HTML5

W

William Gill

This may have already been discussed, or I may be just too thick to
figure it out myself, but since HTML5 is in "Working Draft" status and
HTML4 is at "Recommendation", are any of you doing any production work
in HTML5 yet, and when should I be making the transition?
 
C

C A Upsdell

This may have already been discussed, or I may be just too thick to
figure it out myself, but since HTML5 is in "Working Draft" status and
HTML4 is at "Recommendation", are any of you doing any production work
in HTML5 yet, and when should I be making the transition?

How many of your visitors have browsers which don't support the HTML 5
features that you want to use? And are you willing to exclude that many
visitors from your production site?
 
W

William Gill

How many of your visitors have browsers which don't support the HTML 5
features that you want to use? And are you willing to exclude that many
visitors from your production site?
Duh!

Good point, and good migration index.

Thanks.
 
C

cwdjrxyz

This may have already been discussed, or I may be just too thick to
figure it out myself, but since HTML5 is in "Working Draft" status and
HTML4 is at "Recommendation", are any of you doing any production work
in HTML5 yet, and when should I be making the transition?

First you may use any version of html down to html 2 if it meets your
needs. The w3c still has validators for older versions of html, and in
general newer browsers support earlier versions of html - they just
pile on more code as new versions come out. Unfortunately, as new
versions arrive, they will let older code through even if you use the
correct Doctype and even if the w3c validator shows errors. Some of
the computer companies are pushing html5, but it appears to be mainly
for $ and other interests of the companies, often in the media area.
Google is now using html5 on their home pages. Unfortunately, after
beating the drum for html5 and jumping the gun before html5 is even a
recommendation, they make a huge number of html5 validation errors on
their home pages as revealed at the experimental html5 checker at the
w3c. Apple and Firefox have also jumped on the html5 bandwagon on many
of their home pages. Apple has just a few validation errors, but the
Firefox home pages in html5 usually validate. Many features of html5
are not supported yet by many browsers in very common use. Consider
Microsoft's IE browsers. The IE9-beta is the first IE browser to
support the html5 video and audio elements. Only Vista and Windows 7
OSs can be updated to IE9-beta, and there are still many XP OSs being
used, especially in business. Many are likely to balk at upgrading to
Windows 7 just so they can use IE9 beta. The top Windows 7 OS costs
over $US 300, and the less fancy versions are far from cheap. You
usually can get around this problem concerning audio and video in a
html5 page by using somewhat modified conventional media code that
requires players installed on the computer and not a "house" html5
support built into the browser. In short, if you use html5 now for
writing pages, you need to know html5 and how it is supported on
various browsers very well and check your pages on many browsers still
used. This is a somewhat moving target since changes are still being
made in html5. I have written some media test pages in html5, and find
little reason to do so. The only advantage I have found for video is
that some such as Firefox and Opera, are using Ogg (.ogv) video as a
"house" format. For example, you can use ogv on a html 4.01 strict
page, but many will not have an Ogg player installed, and thus will
not be abe to view it.
 
L

Lewis

In message said:
This may have already been discussed, or I may be just too thick to
figure it out myself, but since HTML5 is in "Working Draft" status and
HTML4 is at "Recommendation", are any of you doing any production work
in HTML5 yet, and when should I be making the transition?

Yes, for certain values of HTML5.

The really sticking point is CSS3 (especially since CSS2.1 is still not
completely supported in most browsers).
 
D

dorayme

Lewis said:
Yes, for certain values of HTML5.

The really sticking point is CSS3 (especially since CSS2.1 is still not
completely supported in most browsers).

So, the strategy for clever but not reckless authors is to use
CSS3 (at least the bits that are widely supported) but only where
it degrades well if it does not work.
 
J

Jukka K. Korpela

dorayme said:
So, the strategy for clever but not reckless authors is to use
CSS3 (at least the bits that are widely supported) but only where
it degrades well if it does not work.

I would say that the statement applies even if you omit the digit "3".
 
