Viewable in Any Browser campain: comments, feedback

[Gérard Talbot]

I personally disagree with the Viewable in Any browser campain. The site
is frankly outdated, seems to be no longer maitained, has several
recommendations that don't make a lot of sense anymore. In my mind, it
is much more important to code according to standards (valid markup and
CSS code) and good coding practices (tableless design, graceful
degradation, etc.) than to code according to the lower common
denominator among browsers.

Let's face it: there is pretty much no one anymore who uses MSIE 4 or NS
4.x. We are already in the era of Firefox 1.5, NS 8.x, MSIE 7 and Opera
8.5. An unanimity of sources indicate that MSIE 5.x is now used less
than Firefox 1.x (eg: world-wide stats of August 2005:
http://www.thecounter.com/stats/2005/August/browser.php ).
Some marginal browsers are no longer being maintained either while some
other marginal browsers (e.g Icab 3, Amaya 9.x) are being upgraded to
support fully HTML 4.01, CSS 2.1 and DOM 1.

In that spirit, I stumble across other campains which I think make a lot
more sense and are more proactive, useful for the web in general:

- Any modern browser: http://merri.net/anymodernbrowser.shtml
- Browse Happy: http://browsehappy.com/
- Alternative Browser Alliance: http://www.alternativebrowseralliance.com/
- Table free campain: http://www.workingwith.me.uk/tablefree/why/

Gérard
--
remove blah to email me

[Guy Macon]

Content-Transfer-Encoding: 8Bit
Gérard Talbot wrote:

I personally disagree with the Viewable in Any browser campain. ....In my mind, it is much more important to code according to
standards (valid markup and CSS code) and good coding practices
(tableless design, graceful degradation, etc.) than to code
according to the lower common denominator among browsers.

Let's face it: there is pretty much no one anymore who uses
MSIE 4 or NS 4.x. We are already in the era of Firefox 1.5,
NS 8.x, MSIE 7 and Opera 8.5.

In my opinion, those few remaining users of ancient browsers
are used to most pages looking bad, and are grateful if the
text is readable at all.

There is, however, one "old" browser that we should all support;
Lynx.

[Chris Morris]

Well, I agree with your general comment that AnyBrowser is a little
unmaintained, and some of the advice (commenting Javascript code) is
out of date, and other bits would benefit from updating to discuss
more recent compatibility issues and more stylesheet issues.

Still, if you do reasonable quality code and remember to hide complex
CSS from older browsers (@import method, usually) then you can support
NS4/IE4 in the same way you support Lynx - content without styling -
without really needing to worry about it much.

*shrug* Why not have it work in any browser? If it works in Lynx, IE
and any modern graphical browser, and you stick to well-known
accessibility principles, it can work in almost any browser without a
huge amount of extra effort. (The big effort for me, usually, is
getting it to work in IE)

Gérard Talbot <ne***********@gtalbot.org> writes:

- Any modern browser: http://merri.net/anymodernbrowser.shtml
"One of the first steps, in my opinion, is to switch to XHTML."

Hmm. The advice to use CSS media types is potentially good, but
there's no need to use XHTML 1.0 (or they even suggest 1.1). HTML 4
Strict has almost exactly the same benefits [1] and disadvantages [2].

[1] If your internal storage format is XHTML 1.0 then there's a case
for using that for display, I suppose. Any other XML format and it can
be XSLTd to HTML about as easily as XHTML.

[2] You need a custom doctype with HTML to check some of the useful
XHTML strictness rules like closing of </p> being required, for
example, but that's not a big deal.
- Browse Happy: http://browsehappy.com/
- Alternative Browser Alliance: http://www.alternativebrowseralliance.com/
I'm not sure these really *replace* anybrowser - which was very much
directed at developers. These seem to be directed mainly at users and
don't give much in the way of detail to developers as to *how* to make
a site work in any browser.

