Talk:Noto fonts

Unicode coverage of Noto fonts

This blog post (An open source font system for everyone) published yesterday states that "Today, Google’s open-source Noto font family provides a beautiful and consistent digital type for every symbol in the Unicode standard, covering more than 800 languages and 110,000 characters". Unfortunately, this is simply not true. If you go to the Noto fonts download page you will find there are big gaps in the coverage of Unicode characters by Noto fonts. For example, there is no font for the Tangut script, added in Unicode 9.0, which comprises 6,881 characters. Ken Lunde, who is responsible for the CJK characters used in Noto fonts recently stated that "Noto Sans CJK covers a little under 30K of the 80K CJK Unified Ideographs in Unicode". There are a total of 128,237 characters in the latest version of the Unicode Standard (v. 9.0 released in June 2016), and if you install all available Noto fonts you get coverage for approximately 64,000 characters, far less than the 110,000 claimed in the blog post, and only approximately 50% of the 128K characters in Unicode 9.0. It is a design goal for Noto fonts to cover all characters in Unicode version 9.0 except for some CJK characters outside the BMP (see What are Google's plans for Noto (so called "Phase 3")?), but this design goal somehow got mistranslated into an actual achievement in the blog post.

I also note that this Monotype article (More than 800 languages in a single typeface: creating Noto for Google) claims that Noto fonts cover "hundreds of thousands of characters", which is even worse hyperbole, and patently impossible given that there are only 128,237 characters in the latest version of the Unicode Standard.

Anyway, both the Google blog post and the Monotype article are unreliable, and cannot be used to support the assertion that Noto fonts now cover all 110,000 Unicode characters which Danny B. added in good faith. Given that the numbers given in these two sources are unreliable I do not see how we can use them as sources for any information about Noto fonts. BabelStone (talk) 14:57, 7 October 2016 (UTC)

I'll second that. If this isn't even the most unicode-comprehensive font ever, then yes, it's still newsworthy and worth having, but there needs to be somewhere on the internet a mention of why it's newsworthy and what is should be seen as.--Mrcolj (talk) 16:34, 11 October 2016 (UTC)

Emoji Fonts not working

There's a line in there saying that the Emoji Color fonts don't work on Mac or Windows, but only on Linux. At least on the machine in front of me, none of the glyphy fonts work on the Mac--not Egyptian, not Deseret, not Aramaic or Javanese. If these are all for the same reason, will someone who understands that reason mention it, maybe even make a list of the ones that don't work and add a quote for when Google expects them to be working. Without counting, it may 50 of the 184 TTFs that won't load on my computer. --Mrcolj (talk) 16:34, 11 October 2016 (UTC)

The Noto Color Emoji font does not work on Windows or Mac by design, and this limitation is documented in the Noto help pages, so we can mention it in this article. However, I have not seen any statement on the Noto help pages that any other Noto font is not intended to work on Mac, so there is not source to specify in the article which fonts don't work on Mac. BabelStone (talk) 22:15, 11 October 2016 (UTC)

Font Family?

Because this is 184 different TTFs, should we use the word "font" or even mention it's misuse? I'm asking the purists here. I can accept that Noto and Noto Bold are the same typeface, but not the same font, but even in this modern world shouldn't we be calling these non-overlapping files of visually-related glyphs "font families"? (My question isn't as much a prescriptive "should" as whether this article should simply refer to Noto as a typeface consisting of 184 fonts, or something like that.) I only cared enough to ask because when i downloaded it, I then had to guess and check whether if I installed all of them they'd add 1, 100, or 184 entries to my "font list," because if they did add more than 1, I'd simply not install the ones I didn't want. Other unicode-large fonts, say Tahoma or Times New Roman or Code 2000, only add one entry to one's "font list."--Mrcolj (talk) 16:34, 11 October 2016 (UTC)

I'm not a purist, but I think that the article is sufficiently careful in not talking about a single Noto font. The opening sentence of the lede calls it a font family, which I think is what you are asking for, and thereafter it talks of fonts in the plural. The table of Noto fonts should make it even clearer that there are a large number of individual fonts in the family. BabelStone (talk) 22:06, 11 October 2016 (UTC)

Table Out of Date

I noticed that the tabe was out of date looking at the devanagari info - it seems the list of fonts hasn't been properly updated since 2016 (as it says), the last update to Noto was October 2017 https://www.google.com/get/noto/updates/ . It's now 2019, so I'm tagging it for update; hope no one minds.- VJ (talk) 16:26, 12 March 2019 (UTC)

