To shorten the lists in each element to describe the
"Elements Allowed Within..." and "Allowed In Content Of..."
various elements are defined by RFC 1866 as members of a group.
These group names are used throughout the RFC 1866.
Since group names can be defined to contain other group names,
this mechanism makes nesting and recursion more obvious.
Note that these element groups only deal with entities defined in RFC 1866
and do not include any of the entities in proposed extensions.
Is allowed to contain the groups
block,
heading,
text
and the entities
HR and ADDRESS
It is recommended that it contain only the groups
block,
heading
and the entities
HR,
ADDRESS and IMG
This recommended change in the members of this group means that
all text within a body should be enclosed in some kind of block.
This means that
<h1>Heading<h1>
<p>Text ...
is recommended over
<h1>Heading<h1>
Text ...
Contains the groups
phrase and font
and the entities
A,
IMG and BR
and parsed character data.
Parsed character data is any valid character data after the data
has been parsed and all
special character
entities have been replaced with their character data.
The internationalization proposal also includes in group text
the entities
BDO,
Q,
SPAN,
SUB,
SUP
The A element brackets (or anchors) a piece of text (and/or image) which
is identified as a hypertext link.
The A element must have either an HREF attribute or a NAME attribute.
The HREF attribute identifies a destination URL,
and the bracketed text is rendered as a hypertext link to the URL.
Browsers will display the contents of an A element with an HREF attribute
in a special manner to indicate that if the contents are selected,
the browser will execute that hypertext link.
The NAME attribute identifies a destination tag, and the bracketed text
is thereby identified as an available hypertext target
within this document.
Browsers do not display the contents of an A element with a NAME attribute
in any special way.
However, an A element with an HREF attribute can now be constructed
by using the document URL suffixed with #name.
This will load the document, but will position the display
starting at the location of this NAME tag.
An A element with an HREF attribute can also be constructed
to jump directly to this destination tag within the same document
by a URL consisting solely of #name
The presence of REL=relation
in document A with HREF to document/object B
identifies a relationship that B has to A that A recognizes/authorizes/verifies.
The presence of REV=relation
of the identical relation
in document B with HREF to document/object A
identifies a desired/expected/claimed relationship that B has to A,
but must be verified by checking with A.
The LANG, DIR and CHARSET attributes are introduced with the
internationalization proposal.
The CHARSET attribute is a hint as to the expected
character set used by the hyperlink.
Earlier proposals suggest changing the NAME attribute to ID,
declaring the NAME attribute as obsolete,
and adding an ID attribute for various elements
including the paragraph and heading elements.
With the ID, MD, and CLASS attributes as part of the style sheet
proposal, this is likely to change.
At present REV and REL are rarely used or supported,
and are Level 1 attributes,
but are of growing interest to automated document environments.
These relationships are more commonly identified in the HEAD
of the document using the LINK element.
REL and REV can be either a comma-separated or white space separated
set of relationship(s) of the HREF link.
One proposal suggests that comma imply "or" and white
space imply "and" for a list of values.
These relationships and their semantics were originally proposed
to be registered with an HTML authority, which was described at
http://www.w3.org/hypertext/WWW/MarkUp/RegistrationAuthority.html,
but that link no longer responds. The entire topic of link relationships
is under active discussion and change.
Early examples of relationship names are
UseIndex, UseGlossary, Annotation, Reply, Embed,
Precedes, Subdocument, Present, Search, Supersedes, History, Made,
Owns, Approves, Supports, Refutes, Includes, Interested.
URN is for a Universal Resource Number, and is not currently
used or supported, and is a Level 1 attribute.
RFC 1866 describes it as a preferred, more persistent identifier
than the value of HREF.
TITLE is little used or supported, and is a Level 1 attribute,
but is expected to be the title of the HREF document.
RFC 1866 suggests TITLE can be displayed as a margin note
or on a small box while the mouse is over the anchor.
(ed. I am unaware of any browser that has implemented this
feature.)
METHODS is little used or supported, but is expected to be
a white-space-separated list of HTTP METHODS supported by the
object and accessible to the user.
RFC 1866 suggests that the content of the A element may be
rendered differently depending upon the HTTP method.
TARGET is a Netscape 2.0 extension to define a window name for use
by the retrieved hyperlink. If the named window is not already
open, Netscape 2.0 will open a new window and assign it that name.
See also the FRAME element for naming a window.
SHAPE is proposed to provide a mechanism to define
multiple A elements and corresponding "hotzones" within the
proposed FIG element, to perform the equivalent function
of ISMAP without the need for writing a responding cgi-bin program.
For an alternate proposal, see the MAP element.
A is a Level 0 element.
Any element that permits members of group
body.content
Variations
ADDRESS text is typically rendered in italics, and may be indented.
The LANG and DIR attributes are introduced with the
internationalization proposal.
The internationalization proposal also includes
ALIGN=center|left|right|justify
The CLEAR and NOWRAP attributes are proposed in Version 3.
ADDRESS is a Level 0 element.
Sandia Requirements
An ADDRESS element is required to identify a "point-of-contact".
For details, see the Sandia Requirements.
The APPLET element replaced the APP element as the mechanism
to identify and invoke a JAVA(tm) application. A browser that
understands this element will ignore everything in the content
of the APPLET element except the PARAM elements. Browsers that
do not understand this element should ignore it and the PARAM
elements and instead process the content of the element.
Thus the content is the alternate HTML if the application is
not invoked. CODE is the name of the file that contains the
compiled Applet subclass. This name is relative to the base
URL of the applet and cannot be an absolute URL. WIDTH and HEIGHT
give the initial width and height (in pixels) of the applet
display area. CODEBASE specifies the base URL of the applet.
ALT specifies parsed character data to be displayed if the brower understands
the APPLET tag but can't/won't run them. NAME specifies a name
for the applet instance, which allows applets on the same page
to commuicate with each other. ALIGN specifies the display
alignment. VSPACE and HSPACE specify the reserved space around the
applet (in pixels).
The AREA element
specifies a single area of an image which, if selected, will
link to the hyperlink identified by HREF.
If multiple AREA elements in the
same MAP define overlapping areas, the first encountered takes precedence.
The AREA element is part of a proposed enhancement to
provide client-side image maps.
COORDS describes the position of
an area (in pixels) of the image in comma-separated x,y coordinates
where the upper-left corner is "0,0".
For SHAPE=rect (the default), it is "left,top,right,bottom".
For SHAPE=circle, it is "center_x,center_y,radius".
For SHAPE=polygon, it is successive x,y vertices of the polygon.
If the first and last coordinates are not the same, then a segment
is inferred to close the polygon. The NOHREF indicates that this
region should generate no links.
The ALT attribute specifies optional parsed character data to describe the area which
could be displayed by a text-only browser as a substitute for the image.
AREA is an extension in Netscape 2.0,
but the only SHAPE recognized is "rect", and ALT is not defined.
Browsers who do not have bold may render in some other manner.
RFC 1866 states that <B> content must be rendered as distinct
from <I> content.
The LANG and DIR attributes are introduced with the
internationalization proposal.
All character definition elements are Level 2.
The BANNER element is proposed for corporate logos, navigation aids,
disclaimers and other information which shouldn't be scrolled
with the rest of the document.
The BASE element provides the absolute
URL
base to be used for any relative URL links in this document.
It must be a complete file name,
and is usually the original URL of this document.
If this file is moved, having the BASE set to the original URL
eliminates the need to also move all the documents which are
identified by relative URL links in this document.
This is a Level 0 element.
Netscape 2.0 defines the TARGET attribute to define a default
named target window for every link in a document that does not
have an explicit TARGET attribute.
The BDO element is a directional override feature needed to deal
with unusual pieces of text in which directionality cannot be
resolved from context in an unambiguous fashion.
It requires the DIR attribute.
The meaning of DIR is different on BDO than on inline text markup elements.
For BDO the DIR attribute
is a bidi override, forcing the directionnality of even those
characters that have strong directionnality. On inline elements,
DIR indicates a new directional embedding level, affecting mostly the
neutrals and the overall layout.
The BGSOUND element will cause an audio file to be presented
as background to the document.
Minimum Attributes
<BGSOUND SRC="...">
All Possible Attributes
<BGSOUND
SRC="..."
LOOP="...">
Elements Allowed Within...
The BGSOUND element is defined as having no content.
Allowed In Content Of...
-In progress, to be determined-
Variations
The BGSOUND element is a MS Internet Explorer 2.0 enhancement.
SRC specifies the URL of the audio
file to be played. LOOP specifies how many times the sound will
be displayed while the HTML document is displayed, and can either
be a number or the string "infinite". The default for LOOP is one.
Considerable opposition to the use of this element has been expressed
on the Web, especially the use of LOOP=infinite, since users currently
have no way to disable the audio.
The BLINK element changes the physical rendering of the
contents of the element to a blinking font.
Minimum Attributes
<BLINK>characters...
</BLINK>
All Possible Attributes
<BLINK>characters...
</BLINK>
Elements Allowed Within...
-In progress, to be determined-
Allowed In Content Of...
-In progress, to be determined-
Variations
This element is a Netscape 1.1 enhancement
and many browsers ignore this element.
Many users find the use of this element annoying.
It should be restricted to short term use for new information.
Netscape 1.1N permits users to disable the rendering of this element.
Any element that permits members of group
block.forms
Variations
RFC 1866 states the contents of the BLOCKQUOTE element is typically
rendered slightly indented both left and right, and/or italic font.
It also states that a single-font browser rendering may
display the contents with a vertical line
of ">" characters down the left margin
to indicate quotation in the Internet mail style.
The LANG and DIR attributes are introduced with the
internationalization proposal.
The internationalization proposal also includes
ALIGN=center|left|right|justify
Version 3 proposes replacing the BLOCKQUOTE element with the
BQ element.
BLOCKQUOTE is a Level 0 element.
