What do you think folks? I’ll hold off on any opinions until we wrap it up. Poll is over in the sidebar.
New Poll: Would you start a new web project today and not use HTML5?
Chris Coyier
on
(Updated on )
DigitalOcean provides cloud products for every stage of your journey. Get started with $200 in free credit!
I tried going with HTML5 a couple of weeks ago. Problem – the then-current stable version for Firefox didn’t recognize it – and let’s be honest, folks, most common people don’t go looking for betas.
Hell, they don’t even go looking for Firefox.
So I had to ditch it.
That’s weird, Firefox recognizes the “simple” HTML5 features and has for a while. It doesn’t even need the JavaScript shim for the new tags, as IE does.
So, either you’re talking about some “advanced” HTML5 feature, or you might have done something wrong.
I dont think its a good idea to use a negative question in a yes/no poll… Could lead to confusion and fake results.
This is clearer for example:
If you started a new web project today would you use HTML5?
Yes/No (Completely clear)
It’s also confusing to leave a comment as a reply that isn’t a reply =)
But I agree, that’s more clear. The reason for the negative is I was hoping people would comment on why they would NOT use it, if that’s the case.
If possible, amending the answers to be ‘No, I would always use HTML5’ and ‘Yes, I would not use HTML5’ would mostly fix that issue.
‘the then-current stable version for Firefox didn’t recognize it’
I’ve been using HTML5 for quite some time now. The ‘not recognizing’ part is incorrect, if you add display block in your css for the html5 block elements everything is working
header, hgroup, section, article, aside, nav, menu{
display: block;
}
I was having the same issue with Firefox… Display Block saved the day…Thanks
It would totally depend on the product. If it was something I expected to launch within 6 months, and not just to the techie crowd, I’d definitely go with older tech.
If it was something that I expected to launch 3 years from now, or only to the techie crowd, I’d probably use HTML5 and all its magic.
Who starts a web project that takes three years to launch?
Yeah seriously, my web projects take 3 months at most…
Web apps and their associated frameworks can take a very long time to mature. 3 years is probably an intentional overstatement, but something like 18 months is not atypical of a large-scale web app.
I work for a retailer with a large ecommerce presence and multi-year projects are not all that unusual.
That said, my answer is yes simply because so many projects are within existing frameworks. Reserving the question simply for projects I’m starting which require a new framework, the answer is no.
I work building websites for the government sector so I would start with HTML5, but legally I must make sure the website works in ie6. So some of the elements I have to skip, for example the new semantic structural features I use divs with classes using the structural names (e.g article and section).
I use the tag with fall back to flash to make video accessible to a wider range of people.
Personally it is all about making the content more accessible to a wider audience.
Much of HTML5 can be easily “made” to work with old browsers, including IE6. In fact, the new semantic structural tags are among the easiest to support across all browsers since they don’t require the browser to provide functionality. Most of the HTML5 resets provide the needed JavaScript to have IE6 reading HTML5 structural tags like a champ.
I definitely don’t fault caution, but it’s important to clearly understand how we can move the web forward while still supporting older browsers.
I would go for it, in fact, I am going for it. Just for using the stuff that already works across browsers, and for the new cool features where they apply.
Learning to effectively use HTML5 is not only good for the web, it’s good for you as a developer.
I’ve already used it for multiple high-profile projects for Staples, Red Hat, America’s Test Kitchen, & more.
Don’t want to learn it? Great! Give me your employer’s contact information & I’ll take over your job.
Dale Cruse, author of the upcoming “HTML5 Rich Media Applications Cookbook” for Packt Publishing.
not to sound like a total d-bag but perhaps you should make sure your site is working in Firefox (3.6.12) before threatening to take other peoples jobs. it’s not working. at all.
True nothing loads in FF you should fix that
That’s interesting. It’s working on Firefox 3.6.12 on both Mac & Windows platforms here. What are you seeing?
nothing but the background shows. when viewing source, it seems as if one of your cond. comments isn’t working/closing correctly because all of the html is “commented out”. i’m on an XP machine using 3.6.12
LOL Dale you came off like some arrogant tool with your comment. Try not trying so hard next time.
I’m willing to admit when I’ve put my foot in my mouth but that doesn’t change the idea that if you don’t want to evolve, sooner or later your employer will find someone who will. None of us want to be a dinosaur, much less a dinosaur with a big mouth.
I can fully agree with the whole “being left behind” thing. no one wants that. but at the same time, we can’t be so fast to abandon tried and true technology that works. the cutting edge is just that, an edge. on one “side” you have fancy functionality and semantic code but on the other “side” of the edge is the risk of having issues like this, even in established, popular browsers.
Oh and about your site… fonts are all unreadable on Chrome… unless I zoom 144%.
Sorry for peeking, but shouldn’t there be some meta tags too? Though with your list of clients you probably don’t need them I guess *shrug*
Dude, if you can convince some of our clients to ditch their IE6 and FF2 browsers and peddle your HTML5 sites then I’ll gladly give you their contact info and I’ll eat my shoe because that’ll be quite the feat. I’ll throw my hat off to you.
his site works fine in chrome 9 beta
it loads slowly, and it isn’t very exciting, but it looks fine.
However, most of the site you feature are only “HTML5” by virtue of the fact that html5 is backwards-compatible. Well-formed and valid html4/xhtml1 is valid html5, as long as you use the new doctype and leave out a few depreciated tags.
That’s more-or-less how I see html5 -for now- anyway. use the parts that work; be ready for the cool stuff later.
Dale’s site displays fine in FF3.612, Chrome 8.0.552.215 and safari 5.0.3 on my Macbook.
html5 works fine for any project you want to undertake now. Shim it for older browsers, and allow the fancy stuff to degrade gracefully and you’re good.
Still haven’t needed to use any of HTML 5 in of my projects as of yet.
I worried though that people will see this poll question and think CSS3 and not just HTML 5
You will never NEED to use HTML5, but the question is, will it benefit you to start using it, and the answer is probably yes.
Sadly, yes. I *might* /not/ use HTML5 in a new web project. There are still some things that you cannot get to work properly in IE6, even if you do use Modernizr (or other) to force older browsers to understand the newer tags. Also, older browsers have older JS engines and there’s just not much you can do about that, except avoid newer JS constructs :(
I see no reason why not to use HTML5. Just because it introduces a whole bunch of new bells ‘n whistles doesn’t mean you have to use them all.