It's a valuable aim, but not necessarily what anybrowser was/is about.
- Table free campain: http://www.workingwith.me.uk/tablefree/why/

Fairly good, for that area. It probably should emphasise a bit more
that data tables *are* still okay, given the number of people posting
here from time to time who have taken 'no tables' too far.

My own attempts to give (reasonably) updated advice on accessible and
compatible web authoring are at
http://www.dur.ac.uk/its/services/web/accessibility/
They also need a bit of work, but I haven't had time to do more than
update links for the last couple of years.

--
Chris

[David Ross]

Gérard Talbot wrote:

I personally disagree with the Viewable in Any browser campain. The site
is frankly outdated, seems to be no longer maitained, has several
recommendations that don't make a lot of sense anymore. In my mind, it
is much more important to code according to standards (valid markup and
CSS code) and good coding practices (tableless design, graceful
degradation, etc.) than to code according to the lower common
denominator among browsers.

Let's face it: there is pretty much no one anymore who uses MSIE 4 or NS
4.x. We are already in the era of Firefox 1.5, NS 8.x, MSIE 7 and Opera
8.5. An unanimity of sources indicate that MSIE 5.x is now used less
than Firefox 1.x (eg: world-wide stats of August 2005:
http://www.thecounter.com/stats/2005/August/browser.php ).
Some marginal browsers are no longer being maintained either while some
other marginal browsers (e.g Icab 3, Amaya 9.x) are being upgraded to
support fully HTML 4.01, CSS 2.1 and DOM 1.

In that spirit, I stumble across other campains which I think make a lot
more sense and are more proactive, useful for the web in general:

- Any modern browser: http://merri.net/anymodernbrowser.shtml
- Browse Happy: http://browsehappy.com/
- Alternative Browser Alliance: http://www.alternativebrowseralliance.com/
- Table free campain: http://www.workingwith.me.uk/tablefree/why/

While the site might not have been updated recently (last update
was a year ago), much of the advice in "Design Guide" is still
valid. To paraphrase some of them:

* Don't create a Web page for a specific window size or
resolution.

* Avoid reserved characters when creating the file names of
pages.

* Observe case-sensitivity when coding links.

* Be sure to have sufficient contrast between text and background
colors.

* If you set the color for unvisited links, set a different color
for visited links.

* Specify the height and width of images.

* Use tables only to make a tabular presentation, not to create
margins or other layout that is best done via style-sheets.

* Be judicious in using frames. Don't use them unless the page
really is MORE useful with them. (See <URL:http://cad.chp.ca.gov/>
for an example of a good use of frames.)

These kinds of guidelines don't change with fashions. They remain
valid without frequent revisions.

--

David E. Ross
<URL:http://www.rossde.com/>

I use Mozilla as my Web browser because I want a browser that
complies with Web standards. See <URL:http://www.mozilla.org/>.

[James Pickering]

For general information:

Here is the current Accessible Site Design Guide on the "Viewable with
Any Browser Campaign" site:

http://www.anybrowser.org/campaign/abdesign.html
--
James Pickering
http://jp29.org/

[Gérard Talbot]

Chris Morris a écrit :

Well, I agree with your general comment that AnyBrowser is a little
unmaintained, and some of the advice (commenting Javascript code) is
out of date, and other bits would benefit from updating to discuss
more recent compatibility issues and more stylesheet issues.

Still, if you do reasonable quality code and remember to hide complex
CSS from older browsers (@import method, usually) then you can support
NS4/IE4 in the same way you support Lynx - content without styling -
without really needing to worry about it much.

I don't use the @import method but I do test my pages with CSS disabled
to see if content can still be accessed, is still readable and if
navigation is still functional. This is a checkpoint in the WCAG list.
I also view/try those pages with Lynx 2.8.5. I never test my pages with
NS 4.x nor IE4/IE5.
*shrug* Why not have it work in any browser? If it works in Lynx, IE
and any modern graphical browser, and you stick to well-known
accessibility principles, it can work in almost any browser without a
huge amount of extra effort. (The big effort for me, usually, is
getting it to work in IE)

Gérard Talbot <ne***********@gtalbot.org> writes:

- Any modern browser: http://merri.net/anymodernbrowser.shtml

"One of the first steps, in my opinion, is to switch to XHTML."

