Intellectual honesty and html5
0xdeadbeef.com
0xdeadbeef.com
Apple is at the forefront of standards adoptions, Safari's got some stuff that other browsers don't, and they want to show it off! What on earth is wrong with this? Firefox has better standards support? So what? Are we claiming Apple has a moral imperative to create demos of Firefox-only standards support?! Or, is it the even crazier claim that only the "best" product is allowed to promote itself? If you're selling Porsches and you say they're fast, is this offensive because Ferraris are faster? Good God.
I've also never before seen anyone so self-righteous about the use of the term html5; there's nothing about Apple's page which deserves special enmity for its "misuse" of this ambiguous term. In fact, I'm pretty sure in the future we'll see plenty of articles talking about html5, just as we have in the past, e.g., scribd converting from flash.
If you can't see how a company might find value in illustrating how quickly it implements emerging standards, I suggest you aren't thinking about it very hard. This page isn't about why web standards are important for cross-platform compatibility. Criticizing it for not demonstrating something which is not is purpose is awfully silly.
The outrage stirred up by the last couple of posts on this subject strikes me as follow-on to other perceived "evils" by Apple, and seems to be heavily biased rather than treating this issue on its own ethically distinct terms. But then, that's what people do.
The term "HTML5" is supposed to be a standard, but the term has become a dumping ground for various parties. They introduce non-standard (or, at least, not-yet-standard) extensions in their browsers and call it HTML5. It's like Microsoft calling in-browser DirectX transformations HTML5.
I don't think a technology may actually be called HTML5 until it's standardized by WHATWG or by W3C, otherwise the term will become meaningless, and it looks like both Apple and Google are quickly on their way towards making that a reality. Once the term has come to mean "arbitrary cool browser-specific extensions" we are only one small step away from the age-old "embrace and extend" practice.
Coincidentally, Safari's recent Webkit 3D CSS transformations support (used everywhere in their demos) is very similar to DirectX transformations. I'm not saying that there's anything wrong with that, but as far as I know Webkit 3D transformations aren't (yet?) standardized by WHATWG or W3C.
The thesis I argued against is simply the best triangulation of the multiple, weakly expressed arguments made to justify the negative reactions.
And the claim that these "extensions" are the same as MS's from years past is, to put it mildly, nonsense. This has been pointed out elsewhere, so I won't rehash it. The aggressive refusal by some folks to not recognize this looks like bias to me.
See this great comment from yesterday: http://news.ycombinator.com/item?id=1403749
For the record, when Google created Chrome Experiments (which used HTML5-ish technologies), they let any browser attempt to run the experiments, but the performance (in some cases) suffered. Apple is doing exactly the opposite.
On the consumer accessible page Apple are requiring Safari because they can guarantee that the demos will work and they can guarantee they will have reasonable performance. And that's what's important. That's an advocacy page, a PR thing, not "look at this cool code we wrote". Presentation is everything.
In other words, different goals - different implementations.
the really important part of html5 at the end of the day is for CONSUMER, not developers.
Here Apple doing exactly what dumb people did when they started the browsers war 10 years ago "this site is best viewed with X", which later degenerated to "if you don't have browser X, go f*ck off" (you know wars ... people pick a side).
A browser is not here to guarantee an experience, it's here to guarantee you to browse content, and wether this browser support whatever version of HTML, this browser should not be blocked in any arbitrary way because the HTML should gracefully degrade, yeah even with Lynx on a command line.
http://www.anybrowser.org/campaign/abfaq.html
that's what people don't get about HTML5 and the article make a good illustration of it.
I'm sorry, how is that supporting standard again? I'm not illuminated yet.
I tried the demos in Firefox and Chrome. Not all of the demos worked in either of them, so much for "could probably". It's not about Safari trying to make itself look better, it's about Apple not letting html5 look bad. If someone not-so-knowledgeable surfed to the demos with browser X and not everything worked then they may proclaim that html5 is crappy and doesn't work. Apple is trying to show-off html5 and that would be the opposite of their goal.
add.: to the downvoters... really, just load the page up in Chrome, check the demos, then load it up in Opera (with User Agent spoofing) and check the demos there, THEN try loading it up in Firefox, and you'll see for yourself whether it's a case of these HTML5 demos being "proprietary", "Safari only" or "broken", or if it's a case of Firefox just not cutting it yet. (Spoiler: it's a case of Firefox just not cutting it yet; almost all of the demos work identically in Chrome and Opera.)
- Let's create some demos of our new HTML5 support!
- Do we want to be responsible for users' bad experience if other browser vendors make mistakes or implement things differently than we did?
- No.
That's it. Nothing nefarious about it.
Did you read the article? Those demos are not even part of html 5 yet (the ones that don't work on chrome and firefox at least). So calling them HTML 5 was not honest and a shameless promotion of Safari. Also, most of the demo do work great on both Chrome and Firefox, except the demos that are not HTML 5.
He highlights a paragraph which apparently is the important bit (and it is) but this, for me, is the really insightful statement (in the last section):
HTML5 is in a dangerous place since everyone wants to own it, but everyone is in a different place in terms of support or even what it means.