Doctype and common elements, header, footer and stuff :)
Yes, me too! I haven’t started a project yet without these simple changes. They are lifechanging.
I submitted Yes [I would start a new project without HTML5] but forgot that even using the new doctype and nothing else would still count.
If the poll is about using new features that have been added since HTML4.x, then I would have to consider it in the context of the project.
I develop a lot of internal web applications for my employer, and they are standardized on IE, so going to HTML5, is fairly pointless at this point. Even the stuff we have customer facing is 95% IE users according to our analytics package.
I still haven’t got around to doing a test project with HTML5 yet but I’m hoping to do that soon.
To be honest I think I would use it for a new project despite IE6 not working properly with it. If people keep trying to cater to users of IE6 it’s just another reason for the users/corporations to leave it in place.
The sooner more things stop working with it the sooner users will be forced to move away from it.
I totally agree with your assessment of IE6! This browser is older than old in the internet age. People just need to upgrade. It seems that the term ‘accessibility’ has almost become synonymous with ‘compatible with IE6’. Why? People can just upgrade their browser to get with the technology.
I am slowly integrating HTML5 into my projects. But until a client specifically requests the site to be compatible with the iPad, I don’t see the point. Since the site code I use tend to work just fine.
I like the features, and will certainly be integrating it more and more.
Well I may not be doing governmental work but I do live just about exactly 3 miles from the middle of nowhere, so I struggle daily with wanting to do progressive web designing and also realizing that the local websites are more being viewed on ie6 and being found on dial up internet. I think it important that we look forward to the newer, and much better, coding practices but we need to keep in mind the end users and the equipment they use to view my designs.
I am a proponent of learning as much as I can about my trade but I still need to make sure I am not alienating my core users by producing something they cannot use.
Kevin
That would make for an interesting article… “Best Practices for Rural and Third-World Web Design: How to deal with dial-up in the New Web Era”
I’ve been developing web apps for the financial sector, our analytics report that i.e still has a 60% market share of our main site, but for our web app, its 33% IE. By implementing HTML5 gives us huge benefits, but we also have to fall back to old techniques.
I start all my projects with html5, and use the basic features for most sites.
Why not use HTML5/CSS3 today?
We always had to cook a separate meal for IE in some cases.
In my opinion we should build websites for current browsers in first place and then use workarounds for older browsers.
If i need to use an image for a shadow in IE678 – fine, but why bother new browsers with an additional (and not needed) request?
Seconded. :D
I am using it everyday thanks to http://html5boilerplate.com It is very easy to do and works in all browsers, at least it is easy to provide fall backs with modernizr.
Cheers.
True! I second that.
+1 for Boilerplate. Easy to implement, well documented, supported and maintained. Unless a project is specialist, or, intranet for the lovely IE6, I’d definitely use Boilerplate and HTML5.
Depends what you mean by HTML5 I guess. If it’s simple HTML 5 elements (section, header, aside, etc) then sure thing. If it’s something more advanced such as web sockets or offline mode, then the answer would probably still be yes, but with more investigation needed.
HTML5 works in all browsers now (even IE6), even if it needs a little shivving, so can’t understand why we’re resorting to the age old “It’s not supported” bollocks.
I was coding in xhtml 1.0 strict until my last project, but I switched to html5, and I think I’m not going back. The code just looks neater, and as for Internet explorer, I use html5.js. If they are using Internet Explorer without Javascript, they don’t deserve to see the site :)
I have HTML5 in my “base” files for starting new sites, and have had it there for the past 3 months. As long as I use the shim for IE, I have no problems at all. Using header, nav, aside, article, section, footer, ect.
Though, I don’t go too deep into it as to create HTML5 elements that only modern browsers support, such as clicker.tv does, where you navigate with arrow keys and such. I just like the semantics of the HTML5 structure for the company I work for and for our clients.
Then again, I develop with LESS/HTML5/CSS3 all at once with pre-defined fixes for IE in my base. It’s just how I like my work flow.
I think HTML 5 is not mature enough to implement it – it’s a draft, and it’s ever changing. I wouldn’t risk it, specially because XHTML still is “correct” (unlike HTML 4) and I don’t really -need- the features of HTML 5.
But hey, that’s just me.
I will use HTML5, but for certain projects. Mostly for web apps and less for “regular” sites.
Yup I agree with Sahan, doctype and common elements, header, footer and stuff
We use the “core” semantic html5 features for nearly every project since they have enough market support. I can’t really see any downside and so far we haven’t had any problems. I will admit we are only scratching the surface of what HTML5 will give us in the future though.
I have to confess, the biggest obstacles for me are rethinking the changes is tags: multiple h1’s (remember to style them so they’re not ginormous), multiple headers and footers, the new hgroups, multiple asides. I know they’re better than div vomit, but right now I’m still getting lost in my own markup. (I keep relying on Chris’ ‘beautiful code’ template: http://bit.ly/beautifulhtml )
As long as there are js shims, I don’t see a downside for older browsers. The only downside is teaching this old dog new tricks. Eek, am I really that old?
I use it. With Modernizr I don’t think there is any reason not to.
All of my new web projects start with a
git pullof html5boilerplate. It just makes everything so easy – it’s the way forward.Hopefully we won’t always need all that boilerplate. I’m looking forward to a version that doesn’t support IE6 – it’ll only need half the code :)
As for the actual poll, maybe keep the question but change the answers to “Would use HTML5” and “Would not use HTML5”? I had to puzzle over it to work out I was definitely giving a double-negative answer.
If I’m creating a web app for iPhone, or for a restricted audience where I know they’ll all be using a compliant browser (eg. company intranets) then absolutely yes – I’ll use all the HTML5 I can get my grubby little hands on (well, as much as the browser will support, at least).
Otherwise, for personal projects I’ll probably accept that it’s okay to annoy a few people with old browsers by using some funky new features that aren’t supported, but that said, even Chrome doesn’t support some of the more bleeding edge features, so it’s early days yet.
