In amongst the flood of details emerging about Windows 8 is the news that the IE 10 browser in the lightweight Metro front-end won’t support plugins. In the scheme of things this might sound pretty small beer, but it’s hugely significant for the long term future of Rich Internet Application (RIA) development and for the web in general.

Most immediately it’s another kick in the teeth for Flash, still reeling from Apple’s iOS ban. It’s not exactly a death blow, as the Windows 8 desktop version of IE will still support the player, but it’s clearly another major disincentive for developers who believed Flash was as universal as HTML.

Understandably all the focus has been on Flash, but even more telling and extraordinary is the realisation that the new no-plugin policy means that the Metro browser won’t even support Microsoft’s own cross-platform RIA technology, Silverlight!

So just what is going on?

Why has Microsoft changed course so dramatically, betraying its Silverlight vision and shafting its developers in the process?

Details on such a major announcement are disappointingly thin on the ground and largely based on an MSDN blog post (Metro style browsing and plug-in free HTML5). However the few reasons given to justify the decision such as they are – “the experience that plugins provide today is not a good match with Metro style browsing and the modern HTML5 web” – are very familiar. Essentially it’s the same argument Steve Jobs gave – “leaving the past behind” – when he outlawed plugins for iOS some 18 months ago. In short, it’s time for the web to move on from old-fashioned “legacy plugins”.

Regular readers will know that I have never bought this argument. More to the point, I know that Microsoft doesn’t either. After all, the company has spent the past five years arguing the exact opposite: namely that page-based HTML is great but that there are certain things that it just isn’t well suited to deliver: little things like high quality media streaming, digital rights management, interactive vector animations, device-based capabilities such as camera and microphone handling and, more generally, the richest possible, desktop-style web experience.

XAML & Silverlight

It’s precisely because Microsoft recognised the limitations of HTML – which remain true for HTML5/ CSS3/JavaScript/SVG – that the company has spent millions rethinking and entirely reworking its application development tools around XAML (eXtensible Application Markup Language). XAML is an open, XML-based markup language for building the user-facing front-end for both full-blown WPF-based desktop applications and, crucially, Silverlight-based lightweight RIAs ready for delivery via its own Flash-style cross-platform in-browser plugin.

So why has Microsoft changed course so dramatically, betraying its Silverlight vision and shafting its developers in the process?

Well of course Microsoft would say that it hasn’t. After all, the beautiful XAML-based technology lives on and thrives in Windows 8, it’s just that the end product won’t be delivered in the browser via Silverlight, but rather as standalone Metro apps. Moreover, with the promised Metro App Store, Microsoft is offering its developers a simple way to get their work out to users and to make real money from it based on the now well-established Apple model.

There’s a lot of truth to this and Metro is undoubtedly an exciting opportunity for XAML-based developers – but why not support Silverlight browser delivery too? How can Microsoft possibly argue that it can’t support its own existing lightweight Silverlight player within its own lightweight Metro front-end? In fact, if you really wanted to help Silverlight deliver on its potential, gain market share and reward your long-suffering developers, why not build Silverlight support into the Metro version of IE10 while relegating Flash to the desktop version?

It’s business – as usual

I think that the real answer to this question is also the real answer behind Steve Jobs’ decision to ban Flash: follow the money. Cross-platform, in-browser RIAs extending the universal browser to deliver rich and protected apps and content directly between producer and consumer aren’t a legacy problem to be solved; rather, they are a leading-edge, cloud-based threat to the platform-dependent empires that Microsoft and Apple have built up, and to the App Store and in-app content empires that they are currently building.

Keep the lid on the universal, browser-based user experience by killing off the in-browser RIA technologies and restricting the web to HTML5 and you get to deliver the full RIA experience outside the browser via your iOS and Metro apps, and via your platform-specific App Stores and in-app subscriptions. Not only is your all-important operating system and software ecosystem protected from third-party, cloud-based, cross-platform alternatives; you also get to take 30% of all paid-for app content, with no possibility of competition within your platform.

Look at it like this and Microsoft’s decision to effectively sacrifice its in-browser Silverlight vision makes absolute sense. The RIA vision behind Flash and Silverlight in which the web delivers on its full potential as a cross-platform, universal, open and truly rich connection direct between producer and consumer is a wonderful dream, but this is business.

