Unicode
|
|
| Unicode |
|---|
| Encodings |
| Bi-directional text |
| BOM |
| Han unification |
| Unicode and HTML |
| Unicode and Email |
In computing, Unicode is the international standard whose goal is to provide the means to encode the text of every document people want to store in computers. This includes all scripts still in active use today, many scripts known only by scholars, and symbols which do not strictly represent scripts, like mathematics, linguistics and APL.
The creation of Unicode is an ambitious project to replace existing character sets, many of which are short in size and problematic in multilingual environments. Despite technical problems and limitations and criticism on process, today Unicode is considered the most complete character set and one of the largest, and has become the dominant encoding scheme in internationalization of software and multilingual environments. Many recent standards such as XML, as well as system software such as operating systems, have adopted Unicode as an underlying scheme to represent text.
| Contents |
Origin and development
It is the explicit aim of Unicode to transcend the limitations of traditional character encodings such as those defined by the ISO 8859 standard, which are used in the various countries of the world, but are largely incompatible with each other. One problem with traditional character encodings is that they allow for bilingual computer processing (usually Roman characters and the local language), but not for multilingual computer processing (computer processing of arbitrary languages mixed with each other).
Unicode in intent encodes the underlying characters and not variant glyphs for such characters. In the case of Chinese characters, this sometimes leads to controversies over what is the underlying character and what is the variant glyph (see Han unification).
Unicode's role in text-processing is to provide a unique code point — not a glyph — for each character. In other words, Unicode is used to represent a character in an abstract way, and leaves the visual rendering (size, shape or style) to another program, such as a web browser or word processor.
This simple aim is greatly complicated by another aim, which is to provide lossless conversion amongst different existing encodings in order to ease the transition.
The Unicode standard also includes a number of related items, such as character properties, text normalisation forms, and bidirectional display order (for the correct display of text containing both right-to-left scripts, such as Arabic or Hebrew, and left-to-right scripts).
In 1997 a proposal was made by Michael Everson to encode the characters of the Klingon language in Plane 1 of ISO/IEC 10646-2. The proposal was rejected in 2001 as "inappropriate for encoding" — not because the proposal was technically faulty, but because users of Klingon normally read and write and exchange data in Latin transliteration. The elvish scripts Tengwar and Cirth from J. R. R. Tolkien's Middle-earth setting were proposed for inclusion in Plane 1 in 1993. The draft was withdrawn to incorporate changes suggested by Tolkienists, and is as of 2004 still under consideration.
Mapping and encodings
Standard
The Unicode Consortium, based in California, is the organization that develops the Unicode standard. It is an organization open to any company or individual willing to pay the membership dues. Members include virtually all of the main computer software and hardware companies with any interest in text processing standards, such as Apple Computer, Microsoft, IBM, Xerox, HP, Adobe Systems and many others.
The Consortium first published The Unicode Standard (ISBN 0321185781) in 1991, and continues to develop standards based on that original work. Unicode was developed in conjunction with the International Organization for Standardization and it shares its character repertoire with ISO/IEC 10646. Unicode and ISO/IEC 10646 are equivalent as character encodings, but The Unicode Standard contains much more information for implementers, covering, in depth, topics such as bitwise encoding, collation, and rendering, and enumerating a multitude of character properties, including those needed for BiDi support. The two standards also have slightly different terminology.
When writing about unicode code points it is normal to use the form U+xxxx or U+xxxxxx where xxxx or xxxxxx is the hexadecimal code point.
Unicode revision history
- 1991 Unicode 1.0
- 1993 Unicode 1.1
- 1996 Unicode 2.0
- 1998 Unicode 2.1
- 1999 Unicode 3.0
- 2001 Unicode 3.1
- 2002 Unicode 3.2
- 2003 Unicode 4.0
- 2005 Unicode 4.1
Storage transfer and processing
So far, it has only been said that Unicode is a means to assign a unique number for all characters used by humans in written language. How these numbers are stored in text processing is another matter; problems result from the fact that much software in the West has so far been written to deal with 8-bit character encodings only, and Unicode support has only been added slowly in recent years. Similarly, in the East the double-byte character encodings cannot even in principle encode more than 65,536 characters, and in practice the limit imposed by the architectures chosen is much lower. This is not enough for the needs of scholars of the Chinese language alone.
