 Thank you so much, Maureen. I'm joining you from Washington, D.C., in the Office of Heritage Preservation, and we're so pleased to be bringing you our next online course, Caring for Digital Materials, Preventing a Digital Dark Age. You've probably joined us on our previous two webinars, and wanted to just, again, thank Learning Times for helping produce these webinars in our website, and also to the Institute of Museum and Library Services for making these available free of charge. So it's a tremendous service that we're able to offer this continuing education at no charge to you, and so we are very appreciative of IMLS for that. Again, this is our third course, our third class story in this course, and we will be talking about metadata today. It's a term you hear a lot in digital preservation, and if you ever had any questions, they will be answered today. Just want to briefly go over a few points. A couple of you in the chat have been asking where you can get the PowerPoint for today, and we have put that up as a PDF handout on our course website. You'll also find links to the homework assignment, and Danielle will talk a little more about homework later today, but she's referenced a few links, and I've also put those hyperlinks up on the course page. If there are questions that we don't get to or there's more information we need to provide, you'll find that there. Links to helpful resources that all of our speakers have pulled together, so I think you will find that very handy. Just note that we will be emailing you the recording from today, not posting it on the site until the course is concluded. You are not required to work towards a certificate, but if you wish to, you will receive a paper certificate in a digital credential, which you should have registered, sent in a permission form, watch each webinar on the course, either live or through the recording, complete all of our five homework assignments. And thank you for all of you who have completed them so far. Danielle gave a little bit of feedback in her talk today. And just make sure all of that is complete by Monday, April 22nd. We also invite you to join the Connecting to Collections online community at www.connectingtocollections.org. If you haven't joined us, this is a great place to also ask your peers about their opinions. We've had a great little conversation going about large flatbed scanners, for example, this week. It's a great way to sort of get further conversation with your peers, and hopefully some experts will also chime in and help you with any questions you might have. And then any other questions about the course, feel free to contact us at our info at heritagepreservation.org or call us, and we'll try to get you what you need. And with that, I want to turn it over to Danielle Plummer. She is our course coordinator. So you've heard from her in our previous two webinars. She's been helping answer questions and being a great resource in the chat. And she's also planned all the speakers for this course and helped us coordinate our schedule. So she's doing a great job with that, but she's also today's future speaker. She is currently a consultant working with cultural heritage institutions who are interested in putting their collections online. From 2005 to 2011, she coordinated the Texas Heritage Online Program at the Texas State Library and Archives Commission. So she's very skilled at helping libraries, archives, museums of all types and sizes, and it really makes her an asset with all of you in trying to coordinate this course. She has a MS in information studies from the University of Texas at Austin and a PhD in medieval English literature from the University of California Davis. She currently serves on the American Library Association's Joint Committee on Archives, Libraries, and Museums. And this is a collaborative group that involves the Society of American Archivists, the American Library Association, and the American Alliance of Museums. It's a great joint group that helps us all kind of communicate on these interesting issues like we're talking about today. So with that, I'm going to close the hello box over here and I don't mean to cut anybody off, but we're going to move to our moderated chat, as Mike mentioned. You can keep asking any questions you might have, technical questions or course questions, and we'll get you an answer. And if you have any course subject matter questions, type those in and we will get to those during our breaks and at the end. So with that, I'm going to turn it over to you, Danielle. Well, hello to everybody. It looks like we are over 300 people who are on the call today and really hope that we can have a great conversation about some of the issues involving metadata for digital materials. I just want to remind everybody about the goals of this series. The first one is that you will have a better understanding of the inherent fragility of digital objects. The second is that you will acquire information to help you select preservation formats, metadata, and backup systems for your digital objects. And the third is that you will be able to identify one or more actions that you can take to improve your institution's digital preservation efforts. There are five sessions. They each have a different topic, but they do kind of tie together. So we love it when you come in to more than one. We started with an overview of digital preservation and then did a session on conversion and digitization. We're here today with metadata, finding aids, and asset management. Then tomorrow, we'll come back with a session on backups, copies, and what can go wrong. And finally, we'll wrap up with a session on partner to preserve digital preservation networks and collaboration. So overall, we're just about at the halfway mark. Stick with us, and we will get through it all. Today, I'm going to do just a bit of an introduction to metadata. In the first session, we actually had some people who were asking, what is metadata? We'll cover that today. We'll go over some types of metadata and talk about collection inventories and finding aids, descriptive metadata. And then we'll go into administrative metadata and talk quite a bit more about that because the focus here is on how to preserve digital content. The administrative metadata becomes much more important. And then finally, if we have time, I'll close with a bit on metadata quality. So the question about what is metadata? You can think of metadata as being anything you use to describe an object, a file, a thing. So metadata is everywhere. It's around us. And it's such a generic term. In some ways, it's useless. But we've tried to come to an understanding about how we can use metadata, and that helps us come