G

Gus Richter

This may have already been discussed, or I may be just too thick to
figure it out myself, but since HTML5 is in "Working Draft" status and
HTML4 is at "Recommendation", are any of you doing any production work
in HTML5 yet, and when should I be making the transition?


It is true that HTML5 is only in "draft" state, just as CSS2.1 and CSS3
are. HTML5 and is the latest defacto standard, just as CSS2.1 and CSS3
are. They are all usable and are being used as well by more and more web
sites. HTML5 is incomplete as far as the new elements are concerned
(specification and browser support). CSS 2.1 and CSS 3 also are not
fully supported by all browsers.

http://caniuse.com/

is an excellent source to check on browsers' support (including mobile)
for HTML5, CSS 2.1, CSS3, etc.

For HTML5 use this doctype: <!DOCTYPE html> All browsers (that
sniff doctype) support it, from IE5/Mac up. You can use the same
HTML4.01 elements you used before because all HTML4.01 elements are
still supported. The bonus is that you can also use the new elements
available in HTML5, in as much as they are supported by all browsers,
which you will have to keep track of until all settles down. There is
nothing new about this - we've been doing this all along. The HTML5
doctype is a reduction of the HTML4.01 Strict doctype down to the
minimum required portions to satisfy legacy doctype sniffing (removes
all the fluff). It also places the browsers into Standards Mode. There
is only the one in HTML5. Nothing new here as well - we've been told to
use Strict doctype for all new documents for quite some time now.

You may use the same markup as for HTML 4.01 Strict if you prefer, or
pick up shortcuts that HTML5 offers. If you want to use some of the new
HTML5 elements, which it will support, just realize that some of them
are not yet supported by all browsers.
 
J

Jukka K. Korpela

Gus said:
It is true that HTML5 is only in "draft" state, just as CSS2.1 and
CSS3 are.

No it isn't. Not "just as".

CSS 2.1 is a working draft that has remained fairly stable for several
years, it is written in the form of a specification to the extent that W3C
deliverables ever are, it is actively promoted by the W3C, and it is
generally cited as the "de facto standard" for CSS (contrary to what it says
about itself, at least nominally).

CSS 3 is a collection of sketchy documents of very varying maturity and
activity around them. They even contain loads of "what should we do about
this?" statements, indicating that they are "working documents" much more
than draft or proposed recommendations are.

HTML5 is vague name for activities both at the W3C and at the WHATWG, and
everyone has his own idea of what topics belong to HTML5 - some people even
count CSS 3 as part of HTML5! The oxymoron "living standard", used by the
WHATWG, is revealing: to them, HTML5 is whatever happens to be in the WHATWG
document today.
HTML5 and is the latest defacto standard, just as CSS2.1
and CSS3 are.

Do I need to repeat the explanation of why that characterization is utterly
wrong?
They are all usable

No doubt about that, though they should only be used by people who
understand the implications and complications. Well, this applies to any
technology, but still.
HTML5 is incomplete as far as the new elements
are concerned (specification and browser support).

HTML5 is largely still just talk about what should be included. Of course,
as the topic is technical, the talk needs to be technical and often rather
detailed.
CSS 2.1 and CSS 3
also are not fully supported by all browsers.

Neither is CSS 4. You cannot even _define_ support when there is no
semi-stable draft for a semi-specification. It would be much more adequate
to say that CSS 2.1 is reasonably well supported by newest versions of
popular browsers when used in "standards" mode (but there are still
essential problems, especially due to use of pre-IE8 versions of IE),
whereas those parts of CSS 3 that have been reasonably well sketched-out and
stablish have _some_ support (usually partial, excluding the difficult
parts) in _some_ modern browsers.
The bonus is that you can also use the new elements
available in HTML5, in as much as they are supported by all browsers,

Which means pretty much none. Well, <wbr> might be counted in, and <nobr>
but oops, that one is "obsolete".

The new elements you can use are those that require no special support, like
<nav>, and those that can be reasonably well simulated in or taught to IE6
through IE8. For example, if you wish to style a <nav>, you should say
document.createElement("nav")
in a script, called before the first CSS reference, to teach that element
(just as an existing element, no features) to CSS. And if you wish to use
If you want to use some of the
new HTML5 elements, which it will support, just realize that some of
them are not yet supported by all browsers.

