Chinese
When the Spring Festival comes, Chinese people have many traditional customs to create a warm and festive atmosphere. One of these customs is to plaster a large ideograph “福” on the door, to wish good fortune and happiness throughout the New Year, and some businesses would like to plaster their rooms with a special character 
to wish plentiful money and treasures to come in the New Year. This glyph is composed of four characters that are “招財進寳”, “財” means money, and “寳”means treasures. Actually, this glyph is not a real or genuine hanzi, and no character set includes it, so we can not enter this glyph directly via standard fonts. Fortunately, SING (Smart INdependent Glyphlets) technology offers us a way to create and use this character. Please follow these DIY (Do It Yourself) instructions for this glyph.
First, let us confirm our design goals:
1. Typeface of this glyph
I prefer a calligraphic style, so I chose Adobe Kaiti Std as the reference font, and built this SING glyphlet as an extra glyph of the same font.
2. Encoding of this glyph
Although the character for this glyph is not in any character set, and thus does not have a code point, we could regard this glyphlet as an alternate form of the ideograph “寳” meaning that this glyphlet will be treated as a variant form of the Unicode code point for “寳” Thus, we can access this glyphlet by searching for “寳” in the Glyph Panel.
Now, let us start building this glyph.
1. Building SING glyphlets
Step 1. Launch Illustrator CS3/4 Traditional Chinese version, and create a new document.
Step 2. Use the Type tool “T” to enter “寳” in a text frame, then apply the Adobe Kaiti Std font to this frame.
Step 3. Highlight this character, then choose Type > Create glyphlet,which will show the following window. Click the OK (確定) button to create a new glyphlet.
Step 4. Using Illustrator to build glyphlets
The following figure illustrates the five guide lines that are used to correctly position the glyph:
EM Box: This is square space for fonts in which ideographs are optically-centered, and you are strongly advised not to design the glyphlet out of its space.
Latin Baseline: Using a different Latin baseline can cause text to become inconsistent, so please make sure this glyphlet and other regular hanzi share the same Latin baseline.
Tip: Open the Layer Panel (Window > Layer), then lock the guide lines to prevent you from shifting or removing them while making the glyph.
The glyph of “寳” is a now a drawing composed of outlines, not text. Besides “”寳” we also need to borrow some components from other characters, such as “辶” “招” and “准”, Enter these characters using Adobe Kaiti Std in the text frame, highlight the text frame, then choose “Type > Create outlines” to convert the text to outlines. Now that
they are regular outline objects, you are free to modify their shapes as you desire.
Tip: If you want to delete, shift, or isolated the strokes of a glyph,you first need to ungroup or release the compound path. (Choose the glyph, click the right button of mouse, or Control-Click on Mac OS X,then choose ungroup or release compound path in the pop-up menu.)
The following figure illustrates the glyph that I designed:
Step 5: Save this glyphlet as a “gai” file.
The default Unicode code point shows the value for “寳” specifically U+5BF3, but it also allows you to change the Unicode code point. For example, you could input “宝” into this field directly, fill in the requirement items in the dialog, then choose “save as glyphlet” in the panel menu.
2. Registering glyphlets
Launch Adobe SING Glyphlet Manager (ASGM), import the new “gai” file, then refresh the glyphlets.
3. Using this glyphlet in ID.
Step 1: Launch ID CS3/4 then create a new document
Step 2: Using the type tool (“T”), enter the character “寳” into the text frame, then apply the Adobe Kaiti Std font to the text frame.
Step 3: Select the character, and when you open the Glyph Panel, you will find that the character is highlighted.
Step 4: Click the lower-right triangle icon to display a pop-up menu of alternate glyphs, double-click this glyphlet so that it is inserted at the text insertion point.
If you want to use this glyphlet in your own document, please click the right button of mouse on this link to save this ID file to your computer. Because SING glyphlets are always embedded in ID documents, you can easily copy and past glyphlets from any document to your own ID document, then apply any point size to this character. As you can see, it is easy to exchange glyphlets with other people. Enjoy SING!