up with a definition that is more practical for our purposes. So according to the National Information Standards Organization, who published a book in 2004 called Understanding Metadata, metadata is structured information that describes, explains, locates, or otherwise makes it easier to retrieve, use, or manage an information resource. And information resource can be just about anything you imagine. But the key concepts for our purposes are structured information, ease of use, and formal standards. And we're going to go over some of those standards today. I want to first address a metadata myth. There are lots of jokes about metadata. One that I learned when I first started library school was that cataloging, library cataloging was for women, and metadata was for men. And that's clearly not true. But one of the ones I've heard from history museums in particular is that they don't have metadata because objects don't have titles. And they think metadata is for art museums and libraries. To answer that, I turn to the Dublin Core. Dublin Core is one of the most ubiquitous schemas or systems of metadata out there. And there are 15 simple elements, and you don't need to know what they are right now, but they're basically the sorts of things you would use to describe a book or a web page. The system Past Perfect, which is used in a lot of history museums to track their collections, puts the information about the objects in structured fields. And they have a product called Past Perfect Online. And this is a screenshot of a collection or an object that is in Past Perfect Online from the Angel Estate University West Texas collection, is where I got it from. So as you can see, there are a number of pieces of structured information on this record. There's a catalog number, an object name, description, date, photographer, collection, et cetera. Well, if we compare those to Dublin Core, we'll see there's a pretty close match. So if we go through, catalog number becomes an identifier. Object name is really the format of it. Description, the residents of Dr. C. E. May's, that's just a textual description of what is shown in the photograph. Date, circa 1890, photographer becomes creator. Collection becomes source. Person becomes a subject. Subject becomes a subject. And image here becomes the title of the object. So, really, it's just about the names we use for things. You use metadata, but you may not ever have called it metadata. You may not think of it as metadata. But really, everybody uses it, and there are only a few basic types of metadata that most of us collect. One resource I want to point you to is the, again, this is from the National Information Standards Organization. They have a publication called A Framework of Guidance for Building Good Digital Collections, and this is in the third edition currently. It's available. I've put the link on the resources for this page, for this course. But one of their principles here is that a good object has associated metadata. And specifically, they say a good object will have descriptive and administrative metadata, and compound objects, objects that have more than one view, more than one image, books, for example, that have multiple pages are compound objects, will have structural metadata to document the relationships between the components of the object and ensure proper presentation and use of the components. So, we'll go through that a little bit more as we go through, but this is a good baseline for us to remember. Back to NYSA's understanding metadata, they broke metadata up into three basic categories, descriptive, structural, and administrative. And within administrative, there's technical metadata, rights management metadata, and preservation metadata. We're going to talk about each of these just a little bit today. Some of the functions of metadata, and you can probably think of more. The first is to discover resources. How do you find that something exists? Often, we use metadata. We look for an object by its title, by its creator, perhaps by subject access. We use metadata to manage documents. The file names that appear when you open your computer folders and look for what do you have on a particular subject are managed by their file names in large part, and that's metadata. We use metadata to control intellectual property rights. We use metadata to identify versions of different documents. We use metadata to certify that an object is authentic. That is, it's the same object that it was five days ago, or it's the same object that we actually got from the person who originally gave it to us. We use metadata to indicate a status in libraries. This might be like a book is checked out, or it might be in the conservation lab instead of in the exhibit. We use metadata mark content structure to situate objects in space and time, so geotagging or putting the geographic coordinates of things in metadata is becoming more and more common. We use metadata to describe processes at our institutions, and I'm sure you can think of a lot of other sorts of things you do with metadata as long as you remember, you know, metadata is pretty much anything you might ever do. So here's some examples about how to use metadata. You use metadata to describe, and in particular you use metadata to, and I say more or less, uniquely identify an item. And we use it to identify the parts of an item and their relationship to the whole. So in this case, I have a book that has in fact been digitized. It has an author, David, Davis LSA. It has a title, The Encyclopedia of Texas, compiled and edited by LSA Davis and Edward H. Grobe. And it has an imprint, Dallas, Texas Development Bureau 1922. And we need that because there were many editions of this book. In fact, some of them have multiple volumes, and so we need to be able to distinguish them, and all that information together helps us to do that. We use metadata for location to show you where to find the object. In this case, it's a book, and we're not talking about the digital object right at the moment. We're talking about the physical object. We include a call number. And in fact, the catalog includes a location. Willis Fourth Floor Texana Collection with an identifier, which is the call number, 976.4D292TB1. So that gives us a lot of information about where to look for it. We use metadata to document the condition of an item. So, and we use it sometimes to record the actions taken with respect to the item's condition. We don't always make that information public, but we do it or ought to do it for our unique collections. So in this case, binding is full leather. The book condition is fair, and the jacket condition, no jacket. So we just record basic information about the item. We