Intelligent Enterprise

I remember (lo these many years ago…) when IBM proved that the elephant could dance, and the name of the dance was PC. An obscure IBM skunk works in Boca Raton, Florida hatched this insanely great idea – a personal computer made out of cheap parts from all over the place – and two years later it became the world standard. Now I'm not saying that IBM's embrace of enterprise mashups is on the same order as the PC; but of all the really big IT companies, you'd think IBM would be among the last to adopt a technology called 'mashup.'

As I explored in this case study about a Defence Intelligence Agency mashup, the technology at its most fundamental addresses one of the oldest IT problems around: Delivering appropriate data in a usable fashion to those who make decisions. What's different now, obviously, is the Internet and the technologies behind Web applications and Web 2.0.

Does program code entirely between curly braces {} ring any bells for you? Probably not. HTML lives between <> pairs, and LISP code (if you're old enough to remember it) was fully parenthesized (), but curly braces? This is the signature of a Web oriented language called Curl. That name you may have seen kicking around online literature. You'll see a lot more of it. As of this week, Curl is taking on Adobe AIR for the rich Internet application (RIA) Web/desktop application development market with a project named Nitro.

Some day we will stop talking about RIAs (Rich Internet Applications) and Web 2.0 apps and Web apps; they will all just be apps. The same goes for "desktop applications." They will just be apps too. In fact, I wish we had a simple name for all these flavors of apps right now. Thankfully, I can see the horizon of that "some day."

Adobe is not the only company working toward that time when we can build rich, interactive, and data-filled applications for the Web or the desktop with equal aplomb. However, making a big step of its own yesterday (January 25, 2008), Adobe released three products: Adobe Flex 3 Framework, Adobe Flex Builder 3 and Adobe AIR 1.0. A fourth related product, Adobe BlazeDS, was also recently announced. Collectively these products move Adobe and its band of developers further in the direction of the seamless Web/desktop application development system.

Companies that have been around a while have legacy software. Every new piece of software has to be considered in light of existing software, even if the latter is ultimately replaced or ignored. This is also true of companies that make software development products; all of the big ones have been around a while and have their own legacy software.

But what about new companies; can they really start from scratch? Specifically I'm thinking about a company like Nexaweb Technologies, whose flagship product, Nexaweb Enterprise Web 2.0 Suite, I recently reviewed for Intelligent Enterprise. Here's a company formed in 2000 that has bitten off a very large chunk of enterprise software development turf. They aim to make Web 2.0 and Rich Internet Applications (RIA) fit for the enterprise.

"What about NetBeans?" A reader gently asked in response to Monday's blog filled with encomiums for Eclipse, but no mention of Sun's premiere IDE, NetBeans. I felt like a Pandora who not only did not wish to open the box, but ignored the existence of the box. That blog was meant to wave a small flag for IT folks to take a look around the shop and see how many projects they have that use the Eclipse Platform in one fashion or another. Then consider how they might better support them. There wasn't a point, or space, for getting into the festering history of Sun (and the Java Community Process) and IBM (and the Eclipse Consortium). Still, it could have been mentioned.

After starting a third review in a row of a software development environment based on the Eclipse platform (PyDev, Adobe Flex 3, and Nexaweb Enterprise Suite), something a friend said to me a year or two ago rang true: "There will be two major players in development software: Microsoft and its tools and Eclipse and its tools." A cynic might say, "Good for Eclipse, not bad at all for a camel built by a committee," but to make a point, cynics are often unfair and inaccurate.

"Commercial open-source software" certainly sounds like a contradiction in terms. The phrase "free and open" is part of the definition of open source software, which translated into real terms means that people can download the software and source code at will and for no charge. In most instances, this is how open-source works. Where it may work less well is for the enterprise.

Here's what Bill Hilf, Microsoft's General Manager of Windows Server Marketing and Platform Strategy has to say in a recent Information Week interview: "When people buy commercial software, really what they are buying is a guarantee. You're buying a guarantee that what you have will perform, and has been tested and there's someone you can call up, and if things go really bad someone's liable if something doesn't work. You're buying this ecosystem of accountability."

