Merging DD4T and DXA

Authors: Nuno Linhares (SDL) and Quirijn Slings (Trivident)

 

If you attended the excellent Tridion Developer Summit 2016, you may recall that we announced a merger of the DXA and DD4T projects during the keynote session, and now we’d like to take this opportunity to dive deeper into what exactly this means to the product and the larger group of implementers that today use one, both or a combination of these solutions to improve the quality of their implementations.

 


Both DXA and DD4T share the same goals of improving the quality of our software’s implementations, allow for newbies to catch up quickly by relying on de facto web application standards, and generally, provide a means for developers to fulfill company objectives faster. This is why the decision was made to use DD4T as the engine for DXA in the first place.
Throughout the development lifecycle of the DXA, the significant architectural changes done in SDL Web 8, and the large improvements brought into DD4T by the community, decisions were taken on both sides that unfortunately resulted in the solutions diverging instead of merging, and we find ourselves in a position where both solutions compete rather than complement each other, and this is certainly not something any of us ever wished for, or anticipated when the “SDL Tridion Reference Implementation” (later renamed to Digital Experience Accelerator) was first developed.

Now we believe the time has come to merge DD4T and DXA into one powerful framework, which brings down the development time even further, while maintaining all the flexibility that our enterprise customers expect.

Steering group


As a first step towards integration, we have decided to create a steering group for this new combined solution, and make sure that the architecture of our solutions evolves together, and provides the implementer community with the best implementation framework out there.

This Steering Group has now been set up, and we’re drafting our initial milestones as we speak. This is the Steering Group:

  • Quirijn Slings, DD4T "Founding Father", representing the community and DD4T .NET
  • Raimond Kempees, DD4T Java Maintainer, representing the community and DD4T Java
  • Will Price, DXA "Founding Father", representing the community and DXA
  • Wouter Kers, Release Manager, representing SDL DevOps
  • Nuno Linhares, Benevolent Dictator, representing SDL Product Management

 

We will start by cataloging and identifying all the various features of the different frameworks, and in particular highlight those where a different solution was chosen for a similar problem. We expect this to keep us busy throughout the summer of 2016, and once it’s done, we’ll sit together and choose the best solutions for it.

Next to that, we’ll appoint an “Architectural Review Group” made of significant community and SDL contributors to the platforms, so that they can validate our choices - it’s not that we don’t trust our own judgment, we are simply striving for the broadest support possible.

As the path becomes clearer, we will make sure that the backlog and roadmap are public, allowing everyone in the world to comment and influence the priority of our tasks. There will be cases where certain priorities cannot be changed. This could be due to new features being added to SDL Web, or a significant requirement to release on a given date. Determining and enforcing those is the role of the Benevolent Dictator – and our goal is to provide an open forum that makes this platform the reference platform of choice for all SDL Web developers out there.

SDL also commits to keeping our solution open source, under the current Apache 2.0 license. As a long-time implementer, I am a firm believer that Web Application Frameworks that are open source tend to have a lot more acceptance by enterprise developers, as well more commitment for long term support. As revealed in a recent Forrester Study (Digital Experience Technology and Delivery Priorities for 2016) 51% of surveyed professionals agree with us.

And last, but definitely not least, we will continue shipping and releasing as often as we possibly can. The pace at which new features get added to web frameworks today needs a new model for delivery and distribution, and our adoption of NuGet and Maven will ultimately deliver the flexibility and speed required.

We are still working on the final timelines for the process, but we expect to be able to release the new unified framework in 2017.

And we’ll call it…
Well, that’s something we haven’t decided on. In line with the open nature of both DD4T and DXA, we are asking the community for help. Please share your suggestions on Twitter using the hashtag #newframeworkforsdl.