Page 2 of 4

Re: HTML5 hype

Posted: Thu Sep 29, 2011 2:02 pm
by flying_circus
matthijs wrote:Sure, when you start using the new HTML5 features, you have to check which support there is for them. Or what the fallback will be in case something is not supported in a browser. Maybe you could use some of the new form elements. So that the people with newer browsers get to see some bells and whistles, while people using older browsers get to see plain form fields. No harm done there.
I though that's what we were talking about?

Re: HTML5 hype

Posted: Thu Sep 29, 2011 4:03 pm
by matthijs
But what are you saying then? Now I don't understand what you mean.

All I'm saying is that it's perfectly safe to use HTML5/CSS3. No reason to wait 10 years. It's just that you always have to check what works and what not. And use what makes sense for each project. But that's something you have to do anyway, with every technology you use. Waiting for a spec to "get finished", you will wait at least 10 years (seriously, it's probably going to be finalized around 2020). And by that time, another new technology or variant of HTML will be the thing to use.

Re: HTML5 hype

Posted: Thu Sep 29, 2011 6:40 pm
by Eric!
Well, that's certainly how things are working in the wild. But it doesn't have to be this way. If they really want people to adopt and use a standard, then it has to be a priority that the standard is exactly that, a standard. Fixed and published not tweaked constantly and begged to be complied to the latest changes, IMHO. Otherwise plug-ins, hacks, work-arounds are going to be the norm and HTML5 will struggle along and it won't live up to the hype. In the hardware engineering world you write the spec then build it. Doing both at the same time is easy for the person writing the spec, but everyone else suffers.

And like you've said, we'll just have to use bits and pieces that we can manage to get working on an acceptable sub-set of browsers. I spend much more time fixing browser related issues than I would like and that task seems to keep expanding. It makes me afraid to try new techniques. Whereas if I want to do something flash related, there's really no compatibility issues I need to worry about at least until HTML5 kills the plug-in....

Re: HTML5 hype

Posted: Fri Sep 30, 2011 12:59 am
by matthijs
Sure in theory it would be nice if someone would write some spec, get all browsers to implement it all at once and then declare that "HTML is finished!". After that developers could learn the spec, build sites and never have to worry about anything changing.

But think about it: it would also mean there would be no development anymore. People/sites/browsers would support spec X. If you want to change anything, you would have to write a complete new spec and again go through the process of completely switching over everybody (browsers+developers) to that new spec. That would never happen, as everybody was using the older spec.

Further, this theoretical situation would not stop anyone (any company) from developing those plugins/proprietary techniques you talk about. If HTML would be declared "done", not long after that new technological possibilities (plus commercial reasons) would cause those companies to try to get their plugin/technology to be used. But then you are stuck with some proprietary technique owned and controlled by one commercial company. That's just not Good for the web. Look at all the microsoft crap that's still out there, only working in IE6, holding back the companies using that from upgrading their software and causing many problems.

It's just the way the web works. It's in continuous development. It will never end. Sure it can be a hassle to keep up. But I see the positive sides:
- web standards are open, not owned by one company
- web standards are supported for a long time. 15 yr old HTML sites still work on any browser nowadays (unless they used some proprietary technique back then!)
- you can slowly add new features one by one. Again, HTML5 is not something completely new.

Personally I have stopped caring for doctypes a long time ago. I don't put a badge on my site "Build with HTML5!". I just put a doctype there (currently the html5 one), write clean html and use the features that seem like a good idea for that specific project. Knowing the website I build will keep working for a very long time in upcoming browsers.

Re: HTML5 hype

Posted: Fri Sep 30, 2011 7:51 pm
by Eric!
Yeah, I'm not talking about declaring anything done. I'm talking about freezing a version of the spec and moving on to the next version. And the public community to be vocal in compliance to the completed specs as a standard requirement.

I've spent many years in design and development and in no way are specifications a limit on development. In fact I would argue the opposite is true. however I've noticed the internet software culture (not necessarily embedded or app development software designs) to be accustomed to constant tweaking in such a way that nothing gets finalized and things are rarely inter-operable because company A's tweak is better than company B's tweak.

Re: HTML5 hype

Posted: Fri Sep 30, 2011 10:37 pm
by califdon
This is surely a topic worthy of discussion, but I don't think we'll ever resolve the different attitudes, any more than we will the ideologies of the two political parties (here in the States), or get everyone to agree on the path to salvation of our souls. Each of us has his or her own history and experiences that shape our underlying view of how the world should work and we can never understand why everybody else doesn't see how right we are. That's human nature. Let's drink to that, OK? :drunk:

Re: HTML5 hype

Posted: Fri Sep 30, 2011 11:08 pm
by matthijs
Eric, I agree with you that the process of creating the web standards/specs is a muddy process. And I agree that it would be nice if it would be different. But I'm afraid that's just how the things work in these open specifications. There is no single company behind the development of the standards. They are developed by active developers, different working groups and different browser manufacturers. Sure that makes the process less efficient. But the alternative would be to follow (and be locked in) by a single company. That's not what you want either.