For client projects I’ll be a lot more conservative, as it’s really important that 99% of the people viewing the site have a great experience. The big problem is IE 8 and below, as they won’t style the new HTML5 tags without using Javascript to add them to the DOM. It’s the large corporate and government offices you have to watch for, as IT policies for these organisations can be draconian, so you may well find you’ve got whole companies that are still on IE7 with javascript turned off.
Using HTML5 doesn’t prevent a great experience. Using HTML5 and not providing fallbacks for specific features you use that are supported in older browsers prevents a great experience.
This is why I don’t quite agree with implementing hard core HTML5 quite yet. Chris, your whole comment about not providing fallbacks is just like the need to provide fallback or hacks for CSS 2. We can use HTML5 all we want, but we need shims/shivs, Javascript libraries, Modernizer, Boilerplate and all this extra stuff in our markup to make fallbacks or older browsers work.
With all of these needs to add to our HTML heads, why would you use HTML. Then we have to go back to all the websites we have developed in another 3 years, assuming HTML5 is fully developed and all browsers use it by then, strip out all of these extra Javascript libraries, shims/shivs, etc and make everything semantic with HTML5. That is assuming you can get a client to agree with that.
At this current moment in time. It almost seems like too much work to implement anything like that in the real world outside of the web development community. Only web developers know what it is and only we know if it will work or not.
If someone has Javascript turned off (corporations, non-profits, schools) due to lack of funding for upgrades, and/or have older browsers at the same time, it is almost useless to do all of this. I will take time to learn it, and with CSS3 in combination, but at this time it is hard to justify widespread use of HTML5 unless you are developing websites for the web development/design community and not intranet or big corporation sites.
I have a very XML-heavy workflow, with a lot of XSLT in there. I am tempted to say your poll is too simplistic, perhaps assuming that all the ‘no’ votes are for (X)HTML4. The CMS design and tools for working with XHTML5 are very different, so perhaps you should take into account users like me with a lot of mixed DITA/DocBook/MathML/SVG content who will certainly move to XHTML5 for the foreseeable future.
Could you explain what you mean by XHTML5? You mean using XML with HTML5 right? If that’s the case, that’s certainly HTML5.
Chris, the article you link to on HTML 5 Doctor describes HTML5 and XHTML 5 as different things . Bruce uses phrases like ‘. . . use HTML 5 or XHTML 5’ and he also points out that XHTML 5 must have the XML MIME type included and has more rules applied than HTML5. So I believe that is what Nicholas is referring to in his comment. What, I think, you are referring to, is that you can write HTML 5 using the more strict XHTML tag syntax if you like.
My boss cares more about cranking sites out quickly than using the best tools available, so I’m stuck with XHTML at work. If I figure out a workflow outside of work, I’d be allowed to use it, but I don’t see that happening, because the tools available at work are different from the ones I have at home.
I use HTML5 for my personal projects and freelancing, and wouldn’t do without in these circumstances.
There is no reason no to use it. If support for older browsers is important: don’t use new elements, but there is no harm in using new doctype, , script and link tags without “type” attribute, wrapping elements in “a”, etc.
I don’t see any reason not to use HTML5…. it ‘future-proofs’ your sites somewhat against having to recode at a later stage; it adds some very funky functionality that was ‘hell on earth’ to implement before, and there are plenty of tools/scripts that give older non-HTML5 compliant browsers a helping hand in rendering HTML5 correctly.
I just can’t see a reason not to embrace it these days….
Of course, HTML5 is great, we all know that, but it’s not right for EVERY single project.
Examples!
You _CAN_ use some of HTML5 parts today, at least use the doctype (which works great in any browser known to man, since it triggers standards in browsers not supporting HTML5).
I agree with Joakin’s comment above. The double negative poll is very confusing and can lead to misleading results.
I’d like to probe deeper here. To everyone saying, “I see no reason not to”, what’s the largest site that you’ve ever worked on? I’ll try my best not to come off as “talking down”, but there’s a lot of things to consider when working with enterprise-scale sites (especially ones that have to make money, and lots of it).
Of course we all *want* to use HTML5, but there are a lot more road blocks the larger your site gets. Are you in full control of managing the site afterward, or do you have to hand it off to the client whose technical team may not be as knowledgeable on HTML5? Does their CMS even support HTML5? Are there other vendors involved in the project that might not be as willing to learn/use it?
As for Modernizr and Boilerplate, they’re great, but they’re also an extra HTTP request and more to transfer just so you can use a few more tags. You also will have to make the sacrifice of putting a script tag in your head. We should all know that this slows the initial display of the site (considerably on IE6/7)—are you/the client okay with that?
Has there been any studies as to the improved SEO and/or accessibility benefits of these new tags? Yes, I agree, they make the semantics better, but if there’s no added value that has yet to be documented, you should tread carefully for these larger sites (taking into account the above questions).
So, a few things to think about. I’ve worked for clients whom I know I would never be able to pass HTML5 through right now. Thankfully, I’m out of that game and working for a company that, like a few of you here, has the luxury of deciding which browsers we are going to support. If users can’t view it in IE6…well, then it’s time for them to upgrade.
So to distill…
HTML5 may not work with “enterprise-scale” sites because 1) the CMS may not support it 2) others on the project may not know it 3) there may be extra HTTP requests 4) there are possible SEO and accessibility concerns.
1) Not quite sure what this means. I could see a CMS using divs and spans instead of articles and figcaptions or whatever, but divs and spans are still valid HTML5
2) For the good of the project it would definitely be good to get everyone on board. I can’t imagine it’s that big of a deal though. It’s not like HTML5 is this entirely new language. It’s backwards compatible.
3) The only really required thing is the shiv which is 4k and could easily be rolled into another JS file.
4) If you asked me months ago about this I probably would have shrugged it off, thinking if anything these things would be improved. However I was involved in an HTML5 switchover where there was some evidence that post-switch there was a bit poorer SEO. Kind of a bummer and not ignorable.
Chris your fourth point is extremely interesting. Was it for multiple h1s or something else?
Hey Chris,
1 & 2.) I suppose I should have rolled this together with point #2. I’ve worked with clients who will actually spit out HTML fragments (not just raw data) from the CMS for their (the client’s) caching reasons. So there’s communication that needs to go on there. Some clients are more than happy to learn and work with us. Others are a mess of bureaucracy in which no one takes ownership of a project and want to put the least amount of work in as possible.
