This week I gave a 30 minute talk at JMP Discovery Summit 2015 in San Diego. Full disclosure-I work for JMP as director of our devops group, meaning I spend my days making sure that our software release process runs smoothly. There are a lot of moving parts in our effort to get releases out the door in a timely fashion, and my team helps make sure the end stage tasks get done and customers get their software. I’ve started using JMP as a customer back in the JMP 5 timeframe for analysis of genomics data, so I’ve seen the product evolve greatly over the years.
But my Discovery talk wasn’t exactly work-related. Instead, I talked about a fun quantified self project I’ve been doing with my workout data that exercised (pun intended) a lot of JMP’s new (and older) features. I posted in some of my early blogs on this site about this project and you can see more in my Fitness and Food series on the JMP blog. I uploaded my Discovery slides to the JMP user community this morning, but you also can see them at the following link: Discovery 2015 9-19-15 for community.
This project has been different than many of the other QS projects I’ve written about since I didn’t start from a set of files automatically generated by a device. Rather, I had entered data on my workouts manually into notebooks over the years and spent some time typing it up so that I could visualize the data using JMP. Yes, this was boring and took some time. But it’s been well worth the effort for the insights I’ve gained (and the bugs I’ve found and reported to JMP’s developers in the process). I love working with new dirty data and real life use cases, and I feel fortunate that part of my day job includes the chance to do just. I get the chance to help improve our software, develop real-life examples, and learn more about myself and my patterns. On a side note, I am glad to now be collecting my data in a more automated way using the Push Strength band, and on the plane on the way to Discovery, I spent some time working on those exported files to get them in shape for a future project with that data.
As with most manually entered data, I had to do quite a bit of cleanup on my workout data table. I used JMP’s Recode platform to consolidate exercise names, group exercises into body parts, and group body parts into body zones for the purposes of examining changes in my patterns over time. I made use of JMP’s local data filter (a list-based filter) and selection filtering (a graph-based filter) to drill down and visualize my workout data dashboards at the body zone and body part level.
I had a lot of questions at the end of my talk, and over the next few days I was approached by many other users who had been at the talk. They asked me a variety of questions about my workout data tracking, using custom maps, and my experience with various devices. I’m always glad to share what I do and what I’ve learned! Although I presented a poster at last year’s Discovery conference in Cary, this was my first time doing a talk at a Discovery conference and I’m really glad I did. I love sharing the power of personal data and the cool tools that JMP can provide for analyzing it. You can see more details about Discovery 2015 here on the JMP website.