class: inverse, center, middle # Publishing Interim Research Products ### Hao Ye ### Health Science Center Libraries, University of Florida ### (updated: 2022-07-19) <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script> --- # Motivations Why should you care about this topic? -- **Academic systems evaluate people** * I dislike this, but that doesn't make it less true -- Success requires self-advocacy * NOT the same as boasting * make it **easy for other people to find and learn about your work** --- # Learning Outcomes * understand the FAIR principles as they relate to scientific data management * explain the benefits of ORCID and Google Scholar profiles * define DOI, its attributes, and uses * identify tools and platforms for publishing interim research products --- # Costs * to my knowledge, all of the tools introduced here are *free* - they may have paid tiers with additional features - funding sources often include large non-profits, funding agencies, etc. * **DO be aware that some platforms are run by for-profit publishing companies that are monetizing data on academics** --- class: inverse, center, middle # FAIR --- # FAIR * Guidelines for *Scientific Data* (but which apply more generally) - Findable - Accessible - Interoperable - Reusable --- # Findable * Can humans and machines find the data? - use metadata that accurately *describes* the content - use metadata that helps people *find* the content --- # Metadata **`[def]`** data that describes or gives information about other data -- <blockquote class="twitter-tweet"><p lang="en" dir="ltr">The ISO country code for Namibia is NA, which has been a hilariously frustrating discovery I made while coding today <a href="https://twitter.com/hashtag/rstats?src=hash&ref_src=twsrc%5Etfw">#rstats</a></p>— Sarah Bowden (@mamabphd) <a href="https://twitter.com/mamabphd/status/1273352107240968193?ref_src=twsrc%5Etfw">June 17, 2020</a></blockquote> * hashtag identifies content as related to R --- # Accessible * there should be minimal barriers to access the data - e.g. no proprietary interfaces or software required - "available upon (reasonable) request" is insufficient! * use established methods to authenticate access to protected or private data --- # Interoperable * use a common, shared format for storage * use a metadata standard for attributes - [FAIR Data Point](https://github.com/FAIRDataTeam/FAIRDataPoint/wiki/FAIR-Data-Point-Specification) is one - some are domain-specific, e.g. [Ecological Metadata Language](https://eml.ecoinformatics.org/) --- # Reusable * add descriptions to help researchers identify whether it is suitable for their usage * use a legal license that enables re-use - e.g. CC-BY or CC-0 - **don't use a code license**, e.g. MIT * include attribution and a description of how to cite the data --- # Summary * FAIR principles are intended to make data more useful - the same principles apply for academic work! * for more on data, see the [Research Data Management Libguide](https://guides.uflib.ufl.edu/datamanagement) --- class: inverse, center, middle # Personal Profiles --- # Your Academic Identity <blockquote class="twitter-tweet"><p lang="en" dir="ltr">Never ceases to amaze me how many postdoc scientists & other early career researchers have basically zero internet presence. Many have no website (beyond an impoverished uni site & even that's often missing), no readily available contact details, no pub list.<br><br>Don't be invisible.</p>— Chris Chambers (@chrisdc77) <a href="https://twitter.com/chrisdc77/status/1090198315617132545?ref_src=twsrc%5Etfw">January 29, 2019</a></blockquote> --- # is it FAIR? * Findable * Accessible * Interoperable * Reusable --- # ORCID * https://orcid.org/ * Open Researcher and Contributor IDentifier * unique ID that stays with you, regardless of changes in name, email, institution * common way to login to journals and identify yourself on publications --- # Google Scholar * https://scholar.google.com/ * IMO, does the best job of automatically identifying your *papers* * allows you to correct errors in the records for your works * citation statistics - these are imperfect, but LOTS of people care and this makes it easy for them to check --- # LinkedIn * https://www.linkedin.com/ * a *must* if you are considering a position in industry * use the right buzzwords for your skills - find people with jobs that you aspire to, and see what buzzwords they use --- # Other * ask around to see what sites are common in your field - not just what people use, but what they use to *look up or find applicants* * consider a personal website: - present more detailed information NOT found in other sites - many free and nice options (custom domains are about $10/year) --- class: inverse, center, middle # DOI --- # DOI = Digital Object Identifier * format - `10.NNNN/{suffix}` - `NNNN` identifies the registrant - `{suffix}` identifies the object - e.g. `10.5281/zenodo.3892184` --- # DOI Properties * DOIs are permanent - the object and metadata record are archived * unambiguous - each DOI links to a unique object, and will only ever link to that object --- # Why is this Important? A given DOI always point to the same object. - (some exceptions for updating the metadata or deleting the object) The scholarly record needs permanence. - someone else needs to be able to look up your sources - DOIs fit this role Most journals issue a DOI for each article. --- # Get DOIs for Your Work * If you have interim products, e.g. + code + datasets + protocols + images/video -- * **A DOI makes that work citable!** + for fellowship/grant/job applications --- class: inverse, center, middle # Publishing Platforms --- # Places that issue DOIs * preprint servers * data repositories * general-purpose repositories * tooling-specific repositories --- # Preprints * papers, before peer-review and journal formatting * very common in math & physics, catching on elsewhere -- why? * share your work more quickly - not delayed until peer review * (sometimes) get feedback --- # Preprint servers * [arXiv.org](https://arxiv.org/) -- primary one for math and physics - uses their own indexing instead of DOI * [bioRxiv.org](https://www.biorxiv.org/) -- primary one for biology * [medRxiv.org](https://www.medrxiv.org/) -- primary one for health sciences * many others - including [OSF-hosted servers](https://osf.io/preprints/) --- # Data Repositories * Zenodo - https://zenodo.org/ - approved by UF IT for data storage (workshops forthcoming) - 50 GB capacity, supported by CERN * Dryad - https://datadryad.org * figshare - https://figshare.com --- # Data Repositories Additional Notes * Use whatever is common for your field / journal * Be aware of sharing restrictions for protected/private data * See the [Research Data Management Libguide](https://guides.uflib.ufl.edu/datamanagement) for details --- # General Purpose Repositories * Open Science Framework - https://osf.io/ - files can be uploaded or linked through github, dropbox, google drive, etc. - projects can be embedded within other projects - can issue DOIs for public projects --- # Tooling-Specific For specific (common) applications: * laboratory protocols and handy app -- https://www.protocols.io/ * computer code -- https://github.com - does not issue DOI, but deposits to Zenodo * pre-register study design and analysis -- https://clinicaltrials.gov/, or https://osf.io/ * [list of journals with registered reports](https://www.cos.io/our-services/registered-reports) --- # Thanks * Let me know what content you'd like to see * Contact me for additional questions or consultation requests! * Check back in on the libguide for more modules and contact info: - https://guides.uflib.ufl.edu/reproducibility