Slashdot Mirror

http://www.w3.org/P3P/

Oops...
working link

403 is not "Permanently moved", 301 is. See http://www.w3.org/Protocols/rfc2616/rfc2616-sec10. html.

403 means "Forbidden". Maybe that mistake is the root of all your problems.

While it would be nice if they supported the optional features, it's actually the developers who continue to use alpha channel transparency PNG that are deviating from the W3C recommendation.

By "Developers" I'm assuming you mean "Web Developers." "Software Developers" (implementors) of the PNG stream format are generating non-compliant crud if they can't read/write the "optional" alpha channel.

It seems that you have incorrectly interpreted the Abstract of the specification. I might direct you to the actual specification itself where in 15.2.3 Conformance of PNG Decoders it says:

e. All types of PNG images (indexed-colour, truecolour, greyscale, truecolour with alpha, and greyscale with alpha) are processed. For example, decoders which are part of viewers running on indexed-colour display hardware shall reduce truecolour images to indexed format for viewing.

So yes, it is optional for an alpha channel to appear in a PNG data stream (a graphic artist may not want to use an alpha channel on his latest piece of "art") but if there is one present in the stream then it *must* be handled by the software reading it. It's optional as to whether that software actually does anything meaningful with it (a given display device may not support it), but a best-effort is expected.

You can be standards compliant and still use target with XHTML strict. One of the bigs things about XHTML is that it is modularized and target was just moved to a module which you can import in your documents:
http://www.w3.org/TR/xhtml-modularization/DTD/xhtm l-target-1.mod

Using JavaScript for this, which people can disable or just selectively cripple, is a terrible idea and you start getting all these non-standard handlers spread around.

what about targeting in frames?

The "target" attribute is specifically for frames. However, if you want to use frames, then you should be using the Frameset DTD (where it is defined) and not the Strict one.

The "target" attribute's actually defined in XHTML 1.0 Transitional, too, but the OP was talking about XHTML Strict.

what about targeting in frames?

The "target" attribute is specifically for frames. However, if you want to use frames, then you should be using the Frameset DTD (where it is defined) and not the Strict one.

The "target" attribute's actually defined in XHTML 1.0 Transitional, too, but the OP was talking about XHTML Strict.

what about targeting in frames?

The "target" attribute is specifically for frames. However, if you want to use frames, then you should be using the Frameset DTD (where it is defined) and not the Strict one.

The "target" attribute's actually defined in XHTML 1.0 Transitional, too, but the OP was talking about XHTML Strict.

I just wish that FF would let us going document.element instead of force us to write document.getElementById("element") in order to reference DOM objects. the former is less typing than the latter...

That's not a feature, it's a bug. It's not FF, it's the w3c. And it's the right decision. Standards and all that, right?

The document object already has a shitload of properties, and this IE idea doesn't cooperate nicely, since the property namespace clobbers the id namespace. Bad browser-maker! No cookie!

To be fair the CSS validates just fine, simple as it is.

To be fair the CSS validates just fine, simple as it is.

I'm not sure if I understand your use of the word "barely". IE supports PNG as per the W3C recommendation, including binary transparency. IE doesn't support optional alpha channel transparency:

http://www.w3.org/Graphics/PNG/

From the first paragraph:

"Indexed-color, grayscale, and truecolor images are supported, plus an optional alpha channel for transparency."

While it would be nice if they supported the optional features, it's actually the developers who continue to use alpha channel transparency PNG that are deviating from the W3C recommendation.

It usually helps to run the HTML Validator before claiming that HTML is "perfectly valid".

I would have to agree, there is no need to read a book on the subject, there is a huge swath of information on HTML, CSS, XHTML, etc. on the internet already. Just a month ago, I decided to learn XHTML+CSS, it was pretty easy with all the great tutorial sites out there. For me, it's alot easier to learn something by actually doing it and learning through trial and error then reading a book.