use metadata to explain how to use an item, and this is tied to intellectual property. So sometimes we say you can use it because this is in the public domain. It was published in 1922. It's in the public domain, so there's no copyright on it whatsoever. Anyone can use it. We also use it to describe the physical object, and we say, in this case, library is only because it's a fragile item and it can't be taken home. Digital is different. We've been talking about a physical object with those previous examples, but digital does make things a little bit different. We need to decide does the metadata we're going to be using describe the physical item or the digital item, or are we going to try to describe both? In terms of the physical item, the metadata becomes a surrogate for the physical item. You manage the physical item by putting it on a shelf, by taking it to the conservation lab, putting it on exhibit, et cetera. But you manage the metadata in terms of to help you maintain your inventory and ensure that you have access to that physical item wherever it may physically be. We can aggregate that metadata for use in union catalogs, which is a common system of bringing information together. Next week, there's going to be a roll out of a new system for bringing stuff together called the Digital Public Library of America. Libraries have been doing this for a long time. It's a newer phenomenon in museums. When we talk about digital items, though, metadata isn't necessarily separate from the physical item. It really is a component of the digital object itself. It can be used as a way of pointing to the digital object when it's used in like a metadata aggregation, such as the Digital Public Library of America. But a lot of what we're going to do here today is talk about how we keep the metadata of an object together with the digital object, because the closer they are together, the harder it's going to be to lose the metadata over time. In digital objects, metadata can be in the digital object. The digital file headers, parts of the object you never see, record metadata. And so for formats like TIFF, image formats, there's a header that you can write metadata about the object into. You can give it its name. You can tell who the creator is, the creation date, all of that. And that's just invisibly to the person who's looking at the photo embedded in the header. Exit information, which you may have seen from your digital cameras, is another example of metadata in the digital object. If you do encoded archival description or use the text encoding initiative to mark up text, those have headers that record metadata about the digital object within the digital object itself. Broadcast wave, a format we talked about in the session on last Thursday, also has a file header in which you can record metadata. So there are places to do that in the object. It will never be separated from that object. You can store metadata near the digital object. So this is, for example, you could create a file in the same folder as the object that has the metadata. You could save it in an asset management system or a database or a spreadsheet on the same hard drive or the same network as the digital object. And that way, when you do your file paths to say where the object is, it's fairly easy to do. You don't have to really do a lot of specification because you just copy the link from your file explorer. But metadata can also be far from the digital object. It can be on another network. It can be in another state, in another country. And there we have a few extra issues that we need to resolve. I referred to the framework of guidance earlier. There are actually six metadata principles, I believe. Good metadata conforms to community standards in a way that is appropriate to the materials in the collection. The users of the collection and current and potential future uses of the collection. So we have to think a little bit about how do people actually use our materials. And the users may be people in your institution. They may be researchers. They may be visitors. They may be children. They may be adults. But you need to think about all of them and how they might want to access the material in question. Good metadata supports interoperability. You're almost guaranteed not to be using the same system, database, asset management system, whatever, in five years that you are today. Either you'll upgrade versions or you'll switch systems. So interoperability is one way to ensure that you can take your metadata from one system and put it in another one. Good metadata uses authority control and content standards to describe objects and co-locate relative related objects. Authority control just basically means you have a list of terms and you always try to use the same terms for the same things. So you're not just making up terms. You're not introducing typos, et cetera. Good metadata includes a clear statement of the conditions and terms of use for the digital object. This is not just copyright, but copyright is part of it. Good metadata supports the long-term curation and preservation of objects and collections, and that's one of the things we're really going to talk about today. And good metadata records are objects themselves and therefore should have the qualities of good objects, including authority, authenticity, or capability, persistence, and unique identification. And we're going to go through a lot of these elements today. So I'm going to pause and we're going to see if there are any questions at this point. Kristen, did you see anything come up in the chat box? Yes, we had a few questions. Actually, the group has been helping each other out quite a lot, which is great to see. And I think basically these were covered. But let me just quickly ask you some questions. Could metadata be considered keywords? And then the answer from various people in the group were that keywords are considered metadata. Can you just clarify? Absolutely. Keywords definitely are metadata. Sometimes this has been called tagging. If you use a system like Flickr, you add tags to it. And those are, again, just words that you use to associate some semantic concept with the item in question. Definitely that's metadata. When we start talking about archival and preservation metadata, when we start thinking, well, we maybe need to use some controlled vocabularies so that we're ensuring that we're not using terms that we don't know what the meaning is or that different people don't agree on a single meaning. So keywords are one form. They're not necessarily the best form. Okay. And you may have just answered this, but Robert Richard said that he saw that the definition of metadata is data about data. And he wondered