Get real. Most of them are not supported by any browser or are supported by
just one or two minority browser. There's still a lot in HTML5 that can be
used even today, and there's a lot more to be learned and experimented for
future use. But the situation is very different from that of HTML 4 - about
_it_ you could say that some of its features are not yet supported by all
browsers.
 
D

dorayme

"Jukka K. Korpela said:
I would say that the statement applies even if you omit the digit "3".

If a visitor has CSS off, the result should be a usable page, not
a dog's breakfast, agreed.

I just meant that among the visitors who have CSS, a few dropped
CSS3 styles might not take them all the way back to completely
basic styling provided by browser (though, it is a wonder to
behold how improved some sites are on doing just this).
 
G

Gus Richter

No it isn't. Not "just as".

CSS 2.1 is a working draft that has remained fairly stable for several
years, it is written in the form of a specification to the extent that
W3C deliverables ever are, it is actively promoted by the W3C, and it is
generally cited as the "de facto standard" for CSS (contrary to what it
says about itself, at least nominally).

CSS 3 is a collection of sketchy documents of very varying maturity and
activity around them. They even contain loads of "what should we do
about this?" statements, indicating that they are "working documents"
much more than draft or proposed recommendations are.

HTML5 is vague name for activities both at the W3C and at the WHATWG,
and everyone has his own idea of what topics belong to HTML5 - some
people even count CSS 3 as part of HTML5! The oxymoron "living
standard", used by the WHATWG, is revealing: to them, HTML5 is whatever
happens to be in the WHATWG document today.


Do I need to repeat the explanation of why that characterization is
utterly wrong?


No doubt about that, though they should only be used by people who
understand the implications and complications. Well, this applies to any
technology, but still.


HTML5 is largely still just talk about what should be included. Of
course, as the topic is technical, the talk needs to be technical and
often rather detailed.


Neither is CSS 4. You cannot even _define_ support when there is no
semi-stable draft for a semi-specification. It would be much more
adequate to say that CSS 2.1 is reasonably well supported by newest
versions of popular browsers when used in "standards" mode (but there
are still essential problems, especially due to use of pre-IE8 versions
of IE), whereas those parts of CSS 3 that have been reasonably well
sketched-out and stablish have _some_ support (usually partial,
excluding the difficult parts) in _some_ modern browsers.


Which means pretty much none. Well, <wbr> might be counted in, and
<nobr> but oops, that one is "obsolete".

The new elements you can use are those that require no special support,
like <nav>, and those that can be reasonably well simulated in or taught
to IE6 through IE8. For example, if you wish to style a <nav>, you
should say
document.createElement("nav")
in a script, called before the first CSS reference, to teach that
element (just as an existing element, no features) to CSS. And if you


Get real. Most of them are not supported by any browser or are supported
by just one or two minority browser. There's still a lot in HTML5 that
can be used even today, and there's a lot more to be learned and
experimented for future use. But the situation is very different from
that of HTML 4 - about _it_ you could say that some of its features are
not yet supported by all browsers.


I'm reminded of all the criticism and naysayers when CSS was
recommended. Perhaps we could agree on this:

1. All new documents should be created to place browsers into Standards
Mode. If preferring HTML 4.01, use the Strict Doctype.

2. The HTML5 Doctype is a minimized HTML 4.01 Strict Doctype to the
point whereby all doctype sniffing browsers will still trigger Standards
Mode. Use <!DOCTYPE html> as the Doctype with only HTML 4.01 vintage
elements and you achieve the same result as 1. above.

3. When using the HTML5 Doctype you are not obligated to use the new
elements, but you could, when you and/or the browser(s) are ready.

4. The WHATWG initiated the HTML5 project and was founded by Apple,
Mozilla and Opera. Google and Microsoft are also on side. They are
working on supporting HTML5. It's a transition period similar to when
CSS was introduced. It took a while for browsers to support the CSS
Properties and for developers to start using them. I recall a lot of
doom and gloom predictions in those days.
 