I recently came across a very interesting bit of analysis on the Macworld site. According to a survey by Net Markets based on usage share across 160 million unique visitors spread over 40,000 websites:

“Apple’s iOS mobile operating system is now the third most popular platform on the internet, with a share nearly six times larger than Android’s… more than enough to shove Linux off its perch as the third-place operating system on the web.”

Now that really does sound impressive, especially in the context of some quotes from Vince Vizzaccaro, a Net Applications vice president, regarding overall mobile share and the iOS percentage: “Mobile’s growth curve is strong and mobile is becoming quite a phenomenon on the internet… That’s massive when you think about it… we’re seeing iOS totally dominate the market on the web.”

So just what are these amazing figures?

You may well be as surprised as I was to discover that the “massive” overall mobile share is 2.6%, and the iOS figure is well under half of that, at an extraordinary… 1.1%!

Now I’m not saying these figures are irrelevant (clearly in demographic terms, it’s a very important 1.1%), nor that they aren’t rising. But “massive”? Surely when you call a share “massive” it ought to at least form the majority. For example, looking at the similar web share figures collected by Stat Owl, I was surprised at the percentage of browsers with Silverlight installed on their system. Although the platform is generally seen as never fully taking off, Stat Owl puts Silverlight penetration at 51%.

Of course the technology with the biggest web share of all is Flash. A while back I wrote a blog questioning Adobe’s claims for Flash penetration. Sure enough the Stat Owl figure is lower than Adobe’s – but it still comes in at 97%.

Now let’s not quibble about 2.6% here or 1.1% there. It’s clear that with a greater share than Windows (91% according to Net Markets) and Mac OS (5%) combined, Flash has by far the most legitimate claim on the title “massive”. Or, as Adobe puts it, Flash is “the world’s most pervasive software platform”.

1.1% v 97%: Steve Jobs v the web

This all casts a very different light on the current battle between Apple and Adobe in which Flash is generally seen as yesterday’s technology, desperately clinging on but about to be steamrollered from the web by the oncoming Apple juggernaut.

This has important practical implications. Based on this popular perception, and Steve Jobs’ dismissal of Flash as yesterday’s technology,  designers the world over are busily redesigning their sites and changing their workflows to avoid using a perfectly legitimate web technology that proves itself extremely useful in a whole host of different scenarios (which is how Flash built up its near-universal penetration in the first place) and which is set to prove even more important going forward in the era of cloud-based computing.

Now it turns out that we are being asked to go to all this effort and to deliberately limit ourselves and our output to reach just an additional 1.1% of web browsers. Moreover there’s little doubt that, if they were given any say in the matter, the overwhelming majority of that 1.1% would choose to see Flash and Silverlight content (presumably including  those who choose to block Flash content by default in their desktop browsers but still install the player).

To top it all, Jobs’ apparent underlying justification that Flash inherently cannot be made to work in the mobile environment – even on the iPad – is spurious. There are challenges to overcome, but every other mobile manufacturer apart from Apple – including Google, Microsoft, Nokia and RIM – is currently working with Adobe to bring the mobile-optimised 10.1 Flash player to their devices. With initiatives such as Nvidia’s Tegra, Flash should eventually become the natural cross-OS and cross-browser web platform for devices, just as it is for the desktop.

Rather than the web at large falling into line with Steve Jobs’ vision for the future (built around iTunes and the App Store), it’s time for Steve Jobs to fall into line with the web, including its player add-ons and its core principles of openness, extensibility and universality.

At least when Bill Gates held the web to ransom he had the decency to first establish a dominant position. In Steve Jobs’ case, with only 1.1% market share, the would-be emperor isn’t even wearing any clothes.

According to Steve Jobs and Apple however, the web needs to be washed clean of this proprietary plague and the future belongs to HTML 5.

So which is it to be? Flash or vanilla? Adobe or Apple? Player or browser?

Based on the iPhone and iPad’s phenomenal sales, it’s clear that there are plenty of users happy to go with Jobs’ no-Flash option. After all, apart from video, is Flash really that integral to today’s web experience? It seems a very small price to pay for such undeniably brilliant hardware.

