LOGIN | REGISTER NOW | SUBSCRIBE
 
 
News on Monday
more>>
SharePoint Tech Report
more>>


   

 
 
Download Current Issue
ISSUE 2/1/2010 PDF

Need Back Issues?
DOWNLOAD HERE

Receive the print Edition?


 
blogs tab
Visual Studio 2010 Release Candidate Available Today
A Visual Studio 2010 release candidate is available on MSDN.
02/09/2010 09:45 AM EST

Is Microsoft eyeing Office subscription pricing?
Microsoft may be preparing to offer a new Office pricing option called "union," which charges the same for cloud as on-premises.
02/01/2010 09:38 AM EST

Facebook rewrites PHP runtime
Facebook is about to open source its own PHP runtime, written from scratch for speed.
01/30/2010 08:53 PM EST

 

Events calendar tab
Macworld 2010
2/9/2010 to 2/13/2010
San Francisco
IDG World Expo

SPTechCon 2010
2/10/2010 to 2/12/2010
San Francisco
BZ Media

PyCon 2010 (Python Conf.)
2/17/2010 to 2/25/2010
Atlanta
Python Software Foundation

Southern Calif. Linux Expo
2/19/2010 to 2/20/2010
Los Angeles
SCALE

IBM Pulse
2/21/2010 to 2/24/2010
Las Vegas
IBM


 
print Printable version 
Most Read Latest News Blog Resources
Facebook brings HipHop to PHP
HipHop compiles PHP to speed up the Web layer at world's most popular social networking si...

Adobe moves LiveCycle application platform to the cloud
Release gives developers just what they need to create targeted business process-oriented ...

Oracle acquires AmberPoint
Today's merger will probably affect most of AmberPoint's existing partnerships with other ...

Digg!  Digg
Reddit  Reddit
Slashdot  Slashdot
Share on facebook  Facebook
Share on Twitter  Twitter

Windows & .NET Watch: Intentional’s chrome (if not silver) bullet


By Larry O'Brien
June 9, 2009 — 
The public unveiling of the Intentional Workbench at Microsoft’s recent DSL DevCon was the most impressive demo I’ve seen in 20 years of reviewing software development tools. There is no doubt that “domain workbenches” will change the way that business applications are developed. This type of tool will increase by several factors the speed by which large companies can evolve their applications.

The value proposition of a “domain workbench” is “projectional editing.” This means programming a GUI with a screen builder, a database query with (say) a query-by-example editor, a state machine with a diagram or a transition table. Or programming with code—text filled with curly brackets and semicolons being a favored format for experts in the programming domain.

But rather than one or two alternative editors developed by the IDE vendor, a system might have a dozen editors in use at different times, some provided by the IDE, some from third parties, some reused from other projects, and perhaps some custom-written for the project at hand. It is the proliferation and evolution of editors from which the productivity speedup derives; this is why the demo, even though jaw-dropping as it stands (view it here), is really just saying “Hello, world.”

As discussed in my last column, Fred Brooks only credited high-level languages with a quintuple increase over assembly language. But you cannot do <asp:label id=”HelloWorld” runat=”server”/> in five lines of ASM. It may have taken decades, but high-level languages can now easily encapsulate hundreds or even thousands of lines of assembly language. Heck, a single line of CIL, the assembly language of the Common Language Runtime, can hide pages (I’m looking at you, newobj). In the same way, it’s going to take years for truly great projectional editors to evolve, especially editors that are broadly appealing.

What will take considerably less time to emerge are valuable editors for complex business domains. In my last column, I spoke of the Domain Specific Language value proposition and how it is most relevant to companies who have a “software factory”-like environment, where they are developing many applications that reuse domain concepts. Sun Microsystems’ Ted Leung accurately described the Intentional Workbench as not being for the development of DSLs, but rather fitting “in the DSL footprint.”

If the expression of business rules is a bottleneck in your development (undoubtedly the case for many, many SD Times readers), then it makes sense to invest in a technology that minimizes the distance between domain experts and the running application.

