@coolbutuseless Not sure if if this answeres your question, but I think following the golem or leprechaun frameworks could be considered good practice. In both frameworks the package exports a run function that returns the shiny interface. The remainder of the shiny app is in internal functions. See for example: https://leprechaun.opifex.org

@joelnitta @andrew yes exactly, I do it for the finer control. I think I couldn't for example choose r2u as a base image, but I could be wrong. I know I had reasons, but I cannot remember what they were ;-)

@joelnitta @andrew
Sure, here are two screenshots of the r script and the resulting dockerfile for a quick minimal example shiny app that loads the RPostgres package and thus needs libpq-dev

@andrew checkout the pak package https://pak.r-lib.org/reference/pkg_sysreqs.html
You can get install_scripts for your platform. I combine it with renv and the dockerfiler package in my CI pipelines to build the docker containers.

@Mehrad @A11yAwareness It is also totally possible and a valid option to add shapes to a heatmap see for instance this article from Fiona Baudner https://fionabaudner.medium.com/designing-accessible-charts-39ab0ff546b6

@tylermorganwall isn't the bar denoted "70.1%" below the 70 on the y axis?
Very ugly plot

@alesegura as an example in one instance we negotiated for the data scientists to get a dev environment where they had a lot of freedom during the experimenting phase while the IT protected the prod environment. When the data scientists were ready to move their data products to production they had an exact idea of what dependencies were needed and had to do the security stuff only once. This reduced the overhead for both IT and data scientists a lot while maintaining security on prod

@alesegura this is always difficult. When we conduct analytics infrastructure projects for customers, we mostly start with workshops bringing people from IT, management and Data Science together in order to find a common ground and plan out the infrastructure and processes. Most of the time the IT departments do not really know how data scientists work while the data scientists do not have the IT background. We would then mediate between both perspectives.