UTC)
• Support. The current setup risks giving the impression that "affricates" is a special category outside regular pulmonic consonants. (Although given the current layout for fricatives, we might need three rows: sibilant, central non-sibilant, lateral?) --Trɔpʏliʊm • blah 09:04, 5 June 2015 (UTC)

We don't need to write "central non-sibilant", as there are no lateral sibilants. Peter238 (talk) 09:42, 5 June 2015 (UTC)

I don't think it is really necessary to have a third row with affricates currently in the template, but I don't mind either. --JorisvS (talk) 09:45, 5 June 2015 (UTC)

Hmm... actually we do have to write "central non-sibilant" if we want to have three rows. The laterals are by definition non-sibilant, but writing "non-sibilant" and "lateral" feels incomplete. I suggest the following sollution (too lazy to make a table now):

Sibilant Central

Non-sibilant Central

Non-sibilant Lateral

The second "Central" and "Non-sibilant" would be merged with the first ones. Peter238 (talk) 10:37, 5 June 2015 (UTC)

IMO "lateral" is sufficient. Adding "non-sibilant" is redundant, and implies a contrast with sibilant lateral affricates. — kwami (talk) 05:46, 18 June 2015 (UTC)

• Support. I think about this every time I look at this chart. Three rows would probably be a bit excessive. If we really want to be conservative, we could even just add one affricate row and leave the lateral affricates where they are. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 14:12, 5 June 2015 (UTC)

IMO, if we're going to add one, we need to add both. Otherwise we'd have lateral affricates separated out, which is even weirder than separating out affricates. — kwami (talk) 05:47, 18 June 2015 (UTC)

• Support. Pulmonic isn't really the reason, though, or we'd need to include the doubly artic'd Cs as well. — kwami (talk) 05:46, 18 June 2015 (UTC)

 Done. Peter238 (talk) 22:34, 19 June 2015 (UTC)

Hey, what about the non-sib affrics [tɹ̠̊˔] and [dɹ̠˔] (maybe w/o the retraction), aren't they how a lot of people pronounce "tree" and "dream"? — kwami (talk) 02:21, 21 June 2015 (UTC)

That depends on the analysis.

- Many scholars analyse these as mere sequences of two consonants. That's the only analysis I've encountered when reading about New Zealand English.

- Wells in his latest LPD (2008) analyses /tr, dr/ as postalveolar affricates, and calls /ʃ, ʒ, tʃ, dʒ/ palato-alveolar.

- The latest (2014) Gimson's Pronunciation of English (pp. 186-192) also analyses these as (phonetic) affricates. Cruttenden says that in /tr, dr/, /t, d/ are often postalveolar. A few pages earlier, on page 177, he says that /t, d/ may be postalveolar in this environment, but alternatively, /r/ can become alveolar.

- So, these may be either postalveolar or alveolar, with the former being more common. I'm going to update the articles accordingly, and create sections for the non-sibilant postalveolars. Peter238 (talk) 11:35, 21 June 2015 (UTC)

 Done. Peter238 (talk) 12:25, 21 June 2015 (UTC)

Semi-protected edit request on 17 June 2015

Change colspan="2" style="border:1px solid; border-width:0px 1px 0"| Lingo­labial to colspan="2" style="border:1px solid; border-width:0px 1px 0"| Linguo­labial because of the inaccurate transcription of "Linguolabial" as "Lingolabial" StixTheMiner (talk) 17:52, 17 June 2015 (UTC)

 Done. Thanks for catching that. --JorisvS (talk) 19:53, 17 June 2015 (UTC)

Questions

Hello all! I had a question about this template. I'm just starting to learn about linguistics and such and barely understand anything about it, so sorry if these are all silly/obvious questions for those of you who are more experienced in these matters. :) Anyways. Here are the actual questions.

Why do some of the symbols ( n̼̊ l̼ ɺ̼ ɴ̥ ɢʁ tɕʼ ) not link to a page? If it's because a page doesn't exist, it would be good to at least have a red link so people can see a page is needed, right? If it's some other reason, what is it?

Why is "ɫ" (Co-articulated consonants, Continuants) not in the table properly? Does it have a meaning, or is there some odd formatting problem (which might then, of course, only be seen on my computer)?

What does it mean when a letter is between 2 "manners"? (Like β, which looks like it's between "Non-sibilant fricative" and "Approximant" of the "labials" section [ɸ and ɸ˕])

Beneath the tables, there are some comments, including: — Shaded areas denote pulmonic articulations judged to be impossible or not distinctive. Why are these "judged to be impossible or not distinctive"? Did the IPA itself say this, or is this something that's obvious to anyone with a basic understanding of linguistics or something like that? I couldn't find an explanation on here, or a link to an explanation. Thanks! JonathanHopeThisIsUnique (talk) 04:11, 31 October 2015 (UTC)

Hello.

We don't create pages about consonants that haven't been found yet in any language. You can always do your own research, and if you find reliable sources, go ahead and create these pages.

The main table contains only consonants that have no secondary articulation, and [ɫ] is [l] that is velarized (alternative symbol: [lˠ]), uvularized (alternative symbol: [lʶ]) or pharyngealized (alternative symbol: [lˁ]). Oh, it's about the placement. No idea.

It means that the article about the symbol [β] covers both the fricative and the approximant.

Yes, it's either IPA or Esling (2010) (full article name: John Esling (2010) "Phonetic Notation", in Hardcastle, Laver & Gibbon (eds) The Handbook of Phonetic Sciences, 2nd ed.) that say this. Peter238 (talk) 13:02, 31 October 2015 (UTC)

The placement of ɫ looks funny on my display as well. It's slightly off from the co-articulated consonants table. — Ƶ§œš¹ _{[lɛts b̥iː pʰəˈlaɪˀt]} 16:40, 31 October 2015 (UTC)

Proposal: order rows by sonority

If we move the nasal row down and the lateral rows up, the chart will be (roughly) ordered by sonority scale. That will allow rows to be grouped into broader categories like the columns are. Example (I would've moved the approximant row too, but it seems linked to the non-sibilant fricative row in a way that I don't have the linguistic knowledge to untangle.) — Preceding unsigned comment added by 69.126.68.38 (talk) 03:57, 17 June 2017 (UTC)

Oppose. The current order is based on the official IPA chart, except adjusted so that it more accurately captures the place of articulation (e.g. [d̞], a lowered plosive, or [ɹ̝], a raised approximant, means a non-sibilant fricative). It is also used by many publications when discussing the phonemic inventory of a language, including many Wikipedia articles on language phonologies. So it is a convention already supported by many reliable sources, I for one see no reason to ditch it. Nardog (talk) 15:40, 19 June 2017 (UTC)

ʩ and ʩ̬

Why are they included in the chart? They are only used in describing disordered speech and don't occur regularly in any language or dialect. The box should be gray; true pharyngeal / epiglottal nasals are impossible because occluding oral airflow at that point of articulation must also occlude nasal airflow. — brine_for_the_brine_god (talk) 8:09, 4 August 2017 (UTC)

Co-articulated

Why is there a section for "co-articulated"? The official IPA chart doesn't have such a section. It only has "double articulations".

More generally, why depart from the official chart at all? --Amir E. Aharoni (talk) 07:21, 8 October 2017 (UTC)