At SeleniumConf 2018, I introduced a setup based on the Selenium project's Selenium Grid Docker images, a Python library named Selene, and the pytest framework that is easy to maintain, version, upgrade, and distribute to members of your development team.
At SeleniumConf 2014, their first year in India, I spoke about three design patterns you should be using once you understand the Page Object: WebForm Pattern, ItemList Pattern, and IframeWrap Pattern.
Tutorial at the Practice & Experience in Advanced Research Computing Conference about capabilities added to the HUBzero Platform to support geospatial researchers. Collaborated with Kevin Wojkovich to develop presentation slides and a tutorial centered around building, running, and deploying R based applications (Shiny, R Markdown, RStudio IDE) on the HUB. Kevin presented the material and ran the tutorial.
Research at the University level is filled with stories of code being passed down from grad student to grad student and referenced in papers but rarely being used by others outside of the original research team. In this workshop participants learned about what resources the HUBzero Platform provides to help disseminate that research code, with a web or desktop style graphical user interface, to the world so it can be used by others and built upon in an Open Source ecosystem.
From 2007 to 2017, I helped run a software bootcamp sponsored by the Network for Computational Nanotechnology (NCN). The 3-5 day workshop was geared toward getting the summer students, participating in the SURF program, familiar with good programming practices, debugging skills, the Rappture Toolkit, source code repositories, and how to publish simulation tools on nanohub.org. I've also run this workshop at severalotheruniversities and parts of it at conferences.
In this talk with Indy UseR, we hear about a first time experience using Quarto, an open-source scientific and technical publishing system built on Pandoc. For this talk, I try using what I learned to write the presentation using Quarto.
In this talk with Indy UseR, we explore the pins package. The pins package allows users to save and retrieve data to and from “boards” that can be reused locally or shared with other when used with external services like Github, Kaggle, AWS S3, Google Cloud, Azure, Digital Ocean, or RStudio Connect. We take a look at the core functions of the pins package.
You’ve imported the data, cleaned it up, and built the model. Your next step is to communicate your findings with others. Typically this would mean writing a report or building a web application so other people could visualize how the model works, but sometime your Communicate step is someone else’s Import step. In this talk, we’ll explore how to expose your analysis to others in a programmatic way by using HTTP APIs built with the Plumber library. Our conversation will include a high level overview of how HTTP works, code based examples of building an API, and how to use that API from different programming languages.
It's something you've been meaning to look into, and every time you're ready to explore, something else comes up. That's life! Luckily, I have put together this quick primer presentation for the May 18, 2019 HackLafayette Thunder Talks that covers where Docker stands in the soup of virtualization setups out there, which commands you should probably know about to be productive, and an example of packaging up an application so it can be shared with others.