EDA limits, Tinned Fish, SQL Style
-
VHS allows you to create terminal GIFs as code -- great for demoing CLI.
-
Canned Fish. What does that have to do with data analytics? Rainbow Tomatoes Garden publishes detailed data on all their offerings, but this is really just an excuse to promote them because I have gone bananas ordering their amazing offerings of canned seafood. Thank you to Jeremy Singer-Vine's Data is Plural.
-
Here is a list of very useful macOS shell commands.
-
Brandon O'Leary's post on what he learned at GitLab that he doesn't want to forget describes three key points that are especially important for distributed teams.
- Write down everything
- Give and accept: ownership, agency, and responsibility
- transparent with a low level of shame
Read the whole post -- it's not as simple as it sounds. Also, it's worth revisiting GitLab's Values.
-
GitLab's SQL style guide is good, especially the noted best practices, conventions, etc.
-
In the Exploratory Data Analysis phase of data science, it's hard to know when to stop. From Erica Gunn via the Data Viz journal Nightingale:
- When it looks like a dead end.
- When you’ve understood what you needed to see.
- When it starts to feel overwhelming.
- When you start to lose focus.
- When the threads start to dissipate, rather than converge.
-
An incredibly straighforward set of Visual Design Rules with broad applicability.
-
We are constantly debating whether we should just use a monolithic repository.
-
We just kicked off a Data Catalog effort in my group. Ananth Packkildurai's article considers the following two questions:
- Why is it so expensive in terms of the level of effort to roll out a data catalog solution?
- Despite the initial energy from the stakeholders, why does the usage of Data Catalogs keep declining?
-
A good reddit discussion on Feature Selection, which I find one of the more tedious and "arty" aspects of model building.