digg

Join DiggAbout

Discover the best of the web!
Learn more about Digg by taking the tour.

HTML5 differences from HTML4

dev.w3.org — This document describes the differences between HTML4 and HTML5 and provides some of the rationale for the changes that have been made to the language.

Bury
show profanity settings
expand all
  • diegoferreyra, on 10/11/2007, -3/+48I like it that it is more strict. And you gotta love tthe new tags!
    • gafasiesornivek, on 10/11/2007, -37/+15HTML5 is so strict that it steps on my balls while wearing high heels.
    • jtrost, on 10/11/2007, -6/+126What's better than the new tags are these dropped tags:
      frame
      frameset
      noframes
    • afx1, on 10/11/2007, -28/+2psh...html is for conformists
    • Seidoger, on 10/11/2007, -1/+46>The 'irrelevant' attribute indicates that an element is not yet, or is no longer, relevant.

      Sweet! Now that can be the top level tag of MySpace and many other sites.
      Mwhaha!
    • Fordi, on 10/11/2007, -5/+6I wonder, is there support for arbitrarily named null tags?

      I mean, I write 'widgets' for fun; essentially HTML tags with scripted behavior that require only the inclusion of a bit of javascript to work. It'd be kinda cool if I could make, for example, a [slideshow][img][img]...[/slideshow] tag for my slideshow widget, rather than [div class=slideshow][img][img]...[/div]
    • LordofShadows, on 10/11/2007, -4/+24Browsers need to inform users that the webpage they're on is in quirks mode to help motivate the 99% of the "web developers" out there still writing netscape html.
    • freddo, on 10/11/2007, -18/+12What I'd really, really, like to see added is the following attributes to anchors, forms and various external sources like embedded objects and other kind of links to another online resource. They would only modify the corresponding HTTP header the browser would send:
      - referer:
      a href="http://somesite.com/" referer="http://anothersite.com"...Click to go to Digg.../a
      a href="http://somesite.com/download/file.php" referer="http://somesite.com/"...Click to download.../a
      a href="http://somesite.com/download/file.php" referer="_host"...Click to download.../a # same as before
      a href="http://somesite.com/download/file.php" referer="http://somesite.com/download/"...Click to download.../a
      a href="http://somesite.com/download/file.php" referer="_parent"...Click to download.../a # same as before

      - user-agent:
      a href="http://www.slashdot.org" user-agent="Lynx/2.8.5rel.1 libwww-FM/2.14 SSL-MM/1.4.1 GNUTLS/1.4.4"...See a light-weight Yahoo!.../a
      a href="http://www.slashdot.org" user-agent="Motorola 1337g"...See a gsm view of Yahoo!.../a

      That way we wouldn't need proxies with funny regular expressions, grease monkey scripts, or extensions like Ref Spoof or User-Agent Switcher... I'm sure those two would be really great tools, and if someone can make something to make those happen... well thanks!

      -- References: --
      RefSpoof: https://addons.mozilla.org/en-US/firefox/addon/4513
      User Agent Switcher: https://addons.mozilla.org/en-US/firefox/addon/59
    • LordofShadows, on 10/11/2007, -4/+10And the obvious question freddo is why do you care what those values hold?
    • freddo, on 10/11/2007, -13/+5may be i forgot the cookie="host:somesite.com/loggedin=true" attribute...

      Add N Edit Cookies: https://addons.mozilla.org/en-US/firefox/addon/573
    • freddo, on 10/11/2007, -9/+5@lordofshadows:

      Because some webmasters rely on those to authorize access on some resources on their website. And sometimes they're just exaggerating it. It's a way to bypass that.

      If i'd have to develop on that i would go about:
      - it's not a "good" way to allow/deny access to someone
      - it can be easily bypassed by methods mentioned in the previous post
      - those propositions would help when you're not allowed to access something by those means and would save time wasted in having to make yet another perl oneliner to access it anyway.
      - the web is for everyone, don't put something online if you are not prepared to see it everywhere
      - what's the point of playing such a cat and mouse game?
    • andr3, on 10/11/2007, -0/+4Check the author's site for more of his views relating to HTML5...

      Anne's blog at: http://annevankesteren.nl/
      (it's in the article, but just wanted to point it out in the open. it's a blog worth subscribing, if you haven't already)

      Kudos to him for his work and for writing this up.