The BODY element contains all the content of the document,
as opposed to the
HEAD, which contains information about
the document. All displayable elements should be within the content
of the BODY.
The HTML, HEAD, and BODY elements were introduced with
RFC 1866, and are currently optional, but are considered
Level 0 elements.
The LANG and DIR attributes are introduced with the
internationalization proposal.
Version 3 proposes the attribute BACKGROUND.
Netscape 1.1 and MS Internet Explorer 2.0
include the BACKGROUND attribute, which is a URL to point
to an image to be reproduced to fill the background of the document.
Netscape 1.1 and MS Internet Explorer 2.0 both document the BGCOLOR extension.
Netscape requires an "#rrggbb" number, while
Internet Explorer also accepts the following color names:
Black, Maroon, Green, Olive, Navy, Purple, Teal, Gray, Silver, Red, Lime,
Yellow, Blue, Fuchsia, Aqua, White.
The TEXT, LINK, VLINK, and ALINK attributes are Netscape 1.1 extensions,
and also appear to work with MS Internet Explorer 2.0.
BGPROPERTIES is an MS Internet Explorer 2.0 extension and currently
only accepts the value of "fixed" which provides a watermark
(non-scrolling) background image.
The BQ element defines a multi-line set of text
to be rendered as quoted text.
The content of the BQ element may optionally contain the
CREDIT element.
The BQ element is proposed in Version 3 as a replacement
for the BLOCKQUOTE element.
Version 3 does not indicate the typical rendering of the contents
of the BQ element, but does indicate that it may not imply
text separation.
The CLEAR attribute is a Netscape 1.1 extension and was added
to force the line break to clear possible floating graphic images.
The standard tables proposal expects the presence of this attribute
since it expects text to flow around a table, if possible,
but does not standardize the attribute's existence as part of the proposal.
Version 3 includes CLEAR, and proposes the remaining attributes.
The BR element is Level 0.
The CENTER element is a Netscape 1.1 extension.
It was provided as an alternative to the HTML+ and Version 3
ALIGN="center" proposed new attribute for all the
text block elements such as the
<P> paragraph and
<H?> header elements and the new
<DIV> element.
Both RFC 1866 and the current Version 3 specification
only include the ALIGN attribute, not the CENTER element,
and Netscape 2.0 has added the ALIGN="center" attribute.
This is typically rendered in italics.
The LANG and DIR attributes are introduced with the
internationalization proposal.
All character definition elements are Level 2.
The CODE element changes the character rendering of the
contents of the element to logically represent computer code.
It is intended for short words or phrases. PRE is recommended
for multiple-line listings.
This is usually rendered in a fixed-width font.
The LANG and DIR attributes are introduced with the
internationalization proposal.
All character definition elements are Level 2.
The COL element is part of the new proposed standard
tables structure. It is not widely implemented in existing browsers.
The order of placement of a series of COL elements within the TABLE
content (or within a COLGROUP) is significant, and describes the columns
in the TABLE DIR presentation order, first to last.
SPAN is a positive integer that specifies how many columns this element
applies to, defaulting to one. SPAN=0 implies all columns from the
current column up to and including the last column.
WIDTH is a decimal number which specifies the width
of each column in the span.
The default units is pixels, but may be specified by a suffix on the number.
The standard defined units suffixes include:
pt=points,
pi=picas,
in=inches,
cm=centimeters,
mm=millimeters,
em=em units,
px=screen pixels.
In addition, the special suffix of an asterisk may be used
to specify a number to be used as a multiplier of a "standard" column width.
This is used to specify the widths of columns proportionately relative
to each other.
If a number does not accompany the asterisk it defaults to one.
Specifying WIDTH="0*" forces the column to its minimum width.
ALIGN and VALIGN define the alignment of text within a table cell.
If ALIGN=char, CHAR specifies the alignment character which defaults
to the decimal point for the current language. CHAROFF is a decimal
number that specifies the offset for the alignment character from the
beginning of the table cell in the DIR direction. Units may be specified
using the standard defined units suffixes plus the special suffix of the
percent sign indicating the percentage of offset within the cell from the
beginning of the cell.
The COLGROUP element is part of the new proposed standard
tables structure. It is not widely implemented in existing browsers.
The end tag is optional and is inferred when a new COLGROUP is encountered
or the THEAD or TBODY element of TABLE.
The order of placement of a series of COLGROUP elements within the TABLE
content is significant, and describes the columns
in the TABLE DIR presentation order, first to last.
ALIGN and VALIGN define the alignment of text within a table cell.
If ALIGN=char, CHAR specifies the alignment character which defaults
to the decimal point for the current language. CHAROFF is a decimal
number that specifies the offset for the alignment character from the
beginning of the table cell in the DIR direction.
The default units is pixels, but may be specified by a suffix on the number.
The standard defined units suffixes include:
pt=points,
pi=picas,
in=inches,
cm=centimeters,
mm=millimeters,
em=em units,
px=screen pixels.
In addition, the special suffix of the percent sign may be used to
indicate the percentage of offset within the cell from the
beginning of the cell.
The CREDIT element is proposed in Version 3.
Version 3 does not indicate the typical rendering for the contents
of the CREDIT element, nor whether that rendering should be different
or separated from the enclosing block quotation or figure.
The DD element identifies the separated multi-line
definition item in a DL definition list.
In a DL list a DD should always be preceded by at least one DT element.
This is typically rendered as normal text, indented.
Most browsers assume that the contents of the element
ends when it encounters a <DT> or
another <DD> or the </DL>
element to end the list, and thus do not require the
ending tag.
The LANG and DIR attributes are introduced with the
internationalization proposal.
All list elements are Level 0.
The DEL element
changes the character rendering of the
contents of the element to logically represent
deleted text, for instance in modifications in legal documents.
The DEL element is a Version 3 proposed element.
While this will typically be rendered by strikethru characters,
the DEL element is preferred over using the S or STRIKE elements.
All character definition elements are Level 2.
The DFN element is not widely implemented, but
is usually rendered bold or bold italic.
It is a Version 3 proposed element and was in an earlier
Version 2 proposal.
The RFC 1866 describes it and recognizes it as existing practice
but does not include it in the standardized elements.
This list type is not commonly implemented, and
is often rendered identically to UL.
RFC 1866 specifies that the content of the LI element of the
DIR list is usually less than 20 characters in length.
These may be arranged in columns across the page,
each column typically as 24 characters wide.
Specifying <UL PLAIN WRAP=HORIZ> is proposed
in Version 3 as a replacement for the DIR element.
The exclusion in RFC 1866 of group block within DIR implies
(among other things) that DIR can contain no nested lists,
nor any paragraphs even though the LI element normally would allow this.
Some browsers do not enforce this exclusion.
The LANG and DIR attributes are introduced with the
internationalization proposal.
The internationalization proposal also includes
ALIGN=center|left|right|justify
All list elements are Level 0.
The DIV element is proposed to be used
with the CLASS attribute to represent different kinds of containers,
e.g. chapter, section, abstract, or appendix.
This is a proposed Version 3 element.
<DIV ALIGN=center> is the proposed replacement
for the non-standard CENTER element.
This is a Netscape 2.0 extension but it only recognizes
ALIGN=left|right|center.
The COMPACT attribute suggests rendering the list in
a physically compact way, but is not implemented by many browsers.
RFC 1866 suggests that the attribute be used if the list items are
small and/or the entire list is large, and may cause the elimination
of blank lines between DT/DD pairs.
The LANG and DIR attributes are introduced with the
internationalization proposal.
Version 3 proposes the CLEAR attribute.
To obtain a specific look, it has been the practice to construct
a DL with DD elements but empty or missing DT elements.
Missing DT elements violates the standard.
Version 3 proposes a new NOTE element as the
preferred alternative to achieve the effect desired.
All list elements are Level 0.
The DT element identifies the separated term item in a DL definition list.
Multiple DT elements may exist prior to a single DD element.
In a DL list a DD should always be preceded by at least one DT element.
This is typically rendered in a bold font, but not indented.
Most browsers assume that the contents of the element
ends when it encounters a <DD> or
another <DT> or the </DL>
element to end the list, and thus do not require the
ending tag.
The LANG and DIR attributes are introduced with the
internationalization proposal.
All list elements are Level 0.
Usually rendered in italics.
RFC 1866 states that <EM> content must be rendered as distinct
from <STRONG> content.
The LANG and DIR attributes are introduced with the
internationalization proposal.
All character definition elements are Level 2.
The EMBED element defines a container that allows the insertion
of arbitrary objects directly into an HTML page. Embedded objects
are supported by application-specific plug-ins. EMBED is defined
to allow arbitrary attributes.
The EMBED element is a Netscape 2.0 extension.
Netscape gives as examples of plug-in applications:
WebFX by Paper Software for viewing VRML
(Virtual Reality Modeling Language) worlds,
Adobe Acrobat for PDF documents, and Macromedia Director and Apple QuickTime
for multimedia.
Netscape's examples do not include the ending element,
but their documentation does not specify what would imply
the end of the EMBED content if the ending element is absent.
The concept of arbitrary undefined attributes violates the underlying
SGML standard of HTML. Change is expected in this element.
See also the APPLET element.
The Netscape documentation does not currently define the units for
HEIGHT and WIDTH but examples would imply that they are pixels.
The object's image will be scaled to fit the specified height/width.
Embedded plug-in applications may be activated by double-clicking
their image.
The FIG element is an advanced form of the
IMG element to define an image, with optional
overlays, text elements and "hotzones", to be inserted
within a document.
The structure of the contents of the FIG element expects
a series of optional overlay images defined by OVERLAY elements,
followed by an optional CAPTION element,
followed by text to be presented as an alternative to the image(s)
and which may contain normal text elements as well as
hypertext links defined by A elements with
SHAPE attributes to identify "hotzones" on the image,
finally completed by an optional CREDIT element.
