Don’t Blame the W3C for World Hunger
This post comes in the wake of several other posts and conversations – notably Molly Holzshlag’s response in A Proprietary Web in Deed and Fact? to a post from Paul Ellis titled A Proprietary Web? Blame the W3C. In response to Paul’s article directly I’d agree to a point, yes Rich Internat Applications (RIA) are the direction the web is headed. I’d argue its not the only direction, and that we bring the past along with us in the form of HTML and other standards (recommendations). Yes I agree that Flash and Silverlight have their own business models and every right to exist. After all, if the web stopped being a viable commercial environment it wouldn’t have a future. However, I don’t see HTML and Flash as occupying the same space. On many levels we share the same view but my perspective is a little off to the side, I guess.
From my view, HTML and others are what I’d consider core technologies which should be universal bottom lines supported (in my dreams) by browser vendors. Its fair to say that’s a pipe dream but its the ambition, at least. That, to me, is the open web. There is also room for the silo web experience as long as the vendor is willing to accept that not everyone can use or is willing to use their product or technology. For example, using Opera I attempted to view Paul’s Silverlight example of the Hard Rock Cafe’s Memorabilia page. Let’s just say that between the makers of Silverlight and the developers of the Hard Rock Cafe site, if they can factor in the loss of business from excluding customers (and not being sued by anyone who is locked out) then that’s their choice. Accessibility and the legal, moral, commercial environment hold sway there.
I embrace RIAs but unfortunately in many cases they’re over-used. Every web project should boil down to the business goals and objectives the site is meant to achieve – Return on Investment. Not simply flash and glamour and the awesome experiences of interactive televisualisation in my living room! As an example, why create an information site that is absolutely artistically swishy and beautiful if people then find it hard to locate and use the information the site exists to disseminate? In other cases RIA’s make more sense, but they’re never going to be the whole web experience. If that makes sense. And whatever we build does need to be usable and accessible and intuitive and understandable and meet our mental models and yada yada. I’m diverging, and I apologise.
My comment on Molly’s post sums up my 2 cents on the W3C issue and where standards (or recommendations) are a stifling bottleneck. We need to rethink our approach to the specifications (and several pedantic argumentative types should be run down and paddle-whacked). We need to stop being our own worst enemies as standardistas. My comment follows:
Mmm while I agree with your three statements about web standards Molly, because you’re right – they aren’t standards (just recommendations), there isn’t interoperability, and no open web (if I understand the concept you mean by that).
I do, however, think that any set of technologies does need to be standardised because it gets very hard for 2 machines to talk to each other consistently without a shared vocabulary. Its fine for some people to comment that it should be a free market driven atmosphere but if things did move too fast in that market how would hardware and even software keep up with the disparate language differences between programming solutions in points A, B, C etc… seriously, that would be even worse for interoperability (which would rely on a common understanding).
Web standards (or recommendations) have their place. Unfortunately I see things another way, I think a lot of web standards progress have been shot in the foot by individual standardista’s personal agendas and sometimes by fame and sometimes by a combination of those forces. Design by committee is always dangerous, pleasing everyone pleases nobody and tends to create products that are overall mediocre in the end.
My opinion, for what its worth. Small group go in and make a core standard around a technology. Then release B expands that to include next step features, and iterative improvements follow. I’d say push the big boys off the table for that process, too. On day X provide a specification they either follow or not – their commercial choice. If they follow everyone gets to interoperate. If not they get a product in a silo.
That’s my 7am before work rant for the 2 cents its worth. We critically need standardisation, and the support between vendors for that in their products. The 2 major problems I see are that W3C is bogged in beaurocracy (partly because standardistas come in a zillion flavours, not just vendors) and its become about trying to please everyone. How about just producing a best model working spec with some beta testing? A little more agile. A little more proactive. A couple of simple plain english objectives on the wall of that Working Group – interoperability, accessibility, ease of implementation.
Then invite everyone in the door at the end, very smart people, and THEN we ask “how do we do it”? We ask “how do we make this new spec happen”?
Cos the current process is a duck + a beaver = platypus… (although I do love the platypus, don’t get me wrong)…
The W3C are easy to blame for all of our woes about the non-perfect web. But this is an international environment and we haven’t achieved world peace yet, or the eradication of curable diseases by injections of money. I don’t think the W3C are the entire fault to any of our issues. It boils down to being human. Being individuals. Having expectations and agendas of our own, within and without committees and working groups. We have to take some responsibility as standardistas for perpetuating pedantic circular arguments. Blaming the W3C is just too easy.
July 12th, 2008 at 4:12 pm
Hi Steven, As project lead for the Hard Rock Memorabilia site that you referred to above, we just deployed the necessary fix to the site for Opera users. It was a one line fix even though Opera is not currently an “officially” supported browser by Silverlight 2. Works great though.
