Several folks pointed out that this last poll question was worded a bit strangely:
Would you start a new web project today and not use HTML5?
I worded it that way to elicit comments on why a person or organization would not opt to use HTML. However, that meant you needed to answer Yes if you would not use HTML5 and No if you would, so, possibly a bit confusing, but whattagonnado? The results show that the majority of people would not use HTML5:
I also wanted people to come to their own conclusion on what “use HTML5” meant. Some folks asked if they just used the doctype and nothing else, is that using HTML5? Yes, I’d say.
I think the results are pretty disappointing. The comments were full of people claiming HTML5 is not yet ready to be used on “real” sites. I disagree. I know that many of you still need to support really old browsers. Even that doesn’t exclude you from using HTML5, as there are shims, polyfills, fallbacks, etc you can implement to make sure those browsers have a good web experience and newer browsers have an even better one (just like CSS3 progressive enhancement).
It’s simply not the case that you need to wait until all browsers fully support every aspect of it to use it. As designers, we’ve never done that before. We’ve always pressed forward while not leaving anyone behind. Let’s keep doing that.
For those people who have and are deploying HTML5 sites, this new poll is for you.
What features of HTML5 have you used on production sites?
Poll is in the sidebar of the site.
I agree. With HTML5 Boilerplate and CSS3pie you can do anything :)
Totaly true, with graceful degradation of course ;)
Pretty interesting. First, I think browser updates should be forced. Anytime I meet a client and we get comfortable, I try to sneak in a browser update for them. I, like most people, hate Internet Explorer. I’m starting to wonder what that company even does nowadays since dragging their feet seem to be the main objective. I would also disagree on HTML5 not being ready, that statement was thrown out too much when browsers actually were not supporting most of the features and people stuck to that statement. Now, we have so much HTML5 code to use therefore, that old saying isn’t relevant anymore nor does it make sense. If it was too early, features wouldn’t work in any browser. But either way, like you said, we always find a way around getting stuff to work.
I agree to.
Why are the results disappointing? You asked, we answered.
There are some very real security concerns about HTML5. Granted they were inadvertent, but if they’re taken advantage of, our customers could be liable.
Further, while it’s easy to say “overlook older browsers”, you’re advocating turning our back on potential customers. Have you tried selling that to a client? We can’t force upgrades as designers – and most of our clients aren’t about to give up potential sales over style or what’s cool or easy for us as designers. Too many of them went through that already with Flash – and then found they needed a non-Flash version as well.
Eventually they will adopt – but I honestly don’t see it happening soon.
I asked, you answered, I comment on results =)
They are disappointing to me because I feel like HTML5 isn’t being used for the wrong reasons. You say you would be “turning your back on potential customers” by using HTML5? How is that? If you did something like include a product video with <video> and included no fallback so customers with browsers that don’t understand <video> couldn’t watch it, that would be stupid and turning your back on potential customers. But you would never do that. You would include a fallback for that so that browsers that didn’t understand <video> would get the video served in another format. Then you’d also be supporting iPhone and iPad while you were at it and actually be opening your doors to more customers.
I work on alot of projects that are very time sensitive, or have been previously quoted and I only have X amount of hours to complete.
In almost all cases I don’t the luxury of time to implement new html5 features well also developing fallbacks for older browsers.
I agree, but the reality is that newer devices and browsers can handle HTML4 more reliably. Simply because HTML4 has been around longer to form predictability. HTML5 advantages are so far on paper only, such is video and microformats.
HTML5 adds more features over HTML4, but there are many technologies and standards arising and changing alongside HTML5, all of which contribute to the difficulty in reliable and valuable support in browsers and devices.
The truth us, the 12 people who are using the address tag are not yet using it in ways that everyone else using a p or div tag aren’t already using it. That is the truth of it.
HTML5 spec is full of good intentions, which by the time they are well and truely taken full advantage of by browsers and devices within 5 years, will have designers and developers angry again with it’s shortcomings (such as no real solution to seperation between content and function) knowing that they will have to go through all this again like we did when moving away from tables and transparent gifs.
