And after coming back, I was talking with some students at the University of Maryland. And they're like, well, what's gray literature? So I was talking to them about it there. I was saying that this coming year, they're going to be talking about data. They're like, well, how can data be gray literature? So I said, you know, I think they need something kind of more fundamental.

So we've had some really good talks today that are fairly specific, maybe a little bit more sophisticated. But I'm really aiming, I'm probably preaching to the choir here, but I'm kind of thinking of an audience beyond this room for people who are really looking for the fundamentals about how

to get started creating and distributing research data that fulfills legal and other requirements. So there's basically four steps in the model that I've come up with.

And it's identifying the different types of data that you are going to be analyzing, collecting, distributing, whatever. Identifying what all the different types of compliance factors are for using this data.

Identifying the technical needs for everything from distributing to curating the data, and everything in between. And identifying resources for assistance, which is always useful in my book.

So there are a lot of different ways that you can sort of create a taxonomy of data. And I'm taking a very maybe liberal definition of data.

And this actually came from a discussion with a couple of STEM librarians and a digital humanist. And the digital humanist referred to a 14th century text that he was analyzing as data. And the STEM librarian, that's not data, that's medieval German. He said, oh, Wolfram Wernischenbach can actually be data.

So when I'm talking about data, the STEM folks were, yeah, data, that's numbers. I'm like, nah, you know, I think literature can be data. I think words can be data. I think media can be data. So I'm taking a very sort of all-encompassing definition

for what constitutes research data. But quantitative versus qualitative is one way that you can divide data. Restricted versus unrestricted, and we'll talk about the various types of restrictions and the reasons for those. And the content of the data, the format that it's in,

all of the other factors that can be involved when you're working with research data and that you need to take into consideration when you're trying to create and disseminate data that's in compliance with legal and other obligations. So some of the compliance factors

include legal, international, national, and regional or local laws or regulations, policy, either from the institution in which somebody is working. There may be policies that are disciplinary or related to a professional or scholarly society.

And there are policies that come from sponsors, granting agencies, and other partners. And then the other issues, access and use, storage, preservation, and curation. So all of these really have to be looked

at sort of holistically when coming up with a plan for creating and distributing research data that's in compliance with these various factors. So some of the technical needs that users and disseminators

of data are going to need to consider include things like access and restriction. And like I said, this conversation began with some students. And they said, well, why do we have to restrict data? I was a freshman, so he was kind of naive. I said, well, your grades are at data.

Are you going to post them all on Facebook? Oh, no, now I get it. So that was an easy one. We had a discussion about HIPAA and all those other things, which we'll talk about in a couple of minutes. Curation and preservation, which

is becoming an increasingly sort of sophisticated and technical issue as the multiplication of formats just continues. How do we deal with those? How do we make sure that those things are preserved? How do we manage future format migration,

things like that? And then other technical support that you might need for distributing data. So in my conversation with students, they said, well, who do we ask? That's why we're asking you. And I said, well, maybe I'll write a paper

and give in New Orleans and give you some answers. Human resources are always, it seems obvious, but you'd be amazed at how often people don't take advantage of human resources that are available at their own institution. I always ask people when they're doing anything at my organization involving distribution of data,

have you run this past council yet? Nah. I'm like, eh, you might want to do that. I always give the footnote that if you have a telephone or in-person conversation with either attorneys or human resources, always follow it up with an email so that you have their response in writing.

It's always useful to have. Law and policy resources, so if you have access to council, to administrators who can either interpret or sometimes create policy, and the various technical resources, whether those be personnel,

software, hardware, that you will use when you're creating and distributing research data. So restricted and open research data, there's a lot of different research, legal and regulatory requirements

for sponsored research. Many government and private sponsors of research require that information, including data, be open and freely available to other researchers. This is not universally true, but it's often true. And the way in which this is done can vary widely,

especially between different granting agencies. So that's an important thing to take into consideration. A lot of sponsored research involves data with a wide range of restrictions on use, access and identification of source. HIPAA and FERPA data are just two

obvious examples in the United States. So some of the resources that I encourage people to use when they're talking about a project involving significant research data is begin with the human and technical resources.

Begin with your institution who's there who can help you. And like I said, researchers, administrators, counsel, technical and clerical support, these things all vary widely by institution and by organization. But the best place to begin usually is, if you don't know, with your supervisor,

with your, if you're in an institution of higher education with whoever your senior administrator is. And like I said, it's often very useful to talk to counsel. That's something else that you should probably discuss with your supervisor. Some institutions have channels through which

they want people to approach legal counsel. University of Maryland, we have a repository drum that gets a lot of heavy use. And we have a good guide at the University of Maryland

for writing data management plans. A lot of people have found this useful. That's a good technical resource. University of Pittsburgh has a really good research data management site with some really useful information too, these are a couple of good technical sources for guidelines.

And SPARK, the Scholarly Publishing and Academic Resources Coalition, which is basically a project of the Association of Research Libraries. They're very heavily involved with open access and they have a lot of actually very useful information

regarding the creation, storage and dissemination of information of all types, including research data. So some of the resources for law and policy, Europe and the EU, I've got a couple of links there,

the EU Open Data Portal and some information from europa.eu. The UK Data Service, the International Standards Organization can be a really useful resource

for a lot of projects actually. And continued several of the major American resources for policy in the United States regarding research data. So of course, FERPA and HIPAA are two of the main ones.

Most of the major granting agencies of the federal government have good resources online as well. The National Science Foundation, National Institutes of Health, National Endowment for the Humanities, National Endowment for the Arts. And you know, there's a huge list actually

of federal law funding agencies and SPARK actually has a directory that will lead you to the resources that can be really helpful for this. So a model for creating and documenting compliant data is actually pretty straightforward.

Identifying the types of data, identifying compliance factors, identifying the technical needs and identifying the resources. So this is pretty straightforward

but it's actually really important. And one of the things that I do when I discuss this with people, especially with grad students or with junior faculty who are just beginning doing real research, hardcore research projects, is the need to think through all of these issues before you begin.

Really coming up with a plan and a model that works for your specific project from the beginning. A lot of junior researchers are eager with the project and will jump right in without really kind of a plan and they'll get bogged down in the middle of it. So I always encourage people to create a plan

for a compliance plan separate from anything like a data curation plan that you might have to have from the very beginning of their project and discuss that plan with everybody participating. So compliant research data will fulfill statutory

and regulatory obligations of the institutions, organizations, sponsors and governments, maximize the impact on scholarship because it meets requirements for access, distribution and reuse, and it documents due diligence on the part of researchers.

Thank you.