教育資料與圖書館學 Journal of Educational Media & Library Sciences Vol. 46, no. 4 (Summer 2009) : 台灣圖書館網頁標記語言正確性之探討 A Study on

Size: px

Start display at page:

Download "教育資料與圖書館學 Journal of Educational Media & Library Sciences Vol. 46, no. 4 (Summer 2009) : 台灣圖書館網頁標記語言正確性之探討 A Study on"

敬胡
4 years ago
Views:

1 教育資料與圖書館學 Journal of Educational Media & Library Sciences tku.edu.tw Vol. 46, no. 4 (Summer 2009) : 台灣圖書館網頁標記語言正確性之探討 A Study on Markup Language Validations of Library Websites in Taiwan 謝建成 Jiann-Cherng Shieh* Associate Professor jcshieh@ntnu.edu.tw 洪范文 Fan-Wen Hung Graduate Student 陳建傑 Chien-Chieh Chen Graduate Student English Abstract & Summary see link at the end of this article tku.e

2 46 : 4 (Summer 2009) : * jcshieh@ntnu.edu.tw W3C Markup Validation Service /3 W3C * 2009/06/092009/07/222009/07/24

3 : 4 (Summer 2009) ㈠ Market Share IE Internet Explore 66% Firefox 22%Safari 8.2% Market Share, 2009 Chen, Hong, & Shen, 2005 DTD Chen, Hong, & Shen, 2005 Max, 2004 The Web Standards Project, HTML XHTML XML CSS XSL extensible style sheet languagejavascript ECMAScript Chen & Shen, 2006 Chen & Shen, 2006 validator checker W3C World Wide Web Consortiummarkup validator

4 471 Flash Breeding, 2006 Chen, Hong, & Shen, 2005 Breeding 2006 ARL W3C % Breeding ILS Google Amazon information ethics accessibilityaccuracy W3C web markup language validation ㈡㈠W3C W3C World Wide Web Consortium TAW3 WAVE MAGENTA ALPHA Bad neighborhood Detect Headers CSE HTML Validator Lite WDG HTML Validator W3C WDG W3C Log Validator WDG 100 W3C Web Quality Tools open source

472 46 : 4 (Summer 2009) MarkUp Validator Feed Validator CSS Validator Link Checker Log

Validator Atom RSS feed CSS Validator (X)HTML CSS Link Checker link anchor referenced object

5 : 4 (Summer 2009) MarkUp Validator Feed Validator CSS Validator Link Checker Log Validator MarkUp Validator HTML XHTML SMIL MathML computer language markup validity Feed Validator Atom RSS feed CSS Validator (X)HTML CSS Link Checker link anchor referenced object Log Validator W3C DTD W3C 1 Encoding Doctype W3CMarkup Validation Service 2 W3C Markup Validation Service

6 473 3 W3C Markup Validation Service ㈡ Chen Alexa.com Google Yahoo Teoma W3C Markup Validator JTidy parser Chen 95% doctype alt type height width URL & body background border form table Chen HTML Breeding 2006W3C Markup Validation Service 123 W3C 21 W3C % Breeding

7 : 4 (Summer 2009) 136 ILS Google Amazon Google 50 ebay 248 Yahoo 270 CNN 68 Flickr 15 Amazon 1292 Breeding HTML ㈠ main page ㈡

8 475 W3C Markup Validation Service MS Excel ㈠ Doctype HTML 4.01 Transitional 53.6%604 HTML 4.01 Transitional 55.7%9 2 XHTML 1.0 Transitional 77.8% 89 6 HTML 4.01 Transitional 63.7% 1 Doctype Doctype % (no Doctype found) XHTML 1.0 Frameset HTMHTML HTML 4.01 Strict XHTML 1.0 Strict HTML 4.0 Transitional XHTML 1.0 Transitional HTML 4.01 Transitional

