Dashboards are dead, long live dashboards!
E145 | Sun 16 Jul 5:30 p.m.–6 p.m.
Presented by
-
James is a software engineer in the OSPO Community Data team at Red Hat. He's passionate about sustainability in open source and about data science technologies. Outside of work, James loves cycling and running, and is a proud graduate of the Seattle Barista Academy.
Abstract
The lifecycle of data projects is involved. Responsibility for data, properly storing and retrieving data, scalably processing data… it can be a bit much. This talk will focus on a later-stage of the data lifecycle: serving data visualizations and analysis with sustainability in mind.
About a year ago, our team had to pick which tool we wanted to use to serve data visualizations and metrics to stakeholders. We had a laundry-list of requirements, some being deal-breakers while others were nice-to-haves. Our final verdict was a project that fit specific needs for us as a data science team, but in the process of choosing, we piloted a diverse variety of other alternative projects.
The framework for this talk is simple: introduce a collection of stand-out data visualization projects and discuss the pros and cons of each as we see them for a variety of use cases.
All considered projects are open source. They will be introduced in ascending order of interface complexity- and perhaps descending order of customizability. For example, the first project provides the user with a UI for doing data analysis- a later project will require a Python back-end.
The intended take-away of this talk is to provide attendees with a survey of projects that could serve them, and to shortcut the attendees own path toward finding a solution that works best for their team, minimizing platform-churn and saving time.
The lifecycle of data projects is involved. Responsibility for data, properly storing and retrieving data, scalably processing data… it can be a bit much. This talk will focus on a later-stage of the data lifecycle: serving data visualizations and analysis with sustainability in mind. About a year ago, our team had to pick which tool we wanted to use to serve data visualizations and metrics to stakeholders. We had a laundry-list of requirements, some being deal-breakers while others were nice-to-haves. Our final verdict was a project that fit specific needs for us as a data science team, but in the process of choosing, we piloted a diverse variety of other alternative projects. The framework for this talk is simple: introduce a collection of stand-out data visualization projects and discuss the pros and cons of each as we see them for a variety of use cases. All considered projects are open source. They will be introduced in ascending order of interface complexity- and perhaps descending order of customizability. For example, the first project provides the user with a UI for doing data analysis- a later project will require a Python back-end. The intended take-away of this talk is to provide attendees with a survey of projects that could serve them, and to shortcut the attendees own path toward finding a solution that works best for their team, minimizing platform-churn and saving time.