3.) Although it will be cached, 4k adds up when you multiply it by a half million unique requests. Probably not a big concern if your site is making enough. Still, my argument for requiring this script to be in the head of the document stands. If you have scripts in the head already and want to include that’s solves a lot, but only in that case.
4.) Good to know, thanks for sharing!
No.
I already use those parts of HTML5 that work in all browsers. Like the DTD.
[also wanted to see how my gravatar looks here inside circles :) ]
HTML5 is a great term but what does it mean when you ask if I use it in any upcoming projects?
The term HTML5 encompass so much.
If you’re talking about the new semantic tags then support in ie6/7/8 is ok if your willing to use HTMLShiv javscript.
But, if you want something that can run on any browser and be supported without the help of javascript you’re going to have to wait for IE9 to go widespread or for a devastating bug to wipe IE out of this world.
Besides there are also many implications for the new Javascript HTML5 features. Most of which don’t run on all modern browsers coherently yet.
I for one will be sticking with HTML 4.01 Transitional for now.
I can not understand why people are still on IE6. Quite remarkable. Maybe it’s time for their IT departments to start upgrading.
I’ll give you two words: Legacy Applications.
In our case, my company’s current standard desktop build installs IE7. An improvement over IE6 but no where near IE9… They do it because there are certain applications that either haven’t been or can’t be upgraded that are critical to the business that will not work (without generating a ton of errors or providing a seriously degraded experience) in a more modern browser.
I could say more, but that is one short answer.
It’s not that IT professionals in these organizations don’t know that the browser has its flaws… it’s more likely because there are applications in the wild that actually *require* that browser… applications that IT can’t simply turn off or modify.
Unfortunately, the status quo where I currently work is to use HTML 4.01 Transitional (yuck!), but it is something I challenge as often as possible without being a total d-bag about it. :)
ya’know, 4.01 Strict *is* valid 4.01 Transitional. ; )
Of course. But 4.01 Transitional is not valid 4.01 Strict. And we use the 4.01 T doctype, so we’re not really holding ourselves to a very good standard.
of course. but my point was that you could write strict if you wanted to, and get away with it. Of course, if you have components or framework that’s out of your control, then that’s out of your control… : )
Too many commonly used browsers don’t implement HTML5 to justify using it.
I just wrote my first HTML5 site not too long ago. At first I wasn’t going to use the new semantic tags because it didn’t seem worth the extra effort.
I decided to do it anyway. It felt better to make something that’s forward thinking instead of backward thinking. Making something that works everywhere today with a few caveats (i.e. requires JavaScript on IE) and works better in the future (potentially better SEO from semantic tags) seemed like a good trade off for me.
Honestly, it’s an excuse for me as a developer to start getting comfortable with it as well.
Sure. After all, HTML 4.01 Strict is still a perfectly valid, widely-supported, stable DOCTYPE. Depending on the client’s budget and the project’s specifications, I’ll use either HTML5 with Flash as a fallback where needed, or use HTML 4.01 Strict. If you’re using a fallback anyway, the choice between HTML 4.01 or HTML5 DOCTYPE is a bit of a toss-up.
Since most sites will be rebuilt or modified in the time it will take Joe Webuser to upgrade his browser, future-proofing is less of an issue IMO.
(*Yes, I know about Modernizr and HTML5 shiv. They make it possible to style elements and degrade gracefully. They *do not* enable support for the tags.)
You are correct that Modernizr does not enable support, but The HTML5 shiv DOES enable support for the tags. That’s the entire point of the shiv.
Modernizr allows you to style elements.
Modernizr has essentially included the entire html5shiv for a long time now.
They enable styling but you still need to display:block a bunch of them. That styling is available in the HTML5 Boilerplate.
There’s no reason not to use the HTML5 doctype – it’s completely valid in all browsers. For many of the elements, older browsers will need a shiv to make them work correctly, but I don’t see that as an issue. There’s still stuff you have to avoid (as far as attributes and certain tags like canvas, audio, video, etc), but there’s no reason not to sprinkle in supported HTML5 now.
I think it’s a great idea to learn how to use HTML 5. If anything it’ll benefit you as a developer. I think it’s great to get in on it while it’s still new, it offers a wonderful opportunity to not only learn the latest and greatest but be on the cutting room floor with new technology. Not to mention it does look “cool” to potential employers to embrace and understand HTML 5.
Would I use it on a new project? Perhaps once I get a better understanding of it! Time to hit the books.
Definitely with HTML5. Even though a lot of features are not supported by IE, that doesn’t mean that we should not implement them. Progressive enhancement is something that will not only make our web projects better, but may also drive some users to upgrade their browsers in order to get those extra features.
I start my every project with http://html5boilerplate.com/
I share Milovan’s opinion.
You should always try to use state-of-the-art solutions, but keep in mind that not everybody is lucky to enjoy the magic.
Nope. Not till it’s more stable across browsers, but I am eager to start. Will probably start introducing it, by drips and drabs, after IE9 has been out for 3-6 months.
I start all projects with HTML5 Boilerplate. I guess I regret not implementing fallbacks to the fullest of my ability — I hardly have access to legacy browsers for testing.
I already use basic html5 stuff (the doctype, for example, and various little bits that make life easier and already work across browsers), but what I write is -basically- very enjoyable html4.01.
Aside from the things that aren’t widely usable yet, some stuff about html5 bothers me- for example, everyone seems to rave over “semantic elements,” but I don’t see them as being particularly helpful. The sheer quantity of new tags means that the markup will be more complex. You still need to use divs as wrappers for layout. And once support for them becomes widespread, it’s just that many more rules that you’re going to have to add to your css reset.
(also, I’m “all for” forward enhancement, but the html5shim doesn’t count as a “fallback.”)
A will now become what its semantics mean, simply a “division.”
Anyone who actually values HTML as HTML instead of as a means to a design, values the added semantic clarity.
semantics is one thing; but the whole reason html exists is to produce webpages -it’s not around for its own benefit. Semantics (and more directly, syntax) is important only because it (can) lead to better, more consistent rendering of the data it represents.
Valuing HTML “as HTML” is something we do as developers, but the rest of the world doesn’t care: they care about what they end up looking at (and it’s not “view source”).
I see the extra syntax as an opportunity for things to get more confused – people confuse things now.
