(1) In my experience, when data management doesn't happen until the end of a project, about 80% of DM is detective work (what are these values, why is data missing, how do we handle this messy data)?

This way of working is tedious, often frustrating, and can lead to lost or misunderstood data.

In research, data without stewardship can quickly spiral into chaos 🌀:

📑 Disorganised files
😵‍💫 Incomplete documentation
👎 Inconsistent naming
✍️ Lost provenance

This is inconvenient and it undermines reproducibility, slows discovery, and can render valuable data unusable.

In the era of data-intensive research, making data findable is essential for impact, reuse, and collaboration. At the heart of this effort lies metadata, and no one understands its power better than a Data Steward. 💪

🌍 𝗔 𝘀𝘂𝗰𝗰𝗲𝘀𝘀𝗳𝘂𝗹 𝗗𝗮𝘁𝗮 𝗠𝗮𝗻𝗮𝗴𝗲𝗺𝗲𝗻𝘁 𝗣𝗹𝗮𝗻 covers:

🔹 Data description
🔹 Storage & backup
🔹 Access & sharing
🔹 Roles & responsibilities
🔹 Long-term preservation

Make it 𝗙𝗔𝗜𝗥: Findable, Accessible, Interoperable, Reusable.
Tools like FAIR Wizard can help.

#FAIRData #DataStewardship #OpenScience

🚀 FAIR data isn’t just for today.

Personal laptops lose data.
Domain repositories preserve it for decades.

Make your data 𝗙𝗶𝗻𝗱𝗮𝗯𝗹𝗲, 𝗔𝗰𝗰𝗲𝘀𝘀𝗶𝗯𝗹𝗲, 𝗜𝗻𝘁𝗲𝗿𝗼𝗽𝗲𝗿𝗮𝗯𝗹𝗲, 𝗥𝗲𝘂𝘀𝗮𝗯𝗹𝗲.

The real question: will anyone reuse your data 10 years from now?

#FAIRdata #OpenScience #DataPreservation

🔹 Metadata mastery – Structuring and standardising metadata for maximum reuse.
🔹 Communication skills – Translating technical detail into stakeholder-friendly language.
🔹 Policy fluency – Navigating legal and ethical frameworks like GDPR, open data mandates, and domain standards.

Learn more: datamgmtinedresearch.com/dmp#dmp-cata...

Template found here: osf.io/e4zqn

#IASSIST2025 Highlights!

Yesterday, our own Maja Dolinar presented a poster, and today she gave a talk on how #OpenAIRE empowers research librarians with tools for data management, discovery, and impact tracking.

From ARGOS to MONITOR, we’ve got the research lifecycle covered.

Our Introduction to anonymisation techniques for social sciences research data workshop starts at 10:00am today.

You can watch the livestream via our YouTube channel:

Sometimes we just need a pep talk.

For anyone who is interested, I've added the document to a @cos.io repository where it can be downloaded. 👇

osf.io/pu3hq

only took several billion dollars to get to “almost as good as a coin-flip”

I am absolutely inspired by the data management systems put together by @utahheartlab.bsky.social. They have built amazing, user-friendly documentation and checklists within Notion to help team members get acclimated to their data management practices and standards. Bravo!
@kjwbaucom.bsky.social

✅ Indexing & Cataloging: Store data in searchable repositories with proper indexing.

✅ Consistent Naming Conventions: Use clear and descriptive file names and folder structures.

#PSDI #Data #DataManagement #FAIRData #FindYourData #Metadata

Excellent AI data open access from @getopendata.bsky.social
Includes criminal justice data sets
youtu.be/8eIHFoMenvI?...

Interesting post on how "research libraries organize research support across a spectrum of organizational configurations" > Examining library structures to scale research support services: Insights from an OCLC RLP leadership roundtable   - Hanging Together

Today, #OpenAIRE 's Outreach & Engagement Officer Gulia Malaguarnera presents the #OpenAIRE Service Catalogue to the UK CORR & JISC community, exploring the tools & strategies shaping the future of #ScholarlyCommunication, #OpenAccess, and #repository management. Grab a taste!

Where to Find the LLMs Datasets for training?

- Kaggle: Hosts many datasets, both domain-specific and general-purpose, for easy download and use.

- Hugging Face Datasets: A repository of datasets for language models, including niche domains.

- AWS Open Data Registry: A collection of public datasets hosted by AWS.

- Google Dataset Search: Helps you find datasets across a wide variety of fields and websites.

- Academic Sources: Websites like ArXiv, PubMed, and Semantic Scholar.

#LLMs #Data #AI