Whatever side of the fence you take, within 5 years, HTML5 sites will still display and function in browsers and mobile devices, but so will HTML4 sites.
For companies, it is simply a question of what is more cost effective. And right now, management systems that optimize time spent in creating such content are typically built around HTML4/XHTML and are responsible for over 99% of content users and companies themselves engage with on the internet. Moving to HTML5 is going to be a hard thing to enforce, in the way that people have not been fooled into replacing their DVD collection with Blueray … if it is already working, people will keep with whatever is more cost-effective.
I started using HTML5 on my own site and projects, but not for work as there are too many issues to overcome. Steven’s argument is good, especially with this:
“Whatever side of the fence you take, within 5 years, HTML5 sites will still display and function in browsers and mobile devices, but so will HTML4 sites.”
The key phrase is the “so will HTML4 sites.” That is hard to argue with.
My problem with HTML5 is with the new semantic tags. The definitions on when and where to use them keep changing, and there are too many ‘ifs’ or ‘you can’ instead of ‘this is exactly where to use this tag’. The address and aside being the most annoying.
Still, I’m using it so that I get used to it. The HTML5 Boilerplate makes starting a project with HTML5 easy.
For our own projects there is no problem, and I’m with Chris. But when it comes to what you can do at work and what is able to be used as far as time*/what custom cms will allow….HTML5 is a couple of years down the road at the very least.
*by time I mean adding in fall backs or actually learning how to use the new feature if you don’t know it.
How come “none” is not an option in the new poll?
If you don’t use any features, don’t vote.
Pretty harsh there, Chris. Haha :)
I’m with you on this one Chris, I’ve been using HTML5 & CSS3 on every single project for the past year or so.
The company I work for has a few huge projects going on and I’ve even been able to convince them to let me do everything in HTML5 & CSS3
While it does take a little more thought, I opt to use HTML5 in every new project. With HTML5 Boilerplate and a little planning it’s easy to implement while staying true to progressive enhancement.
As far as HTML5 features, I focus more on the semantic elements such as header, nav, footer, article etc etc. CSS3 is something that I personally have been using a ton of, it is almost silly not to. Itmakes progressive enhancement almost a non-issue.
Larger companies with larger budgets can, should and will use HTML5 to remain above the cut. Web developers and web design pundits with blogs can and should go with it too. But I feel for the rest of us, the onus has once again being put on the little guy to pull the weight of the browser incompatibilities.
Lets face it, there are two approches to implementing the new technology. 1. the “hardboiled” approach where you don’t or minimally care much about what it looks like in IE, or 2. where you spend a lot of time to match the design using both old/new tech. Well I think anyone with clients who run actual businesses know that option 1 ultimately results in lost business and is rarely going to fly. So that leaves option 2 which (regardless of all of those who claim it is little or no extra work) is a ton of extra work in many or most cases. Who’s gong to pay for it? I don’t think the client will… That should explain the poll results I think.
By the way, I really like the design of the new site. But I can see that the same attention was not given to IE7. I’ve also noticed that’s its pretty buggy even in Firefox 3.5.16. I think this is all fine for a site like CSS Tricks, but IMO not at all fine for an average client’s site.
I definitely think we should at the very least, be experimenting with HTML5 instead of merely reading about it or ignoring it until it’s more widely supported. I think using HTML5 now is especially important if you are just now beginning to learn about web design/development.
I think if we choose NOT to keep up with/utilize these newer features in the world of web design we are only doing ourselves a disservice to our trade. Sometimes it can feel overwhelming as a new web designer to see the trends/standards changing so quickly, but that can also be the best part of it. Every time I experiment with CSS3 or HTML5 and learn something new about it, I have an ‘A-Ha!’ moment. It’s those series of A-Ha moments for me that make this field entirely exciting. I’m really eager to see where HTML5 will be in two years, even one.
I agree with experimenting. I think there are lots of limitations in the HTML5 spec and am eager for a predecessor already, but do use a few HTML5 elements in my XHTML for where they already are making a real substancial benefit to many users, such as form input types for mobile devices.