So true. This is the risk with HTML5 - it is becoming a buzzword for companies such as Apple to jump on and market with. Which isn't a bad thing; unless it is at the detriment of the standard.
The most important aspect of HTML5 isn’t the new stuff like video and canvas (which Safari and Firefox have both been shipping for years) it’s actually the honest-to-god promise of interoperability.
That’s probably their intention, not showcasing interoperability or openness. Those two words do their little buzzword duty and that’s that. (I’m really not all that irate about that. It’s a cheap shot at Flash but many – probably all – of the things Apple uses in their demo actually will be a interoperable and open standard in the near future and all those evil browser prefixes will be dropped.)
I suspect that they simply let their desire to make the demos really impressive distract them from the original point of the demos. They could, I imagine them thinking, make some cool demos that would work in Chrome and Firefox and Safari, or they could make some even cooler demos that showed off things only available in Safari, and have a nice way to promote Safari to boot.
The problem is, once they did that, it went from a showcase of HTML5 standards to a showcase of proprietary Safari features that might be standards eventually. Oops. I'm rarely critical of Apple, but this is one of those times that makes me wonder if anyone there had their brains turned on.
That’s something you can do. I don’t think it’s nice of them to throw around buzzwords while being so completely irony resistant, it’s just that I don’t think it’s a reason to be greatly angered.
This demo requires a browser that supports CSS 3D
transforms.
To view this demo, you’ll need Safari on Mac OS X Snow
Leopard, Safari on iPhone OS, or the latest WebKit
Nightly Build.
I don't think it's realistic to get people to stop saying "HTML5" and start referring to all the little semi-standard pieces individually, so I'm really curious what this "intellectually honest" Mozilla messaging is going to look like.
Marketing using the word "open" is the new fashion.
I see words like these as very useful inflection points; on one side of the curve, a developer that uses them without revealing a more nuanced understanding is a developer I should avoid.
And on the other side of the curve, it's a way for non-technical people to refer to a complex set of technologies they otherwise probably couldn't talk about (or spend money on).
I have a feeling that we're going to have to define a whole bunch of terminology to move ahead. People are confusing the HTML5 "markup additions" to the HTML5 "javascript upgrade" to Apple's HTML5 "hey look Ma, no Flash and lots of shiny!" (And to be fair to Apple, even Chrome doesn't fully support Apple's CSS3 shiny 3D toys, so that's why they had a Safari Only "HTML5" demo.)
EDIT: To the downvoters, I'm sorry that you don't like the truth.
The bad thing about branding them collectively as some sort of "HTML5 standard" is when people start calling other browsers non-compliant with standards that aren't even stable targets yet. Or targeting one browser's draft implementation, without even checking to see which other browsers have comparable support.
This is a major problem for us on the Mobile Firefox team. Since a huge portion of mobile web pages are now targeting WebKit browsers only, it's hard for competing browsers to enter this space. Mobile Safari is the new IE (in the sense that we need to reverse-engineer its non-standard features like meta[name="viewport"] in order to compete with it).
Are you really insistent that you come op with your own "-moz-viewport" crap just because there isn't a draft standard? The vendor prefix bullshit of the last few years is going to be around for at least another decade, with every CSS3 property being set three goddamn times because the vendors will have shipped support for the prefixed version for several major releases before they think about blessing it. I wish y'all would only do that stuff in beta releases to keep it out of the wild.
Setting the viewport doesn't require anything like trampolining through VBScript, loading an ActiveX control, or calling a DirectX function via VML (all things I've had to do recently). MobileSafari is not the new IE.
Have fun being another legacy browser on a slow release cycle that can't ship a decent mobile browser.
No, we're not insistent. We implemented WebKit's "viewport" meta tag. We can be pragmatic too.
I just wanted to point out one of the many of Safari features that Apple is promoting without writing any specs or standards, and the effect this has on the browser market. Others include touch events, link[rel="apple-X"], -webkit-text-size-adjust...
I agree the proliferation of vendor prefixes in the wild has bad effects, and we are feeling the pain of it too.
----
> Have fun being another legacy browser on a slow release cycle that can't ship a decent mobile browser.
Thanks for the kind words. :) We know we're way behind the game on mobile, and the only thing we can do to fix it is ship great software. I think when you get a chance to see what you can do with add-ons for mobile Firefox, you'll be tempted to switch.
We're on a rough six-month release schedule: Fennec 1.0 shipped in January, Fennec 1.1 will ship this month, and we are targeting this October or November for Fennec 2.0. Look for our first 2.0 alpha release for Maemo and Android in a month or two.
[P.S. I see you're in Seattle too! Want to get together for coffee sometime? Email me if you do.]
You said exactly why it is the new IE:
> The vendor prefix bullshit of the last few years is going to be around for at least another decade, with every CSS3 property being set three goddamn times because the vendors will have shipped support for the prefixed version for several major releases before they think about blessing it. I wish y'all would only do that stuff in beta releases to keep it out of the wild.
This is like "Draft-N" Wireless too.