if this definition is the same as archival. So I think what he means is archival metadata. Metadata literally means data about data or the data about the thing itself. And, again, going back to that first definition I gave you, we're focused on things like structured information and ease of use. Metadata that supports the sort of long-term curation of our digital objects. So that's the fundamental difference. Metadata can be anything, but some types of metadata are more useful to us for preservation than others. Okay. So Cassie said she had a lot of Word documents and she had not consciously created metadata for them. Does metadata exist and how do she access it or change it? Well, metadata may be in the document and you don't know it. So in Microsoft Word and most of the Microsoft Suite and in many types of office products, there are something called file properties. And that's where the metadata for the object is primarily stored. So in the newer versions of Word, there are actually ways to review all of those properties and even to clear them out if you don't want them. One of the most common problems is that people start with one document and they modify it extensively and then they decide to use that same template and make it a totally different document and then have to go back and look at the properties. And it records things like who the original creator was. So even if this is a document you got from someone else in your organization and you completely changed it, that creator might still be wrong. And so looking at the metadata properties, file properties is useful in those sorts of documents. There are some other tools to do it, but one by one that would be the way to do it. Okay. And I guess that's sort of the long lines of Maggie's question about how is metadata added to a header. I guess it would just depend on the program they were using. Yeah, and I'll talk more about that later for some types of objects. Okay. Dennis wondered what does non-embedded metadata look like? A text file in the same folder is a digital image file. One text file with metadata for all the files in that folder. That's really one of the ways that people are doing this now. This is something that every institution has to figure out for themselves how to manage this. But one of the ways is to have, if you have an image, to put a text file with the metadata, put those two together in one folder, and then have separate folders for every single digital object. That seems like a lot of folders than it is. But that keeps the metadata and the object together. If you have a book with a lot of different pages, you might have all of the images and one metadata file. But different people do it differently, and we'll talk about a few different approaches to not doing one metadata record for every single object. Okay, great. And then one last question before you go back. Is there recommended software for casual logging that best manages metadata? There is a lot of appreciation for you giving that past perfect example, but others use FileMaker. Well, any system, any asset management system we call them, sometimes people call them digital asset management systems because we have some same dams. But those all manage metadata, and I don't know that there's a best one, but this is something to remember. A lot of what you need to do with metadata in your institution will depend on the asset management system you use because they kind of force you to do things in certain ways. So you could certainly create a FileMaker database that has all of the metadata you want to track or access or MySQL or anything like that. There are lots of different systems. We're not really going to talk about individual systems too much today. And the right system for you depends a lot on the circumstances of your institution. And so that's where I usually say to people, contact me offline. And I'm always happy for any of you to send me an email and we can chat about what would be right for your institution. Okay, I think you can go ahead at this point and I'll just keep tracking questions as they're coming in. Okay, great. So I refer to the idea that we may not always be tracking each digital object separately. And one of the things I want you to start thinking about is the idea that you need to start tracking objects in a global aggregate sense first. There are a lot of digital objects out there and we don't really have enough time to go through and put detailed metadata on every single one of them. A lot of the time the best we can do is to take groups of items and create metadata about them. So Lauren talked about collection inventories. At its most fundamental level, a collection inventory is the list of items in your collection. Now that may be whether it's physical or digital, do you recommend doing the digital collection inventory separately because there may be different pieces of information you want to gather. It could be a Word document, an Excel file, or a database. So collections management software will let you do this sort of inventory. And it includes certain information about the items. So it will probably include a name. It might include a location. And in the digital objects, that's the path to the digital object. So it's the file folder structure in which it lives plus the object name. Or for a group of objects it would just be the folder. It often includes information updates and then notes about rights. Lauren, for Course 1 in this series, provided some information and some sample documents following the session. So that's a good thing to look at. This is her example, and I know you can't read it. So I went through and I made notes about what the fields are here. The first field is collection title. So she's not looking at individual objects. She's looking at whole collections of things. Collection title. The collection number, which is from their main record-keeping system, series information about it. So it may be that this is, some of her examples are things like VHS tapes, audio cassettes, research materials, in a public library it might include things like periodicals, biographies, et cetera. Description. So just a brief textual description, free text of what's in it. A folder description, if there is one, and she doesn't always have that, box and folder information about the collection. And then a note. Does it show an item level description, yes or no? So some of them they've gone through and they've recorded every single item in the collection, but a lot of them they have not. Information on content. So if there are particular notes, especially about condition that goes here, date, rights, and then physical check would just be a field you could use for inventorying the