    • wakkow, on 10/11/2007, -2/+14Dropped tag - "font, although it is allowed when inserted by WYSIWYG editor."

      Why allow it for editors?
    • MisterFlaut, on 10/11/2007, -1/+8freddo, that is a horrible idea.
    • fkr3, on 10/11/2007, -3/+8By the time they finish writing the spec for HTML5 and browsers implement it, we'll all be retired.

      /yawns
    • jhshukla, on 10/11/2007, -1/+4freddo, that defeats their purpose.
    • Spatulas, on 10/11/2007, -0/+1According to http://blog.whatwg.org/faq/#schedule the spec won't even be COMPLETED for 15 years (is that 15 years total? if so it's down to 12 years now), and the recommendation will probably take a bit on top of that. The good news is that browsers are already implementing it, but the bad news is that there is no way to tell when IE will implement a functional (ie. can be used as intended to a reasonable degree) version of it...
  • kingfoot, on 10/11/2007, -13/+33The difference is 1...
  • gafasiesornivek, on 10/11/2007, -22/+1So is this 5 differences from HTML4?
  • bonavistask8er, on 10/11/2007, -14/+91I hope they bring back the <blink> tag!!!
    • Nicksname1, on 10/11/2007, -11/+3Good Riddance
    • JaYBrooks, on 10/11/2007, -4/+12was the blink and marquee tag real html? I think they were more of a feature of the particular browser.. Pretty worthless...
    • Seidoger, on 10/11/2007, -1/+64We all loved the blink tag in conjunction with that blood dripping separator, that spinning email animated gif and that "in construction" icon! Good times for your eyes.
    • Fordi, on 10/11/2007, -6/+2Script one. God's nuts, man, it's easy to do.

      I did it here, but a less-than sign nerfed it.
    • TenebrousX, on 10/11/2007, -1/+39Actually, there is a CSS property for blink:

      element {
      text-decoration: blink;
      }
    • mhearne, on 10/11/2007, -2/+7Whenever I see blinking text (or blinking flash, for that matter), I try to scroll it out of the way, even if it concerns a subject that I am interested in. It just gives me a headache, that's all.

      I suppose that it was real code, but I think we'd have to call it "Marketing Language" rather than "Markup Language".
  • bpapa, on 10/11/2007, -1/+66Cool, now we just have to wait a few years for all browsers to implement it, and then another 2 or 3 until people start paying attention and using the new tags and attributes. :(
    • altjeringa, on 10/11/2007, -1/+34My bet is the you'll see full support for HTML5 and XHTML 5 in Firefox, Safari, KHTML, and Opera long before the W3C every actually ratifies the spec.
    • bpapa, on 10/11/2007, -2/+25Sure, but what about IE? As much as we hate to admit it that's when people will start to actually take note.

      And although I didn't read the list over very thoroughly, these new tags (which are great, bringing a lot of rich semantic meaning to a page) are for the most part going to be looked at with a "huh?" from most web developers. The guys worth their salt will be excited to use the new tags, but most of the poorly skilled drek will just continue to use their tables or their divs.
    • EclipseAgent, on 10/11/2007, -7/+2Opera will fully support it, but sites won't comply so that other browsers like Firefox and IE will still render correctly..
    • crazybrit, on 10/11/2007, -6/+4I think KHTML already partially implements it. Good luck on IE EVER implementing it, though - which makes it practically useless.
    • tizz66, on 10/11/2007, -0/+7bpapa: Actually, I have a feeling they'll implement it quicker than you think. I don't think they'll ever drop the web ball again after realising their mistake the first time around. While I think they'll implement it fairly quick, though, I'm worried they won't implement it well enough for developers to actually make proper use of it.