XHTML
CSS

I would have to agree, there is no need to read a book on the subject, there is a huge swath of information on HTML, CSS, XHTML, etc. on the internet already. Just a month ago, I decided to learn XHTML+CSS, it was pretty easy with all the great tutorial sites out there. For me, it's alot easier to learn something by actually doing it and learning through trial and error then reading a book.

XHTML
CSS

Intuition is good and will get you far.
You should also be familiar with authoritative resources for issues like accessibility.
You can learn HTML, CSS and much more on line.
If you choose to go-it-alone, you might wind up as a clueless website author.

The HTML and CSS specifications make excellent reference material. I don't need an incomplete (and error prone) copy of the original sources on dead trees. Especially if I'm paying for it.

The HTML and CSS specifications make excellent reference material. I don't need an incomplete (and error prone) copy of the original sources on dead trees. Especially if I'm paying for it.

mostly alt tags and & instead of & in uris

This page is not Valid HTML 4.01 Transitional!

SVG provides an easy mechanism to store complex sequences of filter operations and apply them to objects as required. There are filters like blur, color matrix, convolve matrix, phong specular and diffuse lighting and many others. I wrote the implementation of them for librsvg, but it's a shame that more editors don't support their creation. Otherwise it would be great fun for all.

(Kidding, but also completely serious. :D)

Whatever you do, DO NOT try to validate that page...

As for being able to search -- multiple files -- in multiple directories -- what's wrong with grep and (when necessary) find? What's wrong with expecting users to know how to keep their things organized?

I said they were supportive. Discussions on mailing lists, blog entries, that sort of thing. Encouraging.

Please note that Macromedia is being supportive to their own developer community. Also every tool (open source or not) that has a chance to provide added value to their own platform while not being a competitive threat to their own product line is likely to get some (little) support as well. That does not even compare to what other companies do, by really embracing the free software / open source movement. IBM offered a couple of dozzens of programs to the open source community (list) one of them being an Integrated Development Environment (IDE) for Laszlo. Sun open sourced Star Office, Netbeans and will soon open source Solaris. Laszlo Systems open sourced their RIA Platform (OpenLaszlo). These and others are companies being supportive to the open source community. Macromedia however is not one of these companies. On a greed scale they would be somewhere very close to Microsoft.

Flash (which, btw, costs half of what you said)

I don't know where you live but in Germany the half cripple Macromedia Flash MX 2004 costs 694.84 euro (=855.926701 US$) and the full Macromedia Flash MX Professional 2004 costs 973.24 euro (=1,198.868952 US$).

If you don't like it -- don't buy it.

You can bet I won't. I already told about OpenLaszlo. That is what I would use, should I ever consider writing Flash applications again. For now I am a lot better off using SVG and JavaScript for the open source projects to which I contribute. SVG and JavaScript are both open standards while Macromedia's technologies are proprietary. Supporting Macromedia's technologies would help Macromedia more than anybody else, and would surely hurt web standards and interoperability.

With Firefox, the developer toolbar and Quirk SearchStatus, and you can learn all these things yourself.

With Firefox, the developer toolbar and Quirk SearchStatus, and you can learn all these things yourself.

basically it's a rigorous test that ensures that a browser has all the goodies that web developers have been lusting after forever.

No, a "rigorous test" would be a test suite with hundreds of carefully-constructed examples (for example, here's one for CSS2.1). The WSG's Acid tests are more like pop quizzes, and only test a few particular items.

The HTML may be valid but the CSS is slightly broken.

The HTML validates, sure.

But the CSS doesn't.

It certainly validates.

http://validator.w3.org/check?uri=http%3A%2F%2Fwww .webstandards.org%2Fact%2Facid2%2Ftest.html%23top

use some fully compliant XSL/XML and tables or divs to create a stack of layers consisting of images that are wrapped in tags.

You would appear to be far more credible if you actually used the proper names for things. "A stack of layers consisting of images that are wrapped in tags"?

