Hortonworks DataFlow (HDF) 2.0 is now available! HDF is powered by Apache NiFi 1.0.0, which recently underwent a major redesign. Whether you’re a current user or just now planning to try it out, this is exciting news. A lot of new feature content went into this release such as multi-tenancy and zero-master clustering. The purpose of this post, however, is to share a little background on the design of the new user interface (UI). For those of you familiar with a previous version, you’ll notice a big difference. For those who are just getting started, the new UI will hopefully get you on the fast track to reliable and secure data flow for you or your business.
The main goal of the new UI is to provide a fresh look and feel, bringing it up to today’s standards when you think about modern, web-based user interfaces. Reducing heavy and distracting colors, removing excessive gradients and drop shadows, and introducing one-color icons are methods meant to reduce noise and potential distraction. These changes help to reserve the use of heavier design elements and vibrant colors for important messaging and calls-to-action. The ultimate goal is to promote successful interaction with the UI and a more efficient experience overall.
Context Sensitive Palettes Tailored by Job Function
Going beyond the visual design of the interface, some changes have started to help pave ways for improved user experiences, ones that accommodate a wide variety of user types and ever-expanding use cases. The largest change in this regard is the introduction of context sensitive palettes.
Palettes appear directly on the main workspace of the UI, right where one would be working to build, maintain, or monitor a data flow. If you’re familiar with previous versions, palettes bring together navigation (e.g., zoom, pan, etc.) and component controls (e.g., run, stop, group, etc.) closer to one another, and, more importantly, to the data flow itself.
Creating these logical groups of flow-related actions help the UI cater to multiple user types by presenting actions relevant to job function. Palettes can be opened and closed independently, and even tucked away completely for an unobstructed view of components on the canvas. The Operate palette is a great new addition to the UI. It is context-sensitive, meaning information displayed and available actions are directly tied to your selection and location in the data flow.
Move of Menu to Reduce Visual Noise
Another change previous version users will notice is the movement of high-level management and settings related options to a global menu that is positioned in the upper right of the interface. The first reason for this change is a simple one of available real estate and efforts to reduce visual noise. It is scalable in that it can contain any number of potential options versus the previous display method of showing them all on the main UI toolbar. The second can start to be seen when you expand the menu and notice different groupings, again to start to diversify the UI for various user types.
Breadcrumb Trail to Help Guide User Actions
The last significant UI change is the repositioning of the breadcrumb trail to the bottom. The breadcrumbs track your movement through a flow and indicate current location. As you move deeper into a flow, it serves as a navigational control to move back up to parent process groups. The change in position was again part of the effort to reduce visual noise in the upper section of UI, but also a logical move since every action a user takes is relative to their position in the flow. It serves as a foundation upon which user actions occur.
Speaking of foundation, the new HDF 2.0 design represents a big step forward and a great base on which the application will mature. Check out the Hortonworks DataFlow homepage for more information as well as details about Hortonworks’ entire Connected Data Platform.