9 : 4 (Summer 2009) 2 Doctype Doctype % HTML 4.0 Transitional HTML 4.01 Frameset XHTML 1.0 Transitional HTML 4.01 Transitional Doctype Doctype % HTML 4.01 Transitional XHTML 1.0 Transitional Doctype Doctype % XHTML 1.0 Frameset HTML 4.01 Strict XHTML 1.0 Strict HTML 4.0 Transitional XHTML 1.0 Transitional HTML 4.01 Transitional Markup Validator Breeding % % % % / A 863 B2 C197 D11 E1204 F 1 Z 204 Y 3 7 DNS

10 477 w3c 5 w3c 4 DNS 5 % 1~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Total % 1~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Total

11 : 4 (Summer 2009) 3. 7 % 1~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Total % 1~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Total W3C Markup Validation Service There is no attribute X 32Document type does not allow element X here 27Element X undefined 24 There is no attribute X 138Required attribute X not specified 123Element X undefined 88

12 479 There is no attribute X Html- Tag35Html Html Script/Script-Tag 17 7 ㈡ W3C Markup Validator W3C 20 An attribute value specification must be an attribute value literal unless SHORT- TAG YES is specified X not finished but containing element ended Start tag for X omitted, but its declaration does not permit this 8 W3C Markup Validator Doctype Doctype 7 HTML 4.01 Transitional

13 : 4 (Summer 2009) 0 1/ attribute48 Html-Tag 35 Html Html W3C 20 <tr> </tr><td></td> <tr> IE <tr> <td> <html> <head> <body> <head> <body> Cannot continue because of previous errors

14 (1) Breeding, M. (2006). Web 2.0? Let's get to Web 1.0 first. Computers in Libraries, 26(5), Chen, B., & Shen, V. Y. (2006). Transforming web pages to become standard-compliant through reverse engineering. In W4A (Eds.), Proceedings of the 2006 international cross-disciplinary workshop on web accessibility (pp ). New York: ACM Press. Chen, S., Hong, D., & Shen, V. Y. S. (2005). An experimental study on validation problems with existing HTML webpages. In H. R. Arabnia (Ed.), Proceedings of international conference on Internet computing (pp ). Las Vegas. Market Share (2009). Retrieved May 6, 2009, from Max (2004). The benefits of web standards to your visitors, your clients and you! Retrieved May 5, 2009, from The Web Standards Project (2002). What are web standards and why should I use them? Retrieved May 2, 2009, from

15 : 4 (Summer 2009) 1 ( () 1 2 3

16 483 2 an attribute value literal can occur in an attribute specification list only after a VI delimiter 1 literal is missing closing delimiter 1 no document type declaration; will parse without validation 1 non SGML character number X 1 NET-enabling start-tag not immediately followed by null end-tag 1 start tag for X omitted, but its declaration does not permit this 1 unknown declaration type X 1 value of attribute Y invalid: X cannot start a name 1 value of fixed attribute X not equal to default 1 X invalid: only S separators and TAGC allowed here 1 X is not a reserved name 1 character X is not allowed in the value of attribute Y 2 an attribute value specification must be an attribute value literal unless SHORTTAG YES is specified 4 duplicate specification of attribute X 4 ID X already defined 4 invalid comment declaration: found X outside comment but inside comment declaration 4 invalid comment declaration: found character X outside comment but inside comment declaration 4 X declaration not allowed in instance 4 X not finished but containing element ended 4 an attribute specification must start with a name or name token 5 character X not allowed in attribute specification list 6 end tag for X omitted, but OMITTAG NO was specified 6 X is not a member of a group specified for any attribute 6 XML Parsing Error 6 document type does not allow element X here; assuming missing Y start-tag 7 an attribute value must be a literal unless it contains only name characters 8 character data is not allowed here 9 document type does not allow element X here; missing one of Y start-tag 9 no document type declaration; implying X 10 value of attribute Y cannot be X; must be one of %3 10 end tag for X omitted, but its declaration does not permit this 13 end tag for X which is not finished 14 reference to entity X for which no system identifier could be generated 17 end tag for element X which is not open 18 general entity X not defined and no default entity 18 required attribute X not specified 19 element X undefined 24 document type does not allow element X here 27 there is no attribute X 32