C

cwdjrxyz

I'm reminded of all the criticism and naysayers when CSS was
recommended. Perhaps we could agree on this:

1. All new documents should be created to place browsers into Standards
Mode. If preferring HTML 4.01, use the Strict Doctype.

2. The HTML5 Doctype is a minimized HTML 4.01 Strict Doctype to the
point whereby all doctype sniffing browsers will still trigger Standards
Mode. Use  <!DOCTYPE html>  as the Doctype with only HTML 4.01 vintage
elements and you achieve the same result as 1. above.

3. When using the HTML5 Doctype you are not obligated to use the new
elements, but you could, when you and/or the browser(s) are ready.

4. The WHATWG initiated the HTML5 project and was founded by Apple,
Mozilla and Opera. Google and Microsoft are also on side. They are
working on supporting HTML5. It's a transition period similar to when
CSS was introduced. It took a while for browsers to support the CSS
Properties and for developers to start using them. I recall a lot of
doom and gloom predictions in those days.

Microsoft, as of Feb. 10, has moved IE9-beta up to to a Release
Candidate which is now available for download at Microsoft.(Is a
release candidate a purgatory between a beta and plain release? :).
IE9 is greatly different from earlier versions. Even if you hate IE,
most of us who write web pages need to know what it will do with our
pages. See http://msdn.microsoft.com/en-us/ie/ff468705.aspx to see
what IE9 will do. Concerning html5, IE9 supports far more html5
specific code than IE8. Also there is much more css 3 support. A
problem with Javascrapt I was having on IE9 beta, but not on earlier
versions of IE, has cleared up in the release candidate. A few bugs I
have noted on previous posts remain. Besides supporting application/
xhtml+xml, the next surprise I noticed most is that IE9 is using a
html5 "house" video element format of mp4 type called AVC with codec
H264 and uses aac audio for the sound. I was expecting .wmv instead.
It is very interesting to note that the most recent flv/swf also uses
codec H264 and is capable of even top Blu-ray resolution, although the
usual mp3 sound needs to be upgraded for HD. In the past there were
rumors that Microsoft and Adobe might combine, but most did not think
this would happen because of antitrust laws. However, these rumors are
now alive again. Several weeks ago, the CEOs of Adobe and Microsoft
had a meeting, which many thought might concern the media problem
posed by Apple, Google, and others. Since Google, Apple, and several
others are much more important now than in the past, antitrust laws
might not now prevent an Adobe-Microsoft merger. It appears it would
be very easy to merge the new Microsoft H264 codec video support with
that of the latest Adobe flv/swf support. It also is interesting that
recent camcorders from the likes of Sony support a codec H264 format
in HD such as AVCHD with mp4 container files. Of course the importance
of the movie companies, including Sony, is very great, since many
computer companies want to stream or sell audio or video. A major
movie company likely could require that the CEOs of Microsoft and
Adobe dance a tango together, if that is what it took to get access to
their movies and audio :).
 