The internal logic of much 8-bit legacy software typically permits only 8 bits for each character, making it impossible to use more than 256 code points without special processing, and 16-bit software is limited to some tens of thousands of characters, while Unicode is already up to more than 90,000 encoded characters. Several mechanisms have therefore been suggested to implement Unicode; which one is chosen depends on available storage space, source code compatibility, and interoperability with other systems.
The mapping methods are called the UTF (Unicode Transformation Format) and UCS (Universal Character Set) encodings. Among them are UTF-32, UCS-4, UTF-16, UCS-2, UTF-8, UTF-EBCDIC and UTF-7. The numbers indicate the number of bits in one unit, for UTF encodings, or bytes, for UCS encodings. In UTF-32 or UCS-4, one unit is enough for any character; in the other cases, a variable number of units is used for each character. UTF-8 is the de-facto standard encoding for interchange of unicode text with UTF-16 and UTF-32 being used mainly for internal processing.
The Unicode byte order mark (BOM) is specified for use at the beginnings of text files in UCS-2 and UTF-16 encodings. It has been adopted by some software developers for other encodings, including UTF-8, which does not need an indication of byte order. In this case it is an attempt to mark the file as containing Unicode text. The BOM is code point U+FEFF, which has the important property of being unambiguously interpretable regardless of which Unicode encoding is used. The units FE and FF never appear in UTF-8, U+FFFE (the result of byte-swapping U+FEFF) is not a legal character, and U+FEFF is the Zero-Width No-Break Space (a character with no appearance and no effect other than preventing formation of ligatures). The same character converted to UTF-8 becomes the byte sequence EF BB BF.
See also: Mapping of Unicode characters
Ready-made vs. composite characters
Unicode includes a mechanism for modifying character shape and so greatly extending the supported glyph repertoire. This is the use of combining diacritical marks. They are inserted after the main character (it is possible to stack several combining diacritics over the same character). However, for reasons of compatibility, Unicode also includes a large quantity of precomposed characters. So in many cases there are many ways of encoding the same character. To deal with this, Unicode provides the mechanism of canonical equivalence.
The situation with Hangul is similar. Unicode provides the mechanism for composing Hangul syllables with Hangul Jamo. However, the precomposed Hangul syllables (11,172 of them) are also provided.
The CJK ideographs currently are encoded only in their precomposed form. Still, most of those ideographs are evidently made up of simpler elements, so in principle it would be possible to decompose them just as it is done with Hangul. This would greatly reduce the number of required codepoints, while allowing the display of virtually every conceivable ideograph (and so doing away with all problems of the Han unification). A similar idea is used for some input methods, such as Cangjie and Wubi. However, attempts to do this for character encoding have stumbled over the fact that ideographs are not as simply decomposed or as regular as they seem.
Combining marks, like the complex script shaping required to properly render Arabic text and many other scripts, are usually dependent on complex font technologies, like OpenType (by Adobe and Microsoft), Graphite (by SIL International), and AAT (by Apple), by which a font designer includes instructions in a font telling software how to properly output different character sequences. Another method sometimes employed in fixed-width fonts is to place the combining mark's glyph before its own left sidebearing; this method, however, only works for some diacritics and stacking will not occur properly.
As of 2004, most software still cannot reliably handle many features not supported by older font formats, so combining characters generally will not work correctly. Hypothetically, Template:Unicode (precomposed e with macron and acute above) and Template:Unicode (e followed by the combining macron above and combining acute above) are identical in appearance, both giving an e with macron and acute accent, but appearance can vary greatly across software applications.
Also underdots, as needed in Indic Romanization, will often be placed incorrectly or worse. Sample:
Of course, this is in fact not a weakness in Unicode itself, but only uncovers gaps in rendering technology and fonts.
