Is HTML5 Viable Today?

Contributing Editor

We love new technology at SitePoint so it’s easy to be swept along with the excitement and potential of HTML5 — given that “HTML5″ is now a marketing term comprising every cool new web technology. It’s our duty to report the latest techniques so you can adopt them for your own projects.However, our recent poll — “when will you start using HTML5?” — was revealing. Despite the fact the term means different things to different people, the results were:

  • 23% are using HTML5 now
  • 28% intend to use HTML5 soon
  • 46% will use HTML5 when it’s stable
  • and 3% stated they would never use HTML5 (why???)

Risk is perhaps the biggest issue. The specification has not been finalized and could remain a draft for many years to come. No one knows what could happen — that article or header tag you love using could be scrapped. At least HTML4 and XHTML1.0 have been stable for more than a decade.Browser support is another problem. All vendors have jumped on the HTML5 bandwagon, but they have their own interpretations and are yet to implement a consistent set of technologies. Microsoft may have announced comprehensive HTML5 support in IE9, but the browser is unlikely to appear until next year and won’t be available on Windows XP. Developers will need to support IE8 and, to a lesser extent, IE7 and IE6 for some time.Finally, I suspect many developers are reluctant to spend time migrating to a new, largely unproven, and potentially problematic markup language.That said, it is possible to use parts of HTML5 today. For example, you can employ the HTML5 doctype and remove the redundant type=”text/javascript” from script tags. You can also use new tags such as header, footer, nav and article to add semantic richness. IE can be made to understand these elements with a JavaScript shiv or you could avoid styling them altogether.But what benefit does that offer in the real world? Your peers may be impressed and the marketing department can write inaccurate HTML5 promotional BS, but is the additional effort worthwhile? It’s especially difficult when tools such as validators are incomplete or non-existent. I suspect many developers are willing to use HTML5, but what’s the point when you have to implement workarounds and additional code for browsers which don’t support it?However, I do think you should seriously consider HTML5 — especially for new projects. Continue reading “5 Reasons Why You Can Use HTML5 Today”

tip: Learn HTML5 with SitePoint … Live!

If you’d like to get up to date with HTML5 from one of the world’s foremost authorities on the topic, why not sign up to our HTML5 Live course. For only $9.95, you’ll get access to instructional videos, articles, and a vibrant discussion forum where you can learn all of the stuff that matters, without the hype.

Sign up for HTML5 Live today.

Free book: Jump Start HTML5 Basics

Grab a free copy of one our latest ebooks! Packed with hints and tips on HTML5's most powerful new features.

  • http://lukemorton.co.uk Luke Morton

    Can’t understand the nevers… surely the web has to evolve over time. Maybe the 3% are people not expected to see the web evolve in their lifetimes? >.<

  • http://www.hindlist.com afansaad

    Without Any solid assurance it is not wise to adopt the version HTML% right away as it may bring out more worries for webmasters.

  • Wardrop

    “header, footer, nav and article”

    Is this the best the W3C could do to improve semantics? Who was the bright spark who thought “Let’s make the web semantic” yet at the same time thought “but let’s only allow a small handful of semantic tags”? Are the W3C pretending that blogs and news sites are the only websites on the internet, and so they think adding header, footer, nav and article tags are enough? are they not aware of these radical new-age things called “online shops” or “web applications” which have their own semantic requirements?

    My point being, you either embrace the semantic web by providing standardised micro-formats for the most common semantics tags, while allowing custom tags for all other circumstances (XML in other words); or you don’t, and you stick to structural or otherwise “semantic-less” tags like , , and . Clearly, the semantic web is the answer, but I’m over the W3C’s lack of innovation and baby steps when clearly the web is demanding so much more. I wish they’d move away from concepts like “hypertext” and “documents”, and start treating the web in a more abstract way. The less assumptions the W3C make, the more future proof the web will be.

    So it seems that after 10 years (since HTML4), the best the W3C can do is give us a couple more “semantic” tags, do a bit of house cleaning (type=”text/javascript”) and dazzle us with multimedia and the almighty canvas element. Maybe no one’s all that excited about HTML5 because it’s virtually the same as HTML4? Just a crazy thought…

    I’ve lost all hope in the W3C. I guess Google and the Open Source community are now our only real hope for something better…

    • noonnope

      agree.