In the long run, I think a div with a particular class identifier is as semantic (and lots more flexible) than a “semantic” element.
well we are using it in almost all of our projects now. reason’s are simple it’s not that much hard to make it compatible with older browsers. and tags are much more meaningful and CSS are much more managed and it makes much more sense then div.
DocType: yes
For mobile-only themes: quite possibly
For personal experiments and websites where there is proof that the audience will in the majority have supporting browsers: yes with shiv as a fallback
Anything else: still can’t quite see why using < header > then having to fiddle with JS fixes and additional styling to get it to work in IE is preferable to simply using < div id=”header” > It just seems like extra work.
Considering the only thing you need to do to “technically” be using html5 is change the doctype, I use html5 for all of my projects, and have been using html5 for the past few months. Why not?
yup…
plus <style scoped> is no longer invalid – great news for CMS plugins
I stay with a good old xhtml 1.1 and css 2 … possibly with a small amount of CSS3 .. without affecting the functionality of the website in older browsers.
With most of the websites i design i just use a simply WYSIWYG Editor. Don’t really think it would make a difference which one you used but it would be interesting to know if there is a WYSIWYG Editor than codes in HTML5. Any Body?
I think it depends what is project nature, for whom and why! But it best to start learn it now.
I’ll always use the HTML5 tags, the JS would depend on my site’s niche.
P.S: The question is somewhat confusing… “Not use HTML5”, I do wonder how many people voted ‘Yes’ with the intention of voting that they do use HTML5.
Depending on the project, of course! :)
If I’m building a web app, I’d generally be more inclined to use HTML5 due to its wider range of options (you know,
canvas,localStorageand such) but I’d never “force” it onto the users. If I think there’ll be less tech-savvy peeps using the app and the project doesn’t require me to use HTML5 for one reason or the other, I’d probably stick with the old ways.On the other hand, if I’m building a “regular” website I’m more likely not to use HTML5. I just think it’s not yet implemented as well into all browsers and that it doesn’t bring me that many benefits.
My Vote: Maybe. In a few months (or even years), definitely yes. :)
I think it depends on what you mean when you say HTML5. There are a number of levels at which you can participate in using the new hotness.
In my last few projects, I’ve been using the HTML5 doctype, which is one of the things I like most about HTML5.
I’m using the wonderful placeholder attribute to add additional labels to text input fields in browsers that support it. I’m also taking advantage of the ability to wrap anchor tags around block-level elements.
I haven’t quite got up the nerve to use the new structural semantic elements yet. It makes me nervous that JavaScript is required to make them work properly, even in IE8. What I am doing in the mean time though, is learning the semantics of the new elements and applying them by adding classes to divs, so <article> becomes <div class=”article”>, and so on.
I really got excited about taking my first steps with HTML5 after watching Jeremy Keith’s appearance on The Big Web Show, and also reading his book, HTML5 for Web Designers.
A better poll, IMO, would be what specific parts of HTML5 people are actually using.
I’ll follow this poll up with that.
The poll needed more options. ;-)
I would use HTML5 doctype and new tags for any current project that requires ONLY those things.
I would not use the new APIs unless it was explicitly decided in advance that the end-user will be expected to use a modern browser. And even then, it’s a risky proposition. Here’s an example:
We decided for our web application that we would expect end users to have modern browsers with HTML5 and CSS3 support (since it’s not a tool for the public at large, we can afford to be stricter about requirements). This allowed us to tap into emerging technologies like Websockets.
Except, as of today, Websockets have been yanked from Firefox 4 and Opera, and I wouldn’t be surprised if Webkit follows suit.
I’m still glad we’re using it… found a plugin that allows for standard AJAX request on a polling interval when websockets aren’t available… but nonetheless it illustrates that it *IS* a risk to invest in such bleeding-edge technology.
I’m ultimately voting for “no”… I would NOT start a project and not use at least some aspects of HTML5. But the poll isn’t as useful as it could be.
If others are interested in the WebSocket issue, take a look at http://hacks.mozilla.org/2010/12/websockets-disabled-in-firefox-4/
I agree with Greg, adopting early tech is a risk and this is a great example.
IE8 and below doesn’t fully support or recognize HTML tags.
There’s no possible way I will rely on Javascript to enable the very structure of the site to be rendered correctly…that seems extremely counter-intuitive.
Once my Analytics says IE9 is the norm, then of course. Until then, what’s the rush?
I use HTML 5 as much as possible, but the html editer I use (Taco HTML Edit) does not set html 5 as the default, so I often just use 4.0.1
i want to give it a go but i’m scared of new challenges, a bit like getting into a cold pool, wearily walking in genitally avoiding any cold shocks! but, eventually your in and its not that bad, so actually will someone just push me?
No, I’m not coding or designing for HTML5 until it is common place across browsers and general use.
A big Yes! I have to get acquainted with it first before using it on my new web project. Time can tell if HTML5 would become a norm for Web Developers.
Dudeeeeeeeee… people still use IE6… IE6!!!!
When people will stop using IE 6, 7 and 8, maybe I will start learning HTML5.
Just so this is clear, you can create an HTML5 website and support IE 6. You just need a little JavaScript so you can style the new elements, and you use polyfills for lots of other things without doing much more work at all.
Usage of IE 6 is no excuse to stop learning.
I voted no. HTML5 is quickly catching on, as is CSS3 and older browsers are no reason to move on. The W3C released in 2009 that more people used Firefox than IE so it’s rapidly becoming a less important matter.
Perhaps more people used Firefox to visit the W3C website (who would mostly be webdevs) but I think you’ll find that IE6-8 still has the lion’s share of the market across the board.
The issue I have with HTML5 is a restatement. Too many people still use IE7 and 6, and even today’s modern browsers (betas included) do not have full and even HTML5 support.
It would be useful to play with now so you can get a grasp on it, but it can’t be practical until the web consensus is that IE6 is so 2001.
I definitely would, primarily because HTML5 isn’t even a recommendation and isn’t fully supported by much of anything.
I’m content to keep going with HTML4 or XHTML and just read the HTML5 spec over and over.
Stupid negative question! I picked no by mistake, as I have no desire to use html 5 for quite some time.
Oh and I love the look of the new website.