These advantages may initially sound complex and abstract, but imagine what the World Wide Web would be without the <img> tag. The image tag allows domain experts (artists) to use a domain editor (Photoshop) to specify the contents of a document node. Imagine that, instead, artists gave programmers “specifications” in the form of bad photocopies, and then the programmers traced those out, put the page together, and then said, “Here, is this what you meant?” Obviously there could still be a Web, it could still have images and there would even be some great graphical sites. But such sites would be vastly more expensive and slow to build.

Even this metaphor doesn’t do justice to projectional editing, since you can’t arbitrarily switch between a graphical editor and a text-based one. Imagine that your editor allowed you to switch between HTML code, Photoshop and a “throw weight” editor that used a slider to change JPEG compression levels. That’s projectional editing for Web pages.

In Brooks’ terms, the involvement of programmers in the specification of every domain element is an accidental, not essential, difficulty. The most a domain workbench can do is directly involve domain experts in the specification of elements in one domain of a complex and precise model. Programmers are still needed. Building software is still difficult. (Further: building a high-productivity domain editor will be even more error-prone than implementing specific domain elements. Boot-strapping productivity is going to be expensive!)

However, many teams (maybe most among SD Times readership) face huge problems keeping up with domain changes. Month in and month out, the problem is taxes, or regulatory requirements, or… well, any of a million different domains. Many teams build different systems that reuse domain concepts (perhaps customized for a customer, perhaps simply servicing different use-cases and actors).

Simultaneously, many companies (maybe most) can trace the speed by which they can iterate software directly into a competitive advantage. A proprietary domain editor may be expensive to build, but the bottom line will demand it, especially once you get wind your competitors are making the investment.

First-mover or not, commercial success is not wholly dependent on technology. Other projects with similar ambitions, notably JetBrains’ MPS, are nearing fruition. Microsoft’s Oslo, while seemingly years behind, will undoubtedly evolve to become competitive. A demo gives no sense of the working experience, and we all know that perceived difficulties in fit-and-finish can cause even the greatest technology to be bypassed (I’m looking at you, Smalltalk).

Intentional Software has not shipped a silver bullet. But they’ve shipped something awfully shiny.

Larry O'Brien is a technology consultant, analyst and writer. Read his blog at www.knowing.net.


Related Search Term(s): domain-specific languages, Microsoft


Share this link: http://www.sdtimes.com/link/33526
 

Comments

06/10/2009 04:24:58 AM EST

The "view it here" phrase doesn't include a link when I view the article with IE - should it? I'm intrigued to see the presentation because it's hard to get my head around how having multiple grammars to learn (no matter how simplified) will make programming tasks easier. I know many programming/scripting languages. For me, the easier ones are like C#, JavaScript and PHP which share a syntax structure. VB is harder for me. But they are all difficult because their object models are different, differences which reflect their domain specialisms. These differences are not going to go away. It is still important to understand the DOM when using JavaScript and HTML when using PHP. I recently watched the M video showing the DSL 'Murl' created using Microsoft's M, part of the Oslo project. Murl is an M implementation of Curl. But unless you appreciated why Curl exists the benefit of the Murl DSL was not so obvious. Really they'd created a limited grammar that was not more intelligible to the newbie than the options for any other command line tool. That's because there's an implicit assumption that other know the target domain well and share the same mental model of that domain as the DSL author. In my experience, that doesn't happen often. The Java and .NET object models share much in common because they are based on a common understanding of programming fundamentals. Using some other object model is a recipe for time on the sideline, not because the object model is poor but because the model is not familiar (which could mean difficult and alien).

United KingdomBill


06/15/2009 06:42:33 PM EST

The URL should _definitely_ be visible in IE; I don't know why you're having trouble. Although I agree with all your logic, what I perhaps failed to communicate was that the syntax of a projectional editor is (a) often graphical and (b) presumably part of the domain-expert's area. So although it's true that you're introducing multiple syntaxes (and at some level complicating things), the hope is that by minimizing the distance between the expert and the specification of logic, you get a net win.

United StatesLarry O'Brien


Add comment


Name*
Email*  
Country     


  • Comment
  • Preview
Loading