Help talk:Multilingual support/Archive 1 Source: en.wikipedia.org/wiki/Help_talk:Multilingual_support/Archive_1

This is an archive of past discussions about Help:Multilingual support. 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.

[[:Image:Internet Explorer 6.0 SP1 (DejaVu Sans).png]] is being used on this article. I notice the image page specifies that the image is being used under fair use but there is no explanation or rationale as to why its use in this Wikipedia article constitutes fair use. In addition to the boilerplate fair use template, you must also write out on the image description page a specific explanation or rationale for why using this image in each article is consistent with fair use.

Please go to [[:Image:Internet Explorer 6.0 SP1 (DejaVu Sans).png|the image description page]] and edit it to include a fair use rationale. Using one of the templates at Wikipedia:Fair use rationale guideline is an easy way to insure that your image is in compliance with Wikipedia policy, but remember that you must complete the template. Do not simply insert a blank template on an image page.

If there is other fair use media, consider checking that you have specified the fair use rationale on the other images used on this page. Note that any fair use images uploaded after 4 May, 2006, and lacking such an explanation will be deleted one week after they have been uploaded, as described on criteria for speedy deletion. If you have any questions please ask them at the Media copyright questions page. Thank you.

BetacommandBot 01:33, 8 July 2007 (UTC)

There are now some pages that contain classical Mongolian script. However, there seem to be certain issues with the rendering of the "γ" or "gh" (The classical script has two different 'g's. One that comes with e, ö, and ü, and is usually transliterated to 'g', and one that comes with a, o, u, and is usually transliterated to 'γ' or 'gh'). In "Mongγul", the "γ" should display two small dots on the right, but some browsers I checked did not seem to reflect this (Microsoft Vista, IE7 and Firefox), at least not for all fonts. For an example, see the classical script at the intro sentence at Mongolia, which on my browser it wrong. Mongolian Baiti font, OTOH, seems to work better, see the box at Mongolian language. Is this a problem with Unicode or with my computer? Yaan (talk) 16:22, 17 January 2008 (UTC)

Seems to work fine now. Maybe it was really a problem with the computer I am working with. —Preceding unsigned comment added by Yaan (talk • contribs) 00:41, 27 January 2008 (UTC)

My machine is running Windows Vista. When I browse with IE7, the default browser come with Vista, it can't render some languages even after I've installed a "recommended" font. I've found such failure in bug:, got: & my:. Any idea? --Quest for Truth (talk) 19:17, 26 April 2008 (UTC)

I'm not sure but perhaps Wikipedia:Reference desk/Computing#Characters support reveals a shortcoming in the page Nil Einne (talk) 10:41, 20 November 2009 (UTC)

There have recently been some questions raised about browser support for MathML. I think that a table of MathML symbols somewhere in this article might be useful to people who are struggling with browser/font configuration problems. Are there any objections? DavidCBryant 17:18, 15 February 2007 (UTC)

Wikipedia doesn't support MathML (yet), neither as an output not as an input format. We already have Wikipedia:Mathematical symbols, whcih lists some of the symbols supported in HTML. —Ruud 17:33, 15 February 2007 (UTC)

Well, there is some support for MathML, because it's built into browsers. So, for example, if you have the right fonts installed and a browser/OS that supports Unicode, you can probably read all these symbols:

≤ ≥ ϑ ℘ ℑ ℜ ℵ ∅ ⊆ ⊇

and if you don't have the right fonts installed, some of these at least are not going to display very well (probably as an empty box). I coded the line above as HTML entities, specifically, as ≤ ≥ ϑ ℘ ℑ ℜ ℵ ∅ ⊆ ⊇. And today I had a problem with a seasoned editor who started mucking up a bunch of articles simply because he didn't have all the right fonts on his system.

