If you have any interest in model-driven software development you should check out the 3 day openArchitectureWare Power Workshop we are now offering in Cambridge in partnership with the folks from itemis.

These workshops are designed for developers who want to move from simply modelling software to model-driven software development (MDSD).

During the workshop, the concepts and theories of model-driven software development are explained using the openArchitectureWare open-source MDSD Framework. The workshop is hands-on with a strong focus on acquiring practical experience of the various toolsets used.

The workshop covers:

• the basics of model driven development
• an introduction to the Eclipse Modelling Framework (EMF)
• the openArchitectureWare workflow engine
• textual domain-specific languages and Xtext
• code generation from models using Xpand and Xtext
• model validation
• creation and use of generator cartridges
• meta-modelling
• model-to-model transformation
• best practices for Model-Driven Software Development

The next course will be presented in Cambridge from September 22nd-24th by Karsten Thoms from itemis.

The fees for the 3 day course (including training materials, lunch and tea / coffee) are £1200 + VAT but if you book more than 3 weeks ahead of the workshop start date we’ll knock off 25%.

Many of the speakers at June’s Code Generation 2008 event have been kind enough to allow us to put their slides on the conference web site after the event. All slides are copyright their respective owners unless otherwise noted.

Keynotes

• Matching Supply and Demand: Challenges in Model-Based Code Generation for QoS-Constrained Software (Bran Selic) PDF (4.93 Mb)
• The Domain-Specific IDE (Steve Cook) PDF (4.38 Mb), Powerpoint (3.83 Mb)

Other Sessions

• Model-Driven SOA: Synchronize Business Planning with the IT Design Process (Ian Barnard) PDF (2.00 Mb)
• Supporting the Model-driven Development of Component-based Adaptive Systems (Nelly Bencomo) PDF (2.38 Mb)
• Enterprise Level Code Generation (Matthew Fowler) Powerpoint (884 kb)
• Using Eclipse Modeling Tools to develop LEGO Mindstorms Robots (Peter Friese, Frank Zimmermann) PDF (1.85 Mb)
• Evangelizing Code Generation: A Case Study of Incremental Adoption (Brooke Hamilton) PDF (3.00 Mb)
• Strategies for Generating Code from Microsoft DSL Tools and T4 Text Templates (Brooke Hamilton) PDF (224 kb)
• Building industrial sensors with Model-Driven Software Development (Bernhard Merkle) PDF (3.02 Mb)
• MDD: Models, frameworks & code generation: some experiences building business software (Pedro J. Molina) PDF (2.22 Mb)
• Putting It All Together (Michael Osmond) PDF (509 kb)
• A Practical Guide to Code Generation using Model Driven Architecture and Executable UML (Chris Raistrick, Ian Wilkie) PDF (1.71 Mb)
• Using openArchitectureWare for M2M and M2T (Karsten Thoms, Sven Efftinge, Jan Köhnlein) PDF (1.86 Mb)
• Implementation Techniques for Domain-Specific Languages (Markus Völter) PDF (3.14 Mb)
• Building Interpreters with EMF, Xtext and Scala (Markus Völter) PDF (2.15 Mb)

For slides for the sessions by Risto Pohjonen, Steven Kelly and Juha-Pekka Tolvanen - please contact MetaCase directly.

Audio-Recorded Sessions
Several conference sessions were audio-recorded. We will release edited copies of these as time (and energy) allows…

The Mathworks Automotive Conference took place in Stuttgart at the start of June.

Engineers and researchers from Daimler Chrysler AG presented the results of their work on “Functional Variants Handling in Simulink® Models” - work that is partly based on use of the pure::variants software product line management solution.

The presentation (PDF) begins with a rationale for the work - the ever increasing amount of functional variability present in automotive products - so much so that statistically no two Mercedes-C class vehicles produced in any year should be identical.

This variability drives a corresponding need to reuse software across the vehicle product line. To achieve this reuse the engineers argue that a systematic approach is required to variant management.

pure::variants underpins this approach by allowing them to model what the common and variable features of their products are and then using this model to produce variant-specific Simulink® models. (You can find out more about how pure::variants works here.)

In addition to improving reuse of software and models this approach also allows them to improve traceability throughout the life cycle. A valuable benefit in a safety-critical setting such as automotive.

As well as a PDF presentation the presenters also provided an associated whitepaper that describes the work in more detail.

Simulink is a registered trademark of The Mathworks Inc.

The programme for September’s Software Product Lines Conference in Limerick has been online for a few weeks now and registration is now open.

I’m running a panel session titled “Product Line Scoping in Practice” and a demonstration session titled “Jump-Starting Product Lines with Clone Detection” - more details to follow.

The conference runs from September 8th-12th; see the full programme.