What's a stack? I assume you don't mean <layer> elements when you say "layers". And what type of "tag" [sic] (they are element types, BTW)?

i can almost guarantee you if you have more than 2 layers and certainly in a situation where you have differing image widths and more than 2 layers that there will be ENORMOUS gaps between the images that otherwise should be packed with 0px between them. IE by default renders the page as it should.

There should be gaps. The initial value for vertical align is baseline. This means that there should be space for descenders (the tails on glyphs like lowercase 'j').

google provides at least 40 examples of hacks to get around FF/MOZ/SAFARI rendering engine issues

They aren't hacks. They are simply normal CSS. If you don't want room for descenders, you have a number of options. And 40 is an exaggeration. Exaggeration also reduces your credibility. As does claiming multiple independent browsers get the box model wrong and Internet Explorer gets it right.

Tim Berners-Lee and the W3C seems to think they suck.

I tend to agree with them. New TLDs are just a licence to print money. They do not offer new options for domain name holders - everyone with a business or company or product or trademark will just have to go out and get the same permuatations they already own for the existing TLDs and CCs. The name space gets filled up again and people lose out once more. The .mobi domain is harmful for reasons stated in the link above. .xxx is useless because pr0n sites will keep on using other TLDs and CCs. .biz is a ghetto. Can someone explain how a website in .info is more informative than a website in .con? Err... I mean .com?

Maybe it is just that someone at ICANN is a _big_ fan of Pokemon.

The DNS hierarchy needs a complete overhaul, not more random, money making and/or harmful branches.

I expect Microsoft will use these in a manner similar to how they raped^H^H^H^H^Hembraced and extended Java by adding magic code-generating javadoc comment tags like "@dll". For instance:

<?mso-normal some-style load DOTNETCRAP.DLL?>

Microsoft was part of the consortium that designed XML. Processing instructions aren't in there by accident.

To further your point, wouldn't it be better to endorse the use of SVG instead?

I wouldn't get on the hatewagon that that degree. As long as they have schema reference documents available your bet would be a losing one. My bet is that certain documents are encoded using technology similar to this. One of my company's software products has file documents based on encoded XML and for certain situations Microsoft could employ something similar.

You want websites to be rated, add a /rating.txt file that works like the robots.txt that indicates level of content, and have web browsers and proxies respect it.

Or even better, use the PICS standard, which already exists and has for years.

Of course, the fact that not everyone supports PICS on their existing domains, even though it's free to do so, suggests how far .xxx domains are going to get.

There's been a perfectly good W3C standard for voluntarily classifying the content of your own site for years (not just for pornography, but for plenty of other criteria too): It's called PICS, you can classify your own site, and lots of sites do so, and lots of software reads the PICS classification.

The .xxx suffix does the same but less - less efficiently, only one classification, only one degree of classification.

Legislators haven't picked up on PICS, so why should they support an even worse system?

Rich.

I had no problems validating this page via http://validator.w3.org/check. It reported 166 errors on this page.

Isn't SVG an already full-featured and becoming-implemented way to do this kind of thing?

Just so you know, ems are actually vertical units. /* ems, the height of the element's font */

If you're feeling paranoid enough, how about some other questions that same developer has asked around other forums. Try piecing something together.

The site is code verified in XHTML
http://validator.w3.org/check?uri=http%3A%2F%2Fwww .albinoblacksheep.com%2Fflash%2Fsecretofspace.html

There are 7 errors in the page. This page is NOT valid xhtml 1.0 transitional.

That being said, 7 errors is pretty low. Slashdot by comparison returns several hundred -- well, it did, until the w3c was net block banned... wtg taco, ban the w3c so no-one can tell how sloppy slashcode produces invalid html.

7 errors is really good, but microsoft, ibm, and the w3c can all produce valid html. then again sites like tom's hardware can return thousands and thousands of errors...

in terms of open source OSes, FreeBSD's page is valid xhtml 1.0 transitional, and debian.org is valid html 4.01 transitional, but slackware.org returns 69 4.01 errors.

Goatse reurned 7 errors, while tubgirl only returned 3.