However, as the old adverts didn’t quite put it: “before you buy, have you considered the benefits of new improved Flash?”

Currently the Flash debate tends to be based on history. Essentially we’ve all formed a judgment based on our past experience of Flash. Who hasn’t been irritated by flashy banner ads, unnecessary intro pages, criminal resource hogging and other bad usage, such as Flash rollovers that would be better handled via CSS? The one obvious benefit of Flash that everyone recognises is video handling but, as Ian Devlin showed recently, you can now provide native video playback within the browser with HTML 5.

From web page add-ons to standalone RIAs

It looks like Flash is going to be purged from the web page whether Adobe likes it or not, but in fact that’s not really the issue. The argument for Flash as the next chapter of the web isn’t about its old web page enhancement role. The new-improved Flash has much wider ambitions – it wants to provide an alternative (though not replacement) web platform.

But what does this actually mean? The web platform for what? The key concept here is the Rich Internet Application (RIA). Unfortunately it’s an off-putting and nebulous term and Adobe has done a pathetic job of explaining what it actually means and what web authors and end users have to gain from it. So let me have a go. Essentially the benefits of the new Flash RIAs fall into the two main web production camps: designer and developer.

Designer benefits

For the designer, Flash’s new focus on standalone RIAs means that the latest InDesign CS5 lets them quickly convert any print publication for interactive, media-rich web delivery (and the new Flash Catalyst does the same for Photoshop and Illustrator and Fireworks users). Significantly this power is not limited to Adobe as the latest QuarkXPress 8 has the same capability, meaning that the entire professional publishing industry is already tooled up to output to this near-universal format.

The potential benefits aren’t restricted to high-end publishing either as the flourishing ecosystem of PowerPoint-to-Flash converters demonstrates. Jobs’ attack on Flash as an inherently closed and Adobe-only format is way off the mark; instead the new standalone Flash RIA can act as an interactive, media-rich, online-friendly and truly universal ePaper format – a modern PDF-equivalent fully integrated alongside the HTML-based web.

Developer benefits

The potential for web developers is just as exciting. Here Adobe (and Macromedia before it) has recognized the limitations of Flash Professional as an authoring environment and built up the open source Flex framework and dedicated Flash Builder to create a modern IDE for dedicated cross-platform program development.

Moreover with the latest CS5 Web Premium suite and its deep support for MXML-based presentational markup, Adobe has arguably gone further to bridge the designer-developer divide, letting the designer produce the application interface in Photoshop, Illustrator, Fireworks or Catalyst, then use Catalyst to prepare the RIA for further development in Builder. And with AIR, Adobe provides the framework necessary to go the final step and create universal applications that work outside the browser and offline as well as online.

Crucially Adobe’s online projects such as Photoshop.com and Acrobat.com’s Buzzword word processor, and third-party AIR applications such as the New York Times Reader (complete with adaptive layout) show just what can be achieved.

The Dream Team?

It’s an exciting prospect and the potential combination of Flash-based RIAs with the iPhone and especially iPad is mouth-watering. Surely for a device supposedly intended to revolutionise the online magazine reading experience, the tie-in to the latest InDesign/Quark’s in-built Flash output should be just too good to miss.

For designers, developers and end users alike the combination of iPad and Flash/AIR should be a match made in heaven. And does anyone seriously think the iPad can’t be made to successfully run the new mobile-optimised Flash player 10.1 ?

Unfortunately Steve Jobs has clearly thought through the implications of freely-created and freely-distributed Flash RIAs to Apple’s App Store revenue stream and also of the future threat from me-too Android and other OSP devices to Apple’s device sales, and has acted accordingly.

Let’s be clear about this. There is no inherent conflict or even divide between browser and player and we shouldn’t have to choose between the two. The changing nature of Flash and HTML means that they may well lead increasingly independent lives within the browser, but the two can and should work alongside each other. In fact for the web to reach its full potential, they must.

As such, with his refusal to allow Flash on his hardware (or even Flash-derived native applications!), Steve Jobs is not heroically defending the true nature of the open, browser-based web; rather he is deliberately crippling the web’s full potential and undermining its core principle of universality. And he is doing so at the expense of designer, developer and end user alike.