The FIG element is proposed in Version 3 as a sophisticated
alternative to the IMG element, especially to enhance the capabilities
of the ALT text for non-graphical presentations of HTML documents,
as well as a way to accomplish the common use of the ISMAP attribute
of the IMG element without the need for a responding cgi-bin program.
The FN element logically identifies text to be presented
as a footnote. The reference location for the footnote is expected
to be an A element whose HREF attribute references the ID of the
FN element.
The FN element is a Version 3 proposed element.
Version 3 specifies that when practical, the browser should
render the FN element as pop-up notes.
Version 3 does not specify an expected rendering when the footnote
text is simply displayed where it occurs in the document.
However, Version 3 does state that the FN element does not imply
text separation. Therefore, common practice expects that the
contents of the FN element would begin with a markup element
which produces separation, e.g. the P element.
The FONT element changes the font size of the following characters
to one of the seven defined sizes, or plus or minus from the document
BASEFONT
size.
The FONT element with the SIZE attribute is a Netscape 1.1
and MS Internet 2.0 extension.
Netscape 2.0 and MS Internet Explorer 2.0 added the COLOR attribute.
Netscape requires an "#rrggbb" number, while
Internet Explorer also accepts the following color names:
Black, Maroon, Green, Olive, Navy, Purple, Teal, Gray, Silver, Red, Lime,
Yellow, Blue, Fuchsia, Aqua, White.
The FACE attribute is a MS Internet Explorer 2.0 extension to specify
a font style, but the permitted values are not defined.
The FORM element creates a fill-out form.
The browser permits the user to enter information in the fields of
the form and sends this information to a cgi-bin script on a server
identified as a URL by the ACTION attribute.
METHOD=GET (the default) appends the input information to the
ACTION URL which on most receiving systems becomes
the value of the environment variable QUERY_STRING.
METHOD=POST (the preferred) sends the input information in a data body
which is available on stdin with the data length set
in the environment variable CONTENT_LENGTH.
Form data is a stream of name=value pairs
separated by the & character.
Each name=value pair is URL encoded,
i.e. spaces are changed into the plus character and some characters
are encoded into hexadecimal.
At least one of the following is expected inside the FORM
contents: INPUT, SELECT, TEXTAREA.
Any element that permits members of group
block.forms
Variations
The original definition of FORM only defined the default value of
application/x-www-form-urlencoded
for the ENCTYPE attribute.
RFC 1867 adds the value of multipart/form-data
for this attribute to permit a FORM to upload one or more files
in addition to the FORM data. METHOD=POST is required with
this value of ENCTYPE.
This file upload extension is implemented in Netscape 2.0.
Most current browsers only handle ACTION for the
http: access type, but proposals for handling the
mailto: access type are being discussed.
The LANG, DIR, and ACCEPT-CHARSET attributes are introduced with the
internationalization proposal.
The ACCEPT-CHARSET attribute is to provide a hint as to the
character set or sets that the receiving URL is prepared to handle.
Version 3 proposes the SCRIPT attribute to specify a URL which
contains a limited syntax script to be downloaded to the browser
for execution to preprocess the FORM output before sending it
to the ACTION destination.
This SCRIPT proposal may change as part of the new proposals
concerning APPLET and EMBED.
All elements concerning FORM are Level 2.
The FRAME element defines a single frame in a frameset.
The SRC attribute value is the URL of the document to be displayed
in this frame. A FRAME element without a SRC is displayed as
blank space. The NAME element assigns a name to the frame to be used
as a target of hyperlinks. (See the A element)
The SCROLLING attribute is used to define whether the frame should
have a scrollbar, and defaults to the value "auto".
Presence of the NORESIZE attribute
prevents the frame from being resized by the user.
The FRAME element is a Netscape 2.0 enhancement for defining
multiple windows for viewing a document. Netscape 2.0 reserves the
following values for the NAME attribute: _blank, _self, _parent,
_top. These values must begin with the underbar character and
have the following respective meanings: new unnamed window, load in the
same window, load in the parent window (if none use self), load in the
top window (if none use self). Netscape 2.0 defines the units for
MARGINWIDTH and MARGINHEIGHT as pixels.
The FRAMESET element is used instead of the BODY element.
It is used in an HTML document whose sole purpose is to define the
layout of the sub-HTML documents, or Frames, that will make up the page.
The ROWS and COLS values are comma-separated lists describing the
row-heights and column-widths of the Frames.
The FRAMESET element is a Netscape 2.0 enhancement. Netscape 2.0
restricts the ROWS and COLS values to integers with an optional suffix
to define the units. Default units are pixels. A percentsign suffix
indicates the value is a percentage between 1 and 100.
A suffix of an asterisk may be used
to specify a number to be used as a multiplier of a "standard" width/height.
This is used to specify the widths/heights of Frames proportionately relative
to each other.
If a number does not accompany the asterisk it defaults to one.
RFC 1866 recommends that a document use the
header elements in order, without breaks, beginning with H1.
(ed: I have not identified a browser that enforces this.)
The LANG and DIR attributes are introduced with the
internationalization proposal.
The internationalization proposal also includes
ALIGN=center|left|right|justify
Netscape 2.0 implements
ALIGN=center|left|right
The ID attribute was proposed in HTML+ and Version 3
to replace the NAME attribute
in the A element to establish internal hyperlink destinations.
With the expected use of ID for style sheets, this is likely to change.
All other attributes are proposed in Version 3.
RFC 1866 states typical rendering is bold, very large font, centered;
when printed causes a page break.
Headings are Level 0 elements.
RFC 1866 recommends that a document use the
header elements in order, without breaks, beginning with H1.
(ed: I have not identified a browser that enforces this.)
The LANG and DIR attributes are introduced with the
internationalization proposal.
The internationalization proposal also includes
ALIGN=center|left|right|justify
Netscape 2.0 implements
ALIGN=center|left|right
The ID attribute was proposed in HTML+ and Version 3
to replace the NAME attribute
in the A element to establish internal hyperlink destinations.
With the expected use of ID for style sheets, this is likely to change.
All other attributes are proposed in Version 3.
RFC 1866 states typical rendering is bold, large font, flush left.
Headings are Level 0 elements.
RFC 1866 recommends that a document use the
header elements in order, without breaks, beginning with H1.
(ed: I have not identified a browser that enforces this.)
The LANG and DIR attributes are introduced with the
internationalization proposal.
The internationalization proposal also includes
ALIGN=center|left|right|justify
Netscape 2.0 implements
ALIGN=center|left|right
The ID attribute was proposed in HTML+ and Version 3
to replace the NAME attribute
in the A element to establish internal hyperlink destinations.
With the expected use of ID for style sheets, this is likely to change.
All other attributes are proposed in Version 3.
RFC 1866 states typical rendering is italic, large font, slightly indented.
Headings are Level 0 elements.
RFC 1866 recommends that a document use the
header elements in order, without breaks, beginning with H1.
(ed: I have not identified a browser that enforces this.)
The LANG and DIR attributes are introduced with the
internationalization proposal.
The internationalization proposal also includes
ALIGN=center|left|right|justify
Netscape 2.0 implements
ALIGN=center|left|right
The ID attribute was proposed in HTML+ and Version 3
to replace the NAME attribute
in the A element to establish internal hyperlink destinations.
With the expected use of ID for style sheets, this is likely to change.
All other attributes are proposed in Version 3.
RFC 1866 states typical rendering is bold, normal font, indented more than H3.
Headings are Level 0 elements.
RFC 1866 recommends that a document use the
header elements in order, without breaks, beginning with H1.
(ed: I have not identified a browser that enforces this.)
The LANG and DIR attributes are introduced with the
internationalization proposal.
The internationalization proposal also includes
ALIGN=center|left|right|justify
Netscape 2.0 implements
ALIGN=center|left|right
The ID attribute was proposed in HTML+ and Version 3
to replace the NAME attribute
in the A element to establish internal hyperlink destinations.
With the expected use of ID for style sheets, this is likely to change.
All other attributes are proposed in Version 3.
By default, Netscape uses a font size for H5 that
is smaller than default text.
For most other browsers, the font size for all headers
is at least as large as the default text.
RFC 1866 states typical rendering is italic, normal font, indented as H4.
Headings are Level 0 elements.
RFC 1866 recommends that a document use the
header elements in order, without breaks, beginning with H1.
(ed: I have not identified a browser that enforces this.)
The LANG and DIR attributes are introduced with the
internationalization proposal.
The internationalization proposal also includes
ALIGN=center|left|right|justify
Netscape 2.0 implements
ALIGN=center|left|right
The ID attribute was proposed in HTML+ and Version 3
to replace the NAME attribute
in the A element to establish internal hyperlink destinations.
With the expected use of ID for style sheets, this is likely to change.
All other attributes are proposed in Version 3.
By default, Netscape uses a font size for H6 that
is smaller than default text.
For most other browsers, the font size for all headers
is at least as large as the default text.
RFC 1866 states typical rendering is bold, normal font, not indented.
Headings are Level 0 elements.
The HEAD contains general information about the document.
None of the elements authorized to exist
in the contents of the HEAD are displayed;
the displayed material is found within the
BODY.
The HTML, HEAD, and BODY elements were introduced with
RFC 1866, and are currently optional, but are considered
Level 0 elements.
RFC 1866 recommends against permitting the NEXTID element in HEAD.
The LANG and DIR attributes are introduced with the
internationalization proposal.
Version 3 proposes the new elements of RANGE and STYLE
as allowed within a HEAD.
The HP set of elements, with n=1,2,... provided
a mechanism to highlight the characters in a phrase with
one of a set of browser defined highlight mechanisms.
Minimum Attributes
<HPn>characters... </HPn>
All Possible Attributes
<HPn>characters... </HPn>
Variations
HPn was an early idea about highlighting but is seldom used
or implemented.
