Emoji

Please try to keep all the discussions in the main forums on topic! If you have anything else, related to fonts, you want to share, please post it here!
William
Top Typographer
Top Typographer
Posts: 2038
Joined: Tue Sep 14, 2004 6:41 pm
Location: Worcestershire, England
Contact:

Emoji

Post by William »

I read an interesting blog article the other day. I replied and the comment has been added to the blog.

http://blogs.msdn.com/michkap/archive/2 ... 55146.aspx

One link in the blog article leads to the following page.

http://en.wikipedia.org/wiki/Emoji

That page too has various links, including the following.

http://www.nttdocomo.co.jp/english/serv ... ictograph/

That page also has links, two of which lead to detailed information, including Unicode Private Use Area mappings, for lots of pictograms.

An interesting aspect of this is that the colour in which each pictogram is to appear is specified.

It appears from the blog article that the Unicode Consortium are considering what to do about the possibility of encoding such emoji into regular Unicode.

William Overington

27 August 2007
Erwin Denissen
Moderator
Moderator
Posts: 11108
Joined: Fri Oct 04, 2002 12:41 am
Location: Bilthoven, The Netherlands
Contact:

Post by Erwin Denissen »

Hi William,

Thanks for the informative links. Emoticons are extremely popular :wink:
Erwin Denissen
High-Logic
Proven Font Technology
William
Top Typographer
Top Typographer
Posts: 2038
Joined: Tue Sep 14, 2004 6:41 pm
Location: Worcestershire, England
Contact:

Post by William »

Hi Erwin

Thank you for your note.

I am thinking of trying to make an experimental font to contain some of the designs in the linked pages and also to contain some designs of my own.

I am wondering quite how to proceed.

The emoticons are all on a 12 by 12 grid.

I am thinking that the font could also contain a latin alphabet, maybe not at first, yet the font designed with that in mind.

I am wondering about the following questions at present.

Should I make the whole font to be from 0 to 2048 vertical, with no descender space as such, yet making the baseline of the latin characters at 512 font units vertical? Or should I have descender space down to -1024 font units?

If the former, then pixel size is 512/3 font units, which has rounding effects, yet that is not a big issue. If the latter, then pixel size is 256 font units, yet the font does have descender space.

Should I make all latin letters 12 pixels wide or should I make wide m and narrow i?

I am thinking of adding my own designs for pictograms at U+E800 onwards so that they are nearby in the map to the others and a bit later in sequence.

Do you know how fonts for mobile telephones are done please?

William Overington

28 August 2007
William
Top Typographer
Top Typographer
Posts: 2038
Joined: Tue Sep 14, 2004 6:41 pm
Location: Worcestershire, England
Contact:

Post by William »

I have uploaded a first test font of emoji to the web.

http://www.users.globalnet.co.uk/~ngo/EMOJI001.TTF

The glyphs for the emoji in the font are all on the basis of a 12 by 12 grid of pixels.

The font includes items for producing chess diagrams.

They are mapped in the range U+E8C0 to U+E8FF.

The mappings are based on those in the following web page.

http://www.users.globalnet.co.uk/~ngo/chess.htm

For these emoji versions I have added 300 hexadecimal to the values of the code points. For example, in the chess.htm page a WHITE CHESS KING UPON A WHITE SQUARE is U+E5C1, so the emoji version is at U+E8C1.

The chess characters in their original form with the original mappings are in the Quest text font.

http://www.users.globalnet.co.uk/~ngo/QUESTTXT.TTF

Here is a chess diagram. It will probably appear as black boxes in this forum. Yet copying and then pasting into WordPad and then using the Emoji 001 font on the characters should produce the display.










William Overington

29 August 2007
Erwin Denissen
Moderator
Moderator
Posts: 11108
Joined: Fri Oct 04, 2002 12:41 am
Location: Bilthoven, The Netherlands
Contact:

Post by Erwin Denissen »

I haven't installed the font yet.

Mozilla Firefox uses code2000 to display the characters:
Image

Internet Explorer 7 shows the famous black boxes:
Image
Erwin Denissen
High-Logic
Proven Font Technology
William
Top Typographer
Top Typographer
Posts: 2038
Joined: Tue Sep 14, 2004 6:41 pm
Location: Worcestershire, England
Contact:

Post by William »

> Mozilla Firefox uses code2000 to display the characters:

I seem to remember that James Kass, the gentleman who produces the Code2000 font, included ligatures for various scripts in the Unicode Private Use Area so that people without Opentype-aware applications and fonts could nontheless be able to produce printed output of those scripts using ligatures. The ligatures are not encoded as individual characters in Unicode, which is why the glyphs are available mapped within the Unicode Private Use Area.

William
Erwin Denissen
Moderator
Moderator
Posts: 11108
Joined: Fri Oct 04, 2002 12:41 am
Location: Bilthoven, The Netherlands
Contact:

Post by Erwin Denissen »

I've just installed the Emoji font, and it immediately showed up in Firefox. Unfortunately it decides to use code2000 to display character $E8C0, most likely it doesn't like your empty glyph...
Image

IE7 still shows the boxes.
Erwin Denissen
High-Logic
Proven Font Technology
William
Top Typographer
Top Typographer
Posts: 2038
Joined: Tue Sep 14, 2004 6:41 pm
Location: Worcestershire, England
Contact:

Post by William »

> I've just installed the Emoji font, and it immediately showed up in Firefox.