The selection depends upon the fact that the new web project seriously needs HTML5 features in it or not. If it is not required I’ll never use it.
I try it a few month ago
I love html5. I hate how we have to always be at mercy of ie. Hopefully one day we won’t. That’s why I turned to mac :P
yes. html5 needs 2 or more years to be stable. therefore css2.0 is still impressive.
Everyone but is any project (no exceptions).
Only HTML 5. Best regards!
Paweł P.
I’m in a situation where I can’t decide whether turning to HTML5 is the right thing to do.
I want to keep up with the industry, but HTML5 is in its infancy and we’re a few years off a W3C recommendation yet. I’ve read articles that say “GO AHEAD! It’s ready to use” and I’ve read articles with the opposite opinion. I just don’t know which direction to take on future projects.
All I do know is that my next project is going to be for a charity, and it’s users are likely to be using Internet Explorer, so using HTML5 on this project is probably unnecessary.
Yeah, absolutely.
HTML5 is great but it’ll be at least another 10 years until this update is compulsory so any new project today will probably have about five updates and re-designs before then.
Personally speaking I’m still to learn HTML5 properly. I understand most of it and what it can and can’t do but I’m not confident enough to write HTML5 fluently so it’ll take a lot of reading and research for almost everything and I don’t have the time for that.
Also, I actually believe there will never come a time when we have to write HTML5 the way we write regular HTML now. 10 years is a very long time in the online/tech world and its more than likely there will be other technologies that become standard. Maybe a new type of language that combines HTML and CSS into one much shorter and much easier language.
Only time will tell.
No, I would not. Nor would I make a poll without radio buttons and a submit button :)
It’s on the side.
We ARE actually developing an HTML5 product. It took me 4 weeks to build it up and it is a pretty good copy of its Flash-Version. The task was to turn a Flash-Site it into an HTML5-Site.
BUT, we are currently experiencing problems across all browsers. This is performance problems, markup styling problems as well as lack of features concerning css animation. Using a fairly complex mixture of jQuery’s fx chain and the queue/dequeue feature as well as tons of timeouts and intervals we can copy the animation system of Flash fairly good but we are missing the typical motion tweening Flash has to offer. We can fake non-linear animations but it does not look as good as in Flash. So far for the Flash-Killer right now :) Only Safari supports CSS-Animation at this time and with HTML5 you want to support at least Safari, Opera as well as Chrome (Firefox still lacks a lot of features)
The current beta versions of Firefox slowly supports more and more CSS-Transitions and layering but there are a lot of bugs and as one of the first repliers wrote – nobody will download a beta version of a browser :)
Right now Safari and Chrome support GPU-based css transitions. We experienced timing problems here but its an overall good animation system that actually works. (You need to work a lot with small delays done with setTimeout and alike) Firefox and Opera still do their transitions on the main processor and you can clearly see it if you have more then a certain amount of layers on top of each other. (I am talking about a number of at least 10-15 layers for a fully animated site). Opera actually presents a fairly save animation system without the mentioned timing problems like in Safari but it is utterly slow and that kills every product – as soon as it is slower than Flash (even on Windows) you can’t sell that anymore.
I’d say it will take another year or maybe two until we will have a stable, fast, GPU-based CSS-animation system in every browser. At that day we can start developing heavyweight Flash-like sites but right now it’s really an adventure and it has a big chance for failing.
So my opinion is: Right now we can start creating sites that partially use HTML5 features that are save in every browser and that is already a fair amount of features. But I would not recommend going for a only-HTML5-solution. That will most likely fail.
Yes I would use HTML going forward.
I think the problem with these standards is that people waste time debating whether or not to use it when we should ALL use it and enact change as opposed to worrying about supporting IE6, etc.
If most folks took a look at browser usage you’d probably notice mobile browsers on a steady or sharp upward trend over the last couple of years as compared to a certain legacy browser which is either flat or on a downward trend.
Where else but on the web do you hear about legacy apps hindering progress?
HTML5, that is.
I think I’d use HTML5, but only the new tags for semantic, e.g. <árticle> or <section>, because you aren’t risking full damage and unusability in old browser.
It’s safe to use the html5 doctype too.
In addition to and , there are quiet a few other tags you can already use.
Examples:
(with multiple ‘s)
I believe using forms, canvas and media types (,, etc…) are some of the few exceptions to widely supported elements. In those few exceptions you might need to offer fallbacks.
Starting a web project, and am not using HTML5 because of IE. However, I markup my HTML using HTML5 semantics, such as which can easily be converted to HTML5 once IE 6-8 are dead.
ha, stripped the HTML out. Trying again: <div class=”section”>
Yes, I would not use HTML5 on a new web project today, unless it counts that I use the new doctype and have stopped doing some of the XHTML things I was used to. I don’t think that’s “using HTML5” though.
I almost always have to support ie6-8 so no I wont be using it for a very long time. Take your guess at when ie8 will die.
If I was going to work on a project that only had to support modern mobile devices that use Webkit or something like Chrome OS I could see using it there.
Not until the standard has been finalized and all major browsers support it like they supported XHTML.
Right now HTML5 and CSS3 are just buzzwords such as Web 2.0.
Those words mean nothing to anyone but marketers and non-technical interviewers.
I thought about using HTML5 (and use some of the snippets of code, but I am really unsure of using it because not everything is widely accepted. Where I do most of my code, the computer of choice is Windows 7 and IE has lacking capabilities so far. When the start supporting more of the code (and when HTM5 is finally a Recommendation), I will reconsider. Regardless, all of my code is HTML5 ready.
Unfortunately I would love to say I’d build all sites in HTML5, but thats just not the case, a Recent client willing to pay thousands for there online presence were only using IE6 – Seriously.
The clientele of there site are mainly older generation and may not update as frequently… So for pure usability for everything I wouldn’t purely make a HTML5 site..
The sad part is that most of our clients are still hanging on to old browsers like IE6 and FF2 and the like. Most of the pearls that HTML 5 has to offer would be lost on swine.
Yes, in the online gaming world. Flash is a much more mature platform over Canvas.