Here is a design that includes this new glyph. If you wish to enjoy this Chinese traditional custom, please print this design and cut out the red shape, then plaster your place with this special character ,to wish plentiful money and treasures to come in the New Year!
English
设计自己的 “招財進寳”
当新春到来时，在中国有许多传统习俗可以营造浓浓的节日气氛。其中一个习俗就是在门上帖上一个‘福’字来祝愿新年平安幸福，而一些商家更喜欢通过另一个特殊字符期盼来年财源广进。这个特殊字符由“招財進寳”四个字组成，实际上它不是一个真正意义上的字，由于没有任何字符集包含它，所以我们也很难直接使用标准字库输入该字符，但是SING(Smart INdependent Glyphlets) 技术为我们制作和使用该字提供了可能，下面请看DIY该字符的介绍。
首先我们先要确定设计目标：
1> 字的风格
我喜欢书法字体，所以选用Adobe 楷体作为参考字体，并希望该字符以补字的形式与Adobe 楷体字库一起使用。
2> 字的编码
该字符不属于任何字符集，这导致它没有现成的编码可用，但我们可以将其看作”寳”的替代字形，这意味着这个字符将借用”寳”的Unicode值，这样做的好处是我们可以通过输入“寳”字很快在字符调板中找到它。
现在让我们开始制作这个新字。
一、创建字形
1,启动Illustrator CS3/4 中文繁体版并新建一个文档。
2,在工具条中选择“T”标识的文字工具，在文档中新建一个文本框，选用Adobe 楷体输入“寳”字，之所以在这里输入该字，是希望新建的字能借用“寳”的编码，和“寳”字产生关联。
3,选中该字，选择 “文字” > “从选取的文字建立字符档”，这时会出现以下对话框，直接点“确定”按钮即可。
4,利用Illustrator 工具制作字模。
在自动生成的字模中有五类辅助线可以协助字形的定位，请看如下图示：
EM Box 是指字库的字面大小,设计字模时请不要超出该框范围;
Latin Baseline 是指基线,在设计字模时请参考其它汉字与该基线的位置关系，确保与其它汉字在一个水平线上。
提示：建议打开图层调板（“视窗”>“图层”），利用图层的锁定功能锁住辅助线，以防在制作字形时移动、删除这些辅助线。
目前“寳”这个汉字已经被转化为轮廓的形式，除了“寳”字，我们还需要借用其它字的部件。请分别用“T”工具输入了“辶”、“招”、“准”这几个字，然后选中文本框，选择 “文字”>“建立外框” 将其转为轮廓的形式，当这些字符转为轮廓之后，我们就可以随意修改、删除这些轮廓。
提示：在转化为轮廓之后，如果想删除、移动、放大或缩小某些独立的部件，必须先解散群组或释放复合路径，具体方法是用选取工具先选中该字轮廓，单击鼠标右键调出右键菜单（在MAC上是按住Control键再点鼠标），在弹出菜单中选择“解散群组”或“释放复合路径”。
下面图示是该字模的设计效果。
5,保存字模为gai文件。
如果字符档调板没有被打开，可以通过 “文字”>“字符档” 调出字符档调板（如图所示），目前默认的Unicode值是“寳”的U+5BF3，如果你想将该字与其它字关联起来，比如简写的“宝”字，那在这里直接输入“宝”即可。填写必要的项目，选中字符档调板右上角的小三角打开调板菜单，选中“另存为字符档”将其保存为gai 文件。
二、注册字形
启动ASGM(Adobe SING Glyphlet Manager)，导入gai文件，并刷新字模。
三、在ID中使用字形
１,启动InDesign CS3/4，新建文档。
2,在文档中用Adobe 楷体输入“寳”字。
3,选中“寳”字,可以看到该字符在字符调板中被突出显示。
4,点击右下角的小黑角，在扩展字符中可以看到我们新增的字，双击该字形插入字符到文档中。
如果你也想在自己的文档中使用该字，可以点鼠标右键保存 该ID 文件。因为ID会自动嵌入所造的字形，所以只需打开ID文件，拷贝粘贴该字符到你的ID文档中，然后按需要任意设置该字的磅数即可。看，DIY自己的新字并与他人共享自己的作品就那么简单，你也来试试吧！
用该字在ID中设计了一个图案，祝愿新的一年里大家都
Comments
Thank you for your two entries explaining SING glyphlets. Information on this is not easy to find, and with my limited Chinese I especially appreciate your laying all this out in English.