  • http://www.stunningmesh.com stunningmesh

    I really love to use HTML5, only problem is browsers compatibility… so just waiting for its stability…

  • didgy58

    im all up for using it, ive actually started using it on one of the sites im currently developing.

  • powerpotatoe

    I am still a bit ignorant about the new possibilities of HTML5, but the cool new features that have been demonstrated are not available on all browsers. I am not experienced enough with development yet to know which HTML5 features I can add for new browsers and then mimic in older browsers. In other words, I want to move ahead with new web standards, but until new features become standards in all major browsers, particularly IE, I will continue to use older standards and methods. I am still waiting on something as simple as rounded corners to be accepted by IE, let alone HTML5 (why does Microsoft insist on such a inconvenient work around for rounded corners when a simple line of code would do?).

    I like the demos Apple is showcasing. However, I develop sites which are viewed most often on IE. I do not use IE, but my clients and their audience do. I feel limited by “the man”, i.e. Microsoft.

  • RamboJambo

    It seems that it’ll be some time before HTML5 get’s proper support from the major browsers. That said, Safari are already showing off some complex element of the technology that they support presently.

    I wrote an article recently on whether HTML5 could spell the end for flash, but with fully formed HTML5 not expected until at least 2015, it’s all just speculation for now.

  • Manmohanjit Singh

    I used HTML5 in my mobile site, mainly because of its features like input placeholders, validation etc which is fully functionable on Safari for iPhone and the Android browser.

    On my main site, I still use XHTML :D

  • http://www.deathshadow.com deathshadow60

    My problems with HTML 5 are many… some of the highlights:

    1) Won’t work in IE8, and since there’ll be no IE9 for XP/earlier lord knows how long we’ll have to continue supporting 8.

    2) All but two or three of the new tags are redundant to existing tags or pointless bloat. See “VIDEO”, “AUDIO” and “EMBED”…

    3) It undoes a lot of the progress STRICT gave us; namely SIMPLIFYING the language by reducing the number of redundant tags and moving a lot of the crap that shouldn’t be in the markup in the first place into the stylesheets. Admittedly the number of people who actually embraced doing it properly likely numbers in the single digits… but that’s hardly the fault of the specification.

    4) MAYBE the browser makers should get HTML4 and CSS2 working properly BEFORE moving on to a specification NOT EVEN OUT OF DRAFT?

    5) It’s still a DRAFT, that means cute to play with, but NOT FOR USE IN PRODUCTION CODE! Again though, it seems like nobody seems to understand what terms like “beta” or “draft” mean given the number of people who try to use beta’s for daily use then bitch about it’s bugs as if a beta isn’t supposed to have bugs.

    In general though, It’s like HTML5 is the rally cry for all those who, as Dan Schulz used to put it “The people who used to write endles bloated nested tables for no reason, now just write endless nested DIV for no reason”… To which I’d append “and will soon write endless nested ‘meaningful’ tags that add no new semantics and serve no purpose apart from unneccessary bloat.”

    It’s such disorganized train wreck of a specification it makes HTML 3.2 look good… Though I highly suspect that is the real target audience for it.

    • simonbanyard

      1) Yes it does. CSS3 doesn’t work in IE8. It would be more correct to state that HTML5 isn’t fully supported by IE8 or lower and there are js libraries that can sort that out. We, as a community, have to stop this “but won’t you think of the children!” approach and start educating our clients.

      2) These tags are not just about embedding video or audio. It’s about semantics. Semantically speaking, <embed> is meaningless; embed what? And why are semantically accurate tags necessary? Accessibility…

      3)”Admittedly the number of people who actually embraced doing it properly likely numbers in the single digits…” I think you should show a little more faith! There’s doing it properly and doing it well. The biggest sinners are more likely to be these “$59.99 per page” than those that are looking at the HTML5 spec with any serious intent.

      4) I doubt that we’ll ever see a 100% compliant browser. Why? Because standards are not prescriptive, rather descriptive, therefore they are open to interpretation. The same is true of nearly all standards.

      5) So, if something is in draft and isn’t used by those that understand what it means, how does it progress? How does it become a recommendation? If a standard is in draft, it doesn’t mean it isn’t ready for general use, it means that one must accept that it is likely to change until it reaches the recommendation stage. It’s only by using the draft standard that we understand how it’s going to work — and as a community, feedback to the authors of the standard what is working and what isn’t. You can’t get more open-source!

      “and will soon write endless nested ‘meaningful’ tags that add no new semantics and serve no purpose apart from unneccessary bloat.” I think that that is a little short sighted.