I understand the bit about "the information already exists", but that information has to be easily located, or else it's not very useful. I've got a query in at the "Help" page on the same subject, because I can't find a link from the main "help" button into the articles about Unicode and MathML. I don't want to hurt this article, but I think I can make a good case for the proposition that "math" is a language with its own set of symbols. In fact, it's probably the most widely recognized written language in the world. DavidCBryant 18:46, 15 February 2007 (UTC)

Those character entries are plain HTML, not MathML. I don't believe a table of such symbols would make a useful addition to this article. A section on which font you need to have them displayed correctly would be. This help page is still very much under developement, so I suggest you add a section on this. I might do it at a later time. Cheers, —Ruud 15:21, 16 February 2007 (UTC)

Thanks for the feedback, Ruud. I did finally find the Help:Special characters article, and there is in fact a (hard to locate) link to it from the main help page. I think making information available in more than one place, and coding cross links between logically related articles, will improve the chance that someone who's looking for a critical piece of information can actually find it. So what I'll probably end up doing is making this article point to "Special characters", and vice versa. And a little test table in here (so a user can see if his browser supports some Unicode math symbols) is probably a good idea, as well (along with a short blurb about fonts). Have a great day! DavidCBryant 16:35, 16 February 2007 (UTC)

Actually Help:Formula is probably what you want Nil Einne 19:45, 11 March 2007 (UTC)

No, I don't thnk so. I've got Unicode2000 working under Linux now, but my Windows fontsets are still incomplete. The PNG stuff works just fine. DavidCBryant 23:03, 11 March 2007 (UTC)

Doesn't the page discuss fonts etc as well? If not it seems to me that it should so you might want to discuss in in the talk page. It seems to me it's most relevant there since it's only an issue you're likely to encounter if you turn on MathML or HTML rendering for Math Nil Einne (talk) 10:43, 20 November 2009 (UTC)

How do I get support for Burmese/Khmer rendering? —Preceding unsigned comment added by 66.95.123.6 (talk) 18:37, 3 November 2009 (UTC)

Me too. They both have their own countries, as does Ethipoia, they should be automatic by now. Chris (クリス • フィッチュ) (talk) 12:55, 7 December 2009 (UTC)

I got some problems with enabling some fonts. There are scripts such as Tifinagh, Balinese etc that do not have their own section in the Internet Options in fonts but appear in User Defined. However, even if i enable them here, they still are squares. Any help? Kanzler31 (talk) 03:19, 7 October 2010 (UTC)

Are there any suggestions for IPA (and other) font-support for android? Searching a bit it does not appear that non-rooted phones can install extended fonts. Cheers, — sligocki (talk) 05:30, 2 March 2011 (UTC)

Links to Code2000, Code2001, Code2002 are dead. --Bothary (talk) 15:19, 25 March 2011 (UTC)

I was criticized for changing a character from Latin "C" to the correct Unicode character "Ϲ" (a GREEK CAPITAL LUNATE SIGMA SYMBOL U+03F9), because "few people's browsers actually display the technically-correct Unicode character". The section Unicode simply states "Because UTF-8 is backwards compatible with ASCII, and most modern browsers have at least basic Unicode support, most users will experience little difficulty reading and editing Wikipedia." That may be true for articles that use only ASCII, but too optimistic in general. (For example, I have not been able to get texts in the Malayalam script to display properly in my browser.) Is there a way of determining which non-ASCII Unicode characters are generally safe to use, in that they are supported in most browsers?  --Lambiam 19:06, 13 September 2011 (UTC)

For European languages, the only thing that's probably absolutely safe is Windows Glyph List 4. Capital lunate sigma has been around in Unicode since version 4.0 (2003), but unfortunately some popular fonts (including those that have been included as standard preinstalled fonts in most Windows versions) that cover polytonic Greek are older than that, and still don't contain it. Fut.Perf. ☼ 12:59, 23 September 2011 (UTC)

Thanks for the information. I see that the polytonic characters such as ἁ, ῃ and ῶ are also not on that list, yet seem to be relatively safe. I guess you just have to know.  --Lambiam 16:37, 23 September 2011 (UTC)