Issues
Some people, mostly in Japan, oppose Unicode in general, claiming technical limitations and political problems in process, which people working on the Unicode standard claim are simply misunderstandings of the Unicode standard and the process by which it was created. The most common mistake, according to this view, is confusion between abstract characters and their highly variable visual forms (glyphs). On the other hand, whereas Chinese can readily read most types of glyphs used by Japanese or Koreans, Japanese often can recognize only a particular variant. Unicode has been decried as a plot against Asian cultures perpetrated by Westerners with no understanding of the characters as used in Chinese, Korean, and Japanese, in spite of the presence of a majority of experts from all three countries in the Ideographic Rapporteur Group. The IRG advises the consortium and ISO on additions to the repertoire and on Han unification, the identification of forms in the three languages which will be treated as stylistic variations of the same historical character. This unification is one of the most controversial aspects of Unicode.
Unicode is criticized for failing to allow for older and alternate forms of kanji, which, it is said, complicates the processing of ancient Japanese and uncommon Japanese names, although it follows the recommendations of Japanese scholars of the language and of the Japanese government. There have been several attempts to create an alternative to Unicode. [1] (http://www-106.ibm.com/developerworks/unicode/library/u-secret.html) Among them are TRON (although it is not widely adopted in Japan, some, particularly those who need to handle historical Japanese text, favor this), UTF-2000 and Giga Character Set (GCS). It is true that many older forms were not included in early versions of the Unicode standard, but Unicode 4.0 contains more than 90,000 Han characters, far more than any dictionary or any other standard, and work continues on adding characters from the early literature of China, Korea, and Japan.
Thai language support has been criticized for its illogical ordering of Thai characters. This complication is due to Unicode inheriting the Thai Industrial Standard 620, which worked in the same way. This ordering problem complicates the Unicode collation process [2] (http://www-106.ibm.com/developerworks/unicode/library/u-secret.html).
Opponents of Unicode sometimes claim even now that it cannot handle more than 65,535 characters, a limitation that was removed in Unicode 2.0.
Unicode in use
Operating systems
Despite technical problems and limitations and criticism on process, Unicode has emerged as the dominant encoding scheme. Windows NT and its descendants Windows 2000 and Windows XP make extensive use of UTF-16 as an internal representation of text. UNIX-like operating systems such as GNU/Linux, BSD and Mac OS X have adopted UTF-8, as the basis of representation of multilingual text.
MIME defines two different mechanisms for encoding non-ASCII characters in e-mail, depending on whether the characters are in e-mail headers such as the "Subject:" or in the text body of the message. In both cases, the original character set is identified as well as a transfer encoding. For e-mail transmission of Unicode the UTF-8 character set and the Base64 transfer encoding are recommended. The details of the two different mechanisms are specified in the MIME standards and are generally hidden from users of e-mail software.
The adoption of Unicode in e-mail has been very slow. Most East-Asian text is still encoded in a local encoding such as Shift-JIS, and many commonly used e-mail programs still cannot handle Unicode data correctly, if they have some support at all. This situation is not expected to change in the foreseeable future.
Web
Recent web browsers display web pages using Unicode if an appropriate font is installed (see Unicode and HTML).
Although syntax rules may affect the order in which characters are allowed to appear, both HTML 4.0 and XML 1.0 documents are, by definition, comprised of characters from the entire range of Unicode code points, minus only a handful of disallowed control characters and the permanently-unassigned code points D800-DFFF, any code point ending in FFFE or FFFF and any code point above 10FFFF. These characters manifest either directly as bytes according to document's encoding, if the encoding supports them, or they may be written as numeric character references based on the character's Unicode code point, as long as the document's encoding supports the digits and symbols required to write the references (all encodings approved for use on the Internet do). For example, the references Δ Й ק م ๗ あ 叶 葉 냻 (or the same numeric values expressed in hexadecimal, with &#x as the prefix) display on your browser as Δ, Й, ק, م, ๗, あ, 叶, 葉 and 냻—if you have the proper fonts, these symbols look like the Greek capital letter "Delta", Cyrillic capital letter "Short I", Arabic letter "Meem", Hebrew letter "Qof", Thai numeral 7, Japanese Hiragana "A", simplified Chinese "Leaf", traditional Chinese "Leaf", and Korean Hangul syllable "Nyaelh", respectively.