RFC 1866 does not include these elements, and these elements
should be considered obsolete.
Any element that permits members of group
body.content
Variations
RFC 1866 states that the HR element is typically rendered as
a full width horizontal rule or equivalent graphic.
The DIR attribute is introduced with the internationalization proposal
since DIR can have meaning
in those cases where the horizontal rule is not full width.
The internationalization proposal also includes
ALIGN=center|left|right|justify
The CLEAR, SRC, and MD attributes are proposed in Version 3.
SRC is proposed to specify a custom image for the rule.
Some browsers draw the line only within the current text margins
(which may be indented on left and/or right due to lists, etc.)
SIZE, WIDTH, ALIGN, and NOSHADE attributes are Netscape extensions,
but Netscape 2.0 only implements
ALIGN=center|left|right
The HR element is Level 0.
The HTML element is intended to bracket the entire HTML text in
the document. All other HTML elements are inside the start and end
of the HTML element.
The element HTML is the outermost element and should not be
nested inside any element.
RFC 1866 introduced the VERSION attribute.
The LANG and DIR attributes are introduced with the
internationalization proposal.
Variations
The HTML, HEAD, and BODY elements were introduced with
RFC 1866, and are currently optional, but are considered
Level 0 elements.
RFC 1866 identifies the presence of the PLAINTEXT element within
the HTML content as deprecated.
Browsers who do not have italics may render in some other manner.
RFC 1866 states that <B> content must be rendered as distinct
from <I> content.
The LANG and DIR attributes are introduced with the
internationalization proposal.
All character definition elements are Level 2.
The IMG element allows an image file to be inserted within
an HTML document along with the text.
The ALT attribute defines parsed character data
that will be displayed if the image
is not or cannot be displayed by the browser.
The SRC attribute identifies a URL for retrieving the image.
The ISMAP attribute is only meaningful if the IMG element is
within the contents of an A element, and a responding cgi-bin program
has been established at the URL identified by the HREF attribute
of the A element.
If a single A element spans both an image and text,
the cgi-bin program will receive the HREF input values, if any,
if the text is selected, or the x,y cursor pixel coordinates
relative to 0,0 as the the upper-left corner of the image
if the image is selected.
For references to this advanced feature, see
Acknowledgements.
The attributes BORDER, HEIGHT, WIDTH, HSPACE, VSPACE, and LOWSRC
are Netscape 1.1 extensions.
The attributes DYNSRC, START, CONTROLS, LOOP, and LOOPDELAY are
MS Internet Explorer 2.0 extensions.
The LANG and DIR attributes are introduced with the
internationalization proposal and relate to the value of
the ALT attribute.
Version 3 proposes adding the WIDTH and HEIGHT attributes.
RFC 1866 defines only bottom|middle|top
as values for the ALIGN attribute.
The additional values for the ALIGN attribute are Netscape extensions,
and include the capability to define floating images.
Version 3 proposes adding only the (left|right) values for ALIGN
to identify images that imply that text can float around the image.
Some image capable browsers will display the ALT
text until the full image is retrieved.
RFC 1866 states that
Level 0 conformance must accept the element, but Level 1
conformance is required before it displays the image.
Version 3 proposes the UNITS attribute for use by the WIDTH and HEIGHT
attributes to define units other than pixels.
RFC 1866 states that an IMG element with an ISMAP element must
only exist in the content of an A element with an HREF attribute.
Version 3 proposes a FIG element
as an advanced alternative to the IMG element,
for more sophisticated multi-part overlay images,
more control over text that is the equivalent of ALT,
and with a proposed SHAPE attribute on the A element
a method to perform the common use of the ISMAP attribute
without the need for a responding cgi-bin program.
The USEMAP is part of an alternate proposal for client-side
image mapes.
It is implemented as a Netscape 2.0 extension.
USEMAP specifies a URL with a "#NAME" suffix
to identify a file and MAP name, and is used with the MAP element.
DYNSRC is defined by MS Internet Explorer 2.0 as the
address of a video clip or VRML world to be displayed if your
browser is capable, otherwise display SRC.
START is defined by MS Internet Explorer 2.0 as when the
videoclip or VRML world is to be displayed. If START=mouseover,
the SRC image is displayed until the mouse cursor is over that image.
CONTROLS is defined by MS Internet Explorer 2.0 as displaying a set
of controls under the animation.
LOOP is defined by MS Internet Explorer 2.0 as how many times a
video clip will loop when activated.
If LOOP=-1 or LOOP=INFINITE,
it will loop indefinately.
LOOPDELAY is defined by MS Internet Explorer 2.0 as how long,
in milliseconds, a video clip will wait between replays.
Sandia Requirements
The ALT attribute and a value describing the image is required.
If known, the BORDER, HEIGHT, and WIDTH attributes are recommended
as they improve download performance on some browsers.
If used, the HEIGHT and WIDTH attributes shall specify the actual
size of the stored image in pixels.
The INPUT element is used to specify a simple input field
as part of the contents in a FORM element.
TYPE=text is default.
NAME defines the symbolic name of the field
returned to the server on submission and must be present
for all but TYPE=submit|reset.
For TYPE=checkbox|radio, multiple INPUT elements may have the same NAME value.
TYPE=radio insures that exactly one choice amongst
INPUT elements with the same NAME value is selected at all times.
VALUE is used to specify a default.
For TYPE=text|password VALUE defines default text to be returned,
which normally is null.
For TYPE=password the value should be obscured as it is entered.
For TYPE=checkbox|radio VALUE defines the value returned when the
checkbox or radio is selected, which defaults to "on"
For TYPE=submit|reset VALUE defines the label for the pushbutton.
Multiple TYPE=submit should have different NAME values to
identify which submission button was selected.
CHECKED defaults the specific checkbox or radio INPUT to selected.
For TYPE=radio the default element checked is the first among
those with the same NAME value.
TYPE=image defines
an image, whose URLis identified by the SRC attribute,
which, when clicked, performs the form submission and sends
the X,Y coordinates of the click, similar to ISMAP in the IMG element.
SIZE and MAXLENGTH are only
used with TYPE=text|password.
SIZE is the physical size of the
displayed input field expressed in characters
or characters,rows.
MAXLENGTH is the maximum number of characters that are
accepted as input.
RFC 1866 defines the TYPE attribute values of
text|password|checkbox|radio|submit|reset|hidden|image
The <TEXTAREA> element
should be used instead of this INPUT element for multiline input areas.
TYPE=hidden VALUE="..." is recognized by some browsers, and is
used to submit fixed information not entered by the user.
Note that RFC 1866 sets a max limit of 1024 characters on VALUE.
RFC 1866 only defines ALIGN=top|middle|bottom
RFC 1867 defines a mechanism for FORM-based file upload.
It defines the value file for the TYPE attribute
and defines a new ACCEPT attribute.
Specifying TYPE=file
permits attaching one or more local files to the submitted output.
The browser may permit the user to specify multiple file names in
response to a single INPUT element with TYPE=file
ACCEPT is a list of media types or type patterns allowed for input.
The valid ACCEPT values and meaning of those values is platform dependent.
The value for the VALUE attribute specifies a default file name.
The browser must ask for confirmation before sending any file based
on the default file name.
A value for the SIZE attribute of "width,height"
would specify a default filename width for the input display and height
sufficient to show some number of files.
A value for the MAXLENGTH attribute specifies a maximum Content-Length
(in bytes) which the responding server is likely to accept
for transferred files.
Including an INPUT element with TYPE=file requires also specifying
ENCTYPE="multipart/form-data" METHOD=post
on the enclosing FORM element.
If the FORM does not specify this ENCTYPE, the behavior is unspecified
and the file transfer is likely to be rejected by the responding server.
Netscape 2.0 implements FORM-based file upload.
The LANG and DIR attributes are introduced with the
internationalization proposal.
Version 3 proposes the additional TYPE attribute values of
range|scribble.
A beta version of Netscape includes a TYPE attribute value of jot
which is similar to scribble.
Version 3 proposes the additional DISABLED, ERROR,
MIN, MAX, and MD attributes.
The Version 3 proposed TYPE=range
uses the proposed MIN and MAX attributes to specify limits to
numeric (real or integer) input.
The proposed default value is halfway between MIN and MAX.
The Version 3 proposed TYPE=scribble
is to allow the user to scribble with a pointing device on top of
the image specified with the SRC attribute.
The Version 3 proposed DISABLED attribute would display this INPUT,
but prohibit user entry/modification.
The Version 3 proposed ERROR attribute defines text to be displayed
in the event that the entered value for this INPUT is invalid.
All elements concerning FORM are Level 2.
The INS element
changes the character rendering of the
contents of the element to logically represent
inserted text, for instance in modifications in legal documents.
This element appears to be a precursor to the
<FORM> element which has more features.
When placed in the BODY of the document,
it requires the ACTION attribute to point to a cgi-bin
program which can handle the query, and
produces a simple INPUT field with a prompt
of: "This is a searchable index. Enter search keywords:"
When placed in the HEAD of the document, it
informs the browser that the document is an index
document and can be examined using a keyword search.
The ISINDEX element is usually generated
automatically by a server-side script.
The ISINDEX element is defined as having no content.
Allowed In Content Of...
the element
<HEAD>
and any element that permits members of the group
block.forms
Variations
ISINDEX is a Level 0 element.
The new, more sophisticated
FORM
element is now used more widely.
The LANG and DIR attributes are introduced with the
internationalization proposal.
Version 3 proposes the PROMPT attribute.
The PROMPT attribute is a Netscape extension.
Any element that permits members of the group
phrase
Variations
While intended to be distinguishable from CODE, so that
input and output would be different, most browsers render
this the same as CODE, simply as a fixed-width font.
The LANG and DIR attributes are introduced with the
internationalization proposal.
All character definition elements are Level 2.
