The approach we've been taking is to use html5 polyglot. Basically XHTML, but with HTML5 head tags etc, ready for moving over to HTML5.
It's easy to do what you like on your own site, but I have to instruct a lot of third parties so they need a spec to work against. It is easier for us to say XHTML, with parts of HTML5 than HTML5 except x y and z.
I wouldn't even consider HTML 4.01 to be "finished" yet. Not until every browser with an incomplete implementation of the spec has a 0% share.
I stated the requirement up front! I mean hell, if you're going to seriously consider Lynx as a browser, you need to consider everything with 0% market share: Word, Excel, Window Media Player, my toaster ...
Nitpicking, I am not. I'm joking, of course, about the toaster. But, the broad definition of a web browser, one which allows one to include lynx alongside other browsers with a straight face, also includes any piece of software that can establish a TCP connection and relay the return. Even telnet is therefore a web browser.
Point is, only browsers with a significant market share are worth considering. And of those, none fully support any complete standard in a consistent manner.
Nitpicking, I am not. I'm joking, of course, about the toaster. But, the broad definition of a web browser, one which allows one to include lynx alongside other browsers with a straight face, also includes any piece of software that can establish a TCP connection and relay the return. Even telnet is therefore a web browser.
Point is, only browsers with a significant market share are worth considering. And of those, none fully support any complete standard in a consistent manner.
What part of HTML 2.0 do modern browsers not support? *Question of complete curiosity* And question: Can Word/Excel/WMP read HTML? Because Lynx--for the most part--can.
Last edited by Mr Initial Man; 04-13-2012 at 10:51 PM.
What part of HTML 2.0 do modern browsers not support? *Question of complete curiosity* And question: Can Word/Excel/WMP read HTML? Because Lynx--for the most part--can.
Hehe ... OK. Had I the time, I'm willing to bet I could find some compliance issues. Though, I suppose it's possible that all significant modern browsers fully support HTML 2.0.
In any case: Yes, Word and Excel can read some HTML. WMP ... maybe not. Though, I actually wouldn't be at all surprised if it could.
Well, that's encouraging for me. Maybe there will be a use for HTML 4.01 for a while yet.
Considering the fact that: (...)
the "doctype" used for the so called "HTML5" (which is nothing but "HTML") is not as one would imagine "<!doctype html5>" but "<!doctype html>" -and there's no doubt about that - just in case someone got confused about - says that: "There is no other HTML Standard - but HTML!"
That's all.
And what does it mean?
It means that: Whatever browser you are targeting for backwards compatibility - you will be using one and a single exact doctype which is:
<!Doctype HTML> for all targets.
What other news is old?
None! - Everything is new to the latecomers who thought they were doing html while using xhtml syntax and other barbaric doctype nonsense.
For best results as we all know, - all internet coding should be done in standards mode regardless of a browser version and when talking about "which doctype should I use?" - the <!Doctype HTML> does exactly that.
Starting with IE 4 [the first browser who introduced standards], probably not ending with IE10, the <!Doctype HTML> declaration will force them to standards mode and the standards level supported by the given release version.
(...henceforth) that's a pretty dumb decision, since the so called html5 is a: Browser Version Free type of thing. And Google coders were not dumb for using (since at least three or 4 years ago) a simple and plain <!doctype> up until html 5 came in.
They, most probably knew, way back then, this same thing I'm "rediscovering" here today.
Ooooooooooooookay, where'd you read that? Citation please.
Originally Posted by Troy III
... and other barbaric doctype nonsense.
You see, what some people don't seem to realize--right up until they work with XML--is that the Document Type Definition performs a very important function that has never been performed by anything else, not schemas, not the HTML5 doctype.
HTML Code:
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd" [
<!ENTITY deg "°"><!ENTITY oq "“"><!ENTITY cq "”">
]>
That's an SVG 1.1 Doctype that I used in a page--but do you see something odd about it? Here's the original:
Code:
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
Here's the one I used:
Code:
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd" [
<!ENTITY deg "°">
<!ENTITY oq "“">
<!ENTITY cq "”">
]>
Character Entity References!
XHTML 1.0 links to a document type definition, which defines these,;XHTML5 does NOT, so you have to go through the effort of memorizing their numerical equivalents or skip any characters you can't type on your keyboard altogether.
Doing something like
HTML Code:
<!DOCTYPE html [<!--CERs Here-->]>
will cause the validator to throw errors, because now it's expecting the COMPLETE document type definition (I've written them. Even ElementML was surprisingly long). SVG has no character entity references, but because of its doctype's structure I was able to add some in.
HTML5's doctype doesn't allow that, and if you don't have a doctype at all, you're out of luck. Oh, I know, "You don't need a doctype" and yadda yadda yadda, but here's a point I made in my book:
&#8730;
√
√
They all amount to the same thing, but which do you want to use for writing "(X)HTML5=√(All Headaches)"?
Last edited by Mr Initial Man; 04-14-2012 at 04:53 PM.
Ooooooooooooookay, where'd you read that? Citation please.
You see, what some people don't seem to realize--right up until they work with XML--is that the Document Type Definition performs a very important function that has never been performed by anything else, not schemas, not the HTML5 doctype.
HTML Code:
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd" [
<!ENTITY deg "°"><!ENTITY oq "“"><!ENTITY cq "”">
]>
That's an SVG 1.1 Doctype that I used in a page--but do you see something odd about it? Here's the original:
Code:
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
Here's the one I used:
Code:
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd" [
<!ENTITY deg "°">
<!ENTITY oq "“">
<!ENTITY cq "”">
]>
Character Entity References!
XHTML 1.0 links to a document type definition, which defines these,;XHTML5 does NOT, so you have to go through the effort of memorizing their numerical equivalents or skip any characters you can't type on your keyboard altogether.
Doing something like
HTML Code:
<!DOCTYPE html [<!--CERs Here-->]>
will cause the validator to throw errors, because now it's expecting the COMPLETE document type definition (I've written them. Even ElementML was surprisingly long). SVG has no character entity references, but because of its doctype's structure I was able to add some in.
HTML5's doctype doesn't allow that, and if you don't have a doctype at all, you're out of luck. Oh, I know, "You don't need a doctype" and yadda yadda yadda, but here's a point I made in my book:
√
√
√
They all amount to the same thing, but which do you want to use for writing "(X)HTML5=√(All Headaches)"?
This ... "issue" has never been a problem for me. I use → ,←, and so on without problems. And in cases where I need many special characters, I won't be typing them out by hand anyway. I'll be letting PHP or .NET perform the conversion for me -- or I'll just serve my document in UTF-8.
Doctype isn't meaningless altogether. But the "html5" doctype is perfectly fine for most circumstances, if not all. Gmail, which serves my mail with all sorts of fancy characters and other neat tricks does just fine with the doctype. No need to define a list of entities anywhere ...
Reply With Quote
Originally Posted by Mr Initial Man
And question: Can Word/Excel/WMP read HTML? Because Lynx--for the most part--can.
Bookmarks