HTML and XHTML

[Cornel Bicutzi]

Hello,

What is the difference between HTML and XHTML...

Thanks,
------------------------------------------------------------------------
IT Interview Questions : http://www.geekinterview.comIT Tutorials and Articles : http://www.geekarticles.com

[Gus Richter]

Andy Dingley wrote:

HTML must have all elements closed, just as XHTML must.

The HTML 4.01 spec shows 15 end tags that are optional, of which 4 start
tags are also optional.
http://www.w3.org/TR/html4/index/elements.html

--
Gus

[Mark Parnell]

In our last episode, Gus Richter <gu********@net scape.net> pronounced to
comp.infosystem s.www.authoring.html:

Andy Dingley wrote:

HTML must have all elements closed, just as XHTML must.

The HTML 4.01 spec shows 15 end tags that are optional, of which 4 start
tags are also optional.

Which is why Andy went on to clarify:

The difference is that SGML allows elements to be
closed by inference from the DTD and the appearance of some other
element that couldn't be contained inside the current element (implying
necessary closure)

IOW, all elements have to be closed in both HTML and XHTML, but HTML is
capable of closing some of them automatically, whereas XHTML has to be
specifically told to close them.

--
Mark Parnell
http://clarkecomputers.com.au

[Gus Richter]

Mark Parnell wrote:

In our last episode, Gus Richter <gu********@net scape.net> pronounced to
comp.infosystem s.www.authoring.html:

Andy Dingley wrote:

HTML must have all elements closed, just as XHTML must.

The HTML 4.01 spec shows 15 end tags that are optional, of which 4 start
tags are also optional.

Which is why Andy went on to clarify:

The difference is that SGML allows elements to be
closed by inference from the DTD and the appearance of some other
element that couldn't be contained inside the current element (implying
necessary closure)

IOW, all elements have to be closed in both HTML and XHTML, but HTML is
capable of closing some of them automatically, whereas XHTML has to be
specifically told to close them.

Thank you for that, but as I understand the question, it is regarding
document creation in using HTML vs. XHTML and not how the browser is to
parse optional tags within an HTML document, or where/how am I missing
the relevance?

--
Gus

[Mark Parnell]

In our last episode, Gus Richter <gu********@net scape.net> pronounced to
comp.infosystem s.www.authoring.html:

Thank you for that, but as I understand the question, it is regarding
document creation in using HTML vs. XHTML and not how the browser is to
parse optional tags within an HTML document, or where/how am I missing
the relevance?

The OP simply asked what the difference between HTML and XHTML was. The
context wasn't specified. Matt's question, which Andy was responding to,
seems to me to be concerned with the way browsers handle the languages,
rather than authoring.

Either way, Andy's statement was correct, regardless of whether it
answered the question. :-)

--
Mark Parnell
http://clarkecomputers.com.au

[Andy Dingley]

Gus Richter wrote:

Andy Dingley wrote:

HTML must have all elements closed, just as XHTML must.

The HTML 4.01 spec shows 15 end tags that are optional, of which 4 start
tags are also optional.

Thankyou for playing straight man to the obvious question 8-)

In well-formed XML, an element is equivalent to either a pair of tags
(start and end tags) or a single tag (empty-element tag). In SGML as
used by HTML (which is slightly simplified from SGML in general) the
situation is different - there are no empty-element tags and a HTML
element can be delimited by either a pair of start and end tags (like
XML), or by a start tag and implicit closure indicated by context and
the DTD (unlike XML). For elements defined as empty (e.g. <br>) this
happens immediately after the start tag, for other elements it depends
on what follows - a <p> might be closed by the start of another <p>, or
it might even not be closed until the end of the HTML document.

However, and this is rarely grasped by HTML authors, the HTML
"elements" _are_ closed just where their XHTML equivalent would be
closed, even if there's no end tag. The "element" is an item at the
level of the DOM or the Infoset, not just a lexical item like a tag.
That's the _real_ difference between SGML tags and elements, not just a
simple "two tags make an element".

[Harrie]

Mark Parnell said the following on 10/12/2005 08:35 +0200:

In our last episode, Cornel Bicutzi pronounced to
comp.infosystem s.www.authoring.html:

What is the difference between HTML and XHTML...

http://www.w3.org/TR/xhtml1/#diffs

I can't understand why in XHTML exclusions are not possible (chapter 4.9. of the above specified URL):

http://www.w3.org/TR/xhtml1/#h-4.9

The XHTML DTD seems to be SGML, just like the HTML DTD, is this correct? If so, then I see no reason why the XHTML DTD can't use exlusions.

The XHTML DTD seems to be excluding "a" elements by declaring an entity "a.content" which does not include "a" elements:

<!-- a elements use %Inline; excluding a -->
<!ENTITY % a.content
"(#PCDATA | %special; | %fontstyle; | %phrase; | %inline.forms; |
%misc.inline;)* ">
<!ELEMENT a %a.content;>

When I validate an xhtml document with nested <a ..> elements, it fails (it's not valid).