I probably wont use HTML 5 on my next project, as it’s just too new and I don’t see the benefits of using it greater than using “old” XHTML. Definitively I’m diving in just out of curiosity, as it has a lot of cool stuff and potential, but I wont use it just yet. I see HTML 5 as an option for developers more than the future of web development.
cheers
I answered ‘yes’ but wanted to answer ‘probably’. Really, it would depend what I wanted to do and who it was for. I’ve just completed my first HTML5 site (just basic HTML5 markup, nothing fancy), and I think there is still a little too much ambiguity in how elements are defined for me to use it in a corporate, commercial environment.
For personal projects, every time, and if the exact semantic usage of elements evolves or becomes clearer to me later and I got it wrong, well, never mind.
Yes, I would *NOT* use HTML 5 on any new project.
Hacking the code so it works properly on all browsers (mostly on IE6/7) is tedious enough using HTML 4 or XHTML… Why would I add another layer of complexity on top of that?
Besides, here’s how this page looks in the latest (stable) version of Firefox at this time!
However, that doesn’t mean I don’t experiment with the new features, but it’s a long way from experimenting to using in production…
That’s a pretty weird bug, a number of people have reported it. It’s literally ONLY Windows and ONLY Firefox in ONLY that version. Wish I could sort it out but seems like a bug to me.
Just wanted to hop in here and say that i encountered that bug in Chrome 8.0.552.215 and Firefox 4.0 as well.
Dale,
I am the person who hires web developers and we aren’t doing HTML 5 at all other than as learning exercises. Support is very uneven for HTML 5 and CSS 3 (which I’d love to use even more than HTML 5) even among those browsers that claim to support it. I recently created a site redesign for a client using some HTML 5 & CSS 3 which they loved until they looked at it in webkit based browsers and IE. That client uses FF 3.6 as their default. We didn’t even charge the client for the time it took to do the HTML 5 implementation because we knew it wouldn’t “fly” but the reality is that the site was done using XHTML, CSS 2.1 & some bits of JavaScript and the result was fully cross browser/platform.
Our clients are large multinationals some of which use Citrix for all web access with IE 6. We won’t even contemplate rolling out HTML 5 websites until there is at least some agreement on video codecs.
We’ve experimented but given the corporate clients we have to support, some of which even have javascript restrictions it isn’t going to happen anytime soon no matter how much I’d like to use some of the feature set of HTML 5 or equally important CSS 3.
Frankly, I don’t consider loading a javascript framework to stimulate some implementation of HTML 5 to be a valid workaround for older browers which in this case includes IE 8. Every one of our clients is standardized and enforces by group policy IE as THE browser. It generally takes them 2 years or more from release of a new OS or browser before they are deployed.
In 5-10 years we’ll be able to deploy HTML 5/CSS solutions assuming that they reach recommended status sometime in the next 2 years.
I totally agree – the shim is -not- a valid fallback.
No sidebar on me mobile device mean no poll for LazyAndroid. Also scrolling to end of page take forever.
Oddly Enough after reading through half of the comments thus far I’m actually more intrested into shifting doc types to html5 to see how much of an effect it has on existing sites (obviously i.p. blocked so only I see the new doctype) just to see what, if anything changes from XHTML 1.1 to HTML 5.
Having said that I doubt I’ll be officially using it for a while but if it’s backwards compatible in the first place which many people appear to state is the case (+4kb and integration into an existing gzipped js file) It seems a lot more promising a suggestion then say switching from standard javascript to jQuery with it’s overheads.
I answered No, I would not because of the simple reason: even though there are tons of awesome features with both CSS3 and HTML5 our clients are still the majority of the audience that uses older browsers for their own site.
Our clients want sites that are 100% functional in IE7, here is your answer. We do try to upgrade the minimum support for browsers, but it comes with the great cost: once you limit what you can do for businesses, you limit your market. Especially in my state where there are not too many companies that keep up with new technologies every day, it just doesn’t make sense form the financial point of view.
Iam all for world-wide HTML5, heck iam even for an ie6 free world…. but both arent exactly happening anytime soon.
But on topic, i wont be using HTML4/XHTML.. Thats the past
It really depends on the target user groups for the website.
A big percentage of people I work for are still stuck with lame versions of IE (usually 6 and 7). Heck, most of my country is stuck on that cause people aren’t really aware in matters of computer technology and open source stuff – Microsoft has earned their trust and that’s quite hard to overcome.
Even when I install (mostly) Firefox at my clients’ computers and show them how much better it works versus older IE versions, they won’t use it. I visit my clients several times per year, and they’re still committed to their relationship with IE…even if it’s bad for their IT health.
If the websites I design would address a more advanced audience, I’d definitely go for HTML5, but in my case it’s futile and there’s only so much pressure you can place on clients to change their tech habits (and that varies a lot between clients). So while we are asking “HTML5 or not”, I am still working on “IE6 support or not”. :)
Of course I don’t plan to stick to the past forever, so part (and duty) of the job is to try and explain to customers how things are evolving and why it would be good for them to follow up. This way we might all slowly move forward. So for me it’s designing “the old way” while trying to educate people about new features, via discussion, examples, articles and whatnot.
Boilerplate!
I wouldn’t use HTML5 for the stuff I do in my day job, at least not yet, I’m not a big fan of “hacks” which is essentially what has to be done at the moment to make things work correctly on all browsers.
For my own projects, websites tests and the like HTML 5 is just fine, but in all honesty who has the spare time to play browser patch up.
It’s starting to feel like 1999 all over again.
HTML5 is often talked about in conjunction with CSS3. I bet almost everyone who’s posted above uses parts of CSS3 (such as border-radius) to help older browsers see a visual improvement.
Most of my sites/apps are used by regular folks who really don’t know or care about the guts… just whether it works or not.
So, for now, no HTML5
Yes!
Questions: http://html5boilerplate.com/
Chris, when I go to the home page at look at the Poll, it only shows me the results. When I view this post, the poll allows me to vote. Might be something on my end, but I thought it was funky.
I still don’t use HTML5 with new projects but I want to start doing it soon.
I’d use the doctype but that’s pretty much it… I voted no.
I voted no.
Why? Because HTML5 is still something new. Not all browsers have proper support for all it’s features. IE is still an issue (as always).
When IE start to support html5, I’ll give it a go. For now, I use html5 for my own educational purposes.
For me, it’s complicated. I’m about to start an MMORPG, and can’t decide between Flash MX 2004, and HTML5. If anyone has an idea on which I should use, please let me know.