As for general developing though, I am sticking to my plethora of HTML4/XHTML applications and management systems because they already work, I invested alot of time getting them to work for me and to work for my clients. I am in no hurry to replace all that, so any HTML5 items will be hand-coded at my discression.
I agree Chris.
I don’t think that it is necessary to exclude HTML5 from new sites. If you know that it is going to be a problem for some browsers, either give that browser an alternative or decide not to support that version of browser.
We do this all the time, don’t we?
I interpreted the question as, “would you require yourself to use HTML5 on a project,” to which I said, “Not necessarily.”
If a site needs it, then sure. Otherwise, why trouble yourself?
Shouldn’t one of the options be just using the DOCTYPE. I have started using
in my code
sorry < meta charset=”utf-8″ >
I don’t use HTML 5 all that much on client projects, mainly because at this point I am still not as good with it as I could be and therefore it just slows me down.
That being said, I am using it on most of my personal projects that I start, and once I feel like I have enough of a handle on it that it doesn’t impact speed and reliability for clients I will make the switch to HTML5 on all client projects as well.
Those who are saying its not ready yet I really don’t agree with. Sure at this point its not as well supported as HTML4, but I can see good reasons for using it, like being ready for when all the browsers DO support it. and like Chris said, using means you will be supporting the iPad and iPhone which I foresee being a HUGE deal in coming years or even months.
Just my 2c
Has anyone heard what the effect of using HTML5 markup is on search engine rankings?
I mean search engines determine the ranking of information based on how it is marked up, so if we’re changing our markup could this effect the way that a search engine indexes a page?
Just wondering if anyone has done any tests regarding this.
The only thing that happens when you validate the markup is a warning that you’re using an experimental feature. As long as you have no errors in HTML5, you’re fine.
Ian Hickson, the HTML5 editor, works for google…
I’ve been curious too if there is any SEO improvement based on using the more semantic tags in HTML5. Has Google or any other major browser has tweaked their crawlers to notice these new tags or they still read as equal to say, div class=’nav’?
Surely as web designers/developers, it is up to us to define what the standards are with regards to HTML 5.
If we use it more it will force clients to upgrade their browsers and make full HTML5 and CSS3 integration a priority for the browser developers.
Even just using the HTML 5 doctype < !DOCTYPE HTML > is a start.
i think it’s a bit too soon to go with html 5! i am reading for quite a while on HTML 5 and fount out that there might be serious security problems with it!
I still can’t fathom why one wouldn’t use at least the basics of HTML5. The elements aren’t going to hurt anything once you put the shim in place. One of our clients still considers the IE6 market and I used HTML5 elements in the re-design of their site. No problem. It can’t hurt to get used to the structure.
html5boilerplate + css3pie is the way to go and even if you don’t want html5 you can easily change boilerplate to be used with html4.
Currently there is no reason not to use html5.
The problem with HTML5 and CSS3 is that it’s nor really a production environment. And of course you can say it degrades nicely but it’s close to impossible to tell the customer that he’ll have round borders and fancy comment lines only in browser X or Y but not in >=IE8 (like the case of css-tricks and IE7 :P).
Additionally as it was mentioned before, no client is willing to pay for the extra overhead and time required to code HTML5 and CSS3 + all the fixes for unsupported browsers (IE I’m looking in your direction :) )
Nevertheless I think that we all should know html5 and css3 and just wait for the day when it will be production ready.
P.s. have You even bothered to view css-tricks in IE6? I understand that 99,9% of the target audience will be using a modern browser, but still it looks a bit unprofessional.
I absolutely don’t care about IE 6 on this site, sorry. I also absolutely understand the need for making sites work in IE 6. Believe me, I have done and do my share of that.
I’ve used the new semantic elements on a few sites. I’d like to get into some of the other features of HTML5 and see how to work on mobile devices but havn’t found the time.
I’m curious of what people are doing with HTML5 that clicked the “Other” option in the poll.
I had no consent of this voting poll; but anyways, I would most certainly use HTML5… Why not use it? It’s made to make coding easier dude. So much time saving, makes your life easier to enjoy and live with.
Chris I completely agree with you.