The LANG element is used to change the default LANG context
for subsequent elements from the current default.
A LANG attribute on an element overrides this default LANG context
for the content of that particular element.
Minimum Attributes
<LANG>characters... </LANG>
All Possible Attributes
<LANG
ID="..."
CLASS="...">characters...
</LANG>
Elements Allowed Within...
-In progress, to be determined-
Allowed In Content Of...
-In progress, to be determined-
Variations
The LANG element is a Version 3 proposed element.
All character definition elements are Level 2.
The LH element defines a list header used as a title for a list.
Browsers can use this in place of the full list when a mechanism
is provided to fold and unfold nested lists.
The LH element is proposed in Version 3 as an optional element
at the beginning of some lists. It is expected that browsers may
render this element in a different style/font than the list itself.
The LI element defines a list item.
It is rendered differently depending upon the list within which
it appears.
Minimum Attributes
<LI>characters...
All Possible Attributes
<LI
LANG="..."
DIR=ltr|rtl
ALIGN=center|left|right|justify
within UL TYPE=disk|circle|square
within OL TYPE=A|a|I|i|i
within OL VALUE=n>characters...</LI>
The LANG and DIR attributes are introduced with the
internationalization proposal.
The internationalization proposal also includes
ALIGN=center|left|right|justify
The TYPE and VALUE attributes are Netscape extensions.
The validity of their presence and possible values depends on
the particular list entity of which this is a list item.
Most browsers assume the list item ends with the beginning
of the next list item or the end of the list and
do not require the closing tag </LI>
All list elements are Level 0.
The LINK element is used to indicate a relationship between
this document and other documents or objects.
Multiple LINK elements may exist in a document.
A LINK in document A with an HREF to document/object B with
REL=relation
identifies a relationship that B has to A that A recognizes/authorizes/verifies.
A LINK in document B with an HREF to document/object A with
REV=relation that is the identical relation
identifies a desired/expected/claimed relationship of B to document/object A,
but must be verified by checking with A.
For further description of the LINK attributes, see the
A element.
The LANG, DIR and CHARSET attributes are introduced with the
internationalization proposal.
The CHARSET attribute is a hint as to the expected
character set used by the hyperlink.
Some browsers will expect a LINK REV relationship of "made"
to be a "mailto:name@e-mail_address" to facilitate
sending comments to the person that made the document.
Contrary to the specification of the standard, this seems to only
work if "made" is the only relationship in that LINK element.
In addition to those mentioned with the A element,
Version 3 reserves some relations for REL that are expected to
be used for document specific toolbars. Currently these include
Home, ToC, Index, Glossary, Copyright, Up, Next,
Previous, Help, Bookmark.
Version 3 also discusses relations for a document banner
and style sheets.
LINK is a Level 0 element.
Sandia Requirements
A LINK element in the HEAD to identify the owner of
the document responsible for its accuracy is
required by Sandia's automated Web tools: <LINK REV="owns" TITLE="Full Name" HREF="mailto:owner-e-mail@sandia.gov">
LINK elements are also recommended to identify
other individuals and their relationships to this document.
The REV value of made should be used to identify the
author of the document.
Suggested values for REV in additional LINK elements are:
approves, editor, publisher.
Large documents which are separated into smaller HTML subdocuments
should use the LINK element with the REL attribute to identify these
relationships.
The parent document should identify all subdocuments by: <LINK REL="Subdocument" TITLE="Subdoc Name" HREF="link-url">
The subdocument may identify its parent by: <LINK REV="Subdocument" TITLE="Maindoc Name" HREF="link-url">
Any Sandia document which is part of
a set of HTML subdocuments which form a sequence or hierarchy should
include two specific LINK elements identifying the REL values of
next and previous.
Only one next and one previous relationship
may be specified in a document.
Any of the other document relationships mentioned in this Reference Manual
may also be used.
Most browsers use fixed-width characters for LISTING text.
This element existed in Version 0 and has been replaced by the
<PRE> element.
RFC 1866 states that the content of the LISTING element should be
rendered so that at least 132 characters fit on a line.
RFC 1866 has declared LISTING as deprecated
and some current browsers no longer recognize it.
The MAP element is used to name and describe a client-side
image map. This is a set of areas defined on an image which
can be selected for hyperlinks. NAME defines the map name to
be used with the USEMAP attribute on an IMG element.
The MARQUEE element is a MS Internet Explorer 2.0 enhancement.
Based on the description it might be assumed that only parsed
character data may be in the content of the MARQUEE element.
The ALIGN attribute defines the location of the text within the marquee.
The BEHAVIOR attribute determines whether the text will
scroll completely in and completely off (the default),
slide in and stay,
or bounce between alternate sides of the marquee.
The BGCOLOR attribute defines the background color for the marquee,
which is specified as an "#rrggbb" number, or
the following color names:
Black, Maroon, Green, Olive, Navy, Purple, Teal, Gray, Silver, Red, Lime,
Yellow, Blue, Fuchsia, Aqua, White.
The DIRECTION attribute specifies the direction towards which the
text should flow.
The HEIGHT attribute specifies the height of the marquee, either
in pixels or with a percentsign character suffix to define a
percentage of the entire screen height.
The HSPACE and VSPACE attributes are specified in pixels
and define the amount to separate the marquee from surrounding text.
The LOOP attribute specifies how many times the text will loop.
If LOOP=-1 or LOOP=INFINITE,
the text will loop indefinately.
The SCROLLAMOUNT specifies the number of pixels and
the SCROLLDELAY specifies the number of milliseconds
between each successive draw of the marquee text.
The MENU element defines an unordered list consisting of a number of
separated multi-line <LI> elements which
may or may not be marked by a bullet or similar symbol.
This list type is often rendered identically to UL.
RFC 1866 states that while similar to the UL element,
MENU is typically rendered in a more compact manner.
The COMPACT attribute of the UL element is more often used than
this MENU element.
The exclusion in RFC 1866 of group block within MENU implies
(among other things) that MENU can contain no nested lists,
nor any paragraphs even though the LI element normally would allow this.
Some browsers do not enforce this exclusion.
The LANG and DIR attributes are introduced with the
internationalization proposal.
The internationalization proposal also includes
ALIGN=center|left|right|justify
By adding the PLAIN attribute to eliminate the bullets to the
UL element, Version 3 proposes to remove the MENU element.
All list elements are Level 0.
The META element is used within the HEAD element
to embed document meta-information not defined
by other HTML elements.
Such information may be extracted by servers/browsers.
The HTTP-EQUIV attribute binds the element to an HTTP response header.
If not present, the NAME attribute should be used to identify this
meta-information and it should not be used within an HTTP response header.
If the NAME attribute is not present, the name can be assumed equal to
the value of HTTP-EQUIV.
The CONTENT attribute defines
the meta-information content to be associated with the given name
and/or HTTP response header.
The META element was introduced in RFC 1866 and is a
Level 1 element.
RFC 1866 specifies that a series of META elements with the same name
is equivalent to a single element with the combined contents concatenated
as a comma-separated list.
One proposal for specific information in the META element involves
including a
PICS label.
The URL attribute is a Netscape extension.
Netscape 1.1 has added a automatic refresh capability using the META element
by setting the HTTP-EQUIV attribute to "REFRESH",
the CONTENT attribute to a number of seconds, and the URL attribute
to the file to load which defaults to reloading the same file.
Netscape 1.1 also recognizes placing the URL inside the quotes
which define the CONTENT value
by using a semicolon following the number of seconds, then the
URL=http://... text.
Sandia Requirements
Two META elements are required to identify specific
information for Sandia's automated Web tools. <META NAME="REVIEW" CONTENT="DD MMM YYYY"> <META HTTP-EQUIV="KEYWORDS" CONTENT="key1, key2, key3">
For large documents split into multiple HTML subdocuments, these META
elements should only be included in the HTML document that is the parent
of all the subdocuments, usually the Introduction or Table of Contents document.
The single attribute N=
provides the next available identifier
for use by automatic hypertext editors.
If the NEXTID element is manually entered, it should be
alphabetical to avoid conflict with such editors.
Minimum Attributes
<NEXTID N="..."
>
All Possible Attributes
<NEXTID N="..."
>
Elements Allowed Within...
The NEXTID element is defined as having no content.
The NOEMBED element defines content within EMBED content that is
to be ignored by browsers that can activiate the EMBED plug-in
application. Browsers that can't/won't activate the EMBED plug-in
but that understand the EMBED/NOEMBED elements
or browsers that do not understand the EMBED/NOEMBED elements
will display the NOEMBED content.
The NOFRAMES element defines content within FRAMESET content that is
to be ignored by browsers that can define Frames.
Browsers that can't/won't define Frames
but that understand the FRAMESET/NOFRAMES elements
or browsers that do not understand the FRAMESET/NOFRAMES elements
will display the NOFRAMES content.
The NOFRAMES element is a part of the Netscape 2.0 enhancement to
define multiple Frames for viewing documents. It appears that Netscape 2.0
permits any element of group
body.content
in the content
of the NOFRAMES element.
The NOTE element changes the rendering of the
contents of the element to logically represent
separated notational text.
The SRC attribute specifies an image to appear preceding the note.
The NOTE element is a Version 3 proposed element.
It is expected to typically be rendered indented, without a preceding
bullet, symbol, or other graphic. Accompanying style guides are
expected to define renderings associated with specific CLASS values.
NOTE, CAUTION, and WARNING are expected values for the CLASS attribute.
Any element that permits members of the group
list
Variations
The LANG and DIR attributes are introduced with the
internationalization proposal.
The internationalization proposal also includes
ALIGN=center|left|right|justify
Version 3 proposes the CLEAR attribute.
The CONTINUE attribute is proposed by Version 3 to continue the
numbering from where the previous OL list left off.
The SEQNUM attribute is proposed by Version 3 to
define a starting number for the list.