I think you mean allow custom CSS for everything but webpages, since the custom CSS is used to skin the XUL app. The -moz- CSS follows a W3C standard for custom CSS. --css-property. The most well known and abused is likely -moz-border-radius, but that should be replaced eventually with the CSS3 border-radius spec.

I think you mean allow custom CSS for everything but webpages, since the custom CSS is used to skin the XUL app. The -moz- CSS follows a W3C standard for custom CSS. --css-property. The most well known and abused is likely -moz-border-radius, but that should be replaced eventually with the CSS3 border-radius spec.

IE's behavior is fully standards compliant. From the article you referenced:

Media types summary for serving XHTML documents

Media type
HTML4
XHTML1.0 (HTML compatible)
XHTML1.0 (other)
XHTML Basic / 1.1
XHTML+MathML

text/html
SHOULD
MAY
SHOULD NOT
SHOULD NOT
SHOULD NOT

Microsoft won't lend full support to CSS2 because they claim it's a flawed specification...

Yeah, it's not like they helped write it over seven years ago or anything. :-)

for those interested to know more about support for application/xhtml+xml mime type, let me sum it up for you:

basically, every time something loads in your browser, several headers are sent before the content, letting the browser know what to do with it. this is how it knows to display web pages in the viewport versus downloading compressed files. specifically, xhtml pages are to be served to the browser as application/xhtml+xml. now, for xhtml 1.0, you MAY serve them as the old html4.0 way (text/html), but you SHOULD use the newer way.

xhtml1.1 doesn't allow such a variance. they SHOULD be served as application/xhtml+xml (the alternative being application/xml which would be interpreted as a straight xml file)... except IE doesn't support such a mime type, thus making it IMPOSSIBLE to correctly serve a xhtml1.1 document to any IE browser. this has severly limited the ability for the web to transform to support documents within several namespaces (such as xhtml, mathml, svg all integrated into a single web page)

for more info, see http://www.w3.org/TR/xhtml-media-types/

stick that in your pipe and smoke it microsoft

SW 2003 Challenge was in October, W3C-OWL standard wasnt yet finalized (It was a Recommendation Standard in Aug 2003, http://www.w3.org/2001/sw/WebOnt/#L151) and to my knowledge no reasoner (Fact/Racer) supported full OWL/DAML+OIL reasoning. So I guess the 'semantics' aspect was not a big concern then..

Today OWL is formalized. Several OWL based api/reasoners are in place. Using such 'RDF only' applications misguides people and the community. My only request to you all Semantic Web Gurus is to preach right message and best practices :)

Alright, so let us check out a sample from OWL (Web Ontology Language):

Wine Rdf

Look through that RDF with emacs/notepad. You will probably not understand all of it, but you can get the gist of it. It attempts to classify things categorically almost, so finding out context for a word is simple. For instance, the owl:Class of "Wine" is a subClass of "PotableLiquid" with a couple restrictions and properties that wine could have in real life.

Why is this useful? It dramatically increases the level at which computers can understand information. In theory, if you tell a computer with this RDF file, "This here is red wine", it will know by inference that the object is a wine that you drink, not that annoying people do, and it will be able to guess at other properites ( such as maker, year made, etc ).

I am intrested in AI and this applies to my job, so this is fantastic news to me.

Alright, so let us check out a sample from OWL (Web Ontology Language):

Wine Rdf

Look through that RDF with emacs/notepad. You will probably not understand all of it, but you can get the gist of it. It attempts to classify things categorically almost, so finding out context for a word is simple. For instance, the owl:Class of "Wine" is a subClass of "PotableLiquid" with a couple restrictions and properties that wine could have in real life.

Why is this useful? It dramatically increases the level at which computers can understand information. In theory, if you tell a computer with this RDF file, "This here is red wine", it will know by inference that the object is a wine that you drink, not that annoying people do, and it will be able to guess at other properites ( such as maker, year made, etc ).

I am intrested in AI and this applies to my job, so this is fantastic news to me.