I can’t for the life of me understand why anyone would keep holding the crutch of IE6 up unless it’s absolutely necessary (some government sites etc). Proper education/explanation to clients about browser upgrades has never failed me.
To everyone saying they can’t use HTML5 because it’s “not ready”, “not complete” or “not supported in enough browsers” let me put this to you, I’m sure you are familiar with CSS 2.1 and use it on a daily basis without considering if it’s complete. It was in development for over 10 years and just became a candidate recommendation in 2009. It doesn’t have 2 browsers fully supporting it, only IE8. Does that stop you from using it? I highly doubt it.
So if you do the above, there should be no reason that you would not use HTML5.
This thread convinced me to switch over to HTML5 for the small project I was working on. Great post.
My personal sites have visitor metrics that would allow me to use HTML5 and CSS3 to extent, but I don’t. The reasons are many, but the gist is that it’s just not necessary. I want everyone to have the same user experience, regardless of their browser. People underestimate the value of constancy and user experience.
As for my company’s site, there’s no way I could rely on HTML5 or CSS3 for any aspect of the site. You really need to take everything into consideration when 1% of users equals millions of dollars of yearly revenue.
That said, it’s good to keep abreast of emerging technologies (I am a W3C member) so you can leverage them when the time comes, and there’s no better place to do that than CSS-Tricks.
Maybe this speaks more about the audience of CSS-Tricks than the use of HTML5. Be interesting to see this same poll conducted on a site that has similar traffic but all relatively new. Maybe a dribble poll or Forst.
We are growing old with you Chris….
However you definitely seem to be one stand-out and passionate guy about everything, music, writing, shooting web.
Here’s to you mate, along with Jeff Way. My two personal favs in blogosphere.
Merry Christmas, thank you, and keep up the fine efforts.
I’m taking the plunge right now with HTML 5, so it’s something I’ll be using going forward! I’ve had a couple clients that have asked me a couple months ago if I’d be coding in HTML 5, and before I literally didn’t have time between all the work I was doing.
Now that Christmas vacation is here, this gives me the perfect opportunity to dive in! I’m interested in seeing just how great IE 9 will be when it finally comes out! I know…”great” I say? IE = great? Well…not exactly, but we’ll see with IE 9.
IE wont be great man. It’s already a lightweight HTML5 spec it’s adopted. Typically behind the rest and slow to catch up.
We’ll be missing a whole host of features running in tandem. It’s not even 25% of the spec implemented. IE follows the rest.
I’ve been putting off html 5 and css3 (I know I know but work must go on and it’s too easy to stick with what you are comfortable with).
I’m going to take some time to learn more – I just know it is going to be a huge undertaking and once I give it a corner of my mind it’ll nag and nag. Can’t guarantee I’ll use it right away as there are lots of issues to think about regarding browser support – however Amber did put up a very good argument earlier on .
All my new projects will be in HTML5 “!DOCTYPE” or, more exactly, XHTML5.
I’m used to XML syntax and I like clean code and unobtrusive javascrit as well as separated semantics, style and behaviour.
I think that coding is something more than a making-things-works matter! :)
Some people mention security concerns with HTML5, but neither here nor on any Google searches I’ve been able to come up with has anyone demonstrated an HTML5 security issue. The closest I see is that some specific HTML5 feature on some specific version of some specific browser had a security issue. I also see what looks like conjecture, such as an article on Network World stating, “Before HTML5, an attacker may have had to steal cookies off a machine and decode them to get the password for an online e-mail service. Now, the attacker needs only to gain entry into the user’s browser, where Gmail stories a copy of the inbox.” But what exactly does that mean? Has anyone demonstrated this in action? It sounds like the old “If you enable cookies, people can steal your credit card information” canard. I’m not denying HTML5 has some bad security issue, but I haven’t found information specific enough for me to take a specific action yet.
There’s vulnerabilities I think in Safari with
Lol I guess HTML isnt working, I meant to say…
I think theres vulnerabilities in Safari with “canvas”
I saw something like that too in the same article (but with Opera), but again not only is there no actionable advice, there’s no reference even to a document describing the issue in detail. In fact, the entire article (“HTML5 raises new security issues”; InfoWorld has the same article under a different title) has not even one actionable item. HTML5 security issues sound like Bigfoot to me right now (“He’s out there; I‘m telling you, man! Look at this fuzzy photo!”).