Recently, while working on my review of Adobe's development products aimed at Rich Internet Applications (RIA), it struck me that here was a vendor with a brilliant array of creative tools that appeared to be challenged to assemble appropriate backend support (especially for the enterprise) in areas such as data access, security, team development and application management. Not unlike the current situation for the enterprise in general — a Cart and Horse problem.

The horse, of course, is the pull of rich presentation media and highly interactive user interfaces. The horse is lively, interesting to watch and, in practice, leads the way. The cart is the conveyance vehicle, all the backend machinery that enables a Rich Internet Application. The cart is homely and utilitarian, but made of necessary things such as data management, security systems, and communications. The current problem is that enterprise IT, quite naturally, prefers to put the cart before the horse.

I take the topic of this blog literally: Will the enterprise enter the "mashup" maze? IBM thinks so. IBM would not call it a maze. They call it the IBM Mashup Starter Kit and recently put it up for (free) download. It's part of their angle on Web 2.0 applications, in fact one of the leading parts (along with webtwoifying Lotus Notes). What IBM is doing with mashups is interesting for a number of reasons, but perhaps the most significant is that IBM is pushing mashups at all.

It's not that mashups (however defined) aren't being taken seriously in the enterprise – for the most part they're not. It's that IBM, the primus inter pares of enterprise vendors, has chosen mashups as a key product for development.

Every time there’s a new technology, trend or technique in the computer industry, thousands of minds engage with the important questions of IF or WHEN to get into it and the well-known "technology cycle of adoption." I usually characterize the cycle like this: First adopters, early adopters, early majority, late majority, laggards, Luddites. The dynamics are fairly well understood; those who start early may have a competitive advantage but also run a higher risk of failure. Later adopters get more stable technology and better perspective, but may have little or no competitive advantage. Each organization gets to pick its strategy on a continuum from gung-ho to do-nothing.

This is the situation with Rich Internet Applications and the new software development ideas that go with it; and beyond that are the many facets of the so-called Web 2.0. I hear theme and variation on the IF and WHEN questions from IT people all the time. Even those shops that are already doing RIA development continue to re-visit the questions.

Stuff crosses your desk, so to speak. I'm not sure what this means in the Internet age, but simply put, pieces of information come to your attention and usually leave attention quickly. For example, on Wednesday, September 5, there was an e-mail blurb about Microsoft releasing version 1.0 of Silverlight. I don't know how many IT managers and developers are following Silverlight. Perhaps not many. Certainly a 1.0 release of anything, even (or especially) from Microsoft, rarely warrants much attention. I'm not reaching for a megaphone to amp this event, but a little soft-spoken commentary might be useful.

In traditional IT software development circles, at a time say about 10 BW (Before Web), seeing the work of an artist (that is, someone who produces something in a recognized art form) was a matter of a Friday or Saturday night out on the town. Then along came the Web with all its bells and whistles… and pictures, and graphics, and animations, and video… and the Web page started to fill with the work of artists, if not works of art. Eventually applications destined for the Web were expected to have artwork, especially in the form of multimedia, and a generation of IT developers began to work with artists of many stripes.

Now we get to Rich Internet Applications, and the connection to artwork goes up a notch. Almost by definition, RIAs are supposed to be enhanced by graphics design and multimedia. In some cases the multimedia, for example a demonstration video, may be the purpose and majority content of the application. In any case, the roles of design and art work (in the broadest sense) are integral to the production of Rich Internet Applications.

For the two past weeks I've been on a working vacation in Montana, where dial-up Internet access is all there is. No DSL and mountains block satellite connection. A good place to experiment with Rich Internet Applications (RIA) that need to be functional when disconnected from the Internet. Of course, an RIA without the Internet violates the definition of RIA, but reality and Murphy's Law tell us that for certain Web applications, an Internet connection may not always be available when it is necessary to run the application.