17 : 4 (Summer 2009) 3 cannot continue because of previous errors 1 character X invalid: only Y and parameter separators allowed 1 literal is missing closing delimiter 1 marked section end not in marked section declaration 1 Missing xmlns attribute for element X. The value should be: Y 1 no document element 2 value of fixed attribute X not equal to default 3 character X is not allowed in the value of attribute Y 4 NET-enabling start-tag not immediately followed by null end-tag 4 start tag for X omitted, but its declaration does not permit this 4 an attribute value specification must start with a literal or a name character 5 unknown declaration type X 5 X not finished but document ended 6 the name and VI delimiter can be omitted from an attribute specification 6 only if SHORTTAG YES is specified value of attribute X must be a single token 6 no document type declaration; will parse without validation 7 an attribute value literal can occur in an attribute specification list only 8 after a VI delimiter invalid comment declaration: found character X outside comment but 8 inside comment declaration X invalid: only S separators and TAGC allowed here 8 invalid comment declaration: found X outside comment but inside 9 comment declaration syntax of attribute value does not conform to declared value 9 duplicate specification of attribute X 11 an attribute specification must start with a name or name token 14 character X not allowed in attribute specification list 14 X is not a member of a group specified for any attribute 15 character data is not allowed here 20 document type does not allow element X here; assuming missing Y start-tag 22 value of attribute Y invalid: X cannot start a name 24 ID X already defined 25 an attribute value specification must be an attribute value literal unless 31 SHORTTAG YES is specified XML Parsing Error 31 value of attribute Y cannot be X; must be one of %3 35 end tag for X omitted, but OMITTAG NO was specified 37 end tag for X omitted, but its declaration does not permit this 38 end tag for X which is not finished 40 document type does not allow element X here; missing one of Y start-tag 44 an attribute value must be a literal unless it contains only name characters 45 X not finished but containing element ended 45 no document type declaration; implying X 60 end tag for element X which is not open 80 general entity X not defined and no default entity 84 reference to entity X for which no system identifier could be generated 84 document type does not allow element X here 87 element X undefined 88 required attribute X not specified 123 there is no attribute X 138

18 485 4 an attribute value specification must start with a literal or a name character 1 literal is missing closing delimiter 1 Missing xmlns attribute for element X. The value should be: Y 1 no document element 1 NET-enabling start-tag not immediately followed by null end-tag 1 the name and VI delimiter can be omitted from an attribute specification 1 only if SHORTTAG YES is specified unknown declaration type X 1 value of fixed attribute X not equal to default 1 character X is not allowed in the value of attribute Y 2 start tag for X omitted, but its declaration does not permit this 2 syntax of attribute value does not conform to declared value 2 X not finished but document ended 2 an attribute value literal can occur in an attribute specification list only 3 after a VI delimiter invalid comment declaration: found X outside comment but inside 3 comment declaration invalid comment declaration: found character X outside comment but 3 inside comment declaration value of attribute X must be a single token 3 X invalid: only S separators and TAGC allowed here 3 an attribute specification must start with a name or name token 4 no document type declaration; will parse without validation 4 character X not allowed in attribute specification list 6 duplicate specification of attribute X 6 X is not a member of a group specified for any attribute 8 an attribute value specification must be an attribute value literal unless 10 SHORTTAG YES is specified character data is not allowed here 10 document type does not allow element X here; assuming missing Y start-tag 10 end tag for X which is not finished 12 value of attribute Y cannot be X; must be one of %3 12 ID X already defined 13 value of attribute Y invalid: X cannot start a name 13 XML Parsing Error 15 end tag for X omitted, but OMITTAG NO was specified 16 an attribute value must be a literal unless it contains only name characters 17 end tag for X omitted, but its declaration does not permit this 17 X not finished but containing element ended 17 document type does not allow element X here; missing one of Y start-tag 19 no document type declaration; implying X 19 document type does not allow element X here 31 end tag for element X which is not open 34 element X undefined 35 general entity X not defined and no default entity 38 reference to entity X for which no system identifier could be generated 38 required attribute X not specified 47 there is no attribute X 54