      MS have already started talking about IE8, so who knows.
    • stephdau, on 10/11/2007, -0/+3The original requesting for consideration was delivered to the W3C signed by people from the Mozilla Foundation, Opera Software and Apple Inc.
      http://lists.w3.org/Archives/Public/public-html/2007Apr/0429.html

      On the other hand, MS is on the board of the W3C WG, so they can;t be fighting it too hard:
      http://blog.johnjosephbachir.org/2007/04/25/microsofts-involvement-in-the-html-5-working-group/
    • stockjones, on 10/11/2007, -0/+4@ crazybrit You are exactly right about the IE thing and almost rendering this all worthless. The whole strict WC3 thing seems great to about 10% of the web world. The other 90% of users could care less and don't give a damn if you ask them to update their browsers etc. I just ran into this awhile ago deploying a completely new rebuild of a business to business website for a very large company. Boy did I pay the price for trying to force web standards and require people to update to at least IE7. You all would be amazed how many joe average users are on windows 95, win 98 using IE5, IE6. If we politely asked and even offered to assist people with updating some were very adamant about NOT wanting to upgrade and got very pissed off about it.
    • jhshukla, on 10/11/2007, -0/+1and then its time for html6
  • toxicityj, on 10/11/2007, -18/+12gah. XHTML should be replacing HTML...
    • bpapa, on 10/11/2007, -2/+29Well, XHTML is just a reformulation of HTML using XML syntax.

      HTML 5 actually has a bunch of cool new tags that add better semantic meaning to documents. XHTML 5 would be it's XML counterpart.
    • Slapo, on 10/11/2007, -2/+10While I do agree that xhtml should be replacing html, you might want to read these parts:
      "The HTML5 draft reflects an effort, started in 2004, to study contemporary HTML implementations and deployed content. The draft:

      1. Defines a single language called HTML 5 which can be written in HTML (HTML5) and XML (XHTML5).
      2. Defines detailed processing models to foster interoperable implementations.
      3. Improves markup for documents.
      4. Introduces markup and APIs for emerging idioms, such as web applications. "

      and

      "The other syntax that can be used for HTML5 is called XHTML5. This syntax is compatible with XML and correct XHTML1 documents. Documents using this syntax need to be served with an XML MIME type and elements need to be put in the http://www.w3.org/1999/xhtml namespace following the rules set forth by the XML specification."
      So it might not be quite a bad thing.
    • Jugalator, on 10/11/2007, -1/+15"and XML (XHTML5)"

      Ugh, with XHTML 2 already in development too (with its own set of new tags and guidelines), way to cause confusion there...
      • akaroa, on 06/19/2008, -0/+0XHTML5 is actually the replacement for XHTML1 and XHTML2, but unfortunately the W3C doesn't want to admit this just yet. Any good ideas from the XHTML2 spec have already been added to the (X)HTML5 spec so there simply isn't any need for XHTML2 anymore, it's become outdated and irrelevant to the web of today, which is probably a result of it being developed in a closed group without input from the public and browser vendors.
    • mabhatter, on 10/11/2007, -1/+8I agree, they should push HTML5 to be a proper "XHTML5" spec as well. Beating people up to properly close tags is a small thing and promotes good coding. The main problem with XHTML is that one certain older browser refuses to properly handle the XML type for a page in XHTML. The better browsers all work fine. What you gain is the ability to mix and match other types of markup... for instance the XHTML + SVG + MATHML combo is really neat because you can put scriptable drawings and formulas right in the page... 100% pure text markup, no plugins or graphic gimicks!! but nobody uses it because one certian browser refuses to implement the XML data types properlyl
    • diegoferreyra, on 10/11/2007, -9/+5To all of you sayin git should be XHTML5 instead of HTML5.

      None of you have used XHTML and seen the results. When you have the "text/html" the browser treats your page as if it was HTML, even if you define the doctype as XHTML.

      There are several articles saying why you should use well formed HTML instead of XHTML. READ THEM.
    • SocialPoison, on 10/11/2007, -2/+4@diegoferreyra
      Haven't had any trouble here...
    • strictnein, on 10/11/2007, -2/+4@diegoferreyra:
      Are you able to use Digg.com?
      !DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
      "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"

      I'm really confused on what you're trying to say.
    • kday, on 10/11/2007, -1/+5@strictnein

      What he is saying is that 99.999% of websites are served in text/html, and almost nobody uses the XML advantages in xhtml. Browsers like IE prevent developers from using xhtml served as application/xhtml+xml.

      Check out this tag in Digg's source:
      meta http-equiv="Content-Type" content="text/html; charset=utf-8"

      So, yes, digg is using an xhtml dtd, BUT the content is being served in text/html.