physical collection. This is another example, and this is from the American Museum of Natural History, and I just found it online. They've got different sorts of fields. So they have, and again it's hard to read, but you can go to their website and look at it, box number, which they subtitle as 099, local call number. And this is an example of a metadata schema. They're specifically tying their collection inventory to the mark schema, which is one that's used in libraries. Library catalogs are mark-based. So this would be a spreadsheet that they could then just plug into their library catalog. They've got photographer who's mapped to 1XS, 1XS, 1XX, I'm sorry, I can't say that today, creator with a note about authorized heading. And that would be, is that from the Library of Congress name authority list of headings for creators or not. 245 collection title, 245 subfield F or G collection date. They've got subject 520 summary notes, descriptions, 545, biographical historical note, extent 300 physical description, 340A physical medium note. So there are lots of different ways of collecting information about your collection. Decide which ones are important for you and start collecting them. But I do suggest that you not try to do every single item initially. Finding aids are another way to do access collections of information rather than individual objects by themselves. Finding aids are something that are used a lot in archives. And so it may be a term that isn't familiar to you if you come from a library or a museum background. And basically they are just a guide to a single collection. And the focus here is on including information about the collection and about the creator or creators of the collection. It usually includes information about access to and use of the collection. And it may include a list of contents. This is frequently organized as series, sub-series, box, folder, down-to-item level, although you don't have to include every single level. You might just include the series. You might just include the sub-series. It's partly what you need at any given time. This is an example of a finding aid. Again, it's pretty small. But the question is about Archon as an asset management system came up in one of our sessions. And I wanted to just point out that Archon allows you to create finding aids relatively simply. Simply in the fact that it makes the technology easier, it doesn't make the bit about figuring out what to put in the fields any easier. It includes a collection overview, scope and contents of the materials, biographical note, subject index terms, box and folder listings, et cetera. And the website archives.library.illinois.edu slash Archon, A-R-C-H-O-N has lots and lots of examples of how this works and how individual collections can be linked to individual digital objects if you go that route. Encoded archival description is the fundamental method of creating finding aids for interchange amongst different archives. A lot of archives traditionally have made their finding aids using Microsoft Word or other word processing programs. EAD is an XML schema, or sometimes it's a DTD, which is a very technical term we don't need to worry about, that encodes the parts of the finding aid in a specific way so that you can take them and you can share them between institutions. And everybody's kind of on the same page with this. There's still some quirks about it and it's hard to get them to go from one system to another, but it can be done. You can also convert your finding aid to HTML for online display relatively easily. So they can be produced in a variety of ways. You can use XML editors, such as metal or oxygen, both of which are somewhat expensive. You can use a tool like Archon or another one that's out there called the Archivist Toolkit, and those two projects are actually merging into a new product called Archive Space. Or you could even use Excel. For any tiny type, I've put a link to the Orbis Cascade Alliance, who has a tool for instructions online for how to use Excel to create finding aids in EAD. You can use library catalogs for finding aids, and this has been done for quite a long time. You can certainly create all of the descriptive metadata about a collection rather than about a single book in a library catalog. You can also include buildings and location information, item status, and more. You may be able to include links to electronic materials if they're on the web. You certainly can do this. And in some systems, you can include thumbnail images of the items. So increasingly, we're finding library catalogs are tied to systems for managing access privileges. So if there are some materials you have available to internal staff, but others that are available to anybody in the world, your system, your library catalog, can be used to help manage those differences because they track users as well as materials. That's an example of one way these things can be used. So with that, I wanted to just go and ask a poll. I see in the chat that some of you are talking about specific tools you use. So I wanted to find out just a little bit more about what's out there. So if you could respond. Kristen, I'll let you give them the instructions on the poll. Just check the applicable boxes. You can check as many as is applicable at your institution. And we just see the results coming in now. Lots of collections in Venturi. Some of you, the digital collections are not in your collections plan, so you don't have any for them, and that's fine. For those of you who are saying other or not applicable, I'd be a little curious if you want to put any of that information in the chat box. We won't be able to get to everything today, but it's always good to know. And there may be other people in the same position as you. I think we can move on. Great. We're going to take a break for questions at this point. Let's close the poll. And yes, so we can definitely ask some questions now. Okay, great. So do you see Cheryl's question on the right-hand side? She's getting obituary and saving them as PDFs, and just wondered her naming convention, surname, first name, maiden name, date of death, and put each in an alphabetical file. The first letter of the surname. Any red flags on that method? It actually works fairly well. One of the things I usually do recommend with names is to break them into parts. So surname and first name separate fields in your system, and this is because sometimes you want it to appear one way, sometimes you want it to appear another, and that makes it much easier to do it. You can sort by last name, first name. You can sort by first name. You can do all sorts of things that way. It's hard to guess what problems you'll have in the future. Right