I note that you say "immediately".

Did Firefox decide to use the font itself or did you need to tell it to use the Emoji 001 font rather than the Code2000 font please?

The text characters in your diagram do not appear to be Code2000 glyphs so I am wondering quite how Firefox decided which font to use for which character codes.

I have not used Firefox at all.

It does seem strange that a glyph for a space of a specific width should not be used.

William
Erwin Denissen
Moderator
Moderator
Posts: 11108
Joined: Fri Oct 04, 2002 12:41 am
Location: Bilthoven, The Netherlands
Contact:

Post by Erwin Denissen »

I note that you say "immediately".

Did Firefox decide to use the font itself or did you need to tell it to use the Emoji 001 font rather than the Code2000 font please?
I installed the font through MainType, then switched back to Firefox, and it already used the Emoji 001 font to show the chess diagram.
The text characters in your diagram do not appear to be Code2000 glyphs so I am wondering quite how Firefox decided which font to use for which character codes.
Are you sure? I have Code2000, version 1.13.

Image
It does seem strange that a glyph for a space of a specific width should not be used.
It will probably work, when you explicit tell the browser what font to use, although users can override that in Firefox.
Erwin Denissen
High-Logic
Proven Font Technology
William
Top Typographer
Top Typographer
Posts: 2038
Joined: Tue Sep 14, 2004 6:41 pm
Location: Worcestershire, England
Contact:

Post by William »

When I wrote "The text characters in your diagram do not appear to be Code2000 glyphs so I am wondering quite how Firefox decided which font to use for which character codes.", I was meaning the text characters for "Here is a chess diagram. It will probably ...".

I was meaning that the Firefox display seemed to use one font (Arial?) for the text, the Emoji 001 font for most of the Private Use Area characters and the Code2000 font for one of the Private Use Area characters.

I was wondering on what basis Firefox had made those decisions: for example, why did it not use Code2000 for all of the Private Use Area characters? It appears to have used (Arial?) where it could, then where it could not use (Arial?) it appeared to have tried to use Emoji 001 and where it felt, for whatever reason, justified reason or not justifed reason, it tried to use Code2000.

So I was wondering why it tried to use Emoji 001 in preference to Code2000. Indeed I suppose that I am now wondering why Firefox tried to use Emoji 001 at all: was it because it was installed on your system at a later date than when Code2000 was installed on your system?

William
William
Top Typographer
Top Typographer
Posts: 2038
Joined: Tue Sep 14, 2004 6:41 pm
Location: Worcestershire, England
Contact:

Post by William »

There is in the Esperanto language a word Ĉu which is spelled as C circumflex followed by u. To English hearing ears it sounds like "Choo" as in the song "Chattanooga Choo Choo".

http://en.wikipedia.org/wiki/Chattanooga_Choo_Choo

The Ĉu word approximately translates into English as "whether".

Its use is important as the word Ĉu can be put at the start of a sentence that is a statement and the full stop at the end of the sentence changed for a question mark and that turns that sentence into a question, without any changing or reordering of words.

So, I have assigned U+E880 as a stylized version of the word Ĉu which the idea that it changes the next Emoji item into a question.

Thinking further on this I have assigned U+E881 as a stylized version of the word Ĉu with a past tense verb implied, U+E882 as a stylized version of the word Ĉu with a present tense verb implied, U+E883 as a stylized version of the word Ĉu with a future tense verb implied. The letters used for the augmentation of the basic design used for U+E881, U+E882 and U+E883 are I, A, O respectively, those particular letters used because they relate to the way that Esperanto verbs are conjugated.

The font is available as follows.

http://www.users.globalnet.co.uk/~ngo/EMOJI002.TTF

So, using the list of Emoji linked from the
http://www.nttdocomo.co.jp/english/serv ... index.html page there could be the following.

U+E640 Rain

U+E881 U+E640 Has it been raining?
U+E882 U+E640 Is it raining?
U+E883 U+E640 Is it going to rain?

U+E641 Snow

U+E881 U+E641 Has it been snowing?
U+E882 U+E641 Is it snowing?
U+E883 U+E641 Is it going to snow?

These four new characters could be applied before an emoji item where the combined meaning makes sense.

If these emoji get used they might need a colour assigned, so I suggest green as that is the colour used for Esperanto.

These four new characters would not be useful for all question situations.

For U+E665 Post office maybe some other question emoji operator would be needed, such as a WHERE IS THE ... PLEASE?

Maybe the Esperanto word for where could be stylized to make such an emoji item. Maybe other Esperanto words could be stylized to make other emoji items.

http://en.wikipedia.org/wiki/Esperanto_ ... rrelatives

Using such question operators could result in more emoji being defined. For example, one for PHARMACY.

William Overington

1 September 2007
William
Top Typographer
Top Typographer
Posts: 2038
Joined: Tue Sep 14, 2004 6:41 pm
Location: Worcestershire, England
Contact:

Post by William »

An interesting aspect of emoji is that they localize in the mind of the user, into the language of the user.

So, if the emoji were developed then they could be used to facilitate communication between people who do not understand the same languages.

Suppose that mobile telephones were developed such that they have an alternative form of output for text messages which could be used instead of a telephone call when desired. That form of output could be as an infra-red burst of data, using technology similar to that used for television remote control devices.

The signals could be received by another such unit, or by a display unit on a desktop, or by an autoresponding computer-driven information point.

Then a person could ask such questions as WHERE IS THE NEAREST VEGAN RESTAURANT PLEASE? and the question be understood in a tourist information centre even if the tourist and the staff do not understand the same language.

Certainly, there are potentially an extremely large number of potential questions, yet maybe a sequence of two or three emoji could be used so as to convey information.

For example, suppose that WHERE IS THE NEAREST ... PLEASE? is encoded as one emoji item, and RESTAURANT is encoded as another emoji item and VEGAN is encoded as another emoji item. That particular combination of three emoji, in any order, could convey the desired meaning.

If one adds other emoji items for nouns such as HOTEL and CAFÉ and emoji for various adjectives and emoji for a way of indicating directions then one could develop the system into something potentially useful to many people.

There could also be an emoji item to indicate that a specific name follows. For example, suppose that one is in Vienna and wishes to ask the directions to the Café Mozart. There could be an emoji item such as WHERE IS ... PLEASE? and that could be used in conjunction with the emoji item for CAFÉ and the emoji item indicating that a specific name follows and the word Mozart. Maybe the rule would be that the emoji item indicating that a specific name follows is never the last emoji item in a sequence so that the sequence always ends with an emoji item.

There are lots of empty planes in the Unicode map. I feel that there is a great potential to use some of them to develop such systems and other systems. For example, some of the code points in one of the planes could be used to encode colours.

Some time ago I carried out some experiments with encoding graphics using some Unicode Private Use Area characters.

http://www.users.globalnet.co.uk/~ngo/ast03000.htm

http://www.users.globalnet.co.uk/~ngo/ast03100.htm

William Overington

6 September 2007
William
Top Typographer
Top Typographer
Posts: 2038
Joined: Tue Sep 14, 2004 6:41 pm
Location: Worcestershire, England
Contact:

Post by William »

Earlier this morning I posted a suggestion to the following Suggest a Topic page.

http://blogs.msdn.com/michkap/pages/4120528.aspx

Not as early as the 3:31 AM time of the post suggests as that is Eastern Standard Time of the United States of America, whereas I posted some time after 8:00 AM here in England.

In fact it is stranger still as Michael is in the United States West Coast where time is three hours behind the East Coast time. So three time zones are involved in all of this!

The suggestion is as follows.
In various recent articles in your blog you have mentioned lots of postings and discussions about the encoding of emoji and emoticons in Unicode.

Yet those postings are not in the Unicode public mailing list.

Could you please consider writing a blog article about what is in fact happening in this field as I, and maybe some other readers who are not representing an organization which is a member of the Unicode Consortium, would like to become aware of what is the present situation.
Whether an article will appear is not known as I write this post.

If an article does appear it would start off at the following place.

http://blogs.msdn.com/michkap/default.aspx

The article would then include a permalink, so if an article does appear then the permalink could be included in this thread in a later post by anyone who notices it.

William Overington

20 September 2007
William
Top Typographer
Top Typographer
Posts: 2038
Joined: Tue Sep 14, 2004 6:41 pm
Location: Worcestershire, England
Contact:

Post by William »

The blog article has now appeared.

http://blogs.msdn.com/michkap/archive/2 ... 34774.aspx

There has at the time of writing this post been one reply, from Andrew West, and he provides some links.

I am now in the process of gathering the documents for study.

I have noticed that the active links on the blog page have, at present, got a trailing > character in them which makes them difficult to use, particularly for right clicking to save the pdf.

So, I have reproduced them here for the benefit of readers, including myself. Andrew's comments are worth reading at the blog page.

William Overington

26 September 2007

----

Markus Scherer's original emoji mapping table:

http://www.unicode.org/~scherer/emoji/e ... c_pub.html

A zip of all the files in the above:

http://www.unicode.org/~scherer/emoji/0 ... -table.zip

A pdf version of the above:

http://www.evertype.com/standards/iso10 ... tc_pub.pdf

Michel Suignard's proposal for Japanese TV Symbols:

http://www.dkuug.dk/jtc1/sc2/wg2/docs/n3341.pdf
William
Top Typographer
Top Typographer
Posts: 2038
Joined: Tue Sep 14, 2004 6:41 pm
Location: Worcestershire, England
Contact:

Post by William »

I have been looking at the http://www.dkuug.dk/jtc1/sc2/wg2/docs/n3341.pdf document and have found within it a link to another pdf document which I am having difficulty saving to local storage.

So here is a link which will hopefully provide easy to use right click access to it.

http://www.dibeg.org/aribstd/STD-B24/AR ... 5.1_E1.pdf

William Overington

26 September 2007
Post Reply