START is a Netscape extension to do the same thing as SEQNUM.
The TYPE attribute is a Netscape extension.
Version 3 proposes using associated style sheets and the CLASS attribute
to handle these, and other, options to numbering style.
All list elements are Level 0.
The OPTION element identifies a choice in a SELECT element,
which in turn is part of the contents of a FORM element.
SELECTED specifies that this option is selected by default.
If SELECT allows MULTIPLE, then multiple options may be SELECTED.
If the VALUE attribute it not present and the OPTION is selected,
the OPTION contents is returned upon submission of the FORM.
If the VALUE attribute is present and the OPTION is selected,
the value of the VALUE attribute is returned instead of the contents.
The DISABLED attribute was part of an early Version 2 proposal,
but is not defined in RFC 1866.
It is defined in Version 3, and
would display this OPTION, but prohibit user selection/deselection.
RFC 1866 sets a max limit of 1024 characters on VALUE.
Version 3 proposes the ERROR and SHAPE attributes.
The Version 3 proposed ERROR attribute defines text to be displayed
in the event that selectiing this OPTION is invalid.
The Version 3 proposed SHAPE attribute defines the shape of the
"hotspot" on the imgage defined by the SRC attribute
of the surrounding SELECT element.
The proposed values of SHAPE are:
"default", "circle x,y,r",
"rect x,y,w,h", and
"polygon x1,y1,x2,y2,...".
All elements concerning FORM are Level 2.
The OVERLAY element is used to overlay one or more images
on top of a FIG image.
The X and Y attributes identify the offsets from the top left
of the FIG where the top left of the overlay will be placed.
The OVERLAY element is proposed in Version 3 as part of
the FIG construct.
It is designed to take advantage of the caching system of most browsers
for a series of similar images which may only differ by an overlay
and the rest is already in cache.
The P element is used to denote a paragraph break,
and separates two blocks of text.
Many other elements automatically imply a text separation,
such as headings, list elements, blockquotes, etc.
In Version 1, the <P> element was a separator
and the </P> element was not defined.
RFC 1866 changed the concept to a container and introduced the
optional </P> element.
The LANG and DIR attributes are introduced with the
internationalization proposal.
HTML+ introduced the ALIGN attribute
and is the only proposal to mention the indent value.
The internationalization proposal includes
ALIGN=center|left|right|justify
Netscape 2.0 implements ALIGN=center|left|right and
MS Internet Explorer 2.0 implements ALIGN=center|left|right|justify
Version 3 does not currently include indent as an option
for ALIGN.
The CLEAR attribute is proposed in Version 3 to deal with text
able to float around an image defined with the IMG element.
HTML+ introduced the ID attribute
to replace the NAME attribute
in the A element to establish internal hyperlink destinations.
With the expected use of ID for style sheets, this is likely to change.
HTML+ proposed WRAP to turn off automatic word wrap, making it possible
to leave text as it appears in the source.
Version 3 replaces the WRAP attribute and values with the NOWRAP attribute.
The P element is Level 0.
The PARAM element is a mechanism to define general purpose parameters
to be passed to APPLET applications. NAME is the name of the parameter
and VALUE will be obtained by the applet with the getParameter()
method.
Minimum Attributes
<PARAM NAME="..." VALUE="...">
All Possible Attributes
<PARAM NAME="..." VALUE="...">
Elements Allowed Within...
The PARAM element is defined as having no content.
The PERSON element
changes the character rendering of the
contents of the element to logically represent
the name of people
to allow these to be extracted automatically by indexing programs.
The PLAINTEXT element defined a separated multi-line set of text
to be rendered as it exists in the source document with
the same line breaks. It was designed to ignore all subsequent
HTML tags. Therefore, it was always last in a document, allowing
the remainder of the document to be presented as text.
Most browsers use fixed-width characters for PLAINTEXT text.
A few browsers accepted the ending element, which meant that
the ending element could not exist in the text.
The RFC 1866 DTD specifies that the ending element can exist and
implies that a browser should recognize the ending
element but accept everything else as just text.
It also prohibits it from being nested inside any element but the
outer HTML element itself.
Comments in RFC 1866 state that PLAINTEXT has no ending element and
all characters after the start element are data.
This element existed in Version 0 and has been replaced by the
<PRE> element.
RFC 1866 has declared PLAINTEXT as deprecated
and some current browsers no longer recognize it.
The LANG and DIR attributes are introduced with the
internationalization proposal.
The PRE element defines a separated multi-line set of text
to be rendered as it exists in the source document with
the same line breaks and spaces.
Normal text removes multiple spaces and line breaks in the source.
RFC 1866 states that the content of PRE is intended to be
formatted with a monospaced font.
For backwards compatibility with some existing documents,
RFC 1866 encourages browsers to accept the <P> element
in the PRE content and treat it as if it were the <BR> element.
While RFC 1866 states that within PRE content the tab character should
position to the next 8 character boundary, it recommends
that HTML documents not contain the tab character.
Version 3 encourages using the proposed TAB element instead.
The LANG and DIR attributes are introduced with the
internationalization proposal.
One typical use of PRE has been for tables, but the Version 3
TABLE element is proposed to fulfill that requirement.
The CLEAR attribute is proposed in Version 3.
RFC 1866 states that the attribute WIDTH specifies the
maximum number of characters for a line.
Some browsers have interpreted this as specifying
the minimum number of characters that the browser should
try to insure are displayed on a single line, possibly by selecting
a smaller font size.
Many browsers ignore the WIDTH attribute, and Version 3 is discussing
the possibility of removing it.
The PRE element is Level 0.
The Q element was a Version 3 proposed element and is part
of the internationalization proposal.
It is intended to be shown enclosed in the kind of quotations marks
appropriate to the LANG language context.
All character definition elements are Level 2.
The S element is proposed in Version 3.
This replaces the STRIKE element which was in a early Version 2 proposal
but is not in RFC 1866.
All character definition elements are Level 2.
Any element that permits members of the group
phrase
Variations
This is usually rendered with a fixed-width font.
The LANG and DIR attributes are introduced with the
internationalization proposal.
All character definition elements are Level 2.
The SELECT element defines a menu of a series of selectable
input in a FORM, each identified with the OPTION element.
At least one OPTION element is expected within the SELECT contents.
NAME is the symbolic name of the field
returned to the server on submission.
SIZE determines the number of OPTIONS physically visible
when the browser displays the FORM.
The default selection will be the first OPTION in the SELECT contents.
The MULTIPLE attribute, if present, allows multiple selections, and
causes no OPTION to be the default.
element <OPTION>
Note that the elements <INPUT><SELECT><TEXTAREA> are expressly prohibited by RFC 1866 from being nested
within the SELECT element content
The ERROR attribute without a value was first introduced
in an early Version 2 proposal but is not in RFC 1866.
The LANG and DIR attributes are introduced with the
internationalization proposal.
Version 3 proposes the ALIGN, DISABLED,
ERROR (with a value), SRC, MD, WIDTH, HEIGHT, and UNITS attributes.
The Version 3 proposed DISABLED attribute would display this menu,
but prohibit user entry/modification.
The Version 3 proposed ERROR attribute defines text to be displayed
in the event that the current selections for this menu are invalid.
Version 3 proposes a SRC attribute to identify a URL for an image
to be displayed instead of the OPTION texts for this menu,
with each OPTION element identifying its "hotspot" on this image.
If SRC is present, the texts would only be used for non-graphical browsers.
All elements concerning FORM are Level 2.
The STRIKE element changes the physical rendering of the
contents of the element to a font
with a strikeout line through the letters.
Minimum Attributes
<STRIKE>characters...
</STRIKE>
All Possible Attributes
<STRIKE>characters...
</STRIKE>
Elements Allowed Within...
-In progress, to be determined-
Allowed In Content Of...
-In progress, to be determined-
Variations
This was an early Version 2 proposed element but is not in RFC 1866.
(ed: This appears to be an undocumented extension which
works in Netscape 1.1. Anyone try this with other browsers?)
This is replaced in Version 3 with the <S> element.
This is usually rendered as bold.
RFC 1866 states that <EM> content must be rendered as distinct
from <STRONG> content.
<STRONG> is intended to be stronger than <EM>.
The LANG and DIR attributes are introduced with the
internationalization proposal.
All character definition elements are Level 2.
The SUB element was in an early Version 2 proposal and is part
of the internationalization proposal, but is not in RFC 1866.
No font size change was implied in Version 2 nor is any specified
as part of the internationalization proposal since it may not
make sense in non-English languages.
The internationalization proposal explicitly restricts the content
to parsed character data to prevent nesting of SUB and SUP.
This element is a Netscape 2.0 extension which renders the text
in a slightly smaller font.
All character definition elements are Level 2.
The SUP element was in an early Version 2 proposal and is part
of the internationalization proposal, but is not in RFC 1866.
No font size change was implied in Version 2 nor is any specified
as part of the internationalization proposal since it may not
make sense in non-English languages.
The internationalization proposal explicitly restricts the content
to parsed character data to prevent nesting of SUB and SUP.
This element is a Netscape 2.0 extension which renders the text
in a slightly smaller font.
All character definition elements are Level 2.
The TAB element aligns the following text according to
a defined horizontal position.
A TAB position can be defined by the ID attribute.
Text is positioned using the TO and/or ALIGN attributes, or
the INDENT attribute.
The TABLE element defines a series of rows of table cell elements.
The contents of the TABLE element
contains a sequence of elements which describe various parts of the table.
The order in the sequence is important and consists of:
at most one CAPTION element,
possibly some COL or
possibly some COLGROUP elements,
at most one THEAD element,
at most one TFOOT element,
and finally at least one TBODY element
or at least one TR element.
The TABLE element is part of a comprehensive new proposed standard
tables structure. It is partially implemented in some browsers.
The proposal has undergone numerous changes.