  • Laz

    I guess you’re just teasing us, because the real articles are coming. :-)
    You sure know there is no “risk” and what “draft” actually means here.

    • http://www.optimalworks.net/ Craig Buckler

      Now, Laz, would I do that?
      Yes — you’re absolutely right!
      Stay tuned for today’s article…

  • Sphamandla

    Yes html5 is not stable yet but it is still in its early days so the poeple in the 3% i dont really get why they’d never use html5. Its a pity microsoft is no longer supporting win xp anymore that was a stable and good operating system for me and still is. I think html5 has its disadvantages like not bieng stable and browser support but on the other hand this scripting language provides more suffient programing style with beautiful output which attracts poeple to your site. So i would say i approve of html5 and yes it will improve with time to achieve what html4 has achieved.

    • http://members.accesstoledo.com/naujocke/ erlicthemad

      Dos 3.3 was a stable and good operating system too. But do you really want to use it now? Rock solid, fast, small and easy to use. What more could you want.
      Its time to move on and leave the past behind. Learn from it, but move on.
      Windows 7 is the now and the future of the Windows Platform. If you don’t like that try one on the many Unix/Linux variants or Mac OS.
      Windows XP is long past its useful lifespan and needs to exit the stage gracefully. The problem is there are a lot of people that know it and won’t / can’t learn the new things coming along.
      As long as your work gracefully degrades then give the bells and whistles to the new browsers and when Microsoft with IE9 catches up then they too can join the party.
      Of course I tend to prefer living closer (Just not on, that can hurt) to the cutting edge. There is more money there since you can’t find a $5/hour code monkey doing it. Or a free GUI program that will do the job for you.

  • spheroid

    What if…

    A *new* standard was developed? Why must the world wait for these organizations to develop what they think are standards?

  • Web Design Norwich

    At the moment it is best to use XHTML, i do for all my sites and my main site at http://www.design-vibe.co.uk

    • http://www.patricksamphire.com/ PatrickSamphire

      That’s really a little bit of a spammy comment, don’t you think? :) Do you have reasons for this, or are you just trying to get a link to your business?

  • Anonymous

    HTML 5 is just a markup language. I will use it for documents and use CSS3 features when needed. But for Rich Internet Applications, you need to use something that was developed to handle such uses, which is Flash, ActionScript, Flex and AIR

  • momos

    I had my hopes up for XHTML 2.0, but W3C abandoned that spec, now my new hopes are on XHTML 1.2; Who knows maybe we one day evolve to XHTML 2.0 by moving through 1.3, 1.4, …
    The problem of XHTML 2.0 has always been the lack of “baby steps”.

    • http://www.optimalworks.net/ Craig Buckler

      I suspect the XHTML 1.2 specification won’t go anywhere either. HTML5/XHTML5 is more radical and has the support of all the vendors.

      • momos

        What is so radical about (X)HTML5?

      • http://www.optimalworks.net/ Craig Buckler

        It’s more radical than XHTML 1.2 … which just adds more accessibility options and re-introduces the target attribute for links.

  • James Williamson

    Web Design Norwich, you do realize on your own site that you are serving up your XHTML as HTML right? So, you’re not really using XHTML either.

    Truth of the matter folks, HTML5 has to be backwards compatible, and the draconian error-handling that XHTML 2.0 was serving up would have broken half the web. Just because you code well doesn’t mean that someone else, or certain generated code, does.

    HTML5 is, for the most part, a well-thought out specification that tries to address the fact that web sites are more likely to be applications than sites. If not now, than certainly in the future.

    I don’t think it’s ready for prime-time yet, but then again the spec is only at the last call stage as it is.

    If you don’t like it, please join the W3C or the WHATWG and contribute.

  • buggy pop

    Totaly can translate for one simple word. For money!

  • http://www.jdclark.org/ clarky_y2k

    Personally, what I have been doing lately, is writing HTML 4.01 markup with a HTML5 doctype. Then, I have been using bog-standard div elements with class attributes with values that correspond to the new HTML5 elements (e.g. “article”, “nav”, “section” etc.)
    Not that I’m saying this is an original idea — several web design/development sites have discussed similar techniques — but for now, at least, I believe this is the most forward-compatible way.
    If you ask me, Andy Budd hit the nail on the head with his August 2007 blog post, HTML 4.5 Anyone?.