G

Gus Richter

Microsoft, as of Feb. 10, has moved IE9-beta up to to a Release
Candidate which is now available for download at Microsoft.(Is a
release candidate a purgatory between a beta and plain release? :).
IE9 is greatly different from earlier versions. Even if you hate IE,
most of us who write web pages need to know what it will do with our
pages. See http://msdn.microsoft.com/en-us/ie/ff468705.aspx to see
what IE9 will do. Concerning html5, IE9 supports far more html5
specific code than IE8. Also there is much more css 3 support. A
problem with Javascrapt I was having on IE9 beta, but not on earlier
versions of IE, has cleared up in the release candidate. A few bugs I
have noted on previous posts remain. Besides supporting application/
xhtml+xml, the next surprise I noticed most is that IE9 is using a
html5 "house" video element format of mp4 type called AVC with codec
H264 and uses aac audio for the sound. I was expecting .wmv instead.
It is very interesting to note that the most recent flv/swf also uses
codec H264 and is capable of even top Blu-ray resolution, although the
usual mp3 sound needs to be upgraded for HD. In the past there were
rumors that Microsoft and Adobe might combine, but most did not think
this would happen because of antitrust laws. However, these rumors are
now alive again. Several weeks ago, the CEOs of Adobe and Microsoft
had a meeting, which many thought might concern the media problem
posed by Apple, Google, and others. Since Google, Apple, and several
others are much more important now than in the past, antitrust laws
might not now prevent an Adobe-Microsoft merger. It appears it would
be very easy to merge the new Microsoft H264 codec video support with
that of the latest Adobe flv/swf support. It also is interesting that
recent camcorders from the likes of Sony support a codec H264 format
in HD such as AVCHD with mp4 container files. Of course the importance
of the movie companies, including Sony, is very great, since many
computer companies want to stream or sell audio or video. A major
movie company likely could require that the CEOs of Microsoft and
Adobe dance a tango together, if that is what it took to get access to
their movies and audio :).


Thank you for the info and the link.
 
L

Lewis

In message said:
1. All new documents should be created to place browsers into Standards
Mode. If preferring HTML 4.01, use the Strict Doctype.
Absolutely.

2. The HTML5 Doctype is a minimized HTML 4.01 Strict Doctype to the
point whereby all doctype sniffing browsers will still trigger Standards
Mode. Use <!DOCTYPE html> as the Doctype with only HTML 4.01 vintage
elements and you achieve the same result as 1. above.
Yep.

3. When using the HTML5 Doctype you are not obligated to use the new
elements, but you could, when you and/or the browser(s) are ready.

Yep. But also, you should use the appropriate checks to see if a feature
in HTML5 you want to use is supported, and if not, degrade gracefully.
4. The WHATWG initiated the HTML5 project and was founded by Apple,
Mozilla and Opera. Google and Microsoft are also on side. They are
working on supporting HTML5. It's a transition period similar to when
CSS was introduced. It took a while for browsers to support the CSS
Properties and for developers to start using them. I recall a lot of
doom and gloom predictions in those days.

The trouble is that so far, at least, Microsoft is not doing it right
and is still doing its own thing and picking and choosing what to
support and how. Maybe IE9 is better than 7 or 8? Hard to say yet, but
they've done atrociously on the Acid tests in the past. Hell, do they
even support fixed position correctly yet?
 
J

Jukka K. Korpela

Gus said:
I'm reminded of all the criticism and naysayers when CSS was
recommended.

I still remember the time when Internet Explorer 3 had the best available
CSS implementation. A bit later it was said that the best part of CSS
support in IE 3 was that it was easy to switch it off - something that later
versions of IE left off for no good reason.

So at least at some point, there were good reasons for the criticism.
Perhaps we could agree on this:

Well, I guess so, with annotations.
2. The HTML5 Doctype is a minimized HTML 4.01 Strict Doctype to the
point whereby all doctype sniffing browsers will still trigger
Standards Mode.

Rather, it's a special kind of incantation constructed for the purpose of
triggering "Standards Mode" (which isn't really standards-conforming mode,
just each browser's best attempt at that, or not-intentionally-buggy mode).
HTM5 has no concept of document type definition, so its "doctype" is just an
imitation of document type declarations. It has no special relation to HTML
4.01 Strict doctype.
Use <!DOCTYPE html> as the Doctype with only HTML
4.01 vintage elements and you achieve the same result as 1. above.

As far as rendering goes, yes. But if you use the W3C markup checker,
somewhat confusingly called W3C Markup Validator, <!DOCTYPE html> has a very
special effect. Instead of running SGML or XML validation, the checker runs
a completely different type of check. In particular, it will accept some
HTML 4.01 constructs that are invalid under HTML 4.01 Strict (like <menu>)
and reject as errors some constructs that are valid under HTML 4.01 Strict
(like <big>).
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

Forum statistics

Threads
473,744
Messages
2,569,484
Members
44,904
Latest member
HealthyVisionsCBDPrice

Latest Threads

Top