Template talk:IPA/Archive 3

No underlining for
Underlining obscures certain IPA characters, particularly those with descenders that may distinguish them from similar characters, so I've added the style "text-decoration: none;" to the template. This works in IE but not Firefox. Did I do something wrong? Usually Firefox is pickier about the exact format, but I'm pretty sure I've followed this tutorial exactly. —Simetrical (talk) 4 July 2005 23:10 (UTC)


 * What you were doing seems fine, but why is   necessary?  The word is not a link, so it shouldn't be underlined in any case, right?


 * If you want to make it a link, and keep it from being underlined, that won't work without changing Wikipedia's style sheet. There's probably a declaration in the style sheet like , which would be more specific in scope than   applied to the surrounding span. The less specific declaration embedded in the page can't override a more specific declaration.  —Michael Z. 2005-07-5 04:13 Z 

I thought that any specification made in the style attribute of a tag is considered more specific than a specification made in the style sheet? refers to any chunk of text in &lt;a> tags, after all, while the low-level declaration only that specific chunk of text. —Simetrical (talk) 7 July 2005 02:27 (UTC)


 * My current recommendation is to never make IPA characters a link. If you want to link to something, make a nearby word a link instead. Nohat 5 July 2005 05:45 (UTC)

That just isn't practical in many cases. Look at Hebrew alphabet, for instance, over on the right (that really needs to be reworked to be much slimmer, incidentally). What should be done, making each entry a full wikilinked name like Glottal stop? Better to have the descenders obscured but allow anyone to figure out what it means by hovering over it or clicking, than fill the table with characters that will be totally incomprehensible to 95% of viewers. —Simetrical (talk) 7 July 2005 02:27 (UTC)


 * Linking IPA text is not the magic cure to making IPA comprehensible to that 95% of viewers who don't understand it. That table is a huge mess and I'd say that the linked IPA does not add to the table's comprehensibility. IMHO, that table should summarize just the basic information for each letter, and the detailed information for each letter should be on each letter's page, which can spell out the name of the sounds represented. Nohat 7 July 2005 02:37 (UTC)

Possibly, but a single unified table is useful for quick reference. (I've actually split it into two now.) In any case, how does linking a confusing IPA symbol like x or j, or a completely incomprehensible one like ʔ, not aid comprehension if someone wants to know how yod or whatever is/was pronounced? —Simetrical (talk) 7 July 2005 03:18 (UTC)


 * I agree with Nohat. Linking this way won't really help those who don't know IPA and will only annoy those who do. And linking every single usage of the template is really not useful. You can use a separate template or just clever normal linkage for that.
 * Peter Isotalo 11:52, 20 August 2005 (UTC)

Problem
The template is currently not displaying IPA at all. I do not know how to fix this --Vincej 12:55, 20 August 2005 (UTC)
 * It seems to be a problem with the code or something. Try to wait and see if it gets fixed before making edits.
 * Peter Isotalo 13:31, 20 August 2005 (UTC)


 * I have mentioned the problems to the developers on IRC. They are currently looking into it. Func( t, c, @,) 13:45, 20 August 2005 (UTC)

This is annoying...
I'm using Firefox and noticed that where this template is used, the text is displayed in a font called Gentium on my computer. But if it is not used, IPA text is displayed in Arial Unicode MS (which I greatly prefer over any other). Anyway, I just thought I should say that it does affect the font display in other web browsers. On some pages, not all IPA text uses this tag, but some IPA text that were recently added do use this tag. So on pages with a mix of old and new IPA text with this tag, I end up with some IPA text on the same graph displayed in Arial Unicode MS and Gentium. It looks very messy so I had to remove Gentium from my fonts folder just so that all IPA text would appear in Arial Unicode MS. --Hecktor 21:09, 2 October 2005 (UTC)


 * Very annoying; this is a brand-new behaviour affecting Safari too, and presumably all other browsers. It looks like Wikimedia has changed the way CSS comments are being rendered, and the second font declaration is now broken.  Instead of

font-family /**/:inherit;


 * The declaration which negates the fonts (for every browser except MSIE/Win) now renders with either a non-breaking space (in numeric entity format), or a regular space in place of the CSS comment:

font-family &amp;#160;:inherit;
 * or

font-family :inherit;


 * It mixes up these two versions on a single page. I'll try to figure out what happened and fix it.  —Michael Z. 2005-10-2 21:43 Z 


 * I filed this as bug 3588 on Bugzilla. —Michael Z. 2005-10-2 22:16 Z 


 * Just wanted to add that I'm using IE (on a university computer, so I get very little option), and I haven't been able to see IPA characters for a few days now. Very annoying!


 * I'll put the CSS for this template into the style sheet monobook.css, and just leave  in the template.  This should work.  I'll have to do this for Template:Unicode and Template:Polytonic, and any others too.  In-page tables or divs that have this construction will have to be updated.
 * I'll do it sometime today, and report back here. Is someone able to update the documentation on this page, and for the other two templates I mentioned?  —Michael Z. 2005-10-4 14:50 Z 
 * The template will add less page bloat—good.
 * Users can style whole tables or divs by adding —good.
 * Non-admin users won't be able to change the font list—good or bad, depending on your point of view.


 * Done. I'll summarize on the relevant talk pages.  —Michael Z. 2005-10-4 15:23 Z 
 * Thanks for that, it works fine for me now. Grinner 15:32, 4 October 2005 (UTC)


 * Whew. I was just waiting for the bug reports, so I'm glad it works for someone.  Thanks.  —Michael Z. 2005-10-4 15:42 Z