Some broswers implemented features in early versions of the proposal
which did not survive in the final proposal.
Consistent with the nature of HTML, the proposal defines the logical contents
of cells and their row/column location, rather than a physical layout
which is left as a function of the browser.
Details of the appearence of the table,
along with the size of margins around cells, even perhaps ruling and shading,
might be controlled by associated style sheets, proposed in Version 3.
The current proposal expects text to flow around a table, if possible.
It suggests using the BR element with a CLEAR attribute after
the table element if the author wishes to avoid text flowing along side
the table. Although the CLEAR attribute is not currently a standardized
part of HTML, the tables proposal does not include text
to standardize the attribute's existence.
The ALIGN attribute specifies the table position relative to the window.
The WIDTH attribute specifies the entire table width, and defaults
to 100%.
The default units is pixels, but may be specified by a suffix on the number.
The standard defined units suffixes include:
pt=points,
pi=picas,
in=inches,
cm=centimeters,
mm=millimeters,
em=em units,
px=screen pixels.
In addition the suffix of a percentsign may be used to specify the
percentage width of the space between the current left and right margins.
The COLS attribute specifies the total number of columns in the table and
aids the browser in determining the initial display and layout of the table.
The BORDER attribute controls frame width around the table and FRAME defines
which parts of the frame to include.
For BORDER, the default units is pixels,
but may be specified by one of the standard defined units suffixes.
BORDER=0 implies FRAME=none, and BORDER without a value implies FRAME=border.
The RULES attribute defines the presence or absence of rulings
between all rows and columns in this table.
The CELLSPACING attribute defines spacing between cells and CELLPADDING
defines spacing within cells.
The TABLE element has been partially implemented in Netscape 1.1 and
MS Internet Explorer 2.0.
The COL, COLGROUP, THEAD, and TFOOT elements are new and not yet implemented.
The FRAME and RULES attributes are new and not yet implemented.
The values justify|bleedleft|bleedright for the ALIGN attribute
are Netscape 1.1 extensions.
The attribute COLSPEC was from an earlier version of the proposal
and is replaced by the COL and COLGROUP elements.
The attribute DP is from an earlier version of the proposal and
was used with a value of decimal on the ALIGN attribute.
It has been replaced with the CHAR and CHAROFF attributes
used when ALIGN=char which are
defined for all table grouping elements except TABLE.
The attributes CLEAR, NOFLOW, UNITS, and NOWRAP are Netscape 1.1 extensions.
Netscape 1.1 defines WIDTH as having a value of either a number of pixels
or a number followed by a percentsign to define a desired width of the table
which may be overridden by the browser.
It defines percentsign to specify the percentage of the entire window,
instead of the current margins.
The TBODY element encloses a series of table row definitions and
specifies the defaults for all the rows in this group.
Table rows are grouped inside the TABLE
content into at most one THEAD,
at most one TFOOT,
and at least one TBODY sections, in that order.
Depending on the capabilities of the browser, this sectioning can be used to
repeat table head and foot rows when breaking tables across page
boundaries, or to provide fixed headers above and footers below
a scrollable body panel, or other similar rendering schemes.
The TBODY element is a part of the latest table proposal.
This TBODY element is new and not widely implemented.
Both the starting and ending elements are optional for this element.
This implies that a TABLE with only one TBODY does not require
the TBODY element since the first TR element not inside a THEAD
or TFOOT but inside a TABLE will imply the TBODY content.
This also allows this proposal to be backwards compatible with
existing documents which have a simple TABLE element whose content
consists of only a series of row and data elements.
The table proposal states that a browser may display
ALIGN=justify the same as the default
ALIGN=left if it lacks support for text justification.
The CHAR and CHAROFF attributes are used when ALIGN=char.
CHAR specifies a single character to be used for alignment, and
defaults to the decimal point in the current language.
CHAROFF specifies the offset, in the DIR direction, to the first
occurrence of the CHAR character on each line. If a line doesn't
include the CHAR character, it is to end at the CHAROFF position.
The default value for CHAROFF is either 50% or whatever will cause
the CHAR characters in all cells with a declared CHAR in a column to align.
The default units is pixels, but may be specified by a suffix on the number.
The standard defined units suffixes include:
pt=points,
pi=picas,
in=inches,
cm=centimeters,
mm=millimeters,
em=em units,
px=screen pixels.
In addition the suffix of a percentsign may be used to specify the
percentage offset within the cell.
The VALIGN attribute defines the vertical alignment of the cell contents.
The ALIGN, CHAR and CHAROFF attribute values may be inherited from
other table elements in the following order:
cells < columns < column groups < rows < row groups < default.
The VALIGN, LANG and DIR attribute values may be inherited from
other table elements in the following order:
cells < rows < row groups < columns < column groups < table < default.
The TD element defines a data cell as part of the TABLE construct.
Both TH and TD elements can exist in any
one of the three table sections: THEAD, TFOOT, or TBODY.
The TD element is part of the proposed TABLE construct.
A browser is to assume the table cell ends with the beginning
of the next table cell or the beginning of another table row
or the end of the enclosing TABLE, THEAD, TFOOT, or TBODY element
and thus a closing tag is not required.
This element is an extension in Netscape 1.1 and MS Internet Explorer 2.0.
The values justify|char for the ALIGN attribute
are new and are not widely implemented.
The table proposal states that a browser may display
ALIGN=justify the same as the default
ALIGN=left if it lacks support for text justification.
The AXIS and AXES attributes for cells provide a means for defining
concise labels for cells, either for speech rendering, or for
database field names.
The attribute AXIS defaults to the cell contents.
The AXES is a comma separated list of axis names which together
identify the row and column headers that pertain to this cell.
The NOWRAP attribute will suppress word wrap in the cell.
The ROWSPAN and COLSPAN attributes define the integer number of rows and
columns spanned by the cell. A value of zero for either implies the
cell spans from this cell to the end of the table.
The attribute DP is from an earlier version of the proposal and
was used with a value of decimal on the ALIGN attribute.
It has been replaced with the CHAR and CHAROFF attributes
used when ALIGN=char.
CHAR specifies a single character to be used for alignment, and
defaults to the decimal point in the current language.
CHAROFF specifies the offset, in the DIR direction, to the first
occurrence of the CHAR character on each line. If a line doesn't
include the CHAR character, it is to end at the CHAROFF position.
The default value for CHAROFF is either 50% or whatever will cause
the CHAR characters in all cells with a declared CHAR in a column to align.
The default units is pixels, but may be specified by a suffix on the number.
The standard defined units suffixes include:
pt=points,
pi=picas,
in=inches,
cm=centimeters,
mm=millimeters,
em=em units,
px=screen pixels.
In addition the suffix of a percentsign may be used to specify the
percentage offset within the cell.
The VALIGN attribute defines the vertical alignment of the cell contents.
The ALIGN, CHAR and CHAROFF attribute values may be inherited from
other table elements in the following order:
cells < columns < column groups < rows < row groups < default.
The VALIGN, LANG and DIR attribute values may be inherited from
other table elements in the following order:
cells < rows < row groups < columns < column groups < table < default.
The WIDTH attribute is a Netscape extension.
The BGCOLOR attribute is a MS Internet Explorer 2.0 extension.
Internet Explorer accepts either an "#rrggbb" number or
the following color names:
Black, Maroon, Green, Olive, Navy, Purple, Teal, Gray, Silver, Red, Lime,
Yellow, Blue, Fuchsia, Aqua, White.
The TEXTAREA element is used to specify a multiline input field
as part of the contents in a FORM element.
NAME defines the symbolic name of the field
returned to the server on submission.
The characters between the opening and closing tags
define the text to be initially displayed in the textarea,
and the default text to be returned, normally null.
Only ASCII text is allowed as characters, and newlines are respected.
ROWS and COLS
define the physical size of the
displayed input field in numbers of characters.
Only parsed character data is expected in the content of the
TEXTAREA element. Note that the elements
<INPUT><SELECT><TEXTAREA> are expressly prohibited by RFC 1866 from being nested
within the TEXTAREA element content
The LANG and DIR attributes are introduced with the
internationalization proposal.
The WRAP attribute is a Netscape 2.0 extension and controls
word wrapping inside the TEXTAREA.
The value off is default and lines are sent exactly as typed.
The value virtual wraps in the display but are sent
exactly as typed.
The value physical wraps in the display and sends new-lines
at the wrap points as if new-lines had been entered.
The early Version 3 proposal added the
ALIGN, DISABLED, and ERROR attributes.
The Version 3 proposed DISABLED attribute would display this TEXTAREA,
but prohibit user entry/modification.
The Version 3 proposed ERROR attribute defines text to be displayed
in the event that the entered value for this TEXTAREA is invalid.
All elements concerning FORM are Level 2.
The TFOOT element encloses a series of table row definitions
and specifies the defaults for all the rows in this group.
Table rows are grouped inside the TABLE
content into at most one THEAD,
at most one TFOOT,
and at least one TBODY sections, in that order.
Depending on the capabilities of the browser, this sectioning can be used to
repeat table head and foot rows when breaking tables across page
boundaries, or to provide fixed headers above and footers below
a scrollable body panel, or other similar rendering schemes.
The TFOOT element is a part of the latest table proposal.
This TFOOT element is new and not widely implemented.
The ending element is optional for this element and may be implied
when the TBODY element is encountered.
The table proposal states that a browser may display
ALIGN=justify the same as the default
ALIGN=left if it lacks support for text justification.
The CHAR and CHAROFF attributes are used when ALIGN=char.
CHAR specifies a single character to be used for alignment, and
defaults to the decimal point in the current language.
CHAROFF specifies the offset, in the DIR direction, to the first
occurrence of the CHAR character on each line. If a line doesn't
include the CHAR character, it is to end at the CHAROFF position.
The default value for CHAROFF is either 50% or whatever will cause
the CHAR characters in all cells with a declared CHAR in a column to align.
