Posts Tagged ‘responsive images’

In Soviet HTML5…

Wednesday, August 29th, 2012
CSSquirrel #97: In Soviet HTML...

Today’s comic features Mat “Wilto” Marquis (Internet folk hero) in a nightmarish world that I imagine Ian Hickson envisions if HTML were ever allowed to slip from his hands and be created by the common man.

We’ve discussed the issue of responsive images in HTML and the proposed <picture> element before, you and I. Here’s a look back in case you need a refresher. It’s a bit cheeky. I get worked up over things like one smart person thinking his brain is more efficient than the combined powers of dozens of smart people.

<Picture> has had a rough road.

All the way back in December 2011, Bruce Lawson posted this idea for how the syntax of <picture> might work. Several people started a community group chaired by Mat Marquis back in February (based on a WHATWG email suggesting such) to push the idea forward so it could see eventual adoption into the HTML spec. They did a great deal of good work to make a workable, problem-solving piece of markup. Scott Jehl even created a polyfill script to provide support for the proposed markup until browsers implemented it, which you can find here.

Then Hixie stepped in. It’s been well documented in the past before, so I’ll summarize: he decided to ignore the community group and efforts of the developer community and went with srcset. Classic Hixie solo action. Here’s an aptly named summary, WTFWG by Tim Kadlec, and a piece on A List Apart by Mat Marquis.

Developers were mad. Developers were vocal. Then for the most part, the web moved on.

If it wasn’t for the continued effort over the past several months of people like Mat, that would have been that. But he and others didn’t stop, and they kept on pushing, and wouldn’t you know it? Something came of it.

Not in the WHATWG, of course. Someone hoping for that sort of miracle would be essentially signaling to everyone that they finally lost their last marble. (Here’s an example of how things were going over there, AKA, Hixie’s got his blinders on like normal.)

But over at the W3C, Mat and his allies accomplished this: an official w3C Editor’s Draft of a proposal for HTML Responsive Images Extension, featuring both srcset AND <picture> to be included into the HTML5 spec. It’s just a step, but it’s a fairly big one, and a great deal of validation for the developers that worked so hard, despite Hixie’s best attempts at ignoring them, to create a responsive image solution that works for them.

I could tell you how it’s a heroic accomplishment that is leading us towards a better web. But I won’t. Not because I think it isn’t (it in fact is!) but because Marc Drummond already said it better. Go read his article Responsive images, the picture element and the W3C: This is how you deal with Hixie and WHATWG. It’s worth it.

It’s hard to say if <picture> will ever make it into the browsers. The browser makers are mostly (but not entirely) involved in WHATWG, and would prefer to work around the W3C. But I do believe that enough community effort will continue to build on the success that Mat and the others have brought us.

I asked Mat about what he saw as the future of <picture> in light of these recent accomplishments, and what interested developers should do to help. Here’s his response.

I totally stand by that follow-up tweet I posted earlier ( http://twitter.com/wilto/statuses/240486846801514498 ): no matter how things play out from here, the dev community has done something that I don’t believe has ever been done before. You just don’t see specs – proposed or otherwise – with a developer’s name at the top. If nothing else I hope this marks the beginning of that, and that it leads to we authors having a voice in web standards equaling that of the vendors.

To get involved, more folks should join up with the RICG! http://www.w3.org/community/account/request

Let’s not waste the opportunity this has given us. We’re the developers that will have to work with the spec that’s being created, and will be answerable to the clients and site visitors that will use the sites we create. We should and can have a bigger voice in the direction of HTML. What Mat and others have done is show us how.

The Egotistical Puppet King & I

Tuesday, May 15th, 2012
CSSquirrel #93: The Egotistical Puppet King and I

In a way I should be grateful to Ian “Hixie” Hickson for being an egotistical tyrant. Without his inability to acknowledge that a consensus-driven, well-crafted and usable solution built by a group of well-meaning, hard-working people could actually somehow be better than his own personal opinion, he’s pulled me out of my long-hiatus and back to to the drawing board.

