And Another Thing

Jon Honeyball argues that Macromedia offers far more than Microsoft when it comes to building front-end applications.

Maybe the time has come to bite the bullet. I have taken the decision not to build any front-end applications using the Microsoft toolset. By that, I mean that I am going to limit my use of Visual Studio .NET to building middle tier and back-end components only. As far as I am concerned it has no place in the building on front-end applications because it is singularly appalling at performing that task.
      Worse still, given my platform of choice for front-end development from now on, I wonder just how far my use of VS.NET will be relegated to tidying up old code. This isn't a 'flash of light' change of mood, but something that has been nagging me for a long time.
      You see, I think Microsoft has dropped the ball badly. Don't get me wrong - the work in VS.NET is a stunning achievement and it is a great platform for moving forward over the next ten years. The problem is that Microsoft have completely failed to address the needs of those users who want to build websites, or build interactive systems talking to back-end data sources, whether that be SQL Server or something else entirely.
      My first inkling that something was not right in the Microsoft world came last year, when I got hold of Macromedia Ultradev. At the time, it clean blew me sideways with its capabilities for building ASP delivered web applications. The design tools were first-rate, especially when you included Fireworks into the equation. The ASP code it generated was not of the highest quality, but it did work. The writing was on the wall - all we needed was an Ultradev update, with a stronger set of middle tier application development tools, and Microsoft would find itself sidelined.

Macromedia MX steps in
This has now happened. With the release of Macromedia's new MX suite, the Redmond boys have received a massive poke in the eye. The radical re-architecting that Macromedia have done with MX is nothing short of stunning. If you want to deliver an HTML based interface, then Dreamweaver MX is the product for you. If you want to go platform independent and be dramatically graphical in nature, then Flash MX is the tool of choice.
      In the past I have grown to hate Flash with a vehemence that startled even myself. All those splashy opening screens with no 'go away, I don't want this crap' button. They were - and still are - the very definition of 'design over content'. But the writing was on the wall. Macromedia's Generator engine allowed you to build Flash objects on the fly, driving the objects from data sources like database engines. And today, with Flash MX, you can tie Flash objects through the middle tier object model and on to anything you like.
      Yes, it is now perfectly possible to build an order reservation system with the front end entirely built in Flash MX. Best of all, it will work the same on a Macintosh, on a PDA or even a Smartphone if you so wished.
      By ignoring the needs of the front-end developers, Microsoft has backed itself into a corner. The non-appearance of its long-rumoured 'Universal Canvas' technology is but one sign of this. The almost complete silence about Internet Explorer 7 is another - after years of breakneck speed on IE development, we have come to a halt with IE6. Is this the end of the road? If so, what is Microsoft expecting us to develop front ends with?
      DHTML is dead in the water - it is too browser specific even on the same operating system, and it is tied to a scripting language that is archaic at best. XML and XSLT seems like a fine idea - generate the data on the server, and let the client decide how to render it using XSLT stylesheets. But there is not a single decent XSLT development tool out there, and certainly nothing from Microsoft, so this is a dead end. We can drop back to plain old HTML, if we are prepared to put up with version compatibility problems, yawning screen refreshes that blank the whole window, and a level of design tool and capability that makes VB1 look like a godsend.
      Maybe Microsoft has a raft of technologies waiting in the wings to wow us once the DOJ trial comes to an end, but I am not sure of that. What is certainly clear is that it does not have the necessary toolset to deliver an end to end development platform. And given the open nature of XML-interfacing SOAP objects, it has no real grip over us, the development community, when it comes to back end objects either. Yes, VS.NET is a great development platform, but I am just as happy talking to a J2EE environment.
      Microsoft needs to do something dramatic, and to do it now. The only solution that I see making sense is for the company to take an in-perpetuity license to the Flash runtime. Better still, buy a 49 percent holding in Macromedia and bundle the MX toolset in with Visual Studio .NET. Then we might see a future Windows desktop built out of pure Flash components, Internet Explorer subsumed into an on-the-fly HTML-to-Flash rendering library owned by Macromedia, and finally some radical forward thinking movement on desktop and user interface design.
      Maybe that grand Microsoft toolset is just around the corner. In the meantime, I'm going with the Macromedia MX platform for front-end and middle tier development. Sorry Visual Basic, it was nice knowing you...

Nov 2002
Sept 2002
July 2002
May 2002
March 2002