Hmm. The advice to use CSS media types is potentially good, but
there's no need to use XHTML 1.0 (or they even suggest 1.1). HTML 4
Strict has almost exactly the same benefits [1] and disadvantages [2].

[1] If your internal storage format is XHTML 1.0 then there's a case
for using that for display, I suppose. Any other XML format and it can
be XSLTd to HTML about as easily as XHTML.

[2] You need a custom doctype with HTML to check some of the useful
XHTML strictness rules like closing of </p> being required, for
example, but that's not a big deal.

This is interesting. I first developed my pages in XHTML strict and then
retroconverted them to HTML 4.01 strict. I kept all of the XHTML
strictness formatting like quoting attribute value, closing with end
tags (except for a few that must not have a closing tag), lowercasing
elements and attributes, no minimization of attributes, etc.

- Browse Happy: http://browsehappy.com/
- Alternative Browser Alliance: http://www.alternativebrowseralliance.com/

I'm not sure these really *replace* anybrowser - which was very much
directed at developers.

You're right.

These seem to be directed mainly at users and don't give much in the way of detail to developers as to *how* to make
a site work in any browser.

It's a valuable aim, but not necessarily what anybrowser was/is about.

Correct. It is another campain, aiming at users.

- Table free campain: http://www.workingwith.me.uk/tablefree/why/

Fairly good, for that area. It probably should emphasise a bit more
that data tables *are* still okay,

Good point. He's not mentioning that using table to render tabular data
is not only ok but logical and suitable. He mentions though in the
campain page (not in the 7 reasons page) that "Using tables for layout
is bad. They are intended to be used to display tabular data, not to
create complex layouts."
http://www.workingwith.me.uk/tablefree/
given the number of people posting
here from time to time who have taken 'no tables' too far.

My own attempts to give (reasonably) updated advice on accessible and
compatible web authoring are at
http://www.dur.ac.uk/its/services/web/accessibility/
They also need a bit of work, but I haven't had time to do more than
update links for the last couple of years.

Gérard
--
remove blah to email me

[Gérard Talbot]

David Ross a écrit :

Gérard Talbot wrote:

I personally disagree with the Viewable in Any browser campain. The site
is frankly outdated, seems to be no longer maitained, has several
recommendations that don't make a lot of sense anymore. In my mind, it
is much more important to code according to standards (valid markup and
CSS code) and good coding practices (tableless design, graceful
degradation, etc.) than to code according to the lower common
denominator among browsers.

Let's face it: there is pretty much no one anymore who uses MSIE 4 or NS
4.x. We are already in the era of Firefox 1.5, NS 8.x, MSIE 7 and Opera
8.5. An unanimity of sources indicate that MSIE 5.x is now used less
than Firefox 1.x (eg: world-wide stats of August 2005:
http://www.thecounter.com/stats/2005/August/browser.php ).
Some marginal browsers are no longer being maintained either while some
other marginal browsers (e.g Icab 3, Amaya 9.x) are being upgraded to
support fully HTML 4.01, CSS 2.1 and DOM 1.

In that spirit, I stumble across other campains which I think make a lot
more sense and are more proactive, useful for the web in general:

- Any modern browser: http://merri.net/anymodernbrowser.shtml
- Browse Happy: http://browsehappy.com/
- Alternative Browser Alliance: http://www.alternativebrowseralliance.com/
- Table free campain: http://www.workingwith.me.uk/tablefree/why/

While the site might not have been updated recently (last update
was a year ago), much of the advice in "Design Guide" is still
valid. To paraphrase some of them:

* Don't create a Web page for a specific window size or
resolution.

* Avoid reserved characters when creating the file names of
pages.