19 : 4 (Summer 2009) 5 an attribute value must be a literal unless it contains only name characters 1 an attribute value specification must be an attribute value literal unless 1 SHORTTAG YES is specified document type does not allow element X here; missing one of Y start-tag 1 document type does not allow element X here; assuming missing Y start-tag 1 duplicate specification of attribute X 1 end tag for X omitted, but OMITTAG NO was specified 1 end tag for X which is not finished 1 ID X already defined 1 invalid comment declaration: found character X outside comment but 1 inside comment declaration no document type declaration; will parse without validation 1 value of attribute X must be a single token 1 value of attribute Y invalid: X cannot start a name 1 X is not a member of a group specified for any attribute 1 X not finished but containing element ended 1 X invalid: only S separators and TAGC allowed here 1 character data is not allowed here 2 end tag for X omitted, but its declaration does not permit this 2 value of attribute Y cannot be X; must be one of %3 2 XML Parsing Error 2 element X undefined 3 document type does not allow element X here 4 end tag for element X which is not open 5 general entity X not defined and no default entity 5 no document type declaration; implying X 5 reference to entity X for which no system identifier could be generated 5 required attribute X not specified 8 there is no attribute X 8

20 487 6 cannot continue because of previous errors 1 character X invalid: only Y and parameter separators allowed 1 marked section end not in marked section declaration 1 no document element 1 character X is not allowed in the value of attribute Y 2 no document type declaration; will parse without validation 2 start tag for X omitted, but its declaration does not permit this 2 value of attribute X must be a single token 2 value of fixed attribute X not equal to default 2 NET-enabling start-tag not immediately followed by null end-tag 3 an attribute value specification must start with a literal or a name character 4 duplicate specification of attribute X 4 invalid comment declaration: found character X outside comment but inside comment declaration 4 unknown declaration type X 4 X invalid: only S separators and TAGC allowed here 4 X not finished but document ended 4 an attribute value literal can occur in an attribute specification list only after a VI delimiter 5 the name and VI delimiter can be omitted from an attribute specification only if SHORTTAG YES is specified 5 invalid comment declaration: found X outside comment but inside comment declaration 6 X is not a member of a group specified for any attribute 6 syntax of attribute value does not conform to declared value 7 character data is not allowed here 8 character X not allowed in attribute specification list 8 an attribute specification must start with a name or name token 10 value of attribute Y invalid: X cannot start a name 10 document type does not allow element X here; assuming missing Y start-tag 11 ID X already defined 11 XML Parsing Error 14 end tag for X omitted, but its declaration does not permit this 19 an attribute value specification must be an attribute value literal unless SHORTTAG YES is specified 20 end tag for X omitted, but OMITTAG NO was specified 20 value of attribute Y cannot be X; must be one of %3 21 document type does not allow element X here; missing one of Y start-tag 24 an attribute value must be a literal unless it contains only name characters 27 end tag for X which is not finished 27 X not finished but containing element ended 27 no document type declaration; implying X 36 end tag for element X which is not open 41 general entity X not defined and no default entity 41 reference to entity X for which no system identifier could be generated 41 element X undefined 50 document type does not allow element X here 52 required attribute X not specified 68 there is no attribute X 76