About the "freezing of a version" and moving to the next one. Maybe I misunderstand what you mean, but I think they tried that with XHTML2. XHTML1 was "finished", so they started working on XHTML2. But that version was different in many ways from version 1, so no browser maker was going to support it/was implementing it. And developers were not using it because there was no browser support. And no browser support/nobody using it = no future for the spec. Chicken and egg problem. So that's when another group started working on HTML5, being a more gradual evolutionary version, backwards compatible with existing HTML. As I said, change your xhtml1 doctype to a html5 doctype, and assuming your code was written correctly in the first place, you are now using HTML5 even though you didn't change anything else. New features are implemented and supported gradually. Not a very clean/structured way of working, but at least it works.

Califdon: I completely agree with you. We should definitely drink to that :drunk:

Re: HTML5 hype

Posted: Sun Oct 02, 2011 6:17 pm
by Eric!
It's easy to freeze a spec. For example CSS has finally worked its way into a somewhat reliable process after 2.1 with their working drafts, but it's still a bit messy.

I don't know. I've mostly worked on hardware and I don't know the XHTML2 story which doesn't sound like they were thinking right by not making it backwards compatible. But I've worked on large international public open specifications that were committee ruled where we repeatedly and successfully rolled out standards and built multi-billions of dollars of hardware and tons of software around it. Interoperability testing was mandatory to receive certification that hardware and software are complaint to the specs. Manufacturers tripped over themselves to make sure they could declare compliance. I just don't see HTML5 living up to the hype if it is all willy-nilly. And I don't get why it is this way or why people are fine with half the html5 not working, no road map for when browsers will support things, or any plan of compliance to those standards. (And perhaps these things exist but as an outsider I don't know about them.)

Perhaps that's the nature of internet based companies. I have a hard time imagining that any team working on browser development is higher than SEI 3 from the way everything is slapped together. But since I don't see any SEI standards discussed about those teams, maybe they have no quality standards in place...?

No drinks for me...too hung over.

Re: HTML5 hype

Posted: Sun Oct 02, 2011 7:14 pm
by flying_circus
matthijs wrote:As I said, change your xhtml1 doctype to a html5 doctype, and assuming your code was written correctly in the first place, you are now using HTML5 even though you didn't change anything else.
No, now you are using XHTML1 with an HTML5 doctype. When this topic was started, I assumed it would be about the new features of HTML5, as no one really cares that HTML5 is backwards compatible, that is expected.

The reason I refrain from using HTML5 (and by that, I mean new features that were not supported in HTML4 or XHTML1) is browser inconsistency. This is why I make the argument that it is fun for personal or cutting edge technology sites, but I would not use it on a clients site at this point.

http://html5test.com/

Re: HTML5 hype

Posted: Mon Oct 03, 2011 6:31 am
by matthijs
@flying_circus: if I have a HTML5 doctype and my html validates to that, by definition I'm using HTML5. In practice, I also started using some of the new elements like nav, article, footer, section, so there are more differences then only the doctype

Anyway, I don't see the problem. I use what's possible on each project. On some it will be plain "old" html4, on others I'm able to use more advanced HTML5 features. The same with CSS/CSS3. My clients don't even know what spec I'm using for their websites. They don't care, as long as their website works and does what it is supposed to do. And they trust me to build their website in a good way, which I do.

Re: HTML5 hype

Posted: Mon Oct 03, 2011 1:57 pm
by califdon
For a chart of which specific elements and functions of HTML5 that are implemented in various Windows and Mac browser versions, take a look at http://findmebyip.com/litmus. It's looking pretty good for most current and even one-level-down browsers.

Edit: here's an even more up-to-date chart, arranged differently: http://caniuse.com

Re: HTML5 hype

Posted: Mon Oct 03, 2011 3:59 pm
by matthijs
That's a great reference califdon. Thanks for posting.

Re: HTML5 hype

Posted: Wed Oct 05, 2011 7:26 pm
by Eric!
According to that list some of those features date back years for when they were already built into the browsers. It just seems so crazy that they can't even come up with intermediate fixed goals that all the groups could agree to and adhere to when you look at a 3+ year time frame. So instead we have this giant pot of ideas called HTML5 with no completion date or any idea who is going to do what by when. Now the browsers are bloating up to get ready for anything. It's crazy.

Re: HTML5 hype

Posted: Wed Oct 05, 2011 7:44 pm
by califdon
@Eric!: Have you ever had to try to negotiate standards with "all the groups" of any kind? It's much harder than you probably imagine. Would that it were so easy.

Re: HTML5 hype

Posted: Wed Oct 05, 2011 8:51 pm
by Christopher
Eric! wrote:According to that list some of those features date back years for when they were already built into the browsers. It just seems so crazy that they can't even come up with intermediate fixed goals that all the groups could agree to and adhere to when you look at a 3+ year time frame. So instead we have this giant pot of ideas called HTML5 with no completion date or any idea who is going to do what by when. Now the browsers are bloating up to get ready for anything. It's crazy.
I haven't been following this discussion, but I think you have missed the changes in the last decade or two in how things are designed. No one creates giant, monolithic specifications and standards anymore. Everything is evolved through trial and error and the whims of markets. The whole idea that there is some complete standard that all browsers could implement that would be perfect is what sounds crazy. You can't just reject reality because you don't like parts of it.