Depending on what is meant with "reserved" characters, this may be an
advice aiming at supporting old browsers only.

HTML 4.01 section B.2.1 Non-ASCII characters in URI attribute values
http://www.w3.org/TR/html401/appendi...on-ascii-chars

"We recommend that user agents adopt the following convention for
handling non-ASCII characters in such cases: (...) Escape these bytes
with the URI escaping mechanism (i.e., by converting
each byte to %HH, where HH is the hexadecimal notation of the byte value)."
Modern browsers will do the conversion; old may not.
Let's try an example here (coming from a bugzilla bugreport actually
https://bugzilla.mozilla.org/show_bug.cgi?id=293177 ):

http://www.gtalbot.org/BugzillaSection/Bug293177.html (testcase)
http://www.gtalbot.org/BugzillaSection/ÐÃÅ¥ÖãßÜü.html (targeted page in
bug 293177)

Lynx 2.8.5 does not make the conversion but will handle the both links
of the page http://www.gtalbot.org/BugzillaSection/Bug293177.html. I
have not tried with NS 4.x and MSIE 4.x/5x and have no intention of
trying those. Mozilla 1.7+, Seamonkey 1.x and Firefox 1.x convert the 2
links properly. Opera 8.5 handles the links without any conversion in
the address bar; same thing with MSIE 6.

* Observe case-sensitivity when coding links.

* Be sure to have sufficient contrast between text and background
colors.

* If you set the color for unvisited links, set a different color
for visited links.

* Specify the height and width of images.

* Use tables only to make a tabular presentation, not to create
margins or other layout that is best done via style-sheets.

All of these recommendations are justly for all browsers and media.
* Be judicious in using frames. Don't use them unless the page
really is MORE useful with them. (See <URL:http://cad.chp.ca.gov/>
for an example of a good use of frames.)

This one is somewhat debatable. If I do not consider the several invalid
markup code, missing doctype decl., etc.. in that page, then the page
is making a good use of frame. Also, the page still works with CSS
disabled and javascript disabled: so the page is good at achieving
accessibility goals.
The issue with anybrowser.org is that it recommends practices which,
with time, are less and less justified or clearly not making any sense
anymore. What will happen in 5-7 years from now when pretty much
everyone will be using a browser supporting "position: fixed" or which
will support XFrames? Will anybrowser.org still recommend to use
<frame>s when it seems "judicious" to do so?
These kinds of guidelines don't change with fashions. They remain
valid without frequent revisions.

Here's one point where I disagree with the anybrowser.org website: the
graphics format section at
http://www.anybrowser.org/campaign/a...n2.html#images

PNG is not formally recommended. Pretty much any decent visual browser
render PNG nowadays. That section is clearly outdated.
W3C QA tips for webmasters recommends very differently from anybrowser.org.
"PNG also compresses better than GIF in almost every case (5% to 25% in
typical cases)."
http://www.w3.org/QA/Tips/png-gif

Spacer tag is somewhat recommended: "if a non-breaking space won't
fulfill your needs, and your document will be readable without the
space, it may be preferable to use a SPACER rather than a transparent
graphic". That is definitely not what I recommend.

Document colors:
"If you want to specify one of the colors (BGCOLOR, TEXT, LINK, VLINK,
ALINK), specify them all to ensure a pleasant and readable mix."
That fits exactly what I meant. If you code according to standards, you
will not use these deprecated attributes. If you code according to the
lowest common denominator in browsers, you'll use deprecated attributes
all the time, everywhere, forever. That's not what I recommend.

Safe Colors and Color Palettes:
How many people are now using a computer screen/graphic card which can
not render more than 256 colors? Is anybrowser.org really offering truly
reasonable and best recommendation in 2005?

The whole issue is still for which/according to which level to code for:
according to web standards (HTML 4.01, CSS 1, DOM 1, WCAG 1, etc.) or
according to the lowest common denominator among browsers. With upgraded
browser versions being released and with people updating/upgrading their
browsers, is the anybrowser.org site still truly making a point?