21 : 4 (Summer 2009) 7 Html Html- DTD Html- Tag Script / Script-Tag XML Parsing Error X invalid: only S separators and TAGC allowed here X is not a function name end tag for element X which is not open X is not a member of a group specified for any attribute X not finished but document ended X not finished but containing element ended X declaration not allowed in instance end tag for X which is not finished ID X already defined character X is not allowed in the value of attribute Y an attribute value must be a literal unless it contains only name characters general entity X not defined and no default entity invalid comment declaration: found X outside comment but inside comment declaration invalid comment declaration: found character X outside comment but inside comment declaration element X undefined document type does not allow element X here document type does not allow element X here; assuming missing Y start-tag document type does not allow element X here; missing one of Y start-tag unknown declaration type X required attribute X not specified character data is not allowed here an attribute value literal can occur in an attribute specification list only after a VI delimiter cannot continue because of previous errors character X invalid: only Y and parameter separators allowed XML SGML

22 489 Html Html- DTD Html- Tag Script / Script-Tag XML SGML character X not allowed in attribute specification list there is no attribute X no document element no document type declaration; implying X no document type declaration; will parse without validation reference to entity X for which no system identifier could be generated non SGML character number X end tag for X omitted, but its declaration does not permit this end tag for X omitted, but OMITTAG NO was specified an attribute value specification must be an attribute value literal unless SHORTTAG YES is specified duplicate specification of attribute X value of fixed attribute X not equal to default l i t e r a l i s m i s s i n g c l o s i n g delimiter marked section end not in marked section declaration NET-enabling start-tag not immediately followed by null end-tag the name and VI delimiter c a n b e o m i t t e d f r o m a n attribute specification only if SHORTTAG YES is specified Missing xmlns attribute for element X. The value should be: Y start tag for X omitted, but its declaration does not permit this value of attribute Y invalid: X cannot start a name value of attribute Y cannot be X; must be one of %3 an attribute specification must start with a name or name token an attribute value specification must start with a literal or a name character syntax of attribute value does not conform to declared value value of attribute X must be a single token

23 : 4 (Summer 2009) 8 an attribute value Line 41, column 28: an attribute value " specification must be specification must be an attribute value an attribute value literal literal unless SHORTTAG YES is specified unless SHORTTAG <td><div class=newsmodule id=newstxt> YES is specified an attribute value specification must start with a literal or a name character cannot continue because of previous errors character X invalid: only Y and parameter separators allowed Line 372, column 56: an attribute value specification must start with a literal or a name character <font size=2 style="text-decoration:none" color=>97 (30)</font> </a> Line 2, column 63: cannot continue because of previous errors HTML PUBLIC "-// W3C//DTD HTML 4.01 Transitional//EN"> color= > Line 2, column 63: character ">" invalid: > only delimiter ">", delimiter "[", system identifier, "CDATA", "NDATA", "SDATA" and parameter separators allowed HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> character X not allowed Line 427, column 159: character ")" not in attribute specification allowed in attribute specification list list 0" marginheight="0" onload="start()")> document type does not allow element X here document type does not allow element X here; assuming missing Y start-tag invalid comment declaration: found X outside comment but inside comment declaration marked section end not in marked section declaration Missing xmlns attribute for element X. The value should be: Y Line 24, column 49: document type does not allow element "BR" here DC.Date" content= :00> <br> <meta http-equiv="content-type" content= Line 44, column 45: document type does not allow element "META" here me="dc. Date" content= :00> <table summary=" " width="100%" bord Line 39, column 68: document type does not allow element "script" here; assuming missing "body" start-tag s/ac_runactivecontent.js" type="text/ javascript"></script> Line 235, column 42: document type does not allow element "tr" here; assuming missing "table" start-tag Line 119, column 69: invalid comment declaration: found name character outside comment but inside comment declaration " size="40" maxlength="40" value=" " accesskey="s" onfocus="cleart Line 335, column 4: marked section end not in marked section declaration // ]]> Many Document Types based on XML need a mandatory xmlns="" on the root element. For example, the root element for XHTML will look like: <html xmlns=" xhtml" xml:lang="en" lang="en"> onload= "start()")> onload="start()" <br> </head> <body> "------" "" // ]> // ]]> <html xmlns=" w3.org/1999/xhtml">