Today’s comic is in fact three comics. No single idea could encompass everything. In all three Hixie gets top billing as the editor-for-life of HTML’s “living spec”. The first comic features Naepalm, chinchilla alter ego of fellow Mindflier Janae Wiedmaier. The second one includes the irreplaceable Justin McDowell. Lastly we see Matt May, Dylan Wilbanks and Ethan Marcotte joining forces with the Squirrel in a bid to take down the HTML king.

(Today’s comics as per usual aren’t meant to imply that the people represented therein endorse my views. I’m saying it outright today because I’m feeling particularly vitriolic and don’t want my words to reflect on them.)

The Situation

For those of you just tuning in, today’s outrage focuses around Hixie’s decision to adopt a problematic, late-arriving, Apple-proposed attribute of the <img> tag into the HTML standard as the solution to the adaptive images issue. In the process he again reinforces his inability to heed the creed of HTML’s priority of constituencies (end users over authors over implementers) while also tossing away the hard work of a community group of developers that built a very functional, very usable solution to that problem in the form of the <picture> element.

If you’d like a summary, you can check out the aptly titled WTFWG by Tim Kadlec, or take a look at Zeldman’s take on the situation over here, which links into an A List Apart article by Mat Marquis on the topic.

I wish this was a new situation. Or that it was surprising. Or that I didn’t feel like I was repeating myself each time I mention Hixie in blog or comic form. The fact is that as the Editor of HTML, Ian keeps doing this. And we keep letting him.

The Puppet

At one point I attributed this issue solely to his gigantic ego and clearly overwhelming case of not invented here syndrome. Now I’m frankly convinced that although these qualities contribute to the problem, the real issue is that he’s the puppet of the browser vendors, namely the three most involved in WHATWG: Apple, Opera, and Google. Although the priority of constituencies dictates that implementers (aka, browser vendors) should be lower priority than developers (who are in turn answerable to end users), it seems that without fail Hixie will bow to the vendors before considering any work on the part of developers at a solution, no matter how reasonable, well-built and documented that solution is.

That’s not a kind accusation, that a man is a puppet. But clearly every attempt to work with the WHATWG has always resulted in developers being treated as second-class citizens to the browser makers. And let’s make it clear: this is our job. We make websites for a living, and the tom-foolery that Ian is engaged in is directly impacting our present and future workflow. We work on making websites every damn day. We know what works for us, and what doesn’t.

And he doesn’t care in the slightest.

“Work With Us”

At this point, Hixie and his backers are relying on the same smoke and mirrors to distract people. Present use cases. Keep engaged with the WHATWG and let them know your technical objections. Get involved in their IRC. But the fact is we as developers have done this over and over and over. Yet at the end of the day, regardless of the hard work put in and all the proof jammed into the pudding, it all amounts to naught. Hixie spends twelve seconds coming up with his own solution or takes what the browser makers gives him and uses that instead.

It happened with metadata. It happened with the <time> element. It’s happening now with responsive images.

The fact is that Ian doesn’t give a shit.

He’s going to do it his way, or failing that he will do what Google and the other browser makers in the WHATWG tell him to do. He’s not going to look at what the developers have built and give that solution a thumbs-up. As John Foliot said: “Dev community, if you continue to author to the WHAT WG doc, you lend your tacit support to heir hixie. Look where that gets you.”

Enough Is Enough

Being part of their process is being part of the problem. I’ve never seen things resolved by following the WHATWG’s “process”, as it amounts to little more than distracting developers while he goes off and implements a less functional, more complex solution to the problem.

Don’t deal with Hixie. Don’t deal with the WHATWG. Directly object to the browser vendors. “Occupy” HTML by making use of the consensus-built techniques that already have functional polyfills. Do what makes sense, and what works for you.Sooner or later, the browser vendors will be tired of the grief sent their way and tell Hixie to roll over.

<time> wasn’t fixed because we followed the WHATWG’s process.

I’m going to say it: I don’t believe that WHATWG is part of the solution anymore. As I’ve been told by others, democracy isn’t always the best approach. Sure, ok. But so far when it comes to the community-build, consensus-driven approach and Hixie’s brain, the community’s solution has proven more effective more often.

I’m going with the community, not the puppet.