Thread by @helmholtz_hmc

Do you love 😍your data? If so, describe it well and make it future proof. How? We have some advice on how to make it happen. #fivestepstobettermetadata #myfairdata #Loveyourdataweek

Step 1: Include costs for data documentation in your grant budget and incorporate metadata creation into the data development process. Many funding agencies require it anyway. #fivestepstobettermetadata #myfairdata #researchdata #Loveyourdataweek

Step 2: Don't trust your memory, start documenting your #researchdata as early as possible! This will ensure that information is fresh in your mind. This in turn saves time and leads to a higher-quality metadata record #fivestepstobettermetadata #myfairdata #loveyourdataweek

#fivestepstobettermetadata Step 3: Create metadata records. In its simplest form, this can be a readme file attached to your dataset. Think about what others would need in order to find, evaluate, understand, and reuse your data. data.research.cornell.edu/content/readme #loveyourdataweek

#fivestepstobettermetadata Step 4: Use standards and controlled vocabularies. No idea? Look here howtofair.dk/how-to-fair/me…, or here rdamsc.bath.ac.uk
#myfairdata #researchdata

#fivestepstobettermetadata Step 5: Make your metadata discoverable and accessible by publishing in a searchable resource (e.g. trustworthy repository). Assign a persistent identifier (PID) such as a DOI to your metadata record.
#myfairdata #researchdata #helmholtzmetadata