24 491 no document element Line 821, column 8: no document element </HTML> </HT start tag for X omitted, Line 82, column 24: start tag for "TR" <td><tr> but its declaration does omitted, but its declaration does not permit </tr> not permit this this <table width=90%><td> value of attribute Y cannot be X; must be one of %3 value of fixed attribute X not equal to default X declaration not allowed in instance X invalid: only S separators and TAGC allowed here X is not a member of a group specified for any attribute X not finished but containing element ended X not finished but document ended XML Parsing Error Line 167, column 268: value of attribute align="absmiddle" "ALIGN" cannot be "ABSMIDDLE"; must top, be one of "TOP", "MIDDLE", "BOTTOM", middle, right "LEFT", "RIGHT" ght="11" border="0" align="absmiddle" /></a></font></td> align="absmiddle"></font></td> Line 5, column 39: value of fixed attribute "xmlns" not equal to default xmlns=" xmlns= " w3.org/tr/rec-html40"> xmlns= " Line 3, column 3: "DOCTYPE" declaration <!DOCTYPE HTML not allowed in instance <!DOCTYPE PUBLIC "-//W3C// HTML PUBLIC "-//W3C//DTD HTML 4.01 DTD HTML 4.01 Transitional//EN"> Transitional//EN"> Line 256, column 23: delimiter """ invalid: only S separators and TAGC allowed here document.write ("</mar"+"quee>") Line 180, column 258: "a" is not a member of a group specified for any attribute 0,height=300,left=150,top=120 ) " a href=# onkeypress="window.open( readnews ) Line 261, column 5: "HEAD" not finished but containing element ended document.write ("</marquee>") a href=# <head> </head> Line 297, column 8: "HTML" not finished <body> <html> but document ended </ht Line 385, column > 80: XML Parsing "=" Error: AttValue: " or expected vascript" src=" </scri Line 385, column > 80: XML Parsing Error: attributes construct error vascript"

25 Journal of Educational Media & Library Sciences 46 : 4 (Summer 2009) : A Study on Markup Language Validations of Library Websites in Taiwan Jiann-Cherng Shieh* Associate Professor jcshieh@ntnu.edu.tw Fan-Wen Hung Graduate Student Chien-Chieh Chen Graduate Student Graduate Institute of Library & Information Studies National Taiwan Normal University Taipei, Taiwan Abstract Library website is an extended service of library; the correctness of library webpage is certainly related to accessibility and correctness of information ethics and also will manifest the importance of the compliance of library webpage with webpage design standard for the general readers. Markup validation service is one of the webpage design standards. Testing the correctness of library webpage can reveal the extent of the compliance of webpage with the standard in order to assist libraries to develop or maintain webpage conforming to the standard. This study used markup validation service provided by W3C to test the correctness of the codes of library homepage of 158 colleges and 24 public libraries in order to investigate the current status of the correctness of the markup language of university and public library websites. The results showed that 0% of the markup language of university and public library websites passed the test; more than 1/3 of the tested library websites had over 100 errors. This indicated that the webpage designers and the validation mechanism of the libraries being investigated need improvement. For the errors found in the webpage testing, which W3C provided no suggestion for amendment, this study offered some solutions by giving examples to serves as references for producing and maintaining library websites. Keywords: W3C; Web markup language validation; Library website SUMMARY Introduction Library website is an extended service which solves the limitation problem of space and distance. However, not all webpage is suitable for different kinds of browsers even if there are diverse webpage producing software and browsers on * Principal author for all correspondence.