This means that the app now works in IE6/IE7, Safari 2/3, FF 2/3, Opera, Camino on both Mac and Windows. And possibly more, but this accounts for 99.9% of our traffic.
July 13th, 2008 at 8:35 am
Hi Will, cool thanks. I’ll pop over and have a look.
The point I was trying to make in the article was that any time we make the decision to create an “enhanced” experience, whether its with silverlight or flash or whatever, then we are making a decision that excludes some part of the audience. For example, those without the plugin or who don’t want to install it or have no time and rush to the store to buy something but “damn now I have to do this plugin thing!”… Or there might be security restrictions on running something or installing something (like plugins) on their work computer.
Which is fine as long as there’s a fallback plan. What if someone can’t install the plugin? What if they are using a mobile device? What if they are using an abomination of a browser not in the big six?
I guess I was saying all that is really a business decision that comes down to ROI in the end and whether or not it meets legal, moral and commercial obligations of all the stakeholders (including the people who want to use the site). No, not everyone can get an enhanced experience, or wants it, but using graceful degradation and progressive enhancement as guiding measures can help.
I’m getting off track though. In short, silos are alright but one has to accept that silos do require entry (plugin, learning curve, whatever – maybe someone just says “now I have to find out what Silverlight is before I install the plugin!”).
So I’d say there will always be an open web of HTML, with several well trodden silos like your chosen technology.
I’m always concerned though when someone says 99% of our traffic. That being analagous to my broken drawbridge at the castle (not real drawbridge), and 2 of the 3 people who make it over that bridge can get inside my castle. If I’m only counting the three in my statistic then 66% of my visitors can get inside my castle. Great. But what about the 88888 gazillion who can’t get across the drawbridge?
I’ll have to check out your silverlight page though, later in the day. I’m a bit late and installing a plugin isn’t a trivial thing for me, the user. How much easier if I could just pop over and buy something?
Thanks Will, sorry for the long comment reply. The 7.35 AM Sunday morning after effects of Shiraz.
July 17th, 2008 at 6:00 am
“Any time we make the decision to create an ‘enhanced’ experience, whether its with Silverlight or Flash or whatever, then we are making a decision that excludes some part of the audience.”
Good point. Same holds for writing in English, or for creating HTML which assumes a certain viewing screen area, or for assuming an expansive subset of JS/DOM support… true?
If so, then it seems we need to first clearly define the audience, and the message/experience we want to reach them with, and only *then* start evaluating technologies.
Choosing “goodWeb/badWeb” teams ahead of time would make the decision process much simpler (!), but seems to risk a worse match between desired audience and final experience.
(I appreciate your essay, btw… been re-reading it a few times the past few days, along with the other contributors, trying to distill my own thoughts on the subject. Thanks for writing it!)
jd/adobe
July 17th, 2008 at 9:05 am
John, good point. Language, audience and contextual barriers are just as valid and should be taken into account early in the business process of web development. A simple answer for the Hard Rock Cafe would have been to build the Silverlight experience but have a fallback base case of HTML / CSS that achieved functionality, then provided an enhanced experience for those who could (or wanted to) utilise it.
As I still haven’t installed the Silverlight plugin myself I think that might just indicate some of us are reluctant, unlikely or even hostile to having it imposed on us as a barrier. But if that fits their business case, and they are aware of legal / moral / ROI issues then its a valid choice on their part, I guess. But you already know all of that, I’m ranting before breakfast again…
The key is to make web design and development about the business not the technologies. Too many projects get hijacked by forces like ‘its got to be Ajax’ or lose sight of the ultimate mission for the site. Ultimately the site, in this case, would be about sales and experience, but mainly about sales. How do we sell our stuff? How do we ‘make money, and make more money’?
Here’s another question. If someone like me doesn’t have Silverlight then I’ll more likely (unless I’m a fanatic customer already) bail out and buy a competitors product. Now that doesn’t work for the business case at all. There aren’t many times in business that you’ll prefer your competitor make a sale.
I definately think enhanced experiences are great. Just in providing them we need to look at more that the flash and wow and pop solution. Like you pointed out, its a contextual set of decisions – 800 x 600 and 1024 x 768 are no longer reliable measures. Screens are all shapes and sizes now, and its getting more so. I’m not saying every solution needs to be all inclusively perfect, just that in using Silverlight, for example, there must be an appreciation that the technology decision ultimately affects sales.
‘We have a fantastic nightclub, Mr Clark, but unfortunately you can’t come in with your retro footwear’. Damn and I was told they had an awesome bar, lots of hot ladies and a fog machine on the snooker table! = I go to another bar and they don’t make $100.