XHTML and validation

1. if you want a quick web page for personal use that may not be around long, then any code will likely display okay...after all many pages on the web today are plan old html and work fine
2. otherwise write code for the "latest" standard which ensures it will be supported long into the future and looks better for your resume/portfolio

The path is not so clear however since the question is what should be the latest standard you write for as explained below

XML is another markup language with a strict structure meaning if any piece of code had one error then the whole document is deemed bad (not well-formed) and nothing displays. However the biggest differences between html and xml are:

• xml separates content and formatting (presentation) into different files so an xml file by itself has no formatting codes (no color, no images, no fonts etc) and requires a second external file (like .CSS or .XSL) to display.
• xml does not have pre-defined elements so you make up your own elements, for example <p> or <img> have no special meaning in xml. HTML basically is all about pre-defined elements (tags) that tell the browser how to display the content

1. well formed: means the document code has no syntax errors; for example if an element is missing the closing tag the entire document is not well formed
2. validation: means the document (markup and content) conforms to a schema or DTD that indicates what can be in the markup and content

• latest HTML version is 4.01: Original html was loose meaning if code had an error (like closing tag without a start tag) browser would still display the rest of the page okay
• XHTML has different versions
• XHTML 1.0 has different flavors, the 2 most important being:
  • transitional still accepts all HTML formatting elements so is more forgiving to older style web pages; it supports the deprecated elements and attributes, so is popular because it does not require style sheets and displays consistently in browsers; the drawback is it allows code which will not be supported by xml type applications, in other words it is not "future" compatible although the timeline for future is not defined so it works today
  • strict means separate formatting from content, so presentation is done with CSS instead of HTML presentation type elements; further it does not support deprecated elements and attributes so many existing pages won't validate under it; It is considered more modern than "transitional" since it is upward compatible and follows XML concepts
• XHTML 1.1 is the latest standard and is similar to 1.0 strict so it uses CSS and does not support deprecated HTML elements
• XHTML 2.0 is the most recent but is still in draft mode. It breaks backward compatibility so is like a new language that does not support many of the typical HTML elements. It  removes all presentation tags in favor of CSS, and supports logic (just like xml). Indeed 2.0 does not necessarily support XHTML 1.1 either...yikes

When creating new pages your best choice is code for XHTML 1.1 since it is more forward compatible than other versions. Even 1.1 may not even be compatible with the future, however, it will be much easier to convert strict XHTML than html into whatever the future is. Further, companies insist on competent coders even though non-standards compliant code (sloppy and deprecated code ) work just fine with today's browsers. Using XHTML strict code does NOT ensure more browsers today will render the code better, since strict requires CSS which very old browsers cannot handle. But most people use fairly modern browsers so probably not much difference display-wise whether use XHTML or not

Code for XHTML 1.1 (or strict) means

1. must follow the syntax rules, this is no big deal since the rules are not that hard and are the basic XML rules, like use proper nesting, and quite simply are good coding practices in any event
2. use styles (CSS) which is really a good idea since styles are the way to go
3. the hard part is must be careful not to use deprecated elements, actually this would be easy if there were no examples available of old code but the problem is many web pages and examples (like my notes and the book) still show code that is deprecated and sometimes you are not aware what is allowed

For this course, you can use any code you want unless the assignment instructs otherwise. You still should understand HTML code because HTML and XHTML are quite similar in many respects and when you go to update old pages you need to know what the old HTML is. The book does not cover XHTML very well so must rely on web readings.

NOTE: even though xhtml tags must be lowercase the DOCTYPE tag must be uppercase to validate

Display and Validation

Every web page  should be tested. There are 2 potential ways:

1. view in a browser: this is a must and is the real world test...the catch is what browser(s) to use since there are different ones, both old and new, and users have different screen resolutions
2. use a validator: this is optional and checks that the code conforms to the DTD in the doctype. Validators give errors for code that is not valid, although such code may in fact display okay in the browser

So to ensure code is well formed and valid for a certain standard, test it in a validator. But always test whether your page works in browsers. The issue becomes what to do when your page generates lots of crazy validator errors and yet still looks okay in a browser which is often the case

Ideally, test a page in every browser your users might use. However, some browsers are awfully old and for casual use you likely do not have more than the latest IE or Mozilla.

As for validators, there are many software and web sites available to do this, for example

What does this mean...one one hand a lot and on other very little?  If the MIME type is not correct then unexpected results may occur. But for everything except XHTML, the MIME is pretty well defined (it is like a file type association with filename extensions in windows) meaning there is only 1 type you would ever use.  However, for XHTML you can actually use a MIME of either html or xml.

In technical detail...the MIME type (sent in the Content-Type HTTP header) tells the browser that the document is an application of XML if it is application/xhtml+xml, or can also use application/xml or even text/xml although text/xml is not recommended. Most  XHTML documents are served with a MIME type of text/html, which means that they are to be considered as HTML documents. With such a MIME type, you are not using XHTML as far as browsers are concerned but are using HTML.

• use XML declaration (the first line) to state the character encoding, and in fact either (in prolog or in meta) will be overridden by an encoding specified in an HTTP header, but the XML declaration is the proper way to specify the encoding for X(HT)ML documents. A real XHTML document will ignore your meta element.
• XML prologue is most appropriate for stylesheet link
• Content-Type HTTP equivalent not needed in meta cause will be ignored for real XHTML anyway

However some browsers have trouble with a prolog so you can specify character encoding by inserting a Content-Type element into the <head> of your document to avoid troublesome prolog.

The fact that XHTML may be served as HTML or XML makes a difference to the way encoding information needs to be declared. Current browsers may display an HTML file in either standards mode or quirks mode. This means that different rules are applied to the display of the file, one conforming to the W3C standards interpretation of expected behavior, the other to expectations based on the non-standard behavior of older browsers.

In recent browsers such as Internet Explorer 7, Firefox, Opera, and others, a page served with a DOCTYPE declaration will be rendered in standards mode with or without the XML declaration,

With Internet Explorer 6, however, if anything  (like xml prolog) appears before the DOCTYPE declaration the page is rendered in quirks mode. Because Internet Explorer 6 users still count for a many users, this is a significant issue. If you want to ensure that your pages are rendered in the same way on all standards-compliant browsers, you need to think carefully about how you deal with this. It is a good idea to use a DOCTYPE declaration at the top of an HTML or XHTML file so that the document is rendered in standards mode by more recent user agents. The presence of an XML declaration in an XHTML file served as HTML will cause your file to be rendered in quirks mode on Internet Explorer 6 (and therefore for a potentially large proportion of your audience).