Gérard
--
remove blah to email me

[Darin McGrew]

David Ross a écrit :

* Be judicious in using frames. Don't use them unless the page
really is MORE useful with them. (See <URL:http://cad.chp.ca.gov/>
for an example of a good use of frames.)

Gérard Talbot <ne***********@gtalbot.org> wrote: This one is somewhat debatable. If I do not consider the several invalid
markup code, missing doctype decl., etc.. in that page, then the page
is making a good use of frame. Also, the page still works with CSS
disabled and javascript disabled: so the page is good at achieving
accessibility goals.

Well, all the incident links at <URL:http://cad.chp.ca.gov/> give me

Microsoft VBScript runtime error '800a01a8'
Object required: 'rsD'
/iiqr.asp, line 56

so I don't really consider it a very good example of anything...
--
Darin McGrew, mc****@stanfordalumni.org, http://www.rahul.net/mcgrew/
Web Design Group, da***@htmlhelp.com, http://www.HTMLHelp.com/

"There are three kinds of people: those who can count and those who can't."

[Gérard Talbot]

Darin McGrew a écrit :

David Ross a écrit :

* Be judicious in using frames. Don't use them unless the page
really is MORE useful with them. (See <URL:http://cad.chp.ca.gov/>
for an example of a good use of frames.)

Gérard Talbot <ne***********@gtalbot.org> wrote:

This one is somewhat debatable. If I do not consider the several invalid
markup code, missing doctype decl., etc.. in that page, then the page
is making a good use of frame. Also, the page still works with CSS
disabled and javascript disabled: so the page is good at achieving
accessibility goals.

Well, all the incident links at <URL:http://cad.chp.ca.gov/> give me

Microsoft VBScript runtime error '800a01a8'
Object required: 'rsD'
/iiqr.asp, line 56

so I don't really consider it a very good example of anything...

Well, I was able to view the page without a problem. It's some sort of
traffic incidents page for the Los Angeles region.

By the way, Darin, did you get my email... regarding the update of WDG
Web Authoring FAQ Updated ? sent on 2005-09-18. Just asking...

Gérard
--
remove blah to email me

[Darin McGrew]

Gérard Talbot <ne***********@gtalbot.org> wrote:

By the way, Darin, did you get my email... regarding the update of WDG
Web Authoring FAQ Updated ? sent on 2005-09-18. Just asking...

Yes, and I've incorporated some of the suggestions already. Specific,
noncontroversial suggestions are easy to incorporate quickly.

Others get added to various "to do" and "to research" lists, and will take
longer. For example, I plan to replace deprecated presentational markup
examples with CSS, and proprietary DOM examples with the standard DOM. And
at some point, I'll include a "No Longer Frequently Asked Questions" page
with the entries that I've deleted from the FAQ.
--
Darin McGrew, mc****@stanfordalumni.org, http://www.rahul.net/mcgrew/
Web Design Group, da***@htmlhelp.com, http://www.HTMLHelp.com/

"There are three kinds of people: those who can count and those who can't."

[Gérard Talbot]

Darin McGrew a écrit :

Gérard Talbot <ne***********@gtalbot.org> wrote:

By the way, Darin, did you get my email... regarding the update of WDG
Web Authoring FAQ Updated ? sent on 2005-09-18. Just asking...

Yes, and I've incorporated some of the suggestions already. Specific,
noncontroversial suggestions are easy to incorporate quickly.

Great! Just great! I see you've incorporated many changes, improvements.

One typo:
Section 5.11:
"Note that the FOO element is deprecated by HTML 4."
should be
Note that the &lt;font&gt; element is deprecated by HTML 4.
or
Note that the FONT element is deprecated by HTML 4.
Others get added to various "to do" and "to research" lists, and will take
longer. For example, I plan to replace deprecated presentational markup
examples with CSS, and proprietary DOM examples with the standard DOM.
Great!