Well… I can’t call it a security hole in HTML5 but a security hole in browsers.
If we’re talking about a security issue concerning a “default” behaviour of any HTML5 elements we could call it a “HTML5 security hole” (since it were “by design”) but it’s not the case.
Whatever… the secure PC is the one turned off! :)
What do you think about the client factor? From what I’ve read, I think it’s fair to deduce that the main factor of not using HTML5 is the client.
Here’s the case :
Clients pay. They’re (most likely) not gonna pay more if we’re to use new technologies.
(Most) Clients don’t really care whether we are developing with HTML 5 or 6 or CSS3 or 1.
Sure, there’s graceful degradation and stuff like that. But do (most) clients care about that? I don’t think so. They just want their sites done. Not to mention degradation adds to development time.
(Most) Clients only use one browser. If we’re fortunate, it’s not THAT dreaded browser and we’re cool. If we’re not, then say goodbye to HTML5. CMIIW, even XHTML and CSS2 struggles to display well in THAT browser.
So my point is like this. Okay, you have a project. You wanna start using HTML5. Your clients don’t mind but they do about paying extra. Also they still keep the deadline strict. Now, with new technologies to play, we need more time to learn. More time = more development time. Will you risk not being able to meet the deadline for HTML5 that won’t earn you extra bucks?
Just my 2 cents. Cos in my country, unfortunately, most clients are like that. So HTML5 for me at the moment is exclusively for (hopefully) my own portfolio site. lol
For some i do agree for some i don’t. @Hendra you have pointed out practical evidence we come a cross day to day coding HTML. I would like to add few more to it.
One thing we must not forget is the target traffic comes form each browser. I don’t think all of them support HTML5. Now IE9 is out. I know for sure it will cause loads of css errors. Luckily it is not compatible with XP.
what i feel we have to filter exclusive feature of HTML 5 which works with all major browsers and implement on production sites.
I just think HTML5 is the devil anyone who thinks about using it is a silly sally!
Do these people not know it has been proven that HTML5 is going to bring forth the apocalypse!
I’ve heard it takes you twice as long to code up a website, because there is technically less code to write.
I’ve heard its this completely new syntax to write as well, its like what could that possible be used for?
If only there were some some resources like a sort of shiv/shim or a boilerplate style package which makes it such a simple step that you will think: “why have I not done this before?”
Crazy stuff that no-one has come up with anything like this!
If you mean what I think you mean, like a framework that let’s you short code, it would all be the same, something else for you to learn how to implement.
Sometimes, going straight and using the source is better than shortcuts. You may very well, discover functions and uses you could miss out on if you use someone else s short code.
Sorry if I am totally off of what you were talking about.
@Johnathan you are wrong, I was being sarcastic about all the HTML5 haters who repeatedly make the same statements and ask the same questions over and over when they have been answered 1000 times. :D
I’m guessing the people who still use IE6 are mostly the older generation of people who just have a computer to read emails from their grandkids.
That’s why you should weigh up who will be using your site before you decide to use HTML5. A news site would have to support everyone but a blog aimed at web designers like the readers of css-tricks wouldn’t have to support IE6 / 7.
There’s no definite answer to whether you should use HTML5 yet, but it’s a viscous circle because if we start using it then people will be forced to update their browsers making HTML5 the most popular.
You never have to code for IE6 Chris!? I want your clients :)
I developed a series of sites last year for a southern California public college and everyone in that very large school was on IE6! It doesn’t mean the site visitors were, but try developing sites for clients who are viewing it in IE6 and telling them it doesn’t matter because only ‘old people reading email from their grandkids are using it :) It’s not gonna wash. For the client, the only browser that exists is the one they’re using!
I’ve noticed though that this whole conversation re: html5/css3 vs html4/css2 is reminiscent of the foot-dragging over getting folks to move from using tables for layout of (has it only been) 6 or 7 years ago?