Fonts
Free and retail fonts based on Unicode are common, since first TrueType and now OpenType support Unicode. These font formats map Unicode code points to glyphs.
There are thousands of fonts on the market, but fewer than a dozen fonts attempt to support the majority of Unicode's character repertoire; these fonts are sometimes described as pan-Unicode. Instead, Unicode based fonts typically focus on supporting only basic ASCII and particular scripts or sets of characters or symbols. There are several reasons for this: applications and documents rarely need to render characters from more than one or two writing systems; fonts tend to be demanding of resources in computing environments; and operating systems and applications are becoming increasingly intelligent in regard to obtaining glyph information from separate font files as they are needed. Furthermore, it is a monumental task to design a consistent set of rendering instructions for tens of thousands of glyphs; such a venture passes the point of diminishing returns for most typefaces.
Unicode characters which cannot be rendered are most often displayed as an open rectangle only, to indicate the position of the unrecognized character. Some attempts have been made to provide more information about these characters. The Apple LastResort font will display a substitute glyph indicating the Unicode range of the character and the SIL Unicode fallback font will display a box showing the hexadecimal scalar value of the character.
Multilingual Text Rendering Engines
- Uniscribe - Windows
- Apple Type Services for Unicode Imaging - new engine for Macintosh
- WorldScript - old engine for Macintosh
- Pango - open source
- ICU Layout Engine - open source
- Graphite - (open source renderer from SIL)
Input methods
On Windows XP, any Unicode character can be input by pressing Alt, then, with Alt down (and using only the numeric keypad keys), pressing the decimal digits of the Unicode characters one after the other. For example, Alt, then, with Alt still down, 9, then 6 and then 0 yields π (Greek lowercase letter Pi). For values less than 256, precede the digits with a 0, to avoid code page translation (see Extended ASCII), e.g. Alt 0, 1, 6, 5 yields ¥.
Word 2003 also allows for entering unicode characters by spelling out the code first, e.g. 014B for the 'ng'-symbol and then hitting 'Alt' plus 'X' to substitute the string to the left by its unicode character.
Macintosh users have a similar feature with an input method called 'Unicode Hex Input', in Mac OS X and in Mac OS 8.5 and later: hold down the Option key, and type the four-hex-digit Unicode code point. Handling of code-points above 0xFFFF is done by entering a surrogate pair; they will be converted into a single character automatically. Mac OS X (version 10.2 and newer) also has a 'Character Palette', which allows users to visually select any Unicode character from a table organized numerically, by Unicode block, or by a selected font's available characters.
Gnome2 follows ISO 14755. Hold down Ctrl and Shift and enter the hexadecimal unicode value.
The Opera web browser in version 7.5 and over allows users to enter any Unicode character directly into a text field by typing its hexadecimal code, selecting it, and pressing alt+x.
See also
External links
- The Unicode Consortium (http://www.unicode.org)
- Unicode versions: 3.1 (http://www.unicode.org/unicode/reports/tr27/), 3.2 (http://www.unicode.org/unicode/reports/tr28/), 4.0 (http://www.unicode.org/versions/Unicode4.0.0), 4.0.1 (http://www.unicode.org/versions/Unicode4.0.1/), 4.1 (http://www.unicode.org/versions/Unicode4.1.0/)
- new characters (http://www.unicode.org/alloc/Pipeline.html), scripts (http://www.unicode.org/pending/pending.html) and characters and scripts under investigation (http://www.unicode.org/alloc/investigation.html)
- Code Charts (http://www.unicode.org/charts/) (PDF)
- Table of Unicode characters from 1 to 65535 (http://unicode.coeurlumiere.com/)
- UTF-8, UTF-16, UTF-32 Code Charts (http://www.macchiato.com/unicode/charts.html) and a character map (http://www-atm.physics.ox.ac.uk/user/iwi/charmap.html) (JavaScript)
- The Letter Database (http://www.eki.ee/letter/) Uses forms to present groups in list or grid format by hexadecimal.