What a question … and the answere is sometimes; because the world isn’t black and white.
Well, if you build sites in xhtml1-strict change the doctype to html5 and in the most cases it will validate. Yes i will use html5 (not just the doctype) for the next website.
If you focus too hard on the past (legacy browsers), you will fail to build for the present & future (modern browsers). Chrome is doubling it’s market share year over year and mobile browsing is growing at 8 times the rate of desktop browsing. In a few years, the choice is going to be a no-brainer. However, even today with a few lines of code to include Modernizer & HTML5 Shiv you can create for the present and future of browsing.
If you like the look and feel of the big blue lego, go right ahead and create with flash. If you like to liter you markup with id’s and classes which are the same value as the equivalent html 5 element (i.e. id=”header”, class=”section”). Go right ahead. While not as bad as the former, it’s probably a waste of time.
I see no problem with using graceful degradation techniques to achieve backward compatibility and progressive enhancement to get the most out of your projects. It seems that everyone is moving so quickly and already getting the most out of these new technologies. I don’t want to get left behind because I was too lazy to learn how to implement new techniques correctly! Yes clients want a reliable product but they also want the latest and greatest and there is no reason why they can’t have both.
I think there’s a general lack of distinction here between:
1) Learning and/or valuing HTML5
and
2) Using HTML5 for current projects.
Beyond that, there’s a lot that “using HTML5” might mean. I use HTML5. But if you changed my doctype, I’d be using (almost-completely) valid 4.01 Strict.
Yes, I would like to start a new web project and not use HTML5.
chris, the vote options are NO and YES in that order, yet the result displays the opposite order, is that intentional ?
About the poll, YES, I will start a new project and not use HTML5, simply because I’m not that into it at the moment, and because most clients don’t care whether their websites are made with HTML5 or not.
But maybe in 2011, the trend’s gonna start to kick in.
I think that Paul Irish’s HTML5 boilerplate makes the migration to html5 super easy and it also grant a neat old browsers compatibility…
I tried it and in my opinion it works pretty well, so i’d like to suggest every one who is still doubtful of trying this brilliant solution…
I would actually not use it today because I have not used it enough in personal projects or not had enough time in R&D to move to use on a live site.
Would I use HTML5? The answer would be “Heck Yeah!”
I have and continue to use it in projects (both personal and client). I prefer the new elements, and have no problem using the shiv to help hold IE’s little blue monster hand.
I always see loads of comments saying “I’d think of using this if more browsers were supported”. To that I say, help forge the path, not just walk it. Another is “I’d use this but to many people use older versions of IE (such as 6 & 7) that I’m currently supporting” (or something along those lines). IE has to be one of the worst browsers when it comes to individuals upgrading it, I’m a big fan of educating not enabling (this does not mean I’m against graceful degradation), so in addition to educating users on options, taking advantage of new HTML5 and CSS3 features to give those users other reasons to upgrade is more progressive.
Really the long and short of it is, HTML5 is here and it can be used if you’re willing so don’t get left behind :)
I think it’s crucial for developers to be able to markup HTML5 and in many cases I think using it is perfectly fine for new projects given the additional tools available for backwards compatibility. Because we’re in transition phase right now, no one on either side is right or wrong. In many ways, a tool is only as reliable or effective as the person using it. IMHO.
I wouldn’t use HTML 5, nor would I for a while… for me its about browser adaptability it took forever for HTML 4 to be 99% compatible with all browsers and it’ll take a long time for HTML 5 to do the same. I am patient, it allows me to develop my own HTML 5 skills while the browsers of the world catch-up.
…i’m sure microsoft will do their best at being the thorn in your side for many years to come anyway, so taking on html 5 isn’t going to help, well not at least for 10+ years…
Nope, don’t use any HTML5.
Our clients want a site that validates as HTML4 strict, often with valid CSS2. They also want a uniform experience across all browsers. On a large site there’s just no need, or time, to potentially complicate things with graceful degradation.
I do use it on small or personal sites that I know don’t have any corporate standards behind them. You’re not going to get left behind by not using it NOW as many folk seem to want to say. There’s a long time to go until it’s finalised and standard across all browsers. Also, since when was HTML not childs play? I’m sure it’ll be a doddle to pick up for those not wanting to spend the time learning it right now.
I would if I could, but the problem is that I’m currently working on a goverment project.
Yep you guessed it, IE6 and winXP and an IT department that does not realise that they are the only ones left viewing the site in it.
That bieng said, there are plenty of ways to make the basis HTML5 functions work in IE and lower versions of firefox.
Its true that most people don’t care if their browser is the latest and the greatest.
But the tide is slowly turning, I know people who were afraid press an update button are now experimenting with Chrome and other things.
I’ve read a couple of books on HTML5 and have played with the code. I think its great, and a bit too lenient with its rules. XHTML is quite strict, which forces designers to be meticulous with their markup. Sometimes you have to go out of the box for certain effects that will not validate, but you still stick to the rules as much as possible. In HTML5 it seems that you have way more freedom which can lead to the unknown. Most people fear the unknown, as for me, I’m not in rush and probably would not use it. You typical client does not know the difference from HTML and Hotmail. Writing extra markup does not excite me for non-compliant browsers does not excite me.
I’ve read a couple of books on HTML5 and have played with the code. I think its great and a bit too lenient with its rules. XHTML is quite strict, which forces designers to be meticulous with their markup. Sometimes you have to go out of the box for certain effects that will not validate, but you still stick to the rules as much as possible. In HTML5 it seems that you have way more freedom which can lead to the unknown. Most people fear the unknown, as for me, I’m not in rush and probably would not use it. Your typical client does not know the difference from HTML and Hotmail. Writing extra markup for non-compliant browsers does not excite me. (excuse the first comment, my three year-old interrupted).
Yes, in Russia it is still full Ie 6
If I develop a site written in 99.9% HTML4 with one line of HTML5, does that mean the site is written in HTML5?
For personal uses with out a doubt, But for a client I would have to stick to html4 unless they requested it.
HTML 5 is 100% compatible with all browsers TODAY. Those who say it isn’t are ignorant and should not call themselves developers. I pity out industry as it tries to move forward yet is held back by small-minded laggards.
Yes. And I already did for one of my projects.