Despite the many databases I might use to access information, sometimes it can only be found in the logs. In this post, I walk through the process of retrieving old barcodes, which seemed lost forever in history, and developing a script to allow a coworker to do this without asking someone on the systems team each time.
Using data from a single Integrated Library System can be a multi-tool, multi-database journey. In this post, I describe our local system and give a few examples of how I can find myself traversing several databases to get the answers or do the updates I need.
Walking through what it looks like when I get down to problem-solving. Sometimes I use the tools as intended. Sometimes I use the most logical tool. Sometimes I get it done well and quickly by going a little sidways in my process.
Even as an experienced OpenRefine user, I didn't know much about the API. I began investigating it when I needed to repeat operations across a large set of spreadsheets. While I found an alternative that's simpler for now, I wanted to share what I learned in case it's of interest to others.
Even though I've been a regular OpenRefine user for ages, I had never paid attention to a very important pair of buttons in the list of operations performed. As it turns out, it's extremely easy to perform a whole set of operations once and repeat it on other projects with the same data structure.
Blanking down is a great way to remove repeating rows within an OpenRefine record. But if the field's contents are the same between records, a single blank down action can wipe out rows of data you wanted to retain once in each record. You can solve this using row.record.index to add the record number to rows beforehand. This writeup contains the two transformations you'll need and a brief walkthrough on how to use them.
A rethinking of how we approach the peer review process in an era where some authors cut corners by using LLMs to generate all or part of their papers. How can we preserve our time and capacity to engage with authors who put in a good faith effort? And how should publications respond?
The desire to share my powerlifting videos on platforms with comparatively little capacity for large amounts of data led me to finally tackle FFmpeg. I share a bit about my use case, then describe and reflect on my process.
I've tried adopting notetaking tools before, even an earlier version of Joplin. But whether it was circumstances aligning or the tool advancing to provide exactly what I needed, this time it stuck. After using Joplin for over a year, I'm outlining why I find it so useful.
This spring, we tested Summon's new "Research Assistant," a Retrieval Augmented Generation (RAG) tool which queries a subset of their Central Discovery Index. This post outlines the few good things we found along with its many shortcomings. It may still be better than random tools they find online, but the bar is in hell.