Yeah, I think all recent operating systems have come with some font preinstalled that covers the polytonic block (like Palatino Linotype on Windows). Unfortunately, Microsoft was so much ahead of its time in introducing these Unicode goodies early on that they are now rather out of step with the new character additions that have happened in Unicode since, so the core repertoire of polytonic Greek is safe while many of the more outlandish symbols are not. Fut.Perf. ☼ 16:50, 23 September 2011 (UTC)

The link for Sundanese fonts no longer works. Anybody know of another source? — Preceding unsigned comment added by 68.238.228.4 (talk) 04:24, 14 January 2012 (UTC)

Oddly enough, going through an open proxy did it for me. (Living in Turkey, if I can't access a site, that's the first thing I try.) Try HideMyAss. (You'll have to Google it, because spammers use it to hide the targets of their links.) --Quintucket (talk) 22:55, 24 January 2012 (UTC)

For some reason about 1/3 of these sites are blocked in Turkey, including the googlepages and sites.google ones with the Javanese and Lontara fonts. Some of them have the "you're trying to access an evil site you bastard" page, and some simply don't load or give me a "resource not available" message. Not that Turkey needs much of a reason to block websites, but does anybody have any idea why this is?

Usually I get around this by going through an open proxy, however it seems that open proxies (I've tried several) don't like ftp sites. Could someone do me a favor and try to get the font at ftp://ftp.ethiopic.org/pub/fonts/TrueType/gfzemenu.ttf and email it to me? Though I don't really need it, since I already have a Ge'ez font installed, I've got an almost obsessive need to collect free Unicode fonts. On the other hand, if you can't access it either and you're not in a censorship-happy country, it would confirm that the link is dead, and should be removed. Thanks, Quintucket (talk) 23:59, 24 January 2012 (UTC)

For some reason about 1/3 of these sites are blocked in Turkey, including the googlepages and sites.google ones with the Javanese and Lontara fonts. Some of them have the "you're trying to access an evil site you bastard" page, and some simply don't load or give me a "resource not available" message. Not that Turkey needs much of a reason to block websites, but does anybody have any idea why this is?

Usually I get around this by going through an open proxy, however it seems that open proxies (I've tried several) don't like ftp sites. Could someone do me a favor and try to get the font at ftp://ftp.ethiopic.org/pub/fonts/TrueType/gfzemenu.ttf and email it to me? Though I don't really need it, since I already have a Ge'ez font installed, I've got an almost obsessive need to collect free Unicode fonts. On the other hand, if you can't access it either and you're not in a censorship-happy country, it would confirm that the link is dead, and should be removed. Thanks, Quintucket (talk) 00:06, 25 January 2012 (UTC)

Wp sites that include traditional Mongolian Unicode characters (e.g., Ulaanbaatar), currently are not displayed correctly on Apple (OSX or iOS) devices. The cleanest way to fix this seems to be to take some fonts with compatible licenses (e..g. one of those offered by www.mongolfont.com under GPL v2) and embed them via css. Is this possible on wp? I am assuming that the embedded font should be located on a wikimedia foundation server. Yaan (talk) 17:00, 12 September 2013 (UTC)

Please file a report in bugzilla: to include such a font for that language in mw:Universal Language Selector and the font will be used automatically wherever needed. —TheDJ (talk • contribs) 17:17, 12 September 2013 (UTC)

I have undone the addition of <span>...</span> tags that specify a font-family. My understanding is that text in the encyclopedia itself won't be marked with such tags, and that the rendering of this help page should be a guide to what you are going to see in the encyclopedia. -- John of Reading (talk) 08:17, 24 January 2016 (UTC)

Please include Safari among the browsers list - it's now much more widely used than (for example) Opera, thanks to the iPhone and iPad. In August 2015 Chrome accounted for over 50% of users, Internet Explorer and Firefox each with around 15%, Safari around 10%, and Opera less than 5%. Source: http://gs.statcounter.com/. Thanks. 80.47.170.163 (talk) 13:55, 8 September 2015 (UTC)