Even worse, because everyone is still thinking and talking about the old Flash, it looks like he’s getting away with it.

Recently I wrote that what makes RIAs (Rich Internet Applications) different from browser-native web applications isn’t rich functionality or rich content but rich design. Moreover I argued that only a player-based approach (effectively Flash/Flex or Silverlight/WPF) can provide the platform necessary to take web design to the next level.

As I expected, the feedback to the piece centred on the best way of blocking Flash content as it always does whenever I mention the technology. It’s hardly surprising as the first thing that comes to most people’s mind when you mention Flash is irritating banner ads specifically intended to distract you from reading the real content of the page – the absolute definition of bad design.

However in the context of a RIA, Flash/Flex is capable of so much more – producing an end user experience that the browser alone can never hope to match…

To begin with, it’s important to realise that Flash/Flex isn’t inherently “flashy” – it’s up to the designer how they put the platform to use and any decent designer knows that less is usually more.

In fact one of the great strengths of a player-based approach is the smooth and sophisticated streamlining it enables. While technologies like AJAX have allowed browser-based design to paper over the inherently page-based nature of HTML content, this just isn’t an issue for a player. At its best, the browser can offer extraordinary functionality but this will always feel clunky compared to a player-based application – just compare Google Docs to Adobe Buzzword to see the difference in action. Ultimately only the player-based RIA can deliver a truly desktop-style experience.

The design strengths of the player don’t just add to functionality; they are crucial when it comes to dealing with content. Here CSS (Cascading Style Sheets) has worked wonders for the browser enabling it to graft on layout and typographic control to the presentation-neutral HTML, but this is still rudimentary. The difference between the browser and player is made clear when you access the New York Times website and then read the same content via the AIR-based Times Reader application (there’s a free taster version).

At first they might look pretty similar – apart from little things such as the live crossword and the fact that the whole reading experience is tighter and more enjoyable and works offline. The most fundamental difference is hidden unless you happen to resize the application. If you do, you’ll see that the layout automatically updates in real time – increase the application width and the columns will widen and then at the tipping point of comfortable reading another column will appear and any images will intelligently resize themselves accordingly.

Of course this adaptive layout isn’t really intended for such resizing on-the-fly (though it is quite addictive). The real purpose of adaptive layout is to ensure that whatever device you are accessing content by – mobile, notebook, desktop, TV or projector – the layout and design and so end experience are optimised. This isn’t a question of flashy bells-and-whistles it’s absolutely fundamental: it’s automatic and universal good design for any device that supports the Flash player.

As we move on from the early days of the Web, from simple desktop-based browsing to advanced desktop-style reading and doing, the time has come to stop dismissing Flash as just an irritating distraction. Thanks to its adaptive layout capabilities, Flash/Flex sets a new benchmark for the best possible cross-platform cross-device design.

Surely you wouldn’t want to block that?

For example in a comparison between the barebones HTML-based Google Docs and the slick Flash-based Acrobat.com, I’d reserve the term RIA (rich internet application) for the latter and dismiss the former as a mere “web application” (more importantly I know which one I’d prefer to use).

Recently though I have to admit that Google caused my jaw to drop… and made me question the distinction.

Like most people I’ve become so used to Google Maps that I now take its extraordinary power for granted. I was forcefully reminded of just how amazing it is however, when I recently went to print out some directions. For the first time I noticed and clicked on the options in the print preview header to show maps and street views for each individual step (see example).

The latter capability in particular is extraordinary (assuming the area that you are interested in is covered by Street View). Being able to almost-instantaneously load in views of each junction that you are going to come across on your journey is breathtaking – especially as each street view is live and explorable (though thankfully that’s not generally necessary as Google automatically orientates the view based on the direction you’re travelling)

It certainly gave me pause for thought. If an application can cause you to say “oh that’s good, that’s really good” out loud, then surely it deserves to be called “rich”? In fact, if Google Maps isn’t a “rich internet application” then what on earth is?

It’s certainly a useful reminder of just how powerful the browser can be as a platform, but ultimately I think it’s worth preserving the distinction and reserving “RIA” for player-based applications (ie Flash / Silverlight). This isn’t just on practical grounds so that we know what we’re talking about. Google is clearly committed to making the browser as rich an environment as it can, but that’s rich in terms of content and functionality not in terms of design.