The default units is pixels, but may be specified by a suffix on the number.
The standard defined units suffixes include:
pt=points,
pi=picas,
in=inches,
cm=centimeters,
mm=millimeters,
em=em units,
px=screen pixels.
In addition the suffix of a percentsign may be used to specify the
percentage offset within the cell.
The VALIGN attribute defines the vertical alignment of the cell contents.
The ALIGN, CHAR and CHAROFF attribute values may be inherited from
other table elements in the following order:
cells < columns < column groups < rows < row groups < default.
The VALIGN, LANG and DIR attribute values may be inherited from
other table elements in the following order:
cells < rows < row groups < columns < column groups < table < default.
The TH element defines a header cell as part of the TABLE construct.
The browser should render the table to cause some difference between
TH elements and TD elements.
Both TH and TD elements can exist in any
one of the three table sections: THEAD, TFOOT, or TBODY.
The TH element is part of the proposed TABLE construct.
A browser is to assume the table cell ends with the beginning
of the next table cell or the beginning of another table row
or the end of the enclosing TABLE, THEAD, TFOOT, or TBODY element
and thus a closing tag is not required.
This element is an extension in Netscape 1.1 and MS Internet Explorer 2.0.
Most browsers typically render the TH element in bold font,
but otherwise identical to TD elements.
The values justify|char for the ALIGN attribute
are new and are not widely implemented.
The table proposal states that a browser may display
ALIGN=justify the same as the default
ALIGN=left if it lacks support for text justification.
The AXIS and AXES attributes for cells provide a means for defining
concise labels for cells, either for speech rendering, or for
database field names.
The attribute AXIS defaults to the cell contents.
The AXES is a comma separated list of axis names which together
identify the row and column headers that pertain to this cell.
The NOWRAP attribute will suppress word wrap in the cell.
The ROWSPAN and COLSPAN attributes define the integer number of rows and
columns spanned by the cell. A value of zero for either implies the
cell spans from this cell to the end of the table.
The attribute DP is from an earlier version of the proposal and
was used with a value of decimal on the ALIGN attribute.
It has been replaced with the CHAR and CHAROFF attributes
used when ALIGN=char.
CHAR specifies a single character to be used for alignment, and
defaults to the decimal point in the current language.
CHAROFF specifies the offset, in the DIR direction, to the first
occurrence of the CHAR character on each line. If a line doesn't
include the CHAR character, it is to end at the CHAROFF position.
The default value for CHAROFF is either 50% or whatever will cause
the CHAR characters in all cells with a declared CHAR in a column to align.
The default units is pixels, but may be specified by a suffix on the number.
The standard defined units suffixes include:
pt=points,
pi=picas,
in=inches,
cm=centimeters,
mm=millimeters,
em=em units,
px=screen pixels.
In addition the suffix of a percentsign may be used to specify the
percentage offset within the cell.
The VALIGN attribute defines the vertical alignment of the cell contents.
The ALIGN, CHAR and CHAROFF attribute values may be inherited from
other table elements in the following order:
cells < columns < column groups < rows < row groups < default.
The VALIGN, LANG and DIR attribute values may be inherited from
other table elements in the following order:
cells < rows < row groups < columns < column groups < table < default.
The WIDTH attribute is a Netscape extension.
The BGCOLOR attribute is a MS Internet Explorer 2.0 extension.
Internet Explorer accepts either an "#rrggbb" number or
the following color names:
Black, Maroon, Green, Olive, Navy, Purple, Teal, Gray, Silver, Red, Lime,
Yellow, Blue, Fuchsia, Aqua, White.
The THEAD element encloses a series of table row definitions
and specifies the defaults for all the rows in this group.
Table rows are grouped inside the TABLE
content into at most one THEAD,
at most one TFOOT,
and at least one TBODY sections, in that order.
Depending on the capabilities of the browser, this sectioning can be used to
repeat table head and foot rows when breaking tables across page
boundaries, or to provide fixed headers above and footers below
a scrollable body panel, or other similar rendering schemes.
The THEAD element is a part of the latest table proposal.
This THEAD element is new and not widely implemented.
The ending element is optional for this element and may be implied
when a TFOOT or the TBODY element is encountered.
The table proposal states that a browser may display
ALIGN=justify the same as the default
ALIGN=left if it lacks support for text justification.
The CHAR and CHAROFF attributes are used when ALIGN=char.
CHAR specifies a single character to be used for alignment, and
defaults to the decimal point in the current language.
CHAROFF specifies the offset, in the DIR direction, to the first
occurrence of the CHAR character on each line. If a line doesn't
include the CHAR character, it is to end at the CHAROFF position.
The default value for CHAROFF is either 50% or whatever will cause
the CHAR characters in all cells with a declared CHAR in a column to align.
The default units is pixels, but may be specified by a suffix on the number.
The standard defined units suffixes include:
pt=points,
pi=picas,
in=inches,
cm=centimeters,
mm=millimeters,
em=em units,
px=screen pixels.
In addition the suffix of a percentsign may be used to specify the
percentage offset within the cell.
The VALIGN attribute defines the vertical alignment of the cell contents.
The ALIGN, CHAR and CHAROFF attribute values may be inherited from
other table elements in the following order:
cells < columns < column groups < rows < row groups < default.
The VALIGN, LANG and DIR attribute values may be inherited from
other table elements in the following order:
cells < rows < row groups < columns < column groups < table < default.
The TITLE is not part of the document text.
Often the title is used by a browser to label the display window.
Some Web search engines only search the title of Web pages.
Therefore the text of the TITLE should be kept short
but sufficient to identify the document.
Only parsed character data is expected in the content of the
TITLE element. Note that the elements
<META><LINK> are expressly prohibited by RFC 1866 from being nested
within the TITLE element content
RFC 1866 specifies that all HTML documents must contain a TITLE element.
Some browsers will truncate titles to a length they can handle.
RFC 1866 recommends that titles be fewer than 64 characters.
The LANG and DIR attributes are introduced with the
internationalization proposal.
TITLE is a Level 0 element.
Sandia Requirements
The TITLE element is required to include a short, uniquely
descriptive document title.
The TR element defines a table row in one of the three sections
of the TABLE construct: THEAD, TFOOT, or TBODY.
The contents of the TR element is required to contain at least one
of either the TH element or the TD element.
The TR element is part of the proposed TABLE construct.
A browser is to assume the table row ends with the beginning
of the next table row, the next table row group, or
the end of the table and thus a closing tag is not required.
An early Version 3 proposal recommended that browsers should
tolerate the absence of the first TR in a table,
but this option was not included in the formal table proposal.
This element is an extension in Netscape 1.1 and MS Internet Explorer 2.0.
The values justify|char for the ALIGN attribute
are new and are not widely implemented.
The table proposal states that a browser may display
ALIGN=justify the same as the default
ALIGN=left if it lacks support for text justification.
The attribute DP is from an earlier version of the proposal and
was used with a value of decimal on the ALIGN attribute.
It has been replaced with the CHAR and CHAROFF attributes
used when ALIGN=char.
CHAR specifies a single character to be used for alignment, and
defaults to the decimal point in the current language.
CHAROFF specifies the offset, in the DIR direction, to the first
occurrence of the CHAR character on each line. If a line doesn't
include the CHAR character, it is to end at the CHAROFF position.
The default value for CHAROFF is either 50% or whatever will cause
the CHAR characters in all cells with a declared CHAR in a column to align.
The default units is pixels, but may be specified by a suffix on the number.
The standard defined units suffixes include:
pt=points,
pi=picas,
in=inches,
cm=centimeters,
mm=millimeters,
em=em units,
px=screen pixels.
In addition the suffix of a percentsign may be used to specify the
percentage offset within the cell.
The VALIGN attribute defines the vertical alignment of the cell contents.
The ALIGN, CHAR and CHAROFF attribute values may be inherited from
other table elements in the following order:
cells < columns < column groups < rows < row groups < default.
The VALIGN, LANG and DIR attribute values may be inherited from
other table elements in the following order:
cells < rows < row groups < columns < column groups < table < default.
Browsers who do not have teletype font may render in some other manner.
The LANG and DIR attributes are introduced with the
internationalization proposal.
All character definition elements are Level 2.
The U element was in an early Version 2 proposal
but is not part of RFC 1866.
It is often rendered in italic or slanted rather than underlined.
All character definition elements are Level 2.
The UL element defines an unordered list consisting of a number of
separated multi-line <LI> elements, and usually
marked by a bullet or similar symbol.
Any element that permits members of the group
list
Variations
Nested UL lists are usually rendered more indented with
differing bullets or symbols.
The LANG and DIR attributes are introduced with the
internationalization proposal.
The internationalization proposal also includes
ALIGN=center|left|right|justify
Version 3 proposes the CLEAR, PLAIN,
SRC, MD, DINGBAT, and WRAP attributes.
The TYPE attribute is a current Netscape extension.
Version 3 proposes to replace it with the SRC and DINGBAT attributes
to point to general server and browser images respectively.
Specifying <UL PLAIN> to eliminate the bullets
is proposed by Version 3 as a replacement for the
MENU element.
Specifying <UL PLAIN WRAP=HORIZ> is proposed
by Version 3 as a replacement for the DIR list element.
The browser may dynamically determine the appropriate number of columns.
All list elements are Level 0.
Any element that permits members of the group
phrase
Variations
This is usually rendered in italics.
The LANG and DIR attributes are introduced with the
internationalization proposal.
All character definition elements are Level 2.
Most browsers use fixed-width characters for XMP text.
This element existed in Version 0 and has been replaced by the
<PRE> element.
RFC 1866 states that the content of the XMP element should be
rendered so that at least 80 characters fit on a line.
RFC 1866 has declared XMP as deprecated
and some current browsers no longer recognize it.
The LANG and DIR attributes are introduced with the
internationalization proposal.