The Keene View on Web 2.0, Ajax and SaaS

Ben Worthen of the WSJ recently posted an entry about Web 2.0 adoption. He sited a Forrester survey that concluded Enterprise Web 2.0 solutions would gain broad adoption in 2008 despite clear CIO resistance to the siren call of blogs and wikis.

As a strong proponent of Web 2.0 in the enterprise, we at WaveMaker want very much to see a rapid adoption of these technologies at the corporate level. On the other hand, wishing won't make it so - the grab-bag of technologies and ideas that constitute Web 2.0 are bound to confuse the IT community.

There are two possible observations here:

1. Damn the data, full speed ahead. Just like drug companies that sink a great deal of money into disappointing clinical trials that they then attempt to spin into medical break-throughs, analyst firms like Forrester won’t sell many copies of a Web 2.0 report that concludes Web 2.0 ain’t happening in the enterprise anytime soon. Thus there is a strong incentive for the report authors to refute their own data in the survey summary.
2. Web 2.0 is being pulled into the enterprise, not pushed. Since the PC, all client-side technologies have been pulled into the enterprise by business users, not pushed into the enterprise by central IT. The main themes of Web 2.0 – rich interface, collaboration and user-driven content – have more to do with how users interact with their computers than with computer infrastructure.

At my WebGuild panel this week, Rod Johnson of SpringSource observed that open source technologies like Spring and WaveMaker build unstoppable momentum within the IT organization by solving fundamental problems that much bigger players either cannot or will not solve.

For Spring, the tailwind came the intractable complexity of the EJB standard that left developers desperate for a simpler, more lightweight java application server. For WaveMaker, the tailwind comes from the complete dearth of visual tools for web development that would enable MS Access, Lotus Notes and PowerBuilder developers to come out of the client/server dark ages and build the kinds of web-based applications that their end users want.

Labels: access, powerbuilder, spring, WaveMaker

We launched our new WaveMaker product last week in Boston. At its worst, a product launch tour can be a mind-numbing exercise in navigating the Mass turnpike while regurgitating at regular intervals an adjective-laden product pitch to barely conscious industry pooh-bahs. These soporific meetings are puncuated with frequent stops at Dunkin' Donuts (apparently the only chain allowed in suburban Massachusetts) for undercooked donuts and overcooked coffee.

Our tour last week wasn’t like that. Mostly because our story is simple – building web apps is too hard – and our solution is equally simple – a visual development tool that generates pure Java apps that run in any Java platform.

The best meeting of the week was with Judith Hurwitz, who wrote a great blog posting on our product launch, asking Is WaveMaker the Web 2.0 version of Powerbuilder? How often does a seasoned analyst describe a company meeting as fun!? Judith said:

Some meetings are just fun (I can’t always say that..sometimes I just want to run away and hide under my desk). But my meeting today with WaveMaker reminded me of the type of meetings I had in the .com days. I admit I was excited about what I heard.

I also had an interview with Jason Meserve of Network World on the launch of the WaveMaker product. This 15 minute podcast gives a good overview of our strategy and you can listen to it here. Finally, Paul Krill of InfoWorld had a good product write-up here.

Labels: Judith Hurwitz, powerbuilder, WaveMaker

Although the computer industry is still young, we have already seen several waves of development tools. These waves follow technology and architecture trends: new technologies like the PC make new architectures possible, like client/server, which in turn require new tools, such as PowerBuilder.

Each wave of development has had a distinctly character - either logic-based or visually based. The tools to support each of these waves have been tailored correspondingly.

While the last 8 years have been dominated by thin-client architectures and code-based tools for expert users, the rich client technologies and collaborative user expectations behind Web 2.0 argue strongly that a new wave of development tools is beginning.

Here is my take on the four big development waves:

1. Green screen (1960-1990): centralized development with code-based Cobol tools


2. Client/server wave (1991-1997): departmental development with visual-based tools like PowerBuilder, Notes, Access, Oracle Forms.


3. Thin client wave (1998-2007): centralized development with code-based Java tools


4. Rich client wave (2008+): Web 2.0 democratizes development and dramatically opens up both the responsiveness and functionality of the client with an avalanche of Internet-based widgets and services. Can today's code-based tools targeting expert Java and C# developers ride this wave? History would say no.

It is hard to remember today - when all the development tools are code based - that all the client/server development tools were visually based. Ten years ago, visual tools like PowerBuilder and Notes dominated the development world. Now, the programming world has shrunk to just two code-based tools - Eclipse and Microsoft Studio.

We believe that the industry stands on the edge of a new wave of computing, driven by the Rich Internet Application architecture. Web applications are moving from static, limited html pages to much more responsive and powerful widgets and services.

The explosion of widgets and services available on the Internet is driving a demand among business users for more useful business applications. As thing client applications get more visual, coding tools like Eclipse and Microsoft's mis-named Visual Studio are increasingly awkward.

To underline this point, a recent customer was able to re-build an application in our visual tools using 98% less code than in .NET/Visual Studio. Note that it is important for visual tools to support coding for advanced services - we are not advocating code-free development, just code-lite development, particularly for visual web applications.

After nine months of hard work, we are ready to start making waves. Over the next month, we will be introducing a product that will launch the next wave of enterprise computing. Our vision is to be the Powerbuilder of Web 2.0.

To underline our commitment to this vision, we have changed the company name to WaveMaker. And that is exactly what we intend to do!

Labels: Eclipse, powerbuilder, RIA, WaveMaker

Mike Gunderloy of Larkware blogged about our MS Access to MySQL migration here. Mike is well known in the Microsoft community, although he has another blog, A Fresh Cup, that he bills as "notes from a recovering Microsoft addict," where he waxes eloquent on Ajax applications, Ruby on Rails and all things Enterprise 2.0.

The interesting point is that although there is a great deal of enthusiasm around Web 2.0 tools like Ruby on Rails, there are no Web 2.0 equivalents of the trusty drag-n-drop client/server tools like MS Access and PowerBuilder.

Face it, if you have to learn about Rails concepts like meta-programming and scafollding just to build a Web 2.0 app, your average VB developer is just going to continue taking a pass on this whole web thing and hope that it blows over.

Until there are Web 2.0 tools that provide a natural evolution from client/server development to web development, an entire generation of programmers will continue to hang fire. At a minimum, this will require drag and drop web development tools that don't require an entire bookshelf of manuals to get started building useful apps.

Labels: access, powerbuilder, ruby on rails