Re: [WSG] Custom DTD's to allow target attribute? Yuck
The only thing that makes XHTML something slightly different from XML is it's DTD. Kind of like saying: The only thing that makes females something slightly different from humans is their anatomical differences. The whole point of XHTML is that besides being HTML, it's also well formed XML. I attempt to clarify what I meant by that in the discussion, http://discuss.webstandardsgroup.org/archives/14.htm . XHTML and XML are not synonymous, so if someone can articulate what is XHTML's relationship to XML, it would be useful. Take the following sentence: Mark went to the store and bought eggs, milk, bread, chicken, rice, and corn. Doesn't that sentence contain a unordered list? Good point. I'm still not in agreement that ul>/ol>'s should be nested in p>'s though. Maybe someone can clarify in the discussion how to view semantics in the context of this example. if the devise is standards compliant (for the sake of argument, lets assume that's a safe assumption) there should be no problem. Then why is WAP 2.0 a subset of XHTML 1.1 for example? Valid XML is not necessarily device independent markup, is it? Except that mutations are random, You're taking the analogy to far; I didn't say all ways. That's right, you didn't. Sorry. While I agree that it can be taken to far, what does the X stand for but eXtendable? Yes, it stands for eXtensible. Some extensions are legal, some not. It is legal to extend XHTML, but is lists in paragraphs the type of extension they are talking about? Let's get some clarification at http://discuss.webstandardsgroup.org/archives/14.htm . Nelson --- On 30-Apr-04, at 2:29 AM, Mordechai Peller wrote: Nelson Ford wrote: The reason I brought this up was...this is a really interesting and pertinent topic at this point as XHTML and CSS start to become the rule rather than the exception. I wasn't even sure if browsers actually read the DTD to allow this to work. The only thing that makes XHTML something slightly different from XML is it's DTD. Kind of like saying: The only thing that makes females something slightly different from humans is their anatomical differences. The whole point of XHTML is that besides being HTML, it's also well formed XML. The definition of semantics from dictionary.com is: The study of relationships between signs and symbols and what they represent. So p>'s represent paragraphs and ul>'s represent unordered lists. Most people understand a paragraph to be a block of text, not a list, so if my interpretation is correct, making that change to a DTD would be detrimental to the semantics of the markup language. Take the following sentence: Mark went to the store and bought eggs, milk, bread, chicken, rice, and corn. Doesn't that sentence contain a unordered list? Another problem with that is that the purpose of the XML-based XHTML is to allow for the display of the same documents across all kind of platforms and screen sizes. Making changes like that could harm XHTML's ability to achieve that goal as well, because the way XHTML behaves has taken years of tweaking and thought Last I checked, W3C's recommendations include CSS, p>s, and ul>s. (Hey look, another unordered list inside a paragraph.) Since all elements needed are already included in the recommendations, if the devise is standards compliant (for the sake of argument, lets assume that's a safe assumption) there should be no problem. Take 'span class=pet>cat/span>' versus 'pet>cat/pet>'. Since span> has no default style beyond being inline, they both should display the same. They are equally machine readable. But the latter is more human readable, which is one of the reasons behind XML in the first place. In many ways it's similar to mutations in DNA Except that mutations are random, You're taking the analogy to far; I didn't say all ways. [M]y point is that if we want to be changing the DTD's, we shouldn't be pretending it is still XHTML. While I agree that it can be taken to far, what does the X stand for but eXtendable? Mordechai
[WSG] Custom DTD's to allow target attribute? Yuck
Hi all,
This should not be off-topic...
I've seen more and more of this fiddling with DTD's lately, and I'm not
sure it is a wise thing for us to be going off the standard in order to
bring back the target attribute --or any other attribute really... if
we open that can of worms then we can all have our own versions of
xhtml --, especially since the target attribute is not required to open
a link in a new window. I found a little explanation of how to open new
windows with just a little javascript and an existing attribute to
anchor tags, rel. I wish I could remember where I saw this, but the
technique is simple. In all your external links, place rel=external,
and the following javascript, (I prefer to put all my js in an external
document):
window.onload = function() {
externalLinks();
}
function externalLinks() {
if (!document.getElementsByTagName) return;
var anchors = document.getElementsByTagName(a);
for (var i=0; ianchors.length; i++) {
var anchor = anchors[i];
if (anchor.getAttribute(href)
anchor.getAttribute(rel) == external)
anchor.target = _blank;
}
}
So with this, your documents will still validate on the W3C DTD. The
downside is that it won't work on really old browsers, but my personal
opinion on that is we should take them off life-support wherever
possible anyway. I'd really like to see a discussion on this XHTML
extensions/personal DTD issue. I know this is the power of XML, but the
reason why certain attributes were left out of XHTML was because the
W3C felt they were the job of Javascript, not markup, and as we see
here, the javascript can act on any attribute in XHTML, not just
target.
