Slashdot Mirror

Bah! Bah, I say!

You kids never used the WorldWideWeb browser... it had a gui, but no inline graphics.

It'd be nice if Google actually followed the HTML standard and put the <LINK rel="prefetch"> in the HEAD section, instead of in the BODY. Another problem is that most of the time spent downloading a typical page is the graphics, style sheets, scripts, etc. Just caching the page itself doesn't seem to do a whole lot in most cases.

Here's a better question. Remember way back in the day, when search engines were kinda finiky? When we found a cool site, we didn't just bookmark it, we added it to our personal homepage. Along with something to tell people what that site was, and hopefully we made sensible links. How is this better than that?

Google capitalized on that linking, figuring the more people linked to a page/site, the better it must be. Too bad everyone stopped keeping homepages or publishing their bookmarks. Too bad SEO's, spammers, and bloggers figured out there wasn't much linking going on, so the system would be easily tipped. Too bad Google is repeatedly and regularly fooled. For a bunch of guys that are so goddamn smart, they seem to regularly get taken to task...and what are they doing during this? Goofing off with mapping and social communities and webmail and and and and..basically falling into the same trap Apple did many years ago, the same trap HP fell into a few years ago... Overdiversification.

Maybe I'm old, but Netscape stored its bookmarks in an HTML file you could regularly FTP up to your homepage, or something similar. Oh, and back in the day, if you had the time, you could update your homepage a lot. That was kinda like what you kids keep telling me is so "revolutionary"- this whole 'web log' thing.

So pardon while I yawn at this service which..um..does what? Let me post my bookmarks? Which I can do already?

Seriously- the web is supposed to be decentralized. Why do I keep seeing all these people expecting me to put my eggs in their basket? The search engine article earlier today was great- part of the reason Google sucks these days is precisely because we put all our eggs in the Google basket, when there were at least a few other good engines, like Teoma, for example. Google lost the motivation to innovate, because they didn't have to. Frankly, searching these days with Google is like walking down a supermarket baking supplies isle and having people scream at you...and what are those boxes of cereal doing here in the baking supplies?

The first web browser would have been Tim Berners Lee's (along with the first web server) at CERN. This browser was also graphical, so Mosaic wasn't even the first *graphical* web browser. See grandparent.

You might be thinking of the WWW invented by Tim Berners-Lee in 1989 over at CERN. The WWW uses the Internet as a backbone, and is not owned by the US.

All that being said, the UN has proven time and time again to be utterly corrupt and without leadership. How many billions of dollars have dissapeared in one corruption plagued scandal or another? While the US military was busy actually saving lives, establishing safe drinking water and the like right after the boxing day Tsunami, the UN was busy setting up accomodations in luxury hotels for committees to have meetings.

Perhaps you dont like the US or the US military, and this colors your world view. If you don't like it, I have a very simple suggestion for you. Invent your own network, pour billions of dollars of research into it, setup a few international treaties, get the hardware co's to play ball, develop communications protocols, get the telco's on board and dont forget the software companies. Really, if you don't like it, just make your own.

Why do people bitch when we pour billions of dollars into something, spend decades researching it, and get to benifit from it for free? How much of your economy now depends on this thing you have been given access to for free? Have you ever heard of looking a gift horse in the mouth? Is this a case of jealousy, paranoia or just another anti-US rant? Seriously, I want to know.

Or maybe they could use an HTML tag to help.

Oh, wait...there already is one!

US Copyright law no longer requires a notice of copyright on the copyrighted work for it to be protected by copyright law; nonetheless, the XML specification is protected by copyright. Anything someone creates is copyrighted by default under the current copyright regime; one has to explicitly put it into the public domain to disclaim copyright.

However, the documents are licensed under very permissive terms-- see the W3C Document License.

See the W3 IPR FAQ:

2. Who holds the copyright on W3C documents?

The original author of the document. Many documents are created by the W3C and we consequently hold the copyright. Owners who allow their works to be published on the W3C site retain the copyright, but agree to the W3C license for the redistribution of those materials from our site.

US Copyright law no longer requires a notice of copyright on the copyrighted work for it to be protected by copyright law; nonetheless, the XML specification is protected by copyright. Anything someone creates is copyrighted by default under the current copyright regime; one has to explicitly put it into the public domain to disclaim copyright.

However, the documents are licensed under very permissive terms-- see the W3C Document License.

See the W3 IPR FAQ:

2. Who holds the copyright on W3C documents?

The original author of the document. Many documents are created by the W3C and we consequently hold the copyright. Owners who allow their works to be published on the W3C site retain the copyright, but agree to the W3C license for the redistribution of those materials from our site.

Oh, you're saying the *web page* is copyrighted?

Let it go. The *specification* IS copyrighted, irregardless of its presentation, not the freaking web page. Once more for the cheap seats, THE XML SPECIFICATION IS COPYRIGHTED.

See for example this.

If it was not copyrighted, people could, for example, add freaky tag behavior like </BORK/> and call it "XML Extreme". CF., w3c's note: "5.5 I really like the HTML 3.2 specification, but would like to make some changes, may I modify the 3.2 specification in a few places and redistribute it? May I call it HTML 3.2.1? No and no."

And it is NOT copyrighted (according to the W3C).

From the W3C XML specification:

Copyright © 2004 W3C® (MIT, ERCIM, Keio), All Rights Reserved. W3C liability, trademark, document use and software licensing rules apply.

XML is indeed, when used in compliance with the specification, a meta-markup language, like SGML. Using XML DTDs or schemas, one defines a document format for markup. See http://xml.coverpages.org/sgml.html, for instance:

Both SGML and XML are "meta" languages because they are used for defining markup languages. A markup language defined using SGML or XML has a specific vocabulary (labels for elements and attributes) and a declared syntax (grammar defining the hierarchy and other features).

I understand being smacked down sucks, but come on-- stop spouting disinformation to try and defend yourself mmkay?

When people get involved in developing a spec, and at the same time patent things that are necessary to implement that spec ("essential", as the patent lawyers say), and then submerge, wait until the spec is widely adopted, and then announce their patent, this is sometimes called a submarine patent attack.

It's partly to prevent these that we (W3C) have our patent policy, which requires all participants to sign an agreement saying (more or less) they agree to let people implement the spec without paying royalties, even if they own patents that would otherwise apply.

It's all a big mess -- and patents also don't fit well with the GPL, of course, and neither does our patent policy, although FSF participated and we did the best we could: the problem is that you might want to take, say, an HTTP server, and re-use the network code for some other server. But if someone has a patent on servers, to which they have granted royalty free use for HTTP only, you may now have to pay them a royalty for the code.

Patents are intended to encourage innovation by ensuring inventors get royalties. Unfortunately the current system seems to have some disadvantages.

Note: I have no idea whether the slashdot story is correct in this instance about this patent, nor, if the patent is essential to implementing IPv6, whether Microsoft plans to enforce royalties or forbid implementations.

Liam

This page is not Valid HTML 3.2!
..and a few hundred errors

RDF/XML Syntax Specification More here

RDF/XML Syntax Specification More here

RDF/XML Syntax Specification More here

10.5 Until user agents (including assistive technologies) render adjacent links distinctly, include non-link, printable characters (surrounded by spaces) between adjacent links. [Priority 3]

Source: WCAG 1.0

It's even worse when you don't even put any spaces between the links. So stop it.

vector graphics solve the problem of scaling artefacts; they do not, by themselves, adjust the level of detail as you zoom.

Actually, they're starting to. Just look at section 12.3 of the SVG 1.2 spec: "Alternate content based on display resolutions". Not only is it in the spec, but it's shown up on the roadmaps of things such as librsvg. Just two or three levels can replace a slew of bitmap icons.

you say "background: color" rather than "background-color"

Well, I think on /. you could cut him some slack on a typo, given that you missed capitalisation of your third sentence ;-)

To answer your questions:

• IE doesn't support :hover on anything other than links (defined as <a> elements with a non-empty href attribute);
• #eee is a perfectly acceptable colour specification in CSS, being automatically expanded to #eeeeee. See the CSS 1 standard, paragraph 4, sentence 2 (stuff in fixed-width not counting as a paragraph).

I'll accept that very light grey isn't a well-thought out colour if the page has a #fff background...

http://www.w3.org/TR/xhtml11/

If you don't find the answer there, then search around the w3c site - lots of cool stuff - if firefox doesn't conform, submit a bug report, or patch, or a few bucks to one of the maintainers.

Actually, CSS 2.1 has table support. You can have tables completely designed in CSS. Of course, I have this bad feeling that MS will do everything they can to stop this from being supported.

XML is defined as a subset of SGML. From the specification:

"The Extensible Markup Language (XML) is a subset of SGML that is completely described in this document."

#1 -- Sign on if you want to argue anything I say.

Slashdot's anti-bot system prohibits persons with vision disabilities from creating accounts.

what I'm concerned about is finally getting HDTV mainstream... $700 isn't going to cut it.

Exactly. In January 2007, when FCC pulls the plug on analog TV broadcasting, electronics stores and the local TV stations are going to have to deal with a lot of pissed-off viewers who can't get their Super Bowl XLI.

> It's the thing Microsoft fears the most - that
> Javascript will evolve into something powerful
> enough to be able to right [ahem] a
> Microsoft Office clone in.

But it already is powerful enough -- it's just that it's either supported badly or not at all in Internet Explorer.

> It's the thing Microsoft fears the most - that
> Javascript will evolve into something powerful
> enough to be able to right [ahem] a
> Microsoft Office clone in.

But it already is powerful enough -- it's just that it's either supported badly or not at all in Internet Explorer.

> It's the thing Microsoft fears the most - that
> Javascript will evolve into something powerful
> enough to be able to right [ahem] a
> Microsoft Office clone in.

But it already is powerful enough -- it's just that it's either supported badly or not at all in Internet Explorer.

> It's the thing Microsoft fears the most - that
> Javascript will evolve into something powerful
> enough to be able to right [ahem] a
> Microsoft Office clone in.

But it already is powerful enough -- it's just that it's either supported badly or not at all in Internet Explorer.

> It's the thing Microsoft fears the most - that
> Javascript will evolve into something powerful
> enough to be able to right [ahem] a
> Microsoft Office clone in.

But it already is powerful enough -- it's just that it's either supported badly or not at all in Internet Explorer.

Doesn't MS own the patent for CSS, and if so, how does its patent factor in?

To me, saying that "IE does not support CSS2" implies that if you write a site using CSS2, it won't work in IE. The truth is that most of it will work, with some omissions. To me, an accurate example of IE "not supporting" something would be SVG. By itself, IE doesn't implement any of SVG, and if you write a site using it, no matter what you do, it will not display.

Now if you said "IE does not fully support CSS2", I'd say that would be a truthful, accurate statement, not misleading at all. It implies that if you design a CSS2 site, some of it will fail if you use part of CSS2 that falls outside of its support range.

Note that by your definition, Firefox also "does not support CSS2". See here.

I suppose it's the difference between an everyday definition of "supporting" versus a strict definition of "supporting". Seeing your original post, I saw no reason to take it strictly.

When there's no clearly dominant browser to code to, the real standards take on a new gravitas. It becomes an
expectation of the browser developers to fix a page when it looks broken, as it should be. At least... that's what I'm hoping
and expecting. =)

If you can't figure out the format of an XML file (with appropriate Schema or DTD) in 15 years, god help you.

Wow! It's easy to see you've not been in this game for fifteen years.

Let's see, it's easy to figure out the format of a file so long as it's in XML. Provided, of course, that you know the character encoding, of course. And that you've got a device that can actually read the media. And provided you can license the patent on the compression algorithm. And provided the XML doesn't include any binary sections. So, no problems, then.

It's not nesting tables (that's not really bad in my book, it just slowed down old Netscapes) that is the big problem.

It is the fact the Slashdot pages are invalid HTML.

And rather than fix it, or at least address the criticism, Slashdot gives a 403 Forbidden error when trying to use validator.w3.org.

As if that will make us have confidence in the HTML being valid, making it so we can't even see the errors. It would be like buying a car with a sheet over it, and not being allowed to look under the sheet before purchase.

http://validator.w3.org/check?uri=http%3A%2F%2Fsla shdot.org%2F

Undefined Tag and Attribute Names

It is an accepted networking principle to be conservative in that which one produces, and liberal in that which one accepts. HTML parsers should be liberal except when verifying code. HTML generators should generate strictly conforming HTML. It is suggested that where ever practical, parsers should at least flag the presence of markup errors, as this will help to avoid bad markup being produced inadvertently.

The behavior of WWW applications reading HTML documents and discovering tag or attribute names which they do not understand should be to behave as though, in the case of a tag, the whole tag had not been there but its content had, or in the case of an attribute, that the attribute had not been present.

* Camino 0.7
* Mac IE 5.1 & 5.2
* IE 5 Windows (and presumably higher)
* MoZilla 1.3
* OperaSix

Is it actually part of the standard

The class attribute, on the other hand, assigns one or more class names to an element; the element may be said to belong to these classes.

To match a subset of "class" values, each value must be preceded by a ".", in any order.

Example(s):

For example, the following rule matches any P element whose "class" attribute has been assigned a list of space-separated values that includes "pastoral" and "marine":

p.pastoral.marine { color: green }

This rule matches when

class="pastoral blue aqua marine"

but does not match for

class="pastoral blue"

.

* Camino 0.7
* Mac IE 5.1 & 5.2
* IE 5 Windows (and presumably higher)
* MoZilla 1.3
* OperaSix

Is it actually part of the standard

The class attribute, on the other hand, assigns one or more class names to an element; the element may be said to belong to these classes.

To match a subset of "class" values, each value must be preceded by a ".", in any order.

Example(s):

For example, the following rule matches any P element whose "class" attribute has been assigned a list of space-separated values that includes "pastoral" and "marine":

p.pastoral.marine { color: green }

This rule matches when

class="pastoral blue aqua marine"

but does not match for

class="pastoral blue"

.

If that's the case, then it seems extremely unlikely that they would be able to enforce the patent, because of prior art. Check out this link. In particular:

The "style" menu was interesting -- you could load a style sheet to define how you liked your documents rendered. You could also set the paragraph style to an HTML element's style - as lists didn't nest, the user could think of the process as styles (heading1, heading 2, list element, etc) and then this implied an HTML structure when the document was written back.

As this was in the first web browser ever written, it seems rather unlikely they could make much of a claim.

However, CSS was first implemented in IE, if I recall correctly. So it seems as though they had a lot to do with its development at some stage, at least.

Whatever the case, IE's lack of halfway decent CSS support is one of the worst things about the web today. Either MS needs to seriously work on bringing their CSS support up to speed, or a competing, standards-compliant browser needs to take the lead from IE. I would happy with either situation.

> we need a standardize event driven model for the
> web not this hacked together javascript nonsense.

And where have you been for the last 5 years...?

http://www.w3.org/TR/DOM-Level-2-Events

it helps when the guy that invented css is also the cto of the company. :)

On the other hand the w3.org maintains a bang-up bunch of white-papers on all web-related technologies. Nothing say's nerd like volumes of loose-leaf white papers falling out of your attache case . . .

Indeed. One of the W3C's own examples omits the closing TD, TH, TR tags (and of course the TBODY tags).

This is a W3C Candidate Recommendation, which means the specification has been widely reviewed and W3C recommends that it be implemented.

IE chokes on "application/xhtml+xml", which is the problem, because for XHTML 1.0 you're supposed to send it as "application/xhtml+xml", but you can send it as "text/html" if you really want (and most people seem to) and it'll still be OK.

With XHTML 1.1 DTD, you have to use "application/xhtml+xml", and IE is too shit to handle it, which means you either revert back to XHTML 1.0 Strict and send as "text/html", or you incorrectly send XHTML 1.1 with the wrong mime type. To be honest, it's not a massive problem since there are very few differences between the two, so you won't really have to change your code, but still it's rather annoying because neither solution is ideal.

However, using 1.1 and "application/xhtml+xml" is good for the newer browsers, since certainly with Firefox (and probably with others too) it uses the XML engine to render the page, and from my (limited) testing, does seem to be a bit quicker, as well as guaranteeing your page is well-formed, because you will know if it's not!

I use the following piece of PHP to switch between them depending on browser capabilities (note: this is content negotiation, which is very different to evil browser sniffing), and it works perfectly. Unfortunately Slashdot doesn't preserve indenting (and messes up some of the spacing in the strings I'm testing on - weird?), but this should work OK with a minor bit of re-adjustment:

<?php
if(isset($_SERVER["HTTP_ACCEPT"]) && stristr($_SERVER["HTTP_ACCEPT"],"application/xhtml +xml")) {
header("Content-Type: application/xhtml+xml; charset=iso-8859-1");
print "<?xml version=\"1.0\" encoding=\"iso-8859-1\"?>\n";
print "<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.1//EN\" \n\t\"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd \">\n";
print "<html xmlns=\"http://www.w3.org/1999/xhtml\" xml:lang=\"en-gb\">\n";
$doctype = "xml";
} else {
header("Content-Type: text/html; charset=iso-8859-1");
print "<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Strict//EN\" \n\t\"http://www.w3.org/TR/xhtml1/DTD/xhtml1-stric t.dtd\">\n";
print "<html xmlns=\"http://www.w3.org/1999/xhtml\" xml:lang=\"en-gb\" lang=\"en-gb\">\n";
$doctype = "html";
}
?>

The $doctype variable can be used later if there's anything else you need to change based on the type of the page being sent. Feel free to modify the code to suit your needs. Perhaps you want UTF-8, but I like my ISO-8859-1, dammit!

This page also has some useful info on the subject.

The fact is that more and more people are getting to the point that they would rather write for everyone but IE rather than just IE.

True. I usually make sure a layout works in standards-compliant browsers first, then add in CSS hacks to make it work in IE. For personal projects, if a particular feature (e.g. adjacent sibling selectors) isn't available in IE, IE users will just have to live without the extra pretty. For work-related projects, I have no choice but to implement bloated workarounds to mimic what should be simple style declarations.

• The CSS2 recommendation has been out since May, 1998. That's almost 7 years ago.
• The CSS2.1 recommendation has actually been out since February, 2004, more then one year ago.

• The CSS2 recommendation has been out since May, 1998. That's almost 7 years ago.
• The CSS2.1 recommendation has actually been out since February, 2004, more then one year ago.

width = width + margin, not content width = width - margin, as you would expect.
This makes layouts trickey.

The CSS WG realize having only one method is limiting, so box-sizing will likely be in CSS3. Can't wait? Wrap a container around, give that the width and leave padding/border on the inner container.

I want all members of class abc to be the same with, but that width is dynamic dependant on the content

This can't work because CSS has to be able to cascade (inherit values from parent elements). You're asking for elements which could be siblings, ancestors, descendants to all share a common value that, for each, has to be calculated independently. An implementation nightmare, that could be easily done with Javascript (there's nothing that says that CSS must be able to accomodate any whim of a designer).

I want my page body to be the menu width away from the edge and I want the menu width to be the size of the largest entry + a 1em margin. No can do.

Just like table layout, you can use display:table and display:table-cell to do this easily. It was in CSS2.

Dude, if documents can contain any undefined elements and attributes they want and still conform to the spec, then what on earth doesn't conform? You're saying that as long as I include a doctype and a title, I can spew random garbage in tags and it'll still be HTML 3.2 compliant? Where in the reference spec does it say that?

You see the first C in CSS actually means CASCADING, no Microsoft browser has ever handled this (inheritance) properly, none!

The "cascading" part of the name refers to the way the user, user-agent and author stylesheets "cascade" together. Please at least familiarise yourself with the specification before telling others what is and isn't broken.

Internet Explorer has better support for this than Mozilla - you can just pick your user stylesheet in the Accessibility part of the options in Internet Explorer, but in Mozilla, you have to edit a file in ~/.mozilla/[junk]/chrome/whatever and restart the browser.

Nor does the A element specify that it must be closed

Nor does the A element specify that it must be closed