off hand, one that I think of is, how do you deal with people who change the form of their name? And historically, this has been relatively common. I don't know, if you got in early, you heard somebody ask about how to pronounce my last name, which is plumber, my husband's last name. And historically, the spelling of that has two M's in it. For some reason, his family switched it to one and that stuck. And so those sorts of things can make names particularly difficult. And there are some things that are emerging as ways to handle this, but, Cheryl, I think what you're doing now is fine. Okay. And then I've suggested also that the metadata could be included within the document as well. Absolutely. PDS has properties just like the Microsoft Office products. Yeah. There's another suggestion that file name can't do all the heavy lifting. It's good to a point, right? Right. And we'll talk more about file names in just a bit. Okay, great. And there was some conversation about the merger of those two programs. Yes. Archon and Archives Space. Right. Into Archives Space, sorry. Someone posted the link for the project timeline. Okay. They just recently announced some guidelines for how to be a number institution of Archives Space. But truly, these are open source products, so anyone will be able to download them and use them. As far as I know, there's no formal plans to continue supporting Archon and Archivist Toolkit. But because they're open source programs, if the community wants to take them over and continue to support them and make them viable into the future on their own, you can do that. That's the great thing about open source. Great. Melissa had a question. She said, we've talked about putting a mark record at the library across the street, which has collections directly related to ours. Would this work? Can it be done from an EAD or is it something entirely new in terms of the management of metadata? Was that a question which you need to start asking the library? Yeah, exactly. EAD maps fairly closely, and I say the term mapping a lot because, again, we use different terms for the same thing. So title and name, for example, I would say title maps to name. Or in the case of EAD, if you look at that example I gave you from the American Museum of Natural History where they had their field names and the mark numbers, they were basically creating an EAD finding aid in mark. And so it shows some of the mappings. You may have to break the pieces of the EAD up to get them into the library catalog. It's a copy-paste sort of thing for the most part. OK. And then Grace wanted to know, how does the adoption of RDA relate to EAD finding aid creation? That's actually a very good question. Kristen mentioned that I'm on the joint SAAALAM committee, and that's one of the things that we're encouraging the SAAEAD community to address and to specify whether they're going to adopt RDA or not. Right now, EAD, and the press community in general, is tied to ACR2. I'll talk about that in a little bit. And so that's an open question, I guess, is what I would say. OK. I think we can keep going at this point, and I will just keep watching for more questions. OK, great. And there are great things coming in the chat box, so thank you all very much for that. So I'm going to talk a little bit about descriptive metadata first, but that's not going to be the focus of this session overall. Descriptive metadata is by far the most standardized and well understood type of metadata out there. There are different descriptive metadata standards for different needs, different communities. The actual information you put in your descriptive metadata field should be developed according to your community's content standards. A content standard is basically a guide that describes the types of data that you want to record and sometimes it tells you where to find the information to record and how to format it. So the types of information that typically get covered in content standards are titles. The content standard may tell you, OK, we're going to take the title that appears on the first page of the book, the main title page. Don't use the version that's on the cover of the book. Don't use the cover that's on the spine of the book. The version on the title page is the title. Creator, same sort of thing. Edition or version. Publication. Identifier. Terms of availability. Those are the sorts of things that content standards tend to be occupied with. There are a few common ones and many, many, many local ones, but and I see someone is asking about RDA. This is where we cover that. AACR 2 is the Anglo-American cataloging rule. So we're a joint project between the American Library Association, British counterparts, Canada, others to create a content standard for traditional library cataloging. It has been closely tied to Mark, the system that's used in library catalogs. And it includes things like where to put your commas, how to capitalize terms. Those sorts of things. It's very, very detailed. Resource description and access is the new content standard that the American Library Association is promoting. And it's starting to get some adoption. The National Library of Medicine is going to adopt it. The Library of Congress will probably adopt it here fairly soon. It attempts to be independent of a particular syntax. And so it's not as closely tied to Mark as AACR 2 was, which presents a few challenges for libraries who want to use it. It is also much better at dealing with digital content than AACR 2 was. So that's one of the reasons we're interested in RDA. DAX, describing archives, a content standard, is the content standard supported by the Society of American Archivists. It relies on AACR 2 in many areas. And this is where we don't quite yet know to what extent the Society of American Archivists will recommend the archives switch over to RDA. But there will be some changes if that happens. So it's something to keep an eye out for. Cataloging cultural objects is a newer standard. It was developed by the visual arts and cultural heritage community. So you can think of it as the art museum standard. A number of people at the Getty Museum helped work on it. It actually is a very nice standard. It does really great things for digital objects. But again, you don't just go out and choose one of these. You use the one that is the content standard in your community. So try to decide what standard would be in use there and make use of it. Separate from descriptive metadata schemas, from content centers, this is the issue of descriptive metadata schemas. So descriptive metadata schemas are the fields in your particular asset management system and what they allow you to put in them. I mentioned Dublin Core