And at some point, I'll include a "No Longer Frequently Asked Questions" page
with the entries that I've deleted from the FAQ.

I will link the WDG Web Authoring FAQ into
"Using Web Standards in Your Web Pages"
http://www.mozilla.org/docs/web-deve...upgrade_2.html
and into my website at
Website design resources for beginner and intermediate Nvu users
http://www.gtalbot.org/NvuSection/Nv...Resources.html

Gérard
--
remove blah to email me

[Chris Morris]

Gérard Talbot <ne***********@gtalbot.org> writes:

Chris Morris a écrit :

Well, I agree with your general comment that AnyBrowser is a little
unmaintained, and some of the advice (commenting Javascript code) is
out of date, and other bits would benefit from updating to discuss
more recent compatibility issues and more stylesheet issues.
Still, if you do reasonable quality code and remember to hide complex
CSS from older browsers (@import method, usually) then you can support
NS4/IE4 in the same way you support Lynx - content without styling -
without really needing to worry about it much.

I don't use the @import method but I do test my pages with CSS
disabled to see if content can still be accessed, is still readable
and if navigation is still functional. This is a checkpoint in the
WCAG list.
I also view/try those pages with Lynx 2.8.5. I never test my pages
with NS 4.x nor IE4/IE5.

The problem isn't so much CSS-disabled as CSS-randomly-interpreted,
which NS4/IE4/IE5 have a habit of doing. Using the right @import hack
means you don't need to worry about NS4 and IE4, which would make the
worst mess of things, and so don't really have to test in those if you
also test in Lynx. IE5, while a falling minority, I think *is* worth
testing in for at least the next year or so. IE5/Mac probably for even
longer.

--
Chris

[Darin McGrew]

Gérard Talbot <ne***********@gtalbot.org> wrote:

Great! Just great! I see you've incorporated many changes, improvements.

One typo:

Thanks. It's fixed now.

Also, the FAQ is now HTML 4.01 Strict. Now I just need to update all the
answers and examples to match. :-/

What's the best standards-oriented replacement for <base target="_top">?
--
Darin McGrew, mc****@stanfordalumni.org, http://www.rahul.net/mcgrew/
Web Design Group, da***@htmlhelp.com, http://www.HTMLHelp.com/

"It's bad luck to be superstitious."

[Jukka K. Korpela]

Darin McGrew <mc****@stanfordalumni.org> wrote:

What's the best standards-oriented replacement for <base target="_top">?

None. Keep it. It should do no harm, and it may marginally help against
framers.

Formally, it's deprecated markup. In essence, it follows the spirit, if not
the letter, by fighting against framing, with the tools we have.

--
Yucca, http://www.cs.tut.fi/~jkorpela/
Pages about Web authoring: http://www.cs.tut.fi/~jkorpela/www.html

[Guy Macon]

Content-Transfer-Encoding: 8Bit

Gérard Talbot wrote:

I don't use the @import method but I do test my pages with CSS disabled
to see if content can still be accessed, is still readable and if
navigation is still functional. This is a checkpoint in the WCAG list.
I also view/try those pages with Lynx 2.8.5. I never test my pages with
NS 4.x nor IE4/IE5.

I figure that anyone who still uses NS4 is used to having large
portions of the web be unusable.

I don't test my pages with NS4 or IE4, and plan on stopping my testing
with IE5 sometime soon, but I always test with both lynx and links.

Another test which I find to be very useful is with the latest IE
with all features such as images and jsvascript turned off, security
set to the highest setting, and accessibility set to ignore the site's
choice of fonts, colors, and CSS. Quite eye-opening...
Links about Links and Lynx:
http://world.std.com/~adamg/dehanced.html
http://lynx.browser.org/
http://links.sourceforge.net/
http://freshmeat.net/projects/links/
http://www.delorie.com/web/lynxview.html
http://home24.inet.tele.dk/ccadams/se/lynx.html
http://lynx.isc.org/lynx2.8.5/index.html
http://www.subir.com/lynx.html