Anyone Else Sick of HTML5?

UCDVet

Maybe it's just me, but man, the whole thing seems like such a failure.

Atomic Pineapple

A failure in what way?

UCDVet

Atomic Pineapple wrote: »

A failure in what way?

The whole point was to have 'one way' of doing things. And now we've got two different versions of HTML5; we've got the W3C version and the WHATWG version.

I'm reading the official Microsoft exam guide (for exam 70-480) I got as far as Chapter #1, Section #1 before I reached an already depreciated HTML5 tag (hgroup).

So we have two versions, that are constantly changing - not just new things being added, but only things being removed (with individual browsers deciding whether or not to provide backwards compatibility)

Worse, even just among desktop browsers, you can have an HTML5 compliant page (assume it passes both version's semi-official validator tools), and it will render differently.

I'm going through beginner video tutorials and even the things meant to teach people HTML5 are frequently wrong. For example:
http://www.microsoftvirtualacademy.com/training-courses/html5-css3-fundamentals-development-for-absolute-beginners

HTML5 and CSS3 *FUNDAMENTALS* development for absolute beginners. You'd think the fundamentals wouldn't be a problem, right?

By video #3 (the files are available in a .zip file) but the example they give renders correctly in Chrome, but not Firefox.

And that's not even touching all the problems you run into with mobile devices!

At the end of the day, it's pretty unimpressive, imho.

Aswerty

That is just how front end web development is. Without a benevolent dictator the design by comitee approach rules supreme. Lots of technologies were created to deal with the problems of web UI development including: Bootstrap, jQuery, Modernizr, Autoprefixer, and many more.

oscarBravo

Meh. It works for me - certainly better than previous HTML versions did.

EamonnDunne

Aswerty wrote: »

That is just how front end web development is.

Indeed, the original web was never designed to be used how it is today. This is primarily the reason that front end development is such a frustrating mess.

ILikeBananas

UCDVet wrote: »

The whole point was to have 'one way' of doing things. And now we've got two different versions of HTML5; we've got the W3C version and the WHATWG version.

srsly78

http://english.stackexchange.com/questions/45295/why-is-there-confusion-between-depreciated-and-deprecated

Nollog

Maybe they'll finish it soon, and they'll concentrate on making html6 overly complicated and unstandardised, while the rest of us and browser-makers bring html5 to a single standard (and a seperate microsoft standard).

UCDVet

/\/ollog wrote: »

Maybe they'll finish it soon, and they'll concentrate on making html6 overly complicated and unstandardised, while the rest of us and browser-makers bring html5 to a single standard (and a seperate microsoft standard).

Soon?

I've been hearing how HTML5 is the solution to all of our problems since...like 2005. HTML5 is like the Duke Nukem Forever of software solutions.

UCDVet

oscarBravo wrote: »

Meh. It works for me - certainly better than previous HTML versions did.

I dunno. Back in the day, it was just understood and expected that, most of the time, you'd have to do some extra work to get things to render correctly in more than one browser. You'd develop a little bit, check it in whichever browsers you cared about, as you went.

Now, it's pretty much the same thing - except every idiot in a suit whose seen a powerpoint presentation on HTML5 thinks you'll write something once and it will work on all the browsers and mobile devices. At this point HTML5 isn't even a new buzzword - it's been the hottest new tech for, just about as long as I can remember. 2007? 2008?

In 2008, I could write a webpage with Flash and it would work the same in every browser, or not work at all. I could do the same with Java or Silverlight. I still can't manage that with HTML5. And at this point, I think it's painfully obvious that we will NEVER reach that point.

I don't know any other development suite/languages/markup/whatever that has the same problem. If you develop something in Python version 2.X people either can run it or they can't. If they can run it, it will run the same. Sure, if you decide to target Python 3 you'll need code changes...but the consumer of your app doesn't have any problem.

Imagine if you were writing a .NET or Java application and had to check http://caniuse.com/ to see what features of the language you could use based on what web browser the user had installed! It'd be maddening and nobody would praise it, we'd all say, 'Wow, this sucks!'

With Assembly, sure, you've got that problem where you might be targeting a specific piece of hardware. But there isn't an expectation that it will work for anyone but people running that hardware. And Assembly is generally considered something most people don't want to touch with a 10 foot pole.

And rather than getting closer to an accepted standard, we seem to be drifting apart with two bodies and each of the major layout engines picking, choosing, and interpreting whichever specs they want, giving different amounts of emphasis to backwards compatibility and legacy.

With HTML4 - you didn't have a canvas and lots of other stuff. You'd have used a plugin to achieve the same thing. Fair enough. But we're still up to our knees in crap that anyone not doing web development will never have to deal with.

John_C

oscarBravo wrote: »

Meh. It works for me - certainly better than previous HTML versions did.

+1, things are better than they used to be.

Anesthetize

Youtube has just made the switch from Adobe Flash over to HTML5. Anything that can replace the use of Flash can only be a good thing.

bpmurray

It's definitely much better than it was - even supporting IE doesn't add 30% to the cost of development nowadays.

I hadn't realised that anyone followed the WHATWG "standard" - certainly none of the large computing companies would consider them over the W3C for a second. They're well-meaning but with no industry backing, i.e. they're marginalised without impact.

So here's a possible way to approach it:

• Understand that there's a difference between the standard and how the browsers implement that standard.
• If you hit a place where browsers differ, discover why the situation is as it is and report it as a bug to the browser folks if it really is.
• If it's an issue of interpretation, contact the W3C and ask them to clarify how the feature should work, referencing the conflicting implementations as an illustration of how unclear the spec is.
• Create an abstraction that hides the differences, and program against that. Or use one of the many abstractions in things like JQuery, Dojo, Bootstrat, etc. and use that instead.

See - it's not that difficult. Just be glad you don't have to keep referring back to quirksmode.org!