What is a Rich Internet Application? The Wikipedia entry says: "Rich Internet Applications (RIA) are Web applications that have the features and functionality of traditional desktop applications. RIAs typically transfer the processing necessary for the user interface to the Web client but keep the bulk of the data (i.e. maintaining the state of the program, the data etc.) back on the application server. RIAs typically: run in a Web browser, or do not require software installation; run locally in a secure environment called a sandbox; and can be 'occasionally connected'."

My right knee jerks ever so slightly in reaction to Adobe's announcement, on Monday, June 11, of a beta release and a name change for its (once known as) Apollo project, now AIR (Adobe Integrated Runtime). Some coverage is due, even if it is merely to note that as an important player in RIA (Rich Internet Applications), Adobe has thrown some more tidbits to developers – including a beta release of the new open source Adobe Flex 3. However, this is not news that will reverberate to the far corners of the IT shop.

That's if the IT shop knows something about Adobe software development products. In the IT industry media, Adobe Apollo achieved the appellation "much-anticipated" without ringing many doorbells at the enterprise. That needs to change, if only because shops moving into the Brave New World of RIA and Web 2.0 need to know who's who, and Adobe is most definitely one who is.

I've noticed, and perhaps you have too, that spokespeople for Microsoft, Sun and Adobe have a tendency to gingerly disparage Ajax. Conversations with them have a lot of "We like this/that about Ajax, but…" For their respective companies, Ajax is the classic "bag on the side." It doesn't fit their model: proprietary delivery methods (runtime clients, graphics engines), proprietary or semi-proprietary development tools—their own solutions for overcoming the deficiencies of HTTP Web applications. Although they rarely say so, it pains them that Ajax is so popular. They have to deal with it, treat it with kid gloves, even support it; but they don't like it.

Something about Sun JavaFX in a moment, but first a short preliminary expectoration: Rich Internet Applications (RIA) matter. They are not the be-all-end-all of applications, on the desktop or the Web, but they will be the élan vital (the vital force) that drives application development and the growth of the software industry for the next "generation" (figure three to eight years).

Because rich applications feature user interface and media prowess, they (apparently) must have a "graphics engine" of some sort running on each user's computer. Intuitively I think this engine is what counts most in the competition to develop broadly useful RIA. Yes, the underlying language(s) are important. So are development environments and designers. So are standards and security. But the performance, efficiency, and wide distribution across platforms of the graphics engine — because it directly affects both users and developers — will be the edge which is likely to determine winners from also-rans in the RIA marathon.

Not very long ago there were enterprise developers who did Web applications and other enterprise developers who did streaming and graphical Web programming (illustration, animation, videos, etc.). Essentially these were two different groups who only occasionally would meet on specific projects. That's changing. Whether you pack it into the Web 2.0 rubric or consider new offerings by Microsoft, Adobe, IBM and many others, it's apparent that the worlds of Web applications and rich media are melding into Rich Internet Applications (RIA). I know of a lot of enterprise shops that are lining up on this, because at the gut level (at least) we know it's going to be important – maybe even a dominant part of enterprise software development. Right now, however, this business of melding into RIA is messy.

Microsoft let one Cinderella shoe drop (calling it ever so evocatively Silverlight), what's the next shoe going to be? A marriage of Silverlight and Atlas (aka ASP.NET 2.0 AJAX Extensions)? Perhaps not a marriage just yet, but a kind of pre-nup involving the .NET library capability…

Let me translate the blog-speak: On April 15, Microsoft announced the release of Silverlight, which formerly was called Windows Presentation Foundation/E (WPF/E). Based on technology developed for Windows Vista (the graphics portion formerly called Avalon), Silverlight is a streaming media delivery module, a piece of code that plugs into a browser, that Microsoft hopes will open the floodgates for media rich applications. Also, it's an Adobe (Macromedia) Flash killer.

Last May, Rod Smith, VP of Internet Technologies at IBM invited Microsoft to join OpenAjax, the Ajax standardization consortium, saying, "I hope what comes out is the same as what came out of the early Web services meetings" -- meetings in which IBM, Microsoft and others drew up an early blueprint for Web services standards. In this case, about nine months of internal debate later, Microsoft opted to join OpenAjax. Here's last month's announcement by Microsoft's Scott Guthrie.