- DecodeUnicode - Unicode WIKI, Typographic project to explain every unicode character with 50.000 gifs in three sizes (http://www.decodeunicode.org/)
- Example text files using Unicode (http://www.cl.cam.ac.uk/~mgk25/ucs/examples/)
- Unicode special character map (http://www.lazytools.com/unicode-ascii/) is similar to the Windows version. Click a symbol to obtain either the named or numeric code for HTML.
- ConScript Unicode Registry (http://www.evertype.com/standards/csur/) a project to standardize part of the Private Use Area for use with artificial scripts and artificial languages. An explanation of how to propose character names in Unicode is available here.
- The secret life of Unicode (http://www-106.ibm.com/developerworks/unicode/library/u-secret.html) "A peek at Unicode's soft underbelly" Describes problems requiring resolution. Includes links to Unicode resources.
- Tim Bray's Characters vs Bytes (http://www.tbray.org/ongoing/When/200x/2003/04/26/UTF) explains how the different encodings work.
- Alan Wood's Unicode Resources (http://www.alanwood.net/unicode/) Contains lists of word processors with Unicode capability; characters are grouped by type; characters are presented in lists, not grids.
- The strongest denunciation of Unicode (http://www.hastingsresearch.com/net/04-unicode-limitations.shtml), and a response to it (http://slashdot.org/features/01/06/06/0132203.shtml)
- Fonts and tools:
- Unicode fonts and tools (http://www.cl.cam.ac.uk/~mgk25/ucs-fonts.html) for the X Window System
- Unicode TTF fonts: Arial Unicode MS, Code2000: license info and download link (http://home.att.net/~jameskass/), Junicode: license info and download link (http://www.engl.virginia.edu/OE/junicode/junicode.html), Titus Cyberbit Basic: license info (http://titus.uni-frankfurt.de/indexe.htm?/unicode/unitest2.htm) & download link (http://titus.fkidg1.uni-frankfurt.de/unicode/tituut.asp)
- UnicodeChecker (http://earthlingsoft.net/UnicodeChecker/), a Unicode character browser for Mac OS X
- Software engineering:
- International Components for Unicode (ICU) (http://icu.sourceforge.net/) An open source set of libraries that provide robust and full-featured Unicode services for your applications on a wide variety of platforms.
- The Absolute Minimum Every Software Developer Absolutely, Positively Must Know About Unicode and Character Sets (No Excuses!) (http://www.joelonsoftware.com/articles/Unicode.html) by Joel Spolsky of JoelonSoftware.com (this is now outdated, but still a reasonable starting point).
- Freedesktop.Org's Project UTF-8 (http://freedesktop.org/wiki/Software_2futf_2d8)'s purpose is to document and promote proper Unicode support in free and Open Source software.
- Supplementary Characters in the Java Platform (http://java.sun.com/developer/technicalArticles/Intl/Supplementary/) from Sun Microsystems
- Seeing the entirety of Unicode printed out (http://www.ianalbert.com/misc/unichart.php) as a single large poster gives a good feel for the size of the code.
Template:SpecialCharsbr:Unicode bs:Unikod bg:Уникод cs:Unicode da:Unicode de:Unicode es:Unicode eo:Unikodo fi:Unicode fr:Unicode he:יוניקוד hi:यूनिकोड hu:Unicode ia:Unicode it:Unicode ja:Unicode kn:ಯುನಿಕೋಡ್ ko:유니코드 ks:Yunikōḍa ku:Unicode nl:Unicode no:Unicode pl:Unicode pt:Unicode ro:Unicode ru:Юникод sk:Unicode sr:Уникод sv:Unicode ta:யுனிகோடு th:ยูนิโคด vi:Unicode zh:Unicode zh-min-nan:Thong-iōng-bé