List versions of Noto Emoji

Hi ☺
We can see that there are several versions of Noto Emoji. We can see the differences here
https://commons.wikimedia.org/w/index.php?sort=relevance&search=%221f606%22
https://commons.wikimedia.org/wiki/Emoji/Table
We have a least

• Noto Color Emoji KitKat (Android 4.4)
• Noto Color Emoji Lollipop (Android 5)
• Noto Color Emoji Nougat (Android 7)
• Noto Color Emoji Oreo (Android 8)
• Noto Color Emoji Pie (Android 9)

So is it's worth adding a table listing thoses versions?
--Tuxayo (talk) 18:21, 11 August 2019 (UTC)

Removal of facts from reliable source

user:LiliCharlie decided to remove reliably sourced facts [1] 89.12.60.198 (talk) 16:44, 19 October 2019 (UTC)

1. Noto Sans includes an ambiguous lowercase l.
2. Noto Sans displays "U+10f LATIN SMALL LETTER D WITH CARON" (ď) like d+apostrophe, no caron visible.
3. Some projects provide a package for installing Noto fonts, e.g. Debian, Arch Linux, CTAN.
4. Since 6.0 LibreOffice bundles Noto.

89.12.60.198 (talk) 16:48, 19 October 2019 (UTC)

user:Blythwood did it too [2] 89.12.60.198 (talk) 16:49, 19 October 2019 (UTC)

I do actully agree that this is a flaw in Noto Sans, but I'm not an accredited or accepted expert on fonts or linguistics, and it's common in many widely-used fonts. Some random person grousing on a forum isn't a reliable source. Blythwood (talk) 17:14, 19 October 2019 (UTC)

+1. It seems that at least three users removed this. Why? WP:NOTEVERYTHING says: "A Wikipedia article should not be a complete exposition of all possible details, but a summary of accepted knowledge regarding its subject." Note that the word in italics. Love —LiliCharlie (talk) 17:25, 19 October 2019 (UTC)

user:LiliCharlie, the article is not even near to "be a complete exposition of all" relevant details. 89.12.60.198 (talk) 17:28, 19 October 2019 (UTC)

Blythwood: What is your source for "Some random person grousing on a forum isn't a reliable source" in relation to the three facts listed above that you removed? 89.12.60.198 (talk) 17:22, 19 October 2019 (UTC)

user:Blythwood did it again, plus removing a further fact, here is the list:

1. Noto Sans includes an ambiguous lowercase l.
2. Noto Sans displays "U+10f LATIN SMALL LETTER D WITH CARON" (ď) like d+apostrophe, no caron visible.
3. Some projects provide a package for installing Noto fonts, e.g. Debian, Arch Linux, CTAN.
4. Since 6.0 LibreOffice bundles Noto.

and merging two sections [3]. All without any explanation. 89.12.60.198 (talk) 17:24, 19 October 2019 (UTC)

---

Page looks like advertisement - NPOV violation

With the above fact removals, the page looks like advertisement. Facts that *could* lead a user to not use the font, are removed. This is a WP:NPOV violation. 89.12.60.198 (talk) 17:32, 19 October 2019 (UTC)

"uh" versions in table

What does "uh" mean next to the version numbers in the table? Is there a reason it's not explained? I can't add the explanation because I don't know ...--Paleolith (talk) 17:53, 11 October 2020 (UTC)

@Paleolith: "uh" marks unhinted versions. They're in a separate folder on GitHub. Love —LiliCharlie (talk) 18:09, 11 October 2020 (UTC)

@LiliCharlie: Thanks. Is there any reason I should not add a sentence explaining? Paleolith (talk) 20:38, 21 October 2020 (UTC)

If you believe Dcljr's recent edit doesn't do the job, then go for it and add a short explanation. Love —LiliCharlie (talk) 21:12, 21 October 2020 (UTC)

"Near-unicode-compliant" vs "Unicode-compliant"

Obviously, the entire Unicode ecosystem is a little bit messy as a whole, but there is a somewhat pedantic point to be made with noto. Instead of following strict unicode standard glyph names, noto (google) emoji glyphs follow translation-team names, which follow the vendors (google). This is circular, not based on unicode directly. This circular accountability actually allows noto renders to deviate from unicode glyphs, and this does happen in practice beyond a level that likely makes noto "Unicode compliant" by any reasonable description. -- 68.2.171.3 (talk) 13:17, 21 July 2024 (UTC)