Maybe the XHTML DTD doesn't use exclusion like the HTML DTD does;

<!ELEMENT A - - (%inline;)* -(A) -- anchor -->

but the effect seems to be the same.

Therefor, I'm having trouble with wat is said in 4.9 SGML exclusions.

--
Regards
Harrie

[Gus Richter]

Andy Dingley wrote:

Gus Richter wrote:

Andy Dingley wrote:

HTML must have all elements closed, just as XHTML must.

The HTML 4.01 spec shows 15 end tags that are optional, of which 4 start
tags are also optional.

Thankyou for playing straight man to the obvious question 8-)

In well-formed XML, an element is equivalent to either a pair of tags
(start and end tags) or a single tag (empty-element tag). In SGML as
used by HTML (which is slightly simplified from SGML in general) the
situation is different - there are no empty-element tags and a HTML
element can be delimited by either a pair of start and end tags (like
XML), or by a start tag and implicit closure indicated by context and
the DTD (unlike XML). For elements defined as empty (e.g. <br>) this
happens immediately after the start tag, for other elements it depends
on what follows - a <p> might be closed by the start of another <p>, or
it might even not be closed until the end of the HTML document.

However, and this is rarely grasped by HTML authors, the HTML
"elements" _are_ closed just where their XHTML equivalent would be
closed, even if there's no end tag. The "element" is an item at the
level of the DOM or the Infoset, not just a lexical item like a tag.
That's the _real_ difference between SGML tags and elements, not just a
simple "two tags make an element".

I'm not negating your points at all. I think that your points are well
made and are good to know. I do, however, think that in the context of
this discussion, a new author may be mislead or confused regarding the
differences between HTML and XHTML by the statement:

"HTML must have all elements closed, just as XHTML must."

The established rules for HTML clearly state that there are optional
opening/closing tags and empty elements which must not have a closing
tag. XHTML on the other hand dictates that there are no optional tags
and that all elements must be closed.

When attending a play, one goes for the performance and not for what
goes on behind the scenes, nor for the historical behind the scene
changes. Likewise, an author creates a document according to the
established rules as set out and should not have to consider behind the
scene browser parsing methods, nor historical reasons for such rules.

How it is parsed and historical linkage is good information, but not
requisite knowledge to authoring to HTML or XHTML. In the context of the
question by the OP, as I read it, I believe the statement to be
misleading and subsequent information probably only obfuscates the issue
at that level.

--
Gus

[Lachlan Hunt]

Harrie wrote:

I can't understand why in XHTML exclusions are not possible (chapter
4.9. of the above specified URL):

http://www.w3.org/TR/xhtml1/#h-4.9

The XHTML DTD seems to be SGML, just like the HTML DTD, is this correct?
It's an XML DTD. It's similar to an SGML DTD, but not quite the same.
Maybe the XHTML DTD doesn't use exclusion like the HTML DTD does;

<!ELEMENT A - - (%inline;)* -(A) -- anchor -->

but the effect seems to be the same.

Thy this in both HTML and XHTML.

<a><span><a>tes t</a></span></a>

HTML will fail validation, because the -(A) excludes the a element as
being any descendant of another a element, whereas the XHTML will pass
because the DTD only prevents child a elements.

--
Lachlan Hunt
http://lachy.id.au/
http://GetFirefox.com/ Rediscover the Web
http://GetThunderbird.com/ Reclaim your Inbox

[Harrie]

Lachlan Hunt said the following on 10/13/2005 19:24 +0200:

Harrie wrote:

The XHTML DTD seems to be SGML, just like the HTML DTD, is this correct?
It's an XML DTD. It's similar to an SGML DTD, but not quite the same.

It doesn't begin with an xml prologue, or am I too narrow minded? Can you give me some pointers where an XML DTD is explained?
Thy this in both HTML and XHTML.

<a><span><a>tes t</a></span></a>

HTML will fail validation, because the -(A) excludes the a element as
being any descendant of another a element, whereas the XHTML will pass
because the DTD only prevents child a elements.

Hmmm, so it does, interesting. Thanks for the explanation.

--
Regards
Harrie

[Eric B. Bednarz]

Harrie <sp*****@linux. org.invalid> wrote:

Lachlan Hunt said the following on 10/13/2005 19:24 +0200:

It's an XML DTD. It's similar to an SGML DTD, but not quite the
same.

It doesn't begin with an xml prologue, or am I too narrow minded?

Pardon? A DTD is *part* of the prolog -- it's an optional part of the
document type declaration which itself is an optional part of the
prolog.
Can you give me some pointers where an XML DTD is explained?

The XML spec would be a good guess, I guess.
<http://www.w3.org/TR/REC-xml/#elemdecls>
<http://www.w3.org/TR/REC-xml/#sec-entity-decl>

(note that XML markup declarations have just some restrictions compared
to SGML, so you might be able to look at an arbitrary external DTD
entity and possibly determine 'this cannot be an XML DTD', but you
cannot conclude 'this *is* an XML DTD')
--
Goodbye and keep cold