      Also, if you serve xhtml applications as xml, your page will break if there are any errors in the code. While this is probably a good thing overall, and promotes better coding, it takes more effort.

      If you want to see what a website is like served in application/xhtml+xml, check out the Force Content-Type extension for Firefox. You'll see that websites with non-valid xhtml (most websites) simply won't render, and you will see an error.

      https://addons.mozilla.org/en-US/firefox/addon/3207
  • toxicityj, on 10/11/2007, -4/+39"HTML5 is so strict that it steps on my balls while wearing high heels"

    God forbid web standards be strict.
  • enchanterku, on 10/11/2007, -9/+4So they changed how bold and italics work? That blows.
    • eleven, on 10/11/2007, -1/+4No they just changed how they are defined. The "works" part is / was really up a page's author.
    • bpapa, on 10/11/2007, -1/+2Yeah... the new b and i are kinda weird. They sound a bit too presentation-ish when compared to emphasis and strong.
    • enchanterku, on 10/11/2007, -3/+1@eleven:
      I understand that you(as the page author) can control how it works using a stylesheet, but why change the spec? That's what I don't understand--if it's not broke, why fix it?
    • altjeringa, on 10/11/2007, -1/+5Think 508 compliance. Sometimes you want to make something bold for visual presentation but it's just cause it looks good not because it actually needs emphasis.
    • enchanterku, on 10/11/2007, -1/+2@altjeringa:
      Good point there, I try not to gratuitously use the tag, but it could happen
    • digghasnoethics, on 10/11/2007, -11/+1In general a load of incompatible intellectual masturbation by a bunch of out of touch computer scientist types who never really understood reality anyway.

      If you seriously think anyone will be dumb enough to drop the existing and well used meanings of b, i frames, etc. then you think IE is bug free.

      HTML as defined originally work precisely because it did what people thought needed to be done, not playing games to be intellectually correct.
    • tdous, on 10/11/2007, -0/+0@ digghasnoethics
      Your work, if you have ever actually done any web design, must suck.
  • coldskool, on 10/11/2007, -4/+10The article says they are dropping frame and frameset... huh? I rarely ever use frames but dont a lot of websites still use them?
    • bpapa, on 10/11/2007, -1/+12Frames have been "dead" for a long time for a number of reasons. Of course, a number of websites still use them - either OLLLLD sites or those made by people who make poor design decisions.

      You don't have to worry though. It's not like browsers will support only HTML 5 and stop with the previous versions. So all of the old Frame-based sites will continue to work (as poorly as they always have. :P)
    • Darthypoo, on 10/11/2007, -1/+1Web browsers will still render them. I don't think anyone would just drop support for all the previous versions of HTML.
    • chrislongridge, on 10/11/2007, -3/+9They are dropping frame and frameset because they are obsolete and have no place in modern web design.
    • altjeringa, on 10/11/2007, -2/+3I wish they would just drop them completely because Frames have ALWAYS been for the stupid and lazy. But if you want to use them all you'll have to do is declare your document
    • airship, on 10/11/2007, -10/+5When browsers drop support for frames, you'll no longer be able to view my site from 1995, archived on the Wayback Machine:

      http://web.archive.org/web/19980704013043/www2.giant.net/people/mbrown//
    • coldskool, on 10/11/2007, -2/+10Im not making a case for framesets... only time I use them is when Im asked... but still, I dont see the point in dropping them, its just another option right?
    • enchanterku, on 10/11/2007, -2/+4@altjeringa:

      I would have to disagree- I'm not particularly stupid (or so I hope), nor am I particularly lazy when it comes to code. There are a few instances when frames were acceptable-not always preferable, but useful.

      For example, I was told to incorporate a company PowerPoint into a page, as well as including some other content. A framed layout, coupled with some CSS, worked perfectly for the solution.
    • bpapa, on 10/11/2007, -3/+2Well, they are a poor tool. No point in keeping something that doesn't work well as a part of the tool box.
    • gsnedders, on 10/11/2007, -1/+1The UA conformance requirements require UAs to support elements like frames, though they aren't conformant within an HTML5 document.
    • jerbaker, on 10/11/2007, -3/+2The poor design idea is fixed position DIV elements. Although it's a "cool" trick, it suffers HUGE drawbacks over frames:

      1) If you don't want to use SSI (you do know that SSI-generated pages can't be cached, right?), you have to maintain a the DIV on each and every single
      page in your site.
      2) It's annoying trying to work around IE6's complete inability to deal with them.
    • Bamborzled, on 10/11/2007, -0/+7@jerbaker

      You could always use absolute-positioned DIV elements and IFRAMES; they haven't dropped those.
    • jerbaker, on 10/11/2007, -0/+1So you're saying that is better than . It's a distinction without a difference.
  • doubleforte, on 10/11/2007, -10/+1HTML 5...wouldn't that mean that Web 5.0 isn't far off? For that matter, why aren't we using Web 4.0 right not? What's with this Web 2.0 crap?

    In another funny, my boss used the phrase "first generation Web 2.0 websites."

    Now seriously, which browsers are planning to support this? Safari 3? FF 3? (We all know that IE 10 should be in partial compliance.)
  • Derferman, on 10/11/2007, -1/+2When can we expect HTML5 to be complete? How about browsers supporting it?
    • altjeringa, on 10/11/2007, -4/+13Given the Browser vendors wrote the spec everybody but IE will probably support it in the next version of their browser. Then about a year later it will become an official Specification. Once that happens IE will complain that it was inadequate and so they developed their own technologies which are incompatible.
    • gsnedders, on 10/11/2007, -2/+10Given that Chris Wilson, the platform architect of IE, is chair of the W3C HTML WG, and has said several times he will implement it, that's just FUD.
    • altjeringa, on 10/11/2007, -2/+9@gsnedders

      Given that everybody but IE was involved in creating the spec and a major public outcry had to be voiced before the w3c would even consider it, given Microsofts history of ignoring or subverting public standards and that even with it's "Great push for standards compliance" IE 7 is still the most incompliant piece of crap browser on the web, I'd dare say my statement holds. Please feel free point out some verifiable data that Microsoft has ever played well with others, adopted standards and you can say my statement are FUD all you like. But a representative of an organization which has on multiple occasions has said it would support what it then subverted, say thing it will support anything is quite meaningless.
  • jamesallen74, on 10/11/2007, -5/+1Webpages back in 1994...ahhh I miss those days. Mosaic on my Intel 486.
  • Shambla, on 10/11/2007, -7/+1But...we're only on Web 2.0...
  • qbkooky, on 10/11/2007, -5/+4Even though I know "div align=center" is better.... I'm kind of sad to bid farewell to my old friend, the "center" tag ...
    • bpapa, on 10/11/2007, -2/+7Actually just saying "div" is better and then having the alignment in a CSS file. Or, using a more semantic tag than "div", if possible.
    • Mearn, on 10/11/2007, -0/+4Wasn't the center tag depricated a while back?
  • cogent, on 10/11/2007, -1/+6I see some of the tags being a little too vague. I would definitely use the "header","footer" , and "nav" tags but isn't "section" just another "div"?

    But I absolutely love the new form type attributes. If the browser can do the error checking for dates and emails instead of my lazy ass. That would be awesome.
    • HigherLogic, on 10/11/2007, -0/+1Actually, the section tag makes so much sense, it would be great for content and sidebars. The div element, by its definition, is used for content that doesn't have a semantically equivalent tag. It's just for generic items.

      One of the first few posts, Fordi said he'd like to be able to create null tags like <slideshow>. That's generic though, it belongs in a div.

      Really looking forward to this (and CSS 3).
  • jammnrose, on 10/11/2007, -2/+4I can't wait for HTML5. The spec has so much stuff that will make my life easier. Bring on teh canvas.
    • Takteek, on 10/11/2007, -0/+1I wonder if it's bad that I'm more excited about this than if I had won $100....
  • uptown, on 10/11/2007, -3/+2It's job security 101.
    Break things that may be less than desirable, but function (i.e. frames) to ensure that there's a continual stream of new jobs to fix the sites that are now broken.
  • DiggDagger, on 10/11/2007, -4/+5And, as usual, Internet Explorer will support the new version after 5 years of its release date.
    • Dracos, on 10/11/2007, -2/+11IE doesn't fully support anything, not even HTML 1.0. Saying IE will support any new spec within 5 years is highly optimistic.
    • anphanax, on 10/11/2007, -0/+1Dracos, no modern browsers probably support this "HTML 1.0" you speak of. As of HTML 2.0 (the first spec I am aware of), there were deprecated elements like <xmp>, <plaintext> and <listing>.
  • MavRevMatt, on 10/11/2007, -5/+2I'm gonna miss the good old lazy center tag...And the old b and i...That just blows.
    • rnelsonee, on 10/11/2007, -0/+1Yeah, they're good when you're feeling lazy, but it's good in the long run. Separation of content and style is a great idea. It's not like HTML has been able to help with style in any serious manner anyway, so you might as well get rid of any tags that try to mess with the style anyway.
  • vh1`, on 10/11/2007, -1/+3`datalist` elements? easy combo boxes? yes!

    "Likewise, the target attribute for a is no longer deprecated."
    one of my biggest gripes

    autofocus attributes. the HTML5 spec is wicked!
  • uzytkownik, on 10/11/2007, -1/+4What is the XHTML 5.0? I've heard about HTML 5.0 and XHTML 2.0. Did the naming is changed?
    • altjeringa, on 10/11/2007, -1/+6XHTML 2.0 is the next version of XHTML 1.1. It was developed by the W3C. XHTML 5 is the XML compliant version of HTML 5, it was developed by a group of people who believed the W3C had it's head up it's @$$ and then brought the the W3C for standardization. The names are confusing but that will probably be cleaned up before the Specifications are finalized.
      • akaroa, on 06/19/2008, -0/+0XHTML5 is the next version of XHTML.
        Everything that can be done in XHTML2 can be done in XHTML5 like XForms for example, plus you have all the new features like the video, audio, and canvas tags. Also, all of the good browsers already support some of XHTML5 and are working everyday to increase this support.
  • skags, on 10/11/2007, -0/+1From the WHATWG blog faq page:

    " Why do we need both HTML 5 and XHTML 2.0?

    We don’t. What the wider Web community needs is a language which, if implemented by a Web browser, will result in a browser that can render all the existing content on the Web, and which will have new features to make the Web a better place. XHTML2.0 is not such a language — a browser that only supports XHTML2 could not render existing Web content correctly. "

    What? How is throwing another new language out there going to help?
  • skags, on 10/11/2007, -0/+2Also from the WHATWG blog faq page:
    " Is XHTML better than HTML?

    [Some people say XHTML is better because… But see the next question.]
    Is HTML better than XHTML?

    [Some people say HTML is better because… But see the previous question.] "

    Cute, but the confusion is ***** up the interwebs.
  • freddo, on 10/11/2007, -7/+1
  • knugen, on 10/11/2007, -0/+2"The value attribute for the li element is no longer deprecated as it is not presentational. The same goes for the start attribute of the ol element."

    I never understood why they took this out in the first place... It's not allowed in XHTML either, as far as I know.
  • arctic, on 10/11/2007, -4/+4XHTML FTW!
  • jerbaker, on 10/11/2007, -1/+10Great, they're going to implement the "ping" attribute to anchors. Now Web pages can intentionally or unintentionally inform an arbitrarily long list of external URL's just which link you are clicking without your knowledge. Imagine the possibilities of using the "ping" attribute along with query strings to pass information from a compromised Web page. You think phishing scams are bad now??? I can think of a million reasons why this shouldn't be implemented and only one reason why it would be (Google has their hand in this one).
    • thomashauk, on 10/11/2007, -1/+0Cause it not like they use redirects to do this alredy....
    • jerbaker, on 10/11/2007, -0/+1Except you can remove redirects, and you can see them when you hover over the link. PING is a way to take it behind the scenes where you won't know unless you examine the source for every page you visit before you click.

      I smell a Firefox extension coming to scrub the ping attribute from anchors.
  • braziland, on 10/11/2007, -0/+3LOL, you can tell this is a rough draft:
    "...following new values which enable a bunch of new native controls people can use:"

    I am happy that they are moving away from the current DIV hell and working towards building a true semantic structure for HTML documents. I can definitely see these tags improving the experience with screen readers.
  • punkrock4life, on 10/11/2007, -2/+1XHTML is the future. I can't wait for XHTML2.
  • w3bsmith, on 10/11/2007, -2/+1Ping, ping, ping ... [ http://dev.w3.org/cvsweb/~checkout~/html5/html4-differences/Overview.html#new-attributes ]
  • Dracos, on 10/11/2007, -2/+4Tag soup is a mistake, and the excuse of lazy developers everywhere. If this new spec was well-formed and xml-compliant *only*, I would be much happier. Some people complain that XML/XHTML is too bloated, and to them I say: suck it up. 60 years ago, cars didn't have computers, anti-lock brakes, or seat belts, and most didn't have radios or air conditioning. Are today's cars bloated?

    Some of the changes are good, some are bad, some are lame, and some are ripe for exploit by marketers.

    The ping attribute? Please. Firefox better get an extension pretty damn quick to limit pings to same domain.

    Why is the i tag back, and with a subtly different meaning? It's no more semantic than it was before.

    The dialog tag? Come on. This is obviously someone's pet tag. And it makes the dd and dt tags ambiguous.

    Aside? Jeez. Why is small not dropped like big?

    Footer and header? Lame.

    There would be a lot less new tags here if they had adopted the role attribute (pure genius) from XHTML2.

    This spec is ill-thought out, bloated, and inconsistent. I can't believe people smarter than me are buying into this.

    Long live XHTML2!
    • Daedrik, on 10/11/2007, -0/+2A short Greasemonkey script would clear out all ping attributes on links nicely:

      var a = document.getElementsByTagName('a');
      for (var i=0; i<a.length; i++) a[i].setAttribute('ping','');
    • jerbaker, on 10/11/2007, -0/+1I would prefer:

      var a = document.getElementsByTagName('a');
      for (var i=0; i
  • jjb123, on 10/11/2007, -0/+1Just wondering, why are they dropping the center tag and the name attribute?
    • anphanax, on 10/11/2007, -0/+2The center tag does not make any semantic sense (it describes layout, not the purpose of the content). Use "margin: 0px auto" (Firefox, Safari) and "text-align: center" (MSIE) on blocks. There are other methods (especially for the task of getting things vertically centered/middled/whatever), but those seem to be the most common.
    • tybris, on 10/11/2007, -3/+1They think HTML and CSS should separate semantics from style. However no one really cares about your semantics and CSS is such a poor styling language that your HTML documents will still be full of meaningless styling tags.
  • vertigorider, on 10/11/2007, -2/+0seems like theyre trying to let the basement programmers back in the game with "simple" tags for people who can't/dont want to use CSS. a div and a css id/class could replace more than half of those new tags
    • rnelsonee, on 10/11/2007, -0/+2No, that's incorrect. Every one of the new tags deal with defining the content - it wraps around the data so you know what type of data it is and give it context, and therefore you know how important/relevant the information is. The classic example is if you're viewing the page on a laptop, you'll see the full content. But if you're accessing the page on a device with limited real estate, like an iPhone, then the page can dynamically adapt to the device and not show comments or asides on the first load.

      For style, you're supposed to define the style for each type (comment, aside... etc.) in CSS. Having these tags makes it easier so you don't have to use hacks like the "invisible" attribute.
  • fLUx1337, on 10/11/2007, -0/+6I'm more intrested in CSS3...

    I want my damn multiple backgrounds per div NOW!!! Not in 3 yrs when ALL browsers use it...

    Apple Safari has it, but theirs no point using it, because I would be writing extra code, just for 5% of web users...

    Ah well....
  • deadowl, on 10/11/2007, -0/+1Thank god nav, article, aside, header, and footer tags are finally coming around. I think those will GREATLY improve the internet. Let's not forget to mention, the biggest complaint I hear from blind people using the internet is not being able to find the content immediately, and navigation always getting in the way. So yea, this helps make things easier for web developers in, "What the hell tag should I use for my navigation?"

    I mean, essentially, web design takes about the same form as newspapers, but with additional media capabilities. This definitely better represents how people design their pages.