If I understand matters correctly, Adobe’s Glyphlet Creation Tool is included with recent East Asian versions of Adobe Illustrator, although 3rd-party tools (such as Fontlab’s SigMaker) or Adobe’s Glyphlet Development Kit can serve to make glyphlets. Creating a glyphlet is only the first step: actually using it requires ASGM （Adobe SING Glyphlet Manager), which currently is available only as packaged with East Asian versions of Adobe CS2 and later products.
Thomas Phinney raised the possibility that SING technology might be included in western-language CS4 applications in his PDF “Adobe SING Technology” (presented September 28, 2006, p. 16) . If either the Glyph Creation Tool or the Manager came with my English-language Design Premium CS4 suite I can’t find it.
Over the years I have made many hundreds of 造字 (“zaozi,” or custom Chinese characters): I started with 24×24 bitmaps for the Eten Chinese add-on to DOS, moved on to 48×48 and even 64×64 bitmaps before going vector, first with Fontographer (with single-byte encoding) and now Fontlab. I set the type for scholarly material requiring Chinese, with a sub-specialty in archaic characters (that is, those pre-dating the unification of Chinese writing some 2200 years ago). Along the way, I’ve used Chinese versions of PageMaker under CWin3.1, CWin98 and Mac OS 7.5. InDesign 2 and PDF file format 1.3 were great leaps forward, with continuing incremental improvements since.
Many of my zaozi would work well as SING glyphlets, but a fair number might not because they do not fit clearly into any character set, and SING glyphlets are not supposed to use custom codes. The SING Tutorial says: “PUA (Private Use Area) code points shall not be used in glyphlets to the extent that it would allow PUA code points to persist within the workflow.” So maybe I don’t really need to use SING for zaozi, and can continue using Fontlab to encode characters as seems appropriate for each project.
Where I would like to use SING, then, is to supplement alphabetic fonts with the unusual characters scholars need. Phinney’s PDF speaks of SING’s potential to “Extend fonts a glyph at a time without cracking open original fonts” (p. 5). Music to my ears! Say you are handling production for a book on Chinese Buddhism. While the print version is the immediate concern, an on-line version is foreseen, requiring full Unicode compatibility. You want to set the alphabetic text in a typeface other than Times New Roman, but where can you get the dotted consonants (included in Unicode) necessary for setting romanized Sanskrit? You might reach for Fontlab, and start with a nice Opentype font whose EULA allows modification. Unfortunately, you cannot count on Fontlab being able to decompile the font’s features. Note that “feature” has a special meaning in Opentype, and one of the most important is kerning; and while you might have quibbles about particular kerning pairs, simply tossing the original font’s kerning would be foolish. I really don’t want to “crack open fonts,” so I’m hoping SING will become more widely available
Mr. Goodrich,
Thank you for your comments. This specific article was intended to illustrate one possible workflow. One of the ways in which SING distinguishes itself from legacy gaiji solutions is the fact that encoding is not a requirement. This is precisely why PUA code points are not allowed. The cases that you described, specifically glyphs for which there is no suitable base character in Unicode, can still be handled through the use of SING. If you contact me by email, I can provide to you a White Paper that I wrote on this topic. It accompanied my presentation at IUC32 last September. In fact, I would be happy to send to you the presentation and White Paper, both of which I am sure you’ll find to be useful.
ASGM & GCT currently are available only in CCJK version of CS4. You are not able to create new glyphlet and register glyphlet via non-CCJK CS4 applications, but the ID document with embedded SING glyphlets can be opened and edited by non-CCJK version of ID. If you want to exchange glyphlets with other people who uses non-CCJK version of ID, the easy way is embedding these glyphlets into the ID document via CCJK version of ID, and then send this document to them. ID will register these embedded glyphlets automatically without ASGM when opening this ID file.
