Digital Experience Accelerator in depth

Editor's Notes

• #4: Following the golden circle, I'll explain a bit more about the why first!
• #5: Lets start with going back in time a bit...
• #6: The community already had CWA and DD4T while SDL only had DWT SDL Tridion Reference Implementation v1.0 released September 2014
• #7: So now I hear some of you thinking: "But what about Razor Templates, why is that community initiative not embraced?“ Well let me tell you why Razor templates are evil... They are completely following the DWT approach which is flawed (as I just explained) when you publish JSP or ASPX pages
• #9: Not me, I wasn’t a Product Manager yet back in 2013, you know who I’m talking about (Nuno ;o)
• #10: We chose to release something working as soon as possible so we could gather feedback from the field
• #11: We started only focusing on .NET, since that was the knowledge we had in the team. We developed on SDL Tridion 2013 SP1 (the current release at the time). The plan always was both .NET and Java, but building the team and finding the right people was a challenge.
• #12: I'll keep it short since most of you might already know what DXA is by now (and if not, shame on you, you could have downloaded and installed it already ;o) Download location of DXA 1.1 and its Modules: https://community.sdl.com/developers/tridion_developer/m/mediagallery
• #13: We named our project the Tridion Standard Implementation (tsi), that changed into the Tridion Reference Implementation (tri), and when we were ready to release, it became the SDL Tridion Reference Implementation (stri).But wait, it is called DXA right? Release 1.0 only contained Tridion functionality, so we started planning the 1.1 release with a new name, and modules that would make us earn the Digital Experience part in our name.
• #14: Out-of-the-box adaptive design framework to deliver, dynamic, device-independent experiences. Design templates built around semantic content and markup. Production ready web application New template paradigm that decouples dynamic content, presentation and style Built using latest approaches in adaptive design and packed with schema.org content models Supports agency product workflows by managing style and creative content as presentation elements Time to market – 20% reduction is development to go-live Re-engineered the template platform 1st design agency friendly WCM system Usability – WYSIWYG Separation between design & content  localize content very easily; take designs from an agency (different designs for different geographies) Adobe: here’s the design, and apply the design manually on every page 20% reduction – no more manual building of navigation buttons, menus, dynamic experiences No need for custom developers – just .NET or JAVA Context Experiments Native testing and measurement capabilities supporting cross-channel experiments Full context availability supporting audience, session or device specific conditions. Jaime Santos Alcón: We went from nothing to having the home page "ready for live" in barely 3 days, and that includes infrastructure within a DTAP architecture (yes, that's 3 environments)
• #16: Version 1.0 was released September 2014, last month (August 2015) we released DXA 1.1 with some new features.
• #17: I have a complete new team: testers, developers and technical writer (I’m trying to be a Product Manager and not develop anything myself anymore, but some of the 1.1 code was still done by me ;o). We fixed a lot of small issues and improved even more other things.Changed the .NET code to keep in sync with the upcoming Java version. So that the architecture and approaches are as equal as possible in both the .NET and Java release (its interesting to see what is possible in ASP.NET MVC vs Java Spring MVC). Java code under development is available in the develop branch in GitHub https://github.com/sdl/dxa-web-application-java/tree/developWe changed our API to have better extensibility which meant we could created modules in the right way. Then we got started on our first set of modules, ECL/Media Manager, SmartTarget and an updated Search (adding AWS CloudSearch).
• #18: Robert Curlette said: “code is better than PPT” Check the documentation for more details http://docs.sdl.com/sdldxa11 Check out the sources on https://github.com/sdl/dxa-web-application-dotnet
• #20: We intend to follow Gitflow (http://nvie.com/posts/a-successful-git-branching-model/) with the following main branches: master - Stable develop - Unstable release/x.y - Release version x.y Please submit your pull requests on develop. In the near future we intend to push our changes to develop and master from our internal repositories, so you can follow our development process.
• #21: As promised at Innovate San Francisco, my Impress module is available as an example https://community.sdl.com/developers/tridion_developer/m/mediagallery/1173 https://github.com/sdl/dxa-modules/tree/master/webapp-net/Impress When you use DXA you have the freedom to change anything you want since it is open source. But our idea is that you don't need to change our core implementation and can add your additional features by building modules.If you can't do what you need in a module, let us know, we need that kind of feedback. We have used our own modules to validate the core functionality is extendable.
• #23: We are in the years of acceleration, that is how DXA fits in the general SDL roadmap.
• #24: Nothing is cast in stone yet, we will need freedom to make adjustments and do things right which can mean postpone releases Feedback-based progression Working very close to the field, we are continuously gathering feedback on what works well and what doesn’t, building only what brings real value Remaining nimble and flexible early on allows us to quickly adopt changes that are future proof, with the goal to offer a truly usable solution Adopting support for DXA means SDL must provide: A stable, scalable and reliable solution Support infrastructure behind it (building it is easier than maintaining it) We don’t just need to build this right, we also need to build the right things! Focus on building what customers and partners truly need for a successful SDL Web implementation
• #25: Things we are considering for our future release: General idea behind a DXA Model Service is to take a next step in the “Content Delivery as a Service” (“CDaaS”) architecture, that is: the separation of the (custom) CD Web Application and the Content Delivery Platform (a.k.a. Web Delivery Platform). Minimizing roundtrips between Web Application and Content Delivery Platform: aggregation of data (e.g. link expansion, SmartTarget queries) happens close to the CD Kernel in DXA Model Service (which is part of the CD Platform). Minimizing the DXA footprint in the Web Application; maximizing the amount of logic provided as part of the CDaaS service Minimize the amount of third-party libraries required in the Web Application to minimize version conflicts and reduce upgrade impact Maximize the amount of logic that can be versioned (upgraded/patched) by the CDaaS service provider.
• #26: DXA will remain open source, you already knew that right?
• #27: https://community.sdl.com/events/sdl_community_open_sept_2015_amsterdam/c/e/115 If you can’t make it on Friday, contact me for free access to the DXA elearning course available soon.