December 2016: Safari still not in the list of browsers! — Preceding unsigned comment added by 84.13.116.220 (talk) 14:35, 12 December 2016 (UTC)

Broken link in 2019 on a scripts.sil.org URL... --Wisdood (talk) 14:25, 15 April 2019 (UTC)

Since nowadays many people use phones as well as computers to access Wikipedia, I was wondering: should we replace "Your computer" with "Your device" in the scripts' test-rendering tables? Wikilackey (talk) 04:09, 11 July 2017 (UTC)

How about "Your browser"? - dcljr (talk) 23:33, 6 August 2018 (UTC)

Fair warning: Unless someone objects, I plan to replace "your computer" with "your browser" throughout this article. - dcljr (talk) 07:53, 11 August 2018 (UTC)

Let me observe that no browser ships with all the fonts needed to view the scripts on this help page and that many browsers don't ship with the necessary text shaping engine for complex scripts either (such as HarfBuzz, DirectWrite/Uniscribe etc.). So to me "your device" or simply the vague "your system" seems the better choice. Love —LiliCharlie (talk) 09:02, 11 August 2018 (UTC)

But is anyone actually viewing the page with anything other than a "browser"? I suppose you could argue some are using an "app", but that app must have some form of web browser built into it, right? - dcljr (talk) 03:19, 12 August 2018 (UTC)

When I visited my sister half a year ago I was astonished she viewed a Wikipedia article on her TV. This encyclopedia defines (web) browser as "a software application for accessing information on the World Wide Web," and I'm not sure if the "browser" she used isn't part of the built-in hardware that can't be modified/extended at all. I don't really mind if you change "computer" to "browser" though, it's an improvement anyway. Love —LiliCharlie (talk) 09:17, 12 August 2018 (UTC)

I'm pretty sure that TV's "browser" would still be software (held in its ROM or, probably more likely, flash memory). In any case, as you implied, "browser" may still be a preferable term, even if it's not universally accurate for some reason. - dcljr (talk) 00:55, 13 August 2018 (UTC)

And BTW, to address your first reply more completely: the fact that the browser may depend on other, lower-level software to provide font rendering — or, indeed, that it may be completely unable to display the fonts — is beside the point. We're merely saying "this is what it looks like in your browser", regardless of "why" it looks that way. - dcljr (talk) 01:06, 13 August 2018 (UTC)

I have just replaced "Your computer" with "Your browser/device" throughout. - dcljr (talk) 01:35, 16 April 2019 (UTC)

Looks good, seems to cover all the bases. Wikilackey (talk) 02:36, 25 April 2019 (UTC)

It looks like the first letter of that Armenian demo "Հ" is shown as somewhat like Greek "ς", how do we think about it? --Liuxinyu970226 (talk) 04:48, 10 April 2018 (UTC)

According to Wiktionary they are the same. I assume it is like the lowercase letter "a", which has 2 different appearances. --Voidvector (talk) 02:13, 20 December 2019 (UTC)

There is no Section for Samaritan but my PC doesn't display them properly.

Muonium777 (talk) 14:35, 4 February 2021 (UTC)

is it alright if i added unifont to all of the supported scripts (at least half on 'normal' version)? it does a fine job at being a fallback on my portable linux usb (mx). ill probably also add on the cjk (normal supports them just fine), upper and csur later on... FONZACUS (talk) 18:27, 18 July 2021 (UTC)

so, i found this font called "KNE Sans Serif 24" and has different types of support for each language, but a note for arabic: it does not combine the characters like it does in other fonts. link here

--Alefar (talk) 07:10, 19 September 2021 (UTC)

How good is the Noto Sans font for the Bhaiksuki script? --Apisite (talk) 09:36, 30 September 2021 (UTC)

The text sample contains U+170D which is not assigned a character in Unicode 9.0. It looks from the image like it should actually be U+1707 (DA). Nukeqler (talk) 19:10, 18 November 2016 (UTC)