    The one thing I don't like about this however, is that I can't find anything for content flow. I mean, when working with Qt or Java for designing GUIs, You have flow layouts, directional layouts (north, northeast, etc.), column layouts, row layouts. Why no attribute for how different blocks of stuff line up inside of an element?
  • callingshotgun, on 10/11/2007, -0/+0I'm kind of confused by the dialog tag. dt and dd don't appear to use closing tags. Are they optional, or is there a set of tags for which no closing tag exists? I thought HTML was migrating toward a stricter system for what it means to be well-formed. I mean, it makes sense to get away with it for things like paragraph/line breaks, because those can be considered atomic. But Seriously, confusion. What's with the no-closing-tag system?
  • stephdau, on 10/11/2007, -0/+1Interesting: An HTML5 Conformance Checker
    http://hsivonen.iki.fi/thesis/html5-conformance-checker
  • instabil, on 10/11/2007, -0/+2canvas and event-source... sweet!
  • tybris, on 10/11/2007, -0/+1HTML is not the problem... fix CSS!
    • spliffy, on 10/11/2007, -0/+1but this article is about html, not css.
  • betona, on 10/11/2007, -0/+1Whoa, had to look it up, but HTML 4.01 was released on 12/24/99, making it ancient in internet years. I didn't know there was a v5 movement coming--thanks for the heads-up.
  • redwallhp, on 10/11/2007, -0/+1HTML is fine currently. It's the CSS that needs fixing. I do like some of the new elements though. Especially the "figure" element.

    So, Firefox will support it 6 months after the standard is approved, but IE won't for another 5 years. That's how progress happens on the web.
    • spliffy, on 10/11/2007, -1/+1don't be a noob, please see my comment located 2 above this one.
  • ZoeBee, on 10/11/2007, -0/+0Diegoferreyra's actually right on this one. To clarify: an XHTML document is only valid if the HTTP header for its content type is set to "application/xhtml+xml", not "text/html". The reason that, for example, Digg loads in IE while XHTML documents with the correct content type don't is because its content type is for HTML, not XHTML, regardless of what the HTML headers (not to be confused with HTTP headers) say.

    The W3C information is here:
    http://www.w3.org/TR/2002/NOTE-xhtml-media-types-20020430/#summary

    The essay on why you should be using HTML until IE sorts out its XHTML content type recognition is here:
    http://www.hixie.ch/advocacy/xhtml

    You can verify digg.com's HTTP headers using the command "curl -I digg.com". Note the line "Content-Type: text/html; charset=UTF-8".

    Zoe.
  • oooshola, on 10/11/2007, -1/+0This is all nice, but I'm still waiting for browsers, HTML, and CSS to ALL adopt a standard way to make a 3 column, equal height layout with the ability to put up background images. It's sad that after all this time, this extremely prolific layout can only be achieved through hacks and fixes, sometimes even only with the help of javascript.
    • oooshola, on 10/11/2007, -1/+0Before you reply, note that I'm talking about achieving this without the use of tables.
  • starkraving, on 10/11/2007, -0/+0Having an <input/> tag that can also support a linked set of options is a great addition. How the combo box got forgotten when HTML forms were first developed is beyond me.
  • pfwd, on 10/11/2007, -0/+1Sounds good. I'm glad that elements are now starting to become tighter in their uses. Would be good to know when browsers are going to implement this as it looks like a good step in the right direction.
  • yogastore, on 06/27/2008, -0/+0http://astore.amazon.com/holmes.tower.fan-20
    http://astore.amazon.com/honeywell.tower.fan-20
    http://astore.amazon.com/10.cup.rice.cooker-20
    http://astore.amazon.com/zojirushi.10.cup.rice.coo ...
    http://astore.amazon.com/bug.zapper-20
    http://astore.amazon.com/rat.zapper-20
    http://astore.amazon.com/250gb.external.hard.drive ...
    http://astore.amazon.com/500.gb.external.hard.driv ...
    http://astore.amazon.com/surfboard.cable.modem-20
    http://astore.amazon.com/wireless.cable.modem-20
  •  

    Login or register to add a comment

    Create a new account or login to join in the conversation on Digg. You'll also be able to Digg stories to help promote things you like.


© Digg Inc. 2008 — Content posted by Digg users is dedicated to the public domain.
DIGG, DIGG IT, DUGG, DIGG THIS, Digg graphics, logos, designs, page headers, button icons, scripts, and other service names are the trademarks of Digg Inc.