26 Jiann-Cherng Shieh, Fan-Wen Hung, & Chien-Chieh Chen: A Study on Markup Language Validations of the market nowadays. The compliance of library webpage with webpage design standard is positively related to the accessibility and correctness of information, which would also makes it easer for website maintenance. Web administrators can use various validating tools, such as the markup validator of W3C (World Wide Web Consortium), for web design and correctness. The study used the markup validator of W3C to test the correctness and investigate the problems of the web markup language validation of university and public libraries in Taiwan, and tried to propose some solutions as reference for library website management and development. Related Literature Chen, Hong, and Shen (2005) used a number of validating tools to examine the proportion of correct webpage on the Internet and identified the most frequently occurring types of errors. The sample websites of the study were chosen by four different ways: First, designed a program to randomly generate 1,100 websites from different web hosts; second, chose 10,000 most popular webpage from Alexa.com; third, randomly generated up to eight character strings using three search engines (i.e., Google, Yahoo, and Teoma) to search for the results and choosing the top ten results; 31,540 pages were produced this way; forth, used the first method two month later to obtain 1,700 random websites. The researchers used the Markup Validator of W3C and JTidy parser as the validating tools. The research results of Chen et. al. (2005) showed that up to 95% of the sample pages did not comply with the web standards, and the top ten error types were almost the same regardless of their different collecting ways. Chen et. al. (2005) concluded that those errors of webpage were mainly caused by the excessive tolerance to error webpage of the browsers, the allowance of non-html functions and features by the software of website design, as well as continuous duplication and spread of the error pages. Breeding in 2006 used W3C s Markup Validation Service to validate the homepage of the 123 members of the United States Association of Research Libraries (ARL) for their markup correctness. The results showed that only 21 libraries fully complied with W3C s validation, the rest presented some errors, and there were even seven libraries presented more than 100 errors. The proportion of inaccuracy was as high as 82%. In addition, Breeding also tested 136 metropolitan public libraries, the business Integrated Library System (ILS), as well as some commercial websites (such as Google, Amazon), and the results were almost equal.

27 494 Journal of Educational Media & Library Sciences 46 : 4 (Summer 2009) Research Questions Based on the background of the problem mentioned above, the study hoped to investigate the current status of the correctness on the markup language of library websites in Taiwan, to propose some suggestions for amendment, and to serve as a reference for the library website development. The specific research questions are as follows: 1. What is the proportion of the accurate web markup language on library websites in Taiwan? 2. How to improve the accuracy of web markup languages? Methodology This study used the markup validation service provided by W3C (Markup Validator) to test the correctness of the web markup languages. Through checking the codes of the webpage, the Markup Validator examines the websites and the normative framework (e.g., DTD) to assess the correctness of the webpage markup. It can detect the errors of the webpage (according to the norms of W3C), the server types, page sizes, text formats, encoding, file types, etc., and provide some suggestions to the demonstrated errors. This study proposed some solutions based on the analysis of the investigation. Data Collection The subjects of the study were the homepage of university and public libraries in Taiwan, considering that the library homepage is the entrance to all links and its website markup accuracy would affect the most users. The universities and colleges were selected by the Ministry of Education s University and College Directory of the 97th Academic Year (published in 2008), not including military and police school. A total of 163 libraries were selected. To avoid differences in nature, the study divided them into three categories: regular universities, universities of teacher training system, and vocational colleges. The effective samples of each group respectively were 60, 9, and 89. The date of sampling was May 4, In addition, the directory of public libraries was taken from the National Public Library Directory published by the National Taichung Library. Three categories were chosen for this study, which are National Library, Municipal Library, and Libraries of Cultural Affairs Bureau of City and County. The valid samples were 24 libraries in total, and the numbers for each category were 2, 2, and 20 respectively. The date of sampling was May 4, This study used W3C Markup Validator to examine the accuracy of the markup language of university and public libraries websites in Taiwan. Through