Indeed Google clearly prides itself on its cut-down, almost anti-design approach. This minimalist “anti-Flash” design works well for Google’s core applications such as Search and Maps where Google’s job is to help you get where you want to go as quickly as possible (indeed you could make the case that the street map mini-views are a flashy falling away from this principle).  However for those applications – the majority – where you have to spend time consuming or producing content, this barebones approach backfires as it does with Google Docs.

Clearly content and functionality are crucial to the success of any project but ultimately I’d argue that the “rich” in RIA refers to design and that the distinguishing strength of a player-based approach is that it can offer a richer, tighter, more desktop-like user experience than the browser alone. The browser-based Google Maps is brilliant, but a player-based version could be better still.

———————–

PS. A quick practical postscript for web designers: Google recently introduced the ability to add Google Map directions to your own sites . The Directions gadget doesn’t offer street map views (currently), but it’s a seriously useful option to add to your How To Find Us pages. And it’s a great way to get your clients’ jaws dropping with a single line of code.

PPS. A quick impractical postscript: In spite of the directions and street views, I still managed to get hopelessly lost. Google Maps might be brilliant but it’s no replacement for Sat Nav.

Essentially Silverlight is designed to port Windows’ core WPF technology into a cross-platform  browser-based player like Flash. So what will the new version offer?

The highlights include:

• H264 / MP4 video support 
• Pixel Shaders for effects such as blurring
• Perspective Transforms to simulate 3D
• Animation Easing to produce far more naturalistic movements

“Highlights” which will probably sound less-than-exciting as they are all on the Flash 10 checklist. Far more eye-catching and the stand-out feature of the latest release is: 

• Silverlight 3’s ability to break out of the browser and install XAPs directly to the desktop 

It’s a fundamental step forward but again it sounds very familiar. After all, isn’t this exactly what AIR (Adobe Integrated Runtime) already provides?

Moreover the Silverlight 3 beta fails to deliver any of the “Flash-killer” WPF-based features that users might have hoped to see ported such as true 3D handling, print control and, my personal favourite, intelligently adaptive FlowDocuments. And there’s no support for video alpha either.

Overall it’s clear that Microsoft is still playing catch-up with Adobe on the Rich Internet Application (RIA) front and no doubt critics will say that, as there’s already an established standard here with all-important and unbeatable market penetration (Flash has around 97% penetration compared to around 25% for Silverlight), what is the point of Silverlight? 

It’s a subject I plan to return to. In the meantime, there was one feature which might give the Microsoft-bashers some pause for thought. Significantly, the new Silverlight out-of-browser capabilities are built directly into the version 3 runtime (under 5MB). This means that to save an online Silverlight XAP for offline use on either PC or Mac, all the end user needs to do is right-click and hit Install. And it has built-in auto update too (excellent). And you can remove the app just as easily.  

The whole RIA concept is based on the blurring of online and offline, local and remote and Silverlight’s streamlined moving between the two worlds looks like it will set the new standard.

And so Sun, the company that invented Java around ten years ago, has just released JavaFX. I’m not sure I’ve ever felt quite so cynically dismissive of a new software platform. It seems a desperately cack-handed move to get into the rich internet application market, and it comes at least three years too late.

According to Sun, JavaFX is, “an expressive rich client platform for creating and delivering rich Internet experiences across all screens of your life.”

I’m irritated already – whoever came up with the marketing phrase “all screens of your life” (which crops up all over the JavaFX site) needs a good talking to.

Putting it another way, JavaFX is a competitor to Adobe Flash and its Flex RIA platform. Which is much like Silverlight, the Microsoft competitor that currently isn’t really getting anywhere.

To succeed, JavaFX is going to have to displace its massively well-entrenched competitor with some seriously compelling advantages.

How about a compact footprint? Erm, no. JavaFX runs on the standard Java runtime environment, which is still a lumbering beast of a thing; it’s a 15MB download. Starting up a Java applet on a web page is always heralded by a fanfare of hard-disk grinding, and JavaFX is no different. The JavaFX examples still take far longer to come to life than a Flash application, while managing to look less polished and demonstrating none of the silky-smooth fluidity of Flash.