Nelson
---
Vancouver, BC
www.nelsonford.net
*
The discussion list for http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list getting help
*
Re: [WSG] Custom DTD's to allow target attribute? Yuck
The reason I brought this up was not because I had been seeing a lot
of that talk on this list, but more on some forums on the internet
where a standards beginner asks a question and someone pipes up: Just
change the DTD and we can all validate! [insert south park smile
here]... which I find kind of frightening. I think this is a really
interesting and pertinent topic at this point as XHTML and CSS start to
become the rule rather than the exception.
I wasn't even sure if browsers actually read the DTD to allow this to
work.
The only thing that makes XHTML something slightly different from XML
is it's DTD. Take away or alter this DTD and you no longer have XHTML,
but rather (in my case) NFML. Both languages are XML based, but they
have different semantic meanings for the same tags, if mine allows for
p's to contain lists.
The definition of semantics from dictionary.com is: The study of
relationships between signs and symbols and what they represent. So
p's represent paragraphs and ul's represent unordered lists. Most
people understand a paragraph to be a block of text, not a list, so if
my interpretation is correct, making that change to a DTD would be
detrimental to the semantics of the markup language.
Another problem with that is that the purpose of the XML-based XHTML
(and WAP 2.0 which is a slightly stripped-down version of XHTML 1.1) is
to allow for the display of the same documents across all kind of
platforms and screen sizes. Making changes like that could harm XHTML's
ability to achieve that goal as well, because the way XHTML behaves has
taken years of tweaking and thought by a dedicated team.
In many ways it's similar to mutations in DNA
Except that mutations are random, but this is not. It is kind of like
deciding we all start genetically engineering our children, some with 4
legs, some who are 11 feet tall , suddenly none of the standards, like
doorways that are roughly 8-9 feet high, pants with 2 legs etc will
work for the majority. It will breed chaos. Bizarre example, but I
guess my point is that if we want to be changing the DTD's, we
shouldn't be pretending it is still XHTML.
Nelson
---
Vancouver, BC
www.nelsonford.net
On 29-Apr-04, at 3:11 PM, Mordechai Peller wrote:
Nelson Ford wrote:
I've seen more and more of this fiddling with DTD's lately, and I'm
not sure it is a wise thing for us to be going off the standard...
This is and idea I've been toying with recently, and I wasn't even
sure if browsers actually read the DTD to allow this to work. I should
add that I had no intentions of raising the dead, but rather, I wanted
to fix something I see as a minor failing in XHTML in order to
increase the semantic value of my markup. What I wanted to do is allow
ols and uls inside ps, and then style the list as
{display:inline}. But even if left at block level, there are time when
it would still increase the symantic value.
It is, as you put it a can of worms. In many ways it's similar to
mutations in DNA: for every harmless mutation, never mind the even
rarer beneficial ones, there are many thousands of harmful ones. (One
of the problem with Hypothesis of Natural Selection [it's not a theory
since it makes no predictions which have been verified], but I'm
guessing that that discussion might be slightly off topic.)
Mordechai
*
The discussion list for http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list getting help
*
*
The discussion list for http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list getting help
*
Re: [WSG] Your opinion and feedback requested
the sentinare.com layout falls apart pretty badly on my screen with just a single increase in text size. This is true, but it's mainly the way the 30-day free trial link is marked up with all those spans that goes kablooey and pushes the heading below it over. It may end up being easier to make an image link with alt text in a span to replace that. The heading div (metal) could stand to be scaleable as well. Apart from that the layout seems to hold together and the text flows as it should (in safari). Hey Paul: greetings from Mount Pleasant! Nelson --- Vancouver, BC www.nelsonford.net * The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help *
[WSG] Looking for a little peer review
Nelson, Content of the right sidebar spills over into the center obscuring part of the text. MS internet explorer 6. on Win98SE with text set to largest. Oops! Just tried it with largest text in IE 6.0 SP1 on Win98SE, I got the same problem, the right sidebar spills into the center. Different text sizes seemed fine in firfox 0.8 and netscape 7.1 (and of course Opera). Darian Cabot Thanks Brewnetty and Darian for pointing that out. I have explicitly set the text size for my h3's in px, and that seems to be helping the issue, though I have no OS earlier than Win2k to check it with. What are people's opinions about how best to combat this sort of problem with IE/Win and its text resizing function? Mozilla/Safari etc don't resize text the same way and so don't break the structure quite as severely. IE was bursting the right-sidebar to the left, whereas Safari extends it to the right when the text is too large. Is there any prevailing wisdom on this issue? Nelson --- www.nelsonford.net * The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help *
[WSG] Looking for a little peer review
Hi all, this will be my first time posting to this group. Someone just told me that my ever-in-progress personal site is looking strange on their screen using IE6/Win. I have tested it using IE5.0/Win and IE6/Win on VirtualPC, as well as with various Mac browsers, and have not come across any major layout issues (except for the occasional pixel imperfection). Apparently for this person on IE6/Win the text in the white main column is overlapping the right sidebar. I'm wondering if it may be an issue relating to a minor update of IE6/Win? Anyway, I'd appreciate a quick check with as many browsers as possible just to ensure I haven't missed anything. The central column has a right-margin equal to the width of the right sidebar, and the right sidebar is absolutely positioned. http://www.nelsonford.net/ Thank you! Nelson * The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help *