28 Jiann-Cherng Shieh, Fan-Wen Hung, & Chien-Chieh Chen: A Study on Markup Language Validations of the three indicators: webpage formats (Doctype), the number of inaccurate webpage, and the error types of the websites, to assess the accurate level of the library webpage markup and present the problems. On the web page format, both public and university libraries have more than one Doctype, some may even have seven, as found in a public library. Different page formats may lead to different information presentations and even cause information dislocation or lost. In addition, most webpage of regular universities, vocational colleges, and public libraries are using HTML 4.01 Transitional format which can be used as a reference for future website design. Regarding the rate of inaccurate websites, among the 158 valid samples of university libraries and the 24 of public libraries, all of the pages failed the test; more than 1/3 of the tested libraries had over 100 errors. The high rate of inaccuracy showed that the validation mechanism of domestic libraries needs improvement. For the error types, the most common type was use of wrong or undefined attributes, showing that particular attention should be paid to on the use of attributes in designing library webpage. In addition, the study divided the 48 types of errors into 6 categories in accordance with their written languages. Among the categories, Html-Tag possessed the most errors (35), indicating that the majority of those pages have wrong definition or inaccurate usage of Html tag. Therefore, more attention should be paid to on the usage of Html tags. The study also revised 20 kinds of error types to which the W3C did not provide clear definition, and amended one case as a model. Through the amendment, the majority of the errors could be classified into two types: wrong webpage markup and tag structure. For the wrong webpage markup, tag ending possessed the highest rate of error, and as for the tag structure, tag misplacing was the worst one. The markup and structural errors mentioned above would certainly influence information exchange and transmission, thus reduce the information accessibility. Library website is as an important platform for providing 24 hours of library service. The active maintenance of the existing library webpage is expected to ensure its standard and accuracy, which in turn would enhance readers satisfaction toward the library services. ROMANIZED & TRANSLATED REFERENCES FOR ORIGINAL TEXT [Public Library Information Service Network](2009) [Domestic public libraries] [Retrieved May 2, 2009] [from] [Xu Qin-Qin](2007)[Current situation of government information portal and its rebuild in China][Journal of

29 496 Journal of Educational Media & Library Sciences 46 : 4 (Summer 2009) Library and Information Sciences in Agriculture] 19(1) 5-7 [Department of Statistics](2009) 97[2008 xueniandu dazhuanyuanxiao minglu] [Retrieved May 4, 2009] [from] [National Taiwan Library](2009) [Links to public libraries] [Retrieved May 1, 2009][from] h= [National Taichung Library](2009) [Quanguo gongli gonggong tushuguan tongjibiao] [Retrieved May 1, 2009][from] Breeding, M. (2006). Web 2.0? Let's get to Web 1.0 first. Computers in Libraries, 26(5), Chen, B., & Shen, V. Y. (2006). Transforming web pages to become standard-compliant through reverse engineering. In W4A (Eds.), Proceedings of the 2006 international cross-disciplinary workshop on web accessibility (pp ). New York: ACM Press. Chen, S., Hong, D., & Shen, V. Y. S. (2005). An experimental study on validation problems with existing HTML webpages. In H. R. Arabnia (Ed.), Proceedings of international conference on Internet computing (pp ). Las Vegas. Market Share. (2009). Retrieved May 6, 2009, from Max (2004). The benefits of web standards to your visitors, your clients and you! Retrieved May 5, 2009, from The Web Standards Project (2002). What are web standards and why should I use them? Retrieved May 2, 2009, from

2 SGML, XML Document Traditional WYSIWYG Document Content Presentation Content Presentation Structure Structure? XML/SGML 3 2 SGML SGML Standard Gener

2 SGML, XML Document Traditional WYSIWYG Document Content Presentation Content Presentation Structure Structure? XML/SGML 3 2 SGML SGML Standard Gener SGML HTML XML 1 SGML XML Extensible Markup Language XML SGML Standard Generalized Markup Language, ISO 8879, SGML HTML ( Hypertext Markup Language HTML) (Markup Language) (Tag) < > Markup (ISO) 1986 SGML