All that aside, the thing that annoys the hell out of me is the way in which JavaFX is being marketed. It treats people, especially developers, like total idiots.

Says Schwartz:

“We certainly hadn’t up until recently looked at the impact of time-based media, of video playback, of high-quality audio”

It would be more accurate to say that Sun has paid just enough lip service to video and audio to get developers interested in it, and then frustrate the bejeezus out of them by not supporting it properly. The Java Media Framework has been skulking about like an illegitimate child in the backwaters of Sun’s website since the late 1990s.

Now we come to the real doozy. According to Schwartz, JavaFX “enables you to bypass hostile browsers”. This is accompanied by a slide showing us the hostile browsers in question:

I’ve never read such a ridiculous statement.

The argument seems to be that Sun is graciously giving us a way to avoid those terrible money-making browser companies (including, um, the not-for-profit Mozilla Foundation) and paving the way to simplicity and clarity by introducing a competitor to Flash. Words fail me.

Schwartz goes on to say that JavaFX is for “content owners that want to deliver their content without obstruction.”

What? What obstruction? Anyone who wants to deliver an internet application can construct it in Flash and it will run in any one of the ‘hostile’ browsers. Where’s the obstruction in that?

Turning now to Sun’s Eric Klein, giving some insight into the groups of people that JavaFX was apparently designed for. Klein opines:

The third group we thought a lot about was the visual designer, folks who use things like Adobe Illustrator, Adobe Photoshop and other graphics tools. And what they said to us is: ‘let us join the party. Let’s create a designer-developer workflow, so that our content… can seamlessly get into the development process

The absurdity of that statement genuinely does my head in. As if anybody on the planet is going to think for one second that Adobe designers are going to turn to Sun for their tools by preference, rather than Adobe’s own RIA platform, which is aimed at precisely the same usage models.

It all feels horribly like a cynical catch-up ploy from a company whose share price has dropped 84% in the past year, compared to drops of 45% and 54% for Microsoft and Adobe respectively. Which would be fine were it not for the tasteless way in which it’s being marketed.

The nicely ironic twist, of course, is that all those JavaFX videos run in a Flash video player.

8. Use of Your Content. Adobe does not claim ownership of Your Content. However, with respect to Your Content that you submit or make available for inclusion on publicly accessible areas of the Services, you grant Adobe a worldwide, royalty-free, nonexclusive, perpetual, irrevocable, and fully sublicensable license to use, distribute, derive revenue or other remuneration from, reproduce, modify, adapt, publish, translate, publicly perform and publicly display such Content (in whole or in part) and to incorporate such Content into other Materials or works in any format or medium now known or later developed.

You don’t have to be a lawyer to see that basically you were handing over your all rights as originator and giving Adobe free rein to make money from your photos however it saw fit!

Adobe responded quickly to dowse the inevitable flames with John Nack quoting the PSX team as saying “The original terms of service implied things we would never do with the content within Photoshop Express.” More importantly the relevant terms and conditions were quickly updated and are entirely different and much more end-user friendly in all ways.

So is that the end of the matter? Well it certainly means that the conditions are no worse than those commonly found on photo sharing sites such as Flickr and so should be no bar to joining up to Photoshop Express and taking it for a spin (it’s well worth a look if only to see what modern Flex-based RIAs are capable of).

However is this a case of “there’s no smoke without fire”? After all you don’t normally leave lawyers to just come up with something off the top of their heads (expensive) and then post it up to see if anyone complains (very expensive). And did no-one at Adobe think of looking at how other photo sharing sites managed copyright before launching such a major new service?

Moreover, while the Terms and Conditions have been updated, at the time of writing, the Photo Express FAQ still reads:

“What rights do I have to my photos?”#! Excuse me – splutter – but that phrasing is just bizarre. To my mind the implication is that having transferred all rights going to Adobe I now need to be told which rights Adobe is kind enough to let me retain.

Am I the only one that’s not entirely convinced that this all an innocent mistake but smacks of something a little bit deeper like a planned policy?

But before I get accused of an anti-Adobe conspiracy theory, I’ll float another question that I intend to return to shortly in a follow-up post: Are there really no circumstances in which you’d be happy to sign up to those original terms?