A lot of us – as human beings – are a bit anxious to move out of our comfort zones. Technology moves like lightening and we either ride the strike or move safely out of the way!
I appreciate the conversation. Thanks Chris.
Hm… I had two website’s last year with some minor details that did not work in IE6 and in both cases my client phoned to inform me of this: I think I have to code for IE6 until the last XP PC dies.
Clients pay coders to deliver the website as designed by their agency or by yourself.
If you say you’re going to charge more to do the same thing in a different way (with some sparkling added a.k.a. html5 and new stuff) and extra to implement fallbacks so your site is viewable by everyone, what would they say?
a) If they’re into technology, like new stuff and can afford the cost, they will probably say yes, or even better, they won’t ask for the fallbacks, depending on the target audience for their website;
b) If they’re just… “normal people” who just want their website to look like it’s on the prototype, they’ll probably say no.
Most clients are b’s… that’s why the ‘not-HTML5’ won.
I so agree!
No reason not to use the doctype whatsoever.!
As for the APIs, take them one by one, but please don’t put the doctype and all the different APIs under the same ‘Oh, it’s html5 with security issues’ tag!! :)
I’ve been using the new input types whenever possible, since they fall back to good old type=”text” for the old, but give handy feature, like changing the keyboard, on mobile devices. Actually I really don’t understand why most sites don’t take advantage of input type=”email” or “number”. It would make entering data into those inpus much friendlier, while not taking anything away from anyone. And it doesn’t require additional development either!
You’re right, those results are very disappointing. I think it goes to show that HTML5 is still not very well understood. It’s not an all or nothing thing, there are tons of HTML5 features that people can take advantage of now without worry. Some concerns are things like “I don’t have the time to do it” or “I need to support older browsers;” but as a developer who works with clients, within their budgets, and still starts every project with the HTML5 doctype and takes advantage of features where possible, I simply don’t understand how others can just flat-out refuse to use it.
Douglas Crockford once made a point that new, better technologies are often adopted slowly because it takes the next generation of developers to come in before they’re used. The current developers are too set in their ways and too afraid of change to try any of it out for themselves. And then there’s also quite a bit of the “I’ve been doing this for X years, I shouldn’t HAVE to learn anything new” sense of entitlement. Many developers get frustrated with change, because they feel they just got a hang of the previous way of doing things, and now they’re being asked to update their knowledge and learn new things, and they don’t want to.
Is html5 w3c valid?
Being honest here, I’m probably just apart of the anti-bandwagon-bandwagon. I find fanboyism to be extremely annoying. So annoying, in fact, that even if what the fanboys exalt IS cool or revolutionary, I don’t feel like giving it a second glance. It’s probably just an extension of my antisocial personality.
Does HTML5 have the potential to change the web? I don’t know. Maybe? With so many people circle-jerking over it, I don’t really feel like finding out at the moment.
I guess you can just blame my headstrong youthful ignorance.
If you have a job to do, just get after it. Roll up your sleeves, get the tools ready that you already know you can be productive with, and start dishing out some instant productivity.
Then maybe after you get some work done you can read you feeds and do some scholarly thumb twiddling.
Woah. Maybe I’m having Deja vu.
i dont think html5’s w3c valid yet, am i right?
What does w3c valid mean?
The HTML5 specification is regulated and developed by the W3C and it possible to right valid HTML5 so technically then yes HTML5 is W3C valid!
Just because you can use something doesn’t necessarily mean you need to. What is gained by using HTML 5 that is so great that everyone should adopt it even tho it’s labeled as experimental and has the possibility of changing over time?
Also, it’s a little hypocritical to excited about HTML 5, which requires crutches (Shiv, Boilerplate, etc) to work on browsers, yet get on someone’s back for supporting IE using whatever tools they deem necessary.
See through the buzz words. If the whole world shifted to HTML 5 right now, it really wouldn’t make that big of a difference.
Haven’t tried HTML5 yet, but i don’t do anything anymore without the use of css3…. gradients, round corners and box-shadow allow me to reduce 80% of the graphics i would use
I liked the poll on sidebar about what will use from HTML5 in new project. Even I have shared the data on my FB page and given the link here.