Dear all,
The Psych-DS validator grant has officially kicked off! I'm excited to introduce Brian Leonard, who will be working full time on Psych-DS projects including the validator software. Here's a brief note from him:
Hello psych-DS community! My name is Brian Leonard, and I recently joined the Lookit/psych-DS team as a Software Engineer. My core focus will be developing our suite of validator tools for psych-DS, as well as functioning as a liaison and point of contact for psych-DS in general. My background is in Linguistics, and in my career I've functioned in technical support roles within various research programs in the fields of Psychology, Linguistics, Cognitive Science, and others. I'm excited to get to work building the much-need psych-DS validator according to the needs and standards of the research community at large, and the dedicated psych-DS community in particular.Brian has already been hard at work getting up to date on the conversations we've been having about the Psych-DS standard over the last several years, and organizing what we know about the space of related projects, their validator design choices, and related software tools.
We're now reaching out to the broader community who have been thinking on & off about this project for a while, and we'd love to get your feedback on
the direction of these plans, as well as processes for coordinating contributions of any size and kind you may be interested in making!
We'd also love to hear from you about any recent tools/developments that might be helpful as we move into the next phase of this project - machine-readable metadata and research data standards are hot topics with lots of new activity, and it's quite likely you may know about something we don't about best practices or opportunities for building the best set of Psych-DS tools possible.
For instance, here are two projects we've learned about recently that we're hoping will be very helpful for building/using PsychDS:
*
LinkML, a framework for expressing many kinds of schema (including JSON-LD/Linked data) that we're hopeful can be used to express the rules of Psych-DS in a common format for validator tools in R, Python, browser validator, etc. This will ideally allow us to maintain a single implementation of the rules of Psych-DS that can be updated over time, and automatically propagated to the validator software and potentially other tools as well.
*
FAIR Metadata Wizard, a tool created by developers at Maastricht University for creating/autocompleting and editing JSON-LD objects that follow the
schema.org vocabularies. This tool was introduced at the most recent SIPS hackathon as a potentially very useful interface for researchers working with their Psych-DS metadata, both in its current form and potentially as a basis for additional development.
Longtime Psych-DS mailing list members will know my penchant for long emails, so to wrap this one up with some notes on timeframe and what you can do right now to help:
1. Head to the
Validator Requirements doc and comment on the emerging plans for Psych-DS validation software. Especially if you don't think of yourself as a software programmer, please have a look at the first two sections in particular, which deal with how we need to design these tools to be useful to as many researchers as possible. Make sure to include your name as a contributor at the bottom of the doc!
2. If you are interested in attending a Zoom meeting in a few weeks to continue discussion and even start diving into targeted work on the codebase, documentation, or other contributions to the project, please fill out this
when2meet.
3. Have another idea or thought about something that the Psych-DS ecosystem can do? Please feel free to respond by email or
make an issue on the Github repository so we can continue collecting these ideas!
Looking forward to talking more soon,
- Melissa