See Baybayin#Representation of the character "Ra". DRMcCreedy (talk) 21:56, 18 November 2016 (UTC)

User Nukeqler is right. I can't spot any difference between unassigned U+170D (used for "ra") and U+1707 (letter "da") either. Moreover it seems strange to state that "U+010D is becoming the de facto standard for representing the character Ra" when of all the fonts mentioned here only Paul Morrow's Baybayin Fonts contain a glyph for character U+010D while the others are fully Unicode compliant and lead to a faulty display of the sample text that has a legacy instead of a Unicode encoding. This help page should at least warn users or better replace all occurrences of U+170D with U+1707.

On this issue see also page 7 of https://www.unicode.org/notes/tn35/indonesian-philippine.pdf "A new ‹r› letter." Love —LiliCharlie (talk) 14:01, 29 August 2017 (UTC)

I don't see any Unicode proposals to add TAGALOG LETTER RA and since newer Unicode fonts don't include it I'd say U+170D is no longer de facto. My suggestion is to update the sample text to use U+1707 instead of U+170D. A note could be added to state that some fonts include a separate, non-standard R at position U+170D. Page 7 of https://www.unicode.org/notes/tn35/indonesian-philippine.pdf is definitely helpful but TAGALOG LETTER RA didn't seem to make it into any of the Baybayin/Tagalog Unicode proposals. DRMcCreedy (talk) 17:30, 29 August 2017 (UTC)

I have discovered another issue with our Baybayin sample: Only U+1736 PHILIPPINE DOUBLE PUNCTUATION (which looks like a double slash) should be used in this script as a punctuation mark, see p. 5 of http://www.unicode.org/versions/Unicode10.0.0/ch17.pdf. I have decided to produce and upload an SVG that has U+1736 at the end of each line and uses U+1707 instead of U+170D. The text now reads:

ᜀᜅ᜔ ᜊᜏᜆ᜔ ᜆᜂ ᜀᜌ᜔ ᜁᜐᜒᜈᜒᜎᜅ᜔ ᜈ ᜋᜌ᜔ ᜃᜇᜉᜆᜈ᜔᜶
ᜀᜆ᜔ ᜉᜈ᜔ᜆᜌ᜔ ᜐ ᜇᜒᜄ᜔ᜈᜒᜇᜇ᜔᜶
ᜀᜆ᜔ ᜃᜇᜉᜆᜈ᜔ ᜀᜅ᜔ ᜆᜂ ᜀᜌ᜔ ᜊᜒᜈᜒᜌᜌᜀᜈ᜔ ᜅ᜔ ᜉᜄᜒᜁᜐᜒᜉ᜔᜶
ᜀᜆ᜔ ᜃᜇᜓᜈᜓᜅᜈ᜔ ᜈ ᜃᜁᜎᜅᜅ᜔ ᜋᜄ᜔ᜃᜁᜐ ᜐ ᜃᜉᜆᜒᜇᜈ᜔᜶

Any comments are welcome. Love —LiliCharlie (talk) 20:44, 30 August 2017 (UTC)

Sorry for waiting for you all! Unicode 14.0 has got u+170d for “Tagalog letter ra” Baokhang48812002 (talk) 14:38, 8 October 2021 (UTC)

In fact, Unicode 14.0 includes both U+170D TAGALOG LETTER RA and U+171F TAGALOG LETTER ARCHAIC RA. See the Unicode proposal for the details. DRMcCreedy (talk) 16:06, 8 October 2021 (UTC)

I want to know where to download support fonts for unicode 14.0. --Kankachi1980Kan (talk) 23:49, 14 December 2021 (UTC)

The browser support section speaks of Internet Explorer and not of Microsoft Edge, which suggests to me that it hasn't recently been tested in modern browsers. —Octavo (talk) 02:48, 8 April 2022 (UTC)

Google already released Noto Sans Samaritan, I wonder if someone would add that to the page since it would be helpful --Ternatean Libra (talk) 08:56, 16 May 2022 (UTC)