Thanks to Ken Lunde and his note on Typlography. I will definitely try to follow this blog, since from time to time I am in discussions about Asian typography details, and I hope to learn not only technical things here.
I am astonished to learn that special CJK features are not available in my English or German version of InDesign CS4! With today’s modern OSs I see no reason for this, because all of them are Unicode based and come with CJK fonts. The latest release of Adobe FrameMaker makes CJK features available independent of product or OS language. So why is this distinction made with InDesign?
Hi Michael,
Thanks for your comments. I assume that your question is why ASGM & GCT currently only ship with CCJK version of ID & AI. I think that the reason is related to user requirement. The missing glyph problem often occurs in ideographic languages, such as Chinese, Japanese. Since this problem seldom happens in alphabetic languages, ASGM & GCT are not available in non-CCJK version of ID/AI.
Thank you both for clarifying basic points I glossed over rather carelessly. I did manage to pass the the “bao” demonstration glyphlet from one ID file to another in my English-language ID. Having an interest in the matter, I’ve long noticed the SING module loading when my ID starts up, and continue to hope non-CCJK users will achieve full access to the technology. This would continue Adobe’s pattern of increasing InDesign’s language abilities (for example, adding South Asian capabilities with CS4). SING has caused problems for a few non-CCJK users (as discussed in Adobe’s ID Windows forum), so it will be especially nice when we can all make full use of the benefits.
Ken Lunde kindly pointed me to Chapter 6 of the new edition of his CJKV Information Processing (O’Reilly, 2008). I should have known to look there. I had bought the previous versions as soon as they came out and meant to order the new one in December; I’ve done so now, and look forward to reading it. I still have trouble getting my mind around the notion of characters not requiring encoding — though I have occasionally outlined custom characters in electronic publications lest their non-standard encoding confuse searching and indexing.
I had intended my main point to be the potential value of SING glyphlets for alphabetic fonts, and undercut this with a niche example, Sanskrit diacritics. In a CJK context a better choice would have been the problems of romanizing Chinese, Japanese, and Korean: fonts with macron vowels for Japanese are not so hard to find, but breve vowels for Korean are far from common, while full pinyin diacritics remain rare.
Those with CCJK CS4 have the tools to add these diacritics via SING, but those are not the only CS4 folks out there needing to add a few characters without cracking fonts open. For now, non-CCJK users face serious obstacles: a new version of the CS4 package and separate hardware to run it on (since different-language versions of CS4 cannot run on the same system). Throw in the need for CCJK language skill (unlike the old InChinese add-on for InDesign CS2 [or the still older CJK versions of Quark], the CCJK CS4 user interface is hard-wired to just one language) and the current situation will discourage a good many folks who might want to experiment with glyphlets. Those would seem to be the kind of folks Adobe needs to help move SING into the mainstream.
IMO it is a misunderstanding of use cases if certain features are reserved to certain product versions (or OSes on which they run). Nowadays we keep text in Content Management Systems, use those systems to manage translation and have to create the final deliverables on the same systems. Finishing processes (aka DTP) for all languages of the world must be available on a single system, and not be reserved to be used with certain product versions.
Michael,
The nature of SING requires specific infrastructure, because it represents a new ecosystem in terms of text handling. For this reason, it must be tied to specific versions of the product, because previous versions could not handle it. In terms of font handling, the are libraries under the hood that are handling that.
Your comment equally applies to Unicode in its earlier days. In the early 1990s, very few, if any, applications and OSes supported Unicode. If you fed Unicode-encoded data to them, they would choke or crash.
In other words, progress sometimes requires cutting the proverbial umbilical cord, specifically support for specific features in a previous versions of a product.
BTW, and perhaps to help address your comments beter, there is an XML representation of SING, which allows it to persist in harsher environments. Whether a SING glyphlet in XML form can be used by an application is a different matter, but at least the “plain text” aspect of the raw XML allows it to be there.