I’m always happy to attend Symposium; to see familiar faces and meet some new ones. We gathered for a second time in Orlando’s Swan and Dolphin Resort and the weather cooperated once again for our evening adventures. Here’s a quick rundown of some of my highlights of the conference:
Technical Keynote: Emily Freeman
Emily is a world-renown expert on DevOps. My main takeaway from her keynote is that collaboration between teams is key and that no team is more important than another. It’s important to have trust, rapport, respect, and diversity between teams with the freedom to be able to share new ideas.
While some people are touting that DevOps is dead; Emily is taking a firm stand that it is not. “DevOps is a journey” and there’s always room for improvement. For instance, after delivery, focus on reliability via availability, latency, throughput, fidelity and durability.
Balance for Better: Mentoring: Kelly Brennan and Amy Turrin
I had a chance to meet with Kelly and Amy a few days before the conference started, to get an understanding of what their session on mentoring is about. It sounded so interesting that I attended. As part of the “Balance for Better” initiative, these women are providing resources for gender minorities to have a better chance at succeeding in tech.
They are betting that having an application to support matching mentors from the community with newer people joining the tech ranks will promote less drop-out in tech by women and minorities. Mentees can seek advice, develop their leadership and tech skills, network, and it can help with career advancement. Soon, Amy and Kelly will be releasing their mentor/mentee application.
Netflix Next with Sitecore Cortex: Deepthi Katta and Richard Leiva
(My co-workers!) Deepthi and Richard introduced Cortex in a tangible way, by using Machine Learning to choose recommended movies from within Netflix. Primarily, they touted that if they can use Cortex in a practical application, there’s no stopping you from doing the same.
They explained that Handlers order the tasks and sort the dependencies of the tasks, Agents organize and call the proper workers to perform the task at hand. Workers perform the task either as a Distributed worker (in parallel) or Deferred (just called once, no parallel actions). They chose to use Sitecore goals to determine their Netflix recommendations.
New Editing Experience: What it means for Developers
The focus of this year’s conference was definitely Content. Sitecore asks … how can we improve the experience for content editors?
…by introducing new tools, of course:
- Horizon – new page editing
- New Launchpad – user personalized
- Zenith – accessing Sitecore content in a headless manner
- Auto-personalization Insights – AI-based insights using xDB (and Sitecore AI SaaS)
The focus of this session was on Horizon’s features. Horizon will be released with XP9.3 and will be an opt-in, SIF install. It will include simple page editing, preview device simulator, and analytics insights. 9.3 will allow use of Horizon on single sites. In the Summer of 2020, Horizon will support the ability for developers to extend Horizon. Examples of extensions include switching content from left to right, to right to left or using an instant translator via client-side javascript.
Future releases will include the ability to use Horizon with multi-site configurations.
Facial Recognition and AI
Why facial recognition? It is a key way to identify a person in a non-intrusive format; it’s cheap and simple and the person is recognized as an individual, not a login or group. Is it controversial? Sure. It is important to get consent from the user. It can be used online but also in brick-and-mortar. As an example, imagine a smart bookshelf that provides you personally with book suggestions because it recognizes your face and has a history of your preferences.
Your machine learning model needs to be trained. It can be trained through typical event tracking but with facial recognition, it can also be trained via emotion analysis. We can now use text analysis and also facial reactions to to match the emotion of the face.
You see in the near future, instant recommendation feedback, bookstores with heat maps that prompt the store owner to re-organize the flow the the store based on ML and using emotion to drive the customer to products that align not only with their identified purchases but also their emotion.
Sitecore AI
Consumers are demanding that companies know when to contact them and ALSO when to NOT contact them. This requires machine learning. At present, Cortex is the only supported way to connect Sitecore 9.x to a machine learning service. Coming this December, there will be one more way. Sitecore AI, an auto-personalization, machine learning SaaS will be available as an add-on.
This service, independent of Cortex, provides machine learning and auto-segmentation customized to Sitecore. Sitecore AI has the ability to create marketing segments or use existing ones. Site content is auto-tagged using text and image analysis.
The site visitor patterns and tagged content are used to create the segments. All page components with auto-personalization turned on automatically adjust to the visitor based on segmentation. Sitecore AI is best used with sites that have at least 12 million unique visits per year. As of this writing, there are no announced plans to phase out Cortex.
Chicago Sitecore Symposium 2020
Looking forward to a midwest Symposium next year!
Check out some online content from Symposium 2019.