before, and we're going to talk about that a little bit more. It is one of the most ubiquitous metadata schemas out there for digital asset management systems and that's supporting it. But there are others. IPTC Core, and I've forgotten what it stands for, but it is basically a press standard, newspaper photographers and others who have to share their photographs. So it includes information relevant to digital photographs, including a lot of copyright information. Mark used in library catalogs. Mods is an XML schema developed mostly by the Library of Congress that basically takes the same information in Mark but tries to make it a little bit more interchangeable outside of library catalogs. PB Core is a type of metadata schema that's used for audio-visual assets. And all of these, we're not going to go into depth. So if you can find more examples of those sorts of metadata schemas out there, we'll talk a little bit more about Dublin Core today, though. I told you before that there are 15 elements in simple Dublin Core, and many of those elements are what we would call descriptive metadata and purely descriptive metadata. So the title of the object or the title of the collection, the creator or author, any contributors. So for this presentation, I would be the creator, but heritage preservation is definitely a contributor. And for the recording, we would say that learning times is a contributor because all of these things are part of the final product. Publisher, so who put it online in the case of the recording of the webinar? The date, description, what happened in there. Subjects are often tied to things like the Library of Congress subject headings system, but they could just be keywords. Coverage, what is the spatial or the temporal period that this covers and identified? I'm going to talk a lot about identifiers. So here I have another poll, and I'm curious to know what kind of descriptive metadata elements does your institution regularly collect? I'm especially interested in digital objects, of course, but if you have these sorts of things, other items in your collections, please tell us about those too. Okay, so most of you collect some form of identifier. You collect a title, creator or author, date, and description. And then we start seeing it fall off a bit. Not all of you collect subjects, and other sorts of metadata, and not all of you collect anything else. So that's good for me to know. Thank you. We're going to pause briefly for questions, and I'm going to have to start really speeding up here, I think. Okay, again, lots of people helping each other in the chat. A good question we had was, let's see, Sarah was specifically interested in metadata for digital video files. She's learning about MXF, but she's not sure we'll end up using that software. She wondered if there are better options, or is that just something you should handle offline? That's really an offline question. One I would suggest you look at is PB Core. It's the public broadcasting core that is very useful for video materials. Okay. And another great question from Wilmer and Baltimore. How are people writing descriptions to make them accessible to deaf or blind patrons? Very good. Accessibility, metadata is how we provide access to a lot of these materials for people who cannot see or cannot hear. And full descriptions are really good ways of providing that access. Some of us are legally obligated to do it, and for others it's just a good idea. Among other things, it makes sure things show up better in Google. So descriptions are always good, and the fuller the better. Okay. Connie wondered if you can use two standards. She's in a history museum, and she'd like to use DAX and PCO. You can. Typically institutions use one content standard, but there's also something called a local cataloging guide. Later on I'll refer to it as an application profile. But it's basically you create the set of rules that you use, and you may say, for these fields we're using DAX. For these fields we're going to use CCO because DAX doesn't talk about them, or we just like CCO better. The big important thing there is just to document what you do. And a couple of people asked about that nomenclature 3.0, and it looks like that's the draft. Nomenclature is a controlled vocabulary, and it is particularly useful for providing subject access to collections. So as a controlled vocabulary for objects and artifacts, it's really the only one out there. It's certainly the best one out there. But it is not a complete metadata system in and of itself. Okay. Kathy had a really good question. Do you aware of maps that would map very common museum cataloging terms with DevlinCore, not sort of subjects again? Yeah. After the webinar I'll provide some more resources. And if we don't get to your questions today, we'll try to answer them after the webinar as well. I'll try to provide some links to those crosswalks and some of the other things that I've referenced in this course today. Okay, great. I think we can go back to... Okay. ...which I'm not, and I'll keep looking for questions. Okay. I want to talk about structural metadata very, very briefly. I talked about compound objects and how structural metadata supports how those objects are put back together. So it really is intended to be used to help put them together in an online presentation for digital materials. In Jake's session on Thursday, we talked about PDF, and he said he made the point that PDF can be thought of as structural metadata, and that's absolutely true. And many asset management systems have their own proprietary ways of pulling all these pieces together. Metz is another one. It's an XML schema, and it's used in a lot of digital library systems. So basically, your system will probably help you take care of this, but most smaller institutions use PDF, and so PDF is going to be your structural metadata. This is an example of what it looks like in XML. We don't really have time to go into it much, but it's the metadata encoding and transmission schema, again supported by the Library of Congress. They do a lot of initiatives in this area. Administrative metadata is the information we want to help us take care of materials over the long term. So it's the information to help us manage the resource. It includes preservation metadata, which is information about the content of a file. Fixity, which I'll talk about in a bit, provenance, where it comes from, the context, what other things does it need to be read, or what does it go with in terms of collections. And it also records information about actions on an object. So this is the same for digital as it is for physical objects. When you do conservation treatments on an object, you generally record