I’ll be publishing my own reflections on June’s Code Generation 2008 event real soon now but a number of the other participants have already posted their thoughts:

Peter Kovari was first off the mark with postings recording his highlights for each conference day. Peter summarised by saying that “it was an excellent conference with a lot of great presentations from the top experts on this field, great discussions during panels, BoFs and Goldfish Bowls. A lot of new ideas, thoughts and questions to take away to consolidate and look for answers until next year’s event.” 

Clemens Reijnen enjoyed the conference and was also taken with the goldfish bowl sessions we ran. He’s also included a few photos and a short video from the conference.

Jan said it was“…probably one of the most inspiring conferences I’ve been to. The unique combination of a beautiful surrounding with highly intellectual people, created a steady flow of ideas.” and the event gave him plenty to think about.

In CG2008 - more than just content one participant noted that one of the big benefits of the event is the chance to meet with many people with similar and different experiences of code generation. In fact that author was one of a handful of people who were still at the event as we packed up and left, maybe they’re still there

Jan Köhnlein had ‘fun’ at the conference mixing with a lot of people with interesting views and news.

Leonardo Gomes provides a nice write-up of the main conference themes and notes “During the quite intensive three days of conference, we had the opportunity to listen, learn, discuss, meet great people and have a lot of fun.“

Finally, MetaCase have also mentioned the event in their July 2008 newsletter noting “[This event] is cementing itself as a key event for thought leaders in the fields of code generation and model-driven development.”

With these and several other participants also commenting on the intensity of the event maybe we’ll have to ease the pace for Code Generation 2009, after all, some of us aren’t getting any younger

Thanks to everyone who shared their experiences and contributed to making it a memorable event.

Update 14th July 2008

Sean Kavanagh has also written about the event. In his posting he highlights the Goldfish Bowl discussion: Modelling Standards: what exists, what’s missing and what failed?  as the most interesting panel session. Sean has a long background in modelling and Code Generation and also makes some interesting observations about the state of executable UML.

The Methods & Tools Summer 2008 edition has just been published and contains a short article by Matthew Fowler of NT/e and yours truly on the topic of using UML and / or DSLs for code generation.

We introduce both approaches and give some basic advice about their respective usefulness. We also make the mistake of predicting the future by claiming that UML tools will gradually lose market share to DSLs. Come back in five years time to see if we have to eat our words…

View the Methods & Tools Summer 2008 Edition (PDF 1.1Mb).

Well, the title says it all. Booking closes for next week’s Code Generation 2008 event on Monday.

Join 120+ developers, architects and others from around the world for three days of intensive learning and discussion on the practical aspects of code generation tools and technologies in Cambridge, UK.

With just 5 weeks to go until Code Generation 2008 we’ve been pretty busy tying down the final details…

Early-bird booking period closed at the end of March with bookings up 150% on last year’s levels - looks like word is spreading about the conference.

The event programme is now complete. We added details of the two remaining sessions a few weeks ago. Jean-Marc Prieur of Microsoft will run a tutorial on “Bidirectional DSLs : Synchronizing Model and code in Visual Studio” and Jos Warmer of Ordina will lead a goldfish bowl session entitled “Microsoft DSL Tools: User Experiences”.  Our Thursday night social event will see us take a trip to Cambridge Arts Picturehouse for a mystery movie and some time in the bar.

Places on our hands-on session ‘Using openArchitectureWare for M2M and M2T’ are limited and are going fast so if you do want to take part in that specific session please book as soon as possible and let us know you want to take part in that session. We have added an extra speaker to this session but that only increase capacity to 18.

Peter Bell is doing good work collating suggestions for Birds-of-a-Feather sessions. I’ve seen 5 Birds-of-a-Feather session ideas so far and Peter tells me that more are on the way.

We also have a couple more event supporters. InfoQ.com have joined as a Media Partner for CG2008. In case you didn’t know it’s an independent online community focused on change and innovation in enterprise software development. JPMorgan have also joined as a supporting partner.

Homerton College have also made some bedrooms available in case you haven’t yet booked your accommodation. See here for details on how to book these with the college.

More news on the event as time permits…

I’ve just published my interview with NT/e’s Matthew Fowler on the Code Generation Network.

Matthew talks about his background in CG, NT/e’s JeeWiz toolset and MDA vs. Domain-Specific Languages.

The interview is available here:

http://www.codegeneration.net/tiki-read_article.php?articleId=87

Steven Kelly has just written an interesting post about the performance of various code generators.

The basis for the post is an IEEE article comparing the performance of Eclipse MDD tools and Ruby DSL tools on a model with 40 classes and 50 inheritance relationships with each class having around 6 attributes.

He then compared the results with MetaEdit+ (his company’s toolset) and found that MetaEdit+ is over 30 times faster than Eclipse, and over 20 times faster than Ruby. He atrributes this difference largely to MetaEdit+ not having an M2M phase.

Interesting reading for code generator designers…