the information that it was sent for conservation treatment, and these are the things that happened in that treatment. In the digital world, we do the same thing. If you convert a file format from TIF to JPEG or JPEG 2000, you make a note of that in the metadata record. Technical metadata is the format and extent of the object, and it may be information about the object's creation that is recorded by the tool you use to create it. So digital cameras create exif files, metadata that's put into the file header of the photograph. Rights metadata for preservation is specifically looking at what permission do you have as an institution to do preservation on this particular item. And that's a little bit different from access, but you should do both. So if you should say, can the public see this? And even if the public can't see it, why are we doing things to it? It's your justification. You're get out of jail card. So Dublin Court also gives you some room for things that I consider to be administrative metadata, and sometimes they overlap. So date can be both a descriptive metadata element and an administrative metadata element, especially if you add qualifiers to it. So the date of the original item, the date of the digital item, the date that the item was converted from one format to another. Those are different uses, but each of these fields can be repeated multiple times in the Dublin Court schema, and what you put in it is up to your institution. The source of the item can be administrative. It's provenance information, and you got it from how you know it's authentic. A relation, what is its relationship to other materials in your collection? The identifiers, and then things like language type, format, and rights, and we'll talk more about each of these. So back to the framework of guidance for good digital collections again. I wanted to talk about identifiers because they're one of the most challenging aspects of digital objects. I know many institutions who spend months, months, if not years, trying to figure out what their file naming conventions will be and how they're going to use identifiers. So if you've been there, you're not the only ones. NISO says that a good object will be named with a persistent globally unique identifier that can be resolved to the current address of the object. We're going to talk about that a lot more. But good identifiers will at a minimum be locally unique so that resources within the digital collection or repository can be unambiguously distinguished from each other. Global uniqueness can then be achieved through the addition of a global unique prefix element, such as a code representing the organization. So unique identifiers make it a lot easier to distinguish one digital object from another one. If you have downloaded photos from a digital camera and seen the standard way that the digital camera is named object, you'll know what I mean here. So object PC500001 and PC5002 may have absolutely no relationship to each other. But the names are the same. I apologize for my dog there if you heard that barking in the background. The link between a digital object and its metadata has to be kept over the lifespan of the object. And persistent identifiers are essential for that task. So some ways to do identifiers. In archives and museums, you frequently have something called an accession number. The number that you put on an item when you agree to take it into your collection. And that's a great identifier. They're unique within your institution or they ought to be unique within your institution. They may only be unique within a collection when you have some other issues to worry about. But libraries don't typically do accession numbers. Call numbers are not at all the same thing. There may be a control number. Any time you use an asset management system or a database, the database tends to have an internal control number. And that's unique within the database. But if you use multiple databases to manage different parts of your collections, you might get some overlap. The UUID, universally unique identifier, is another way of doing it. And one of the things it uses is checksums. We're going to talk a lot more about checksums tomorrow. So I'm not going to get into that too much today. Or URIs, uniform resource identifiers. You've seen these most often as URLs, the addresses in your web browser. They resolve to locations on the Internet. Ideally, for your digital objects, they would be static and persistent. They're not going to change over time. They don't have session IDs or other information that means you can't share them with someone else and still expect them to work. But URI is a more general term. And it also includes things like uniform resource names, which are independent of locations. So an ISBN, the number on a book, is a formal uniform resource name. And there are others out there as well. There's some simple rules that I think everyone can benefit from for identifiers and specifically the file naming aspect of identifiers. Use only alphanumeric characters for both files and folders. You can use dashes. You can use underscores. But don't use any other special characters. From operating to operating system, it varies. And some of those characters are reserved for use by the operating system. And so you can cause future problems by introducing them in your object names. Whenever possible, use a three-character file extension. So here I said TIFF. You'll see elsewhere I've abbreviated it T-I-F-F. But for your file extension, you would just say T-I-F. And I put a link to some common file extensions and the programs used to open them. Finally, don't use spaces in file folder names for digital objects. Use dashes or underscores, because space is another one that doesn't always translate well from system to system. This is taken from the draft minimum digitization capture recommendations from the Association for Library Collections and Technical Services Preservation Administration group. So they should be coming out with a final version here fairly soon. And Jake mentioned this as well. Check sums again. We're going to talk about this a lot more tomorrow. So I'm not going to go too much. But it's basically an identifier that's computed algorithmically. That is really, you know, that your computer generates for you based on the contents of a digital file. And it has nothing to do with the name of the file whatsoever. The data fixity check allows you to take a snapshot of the contents of the digital file. And then later on, you might take another snapshot and compare the two of them to see is it changed. And this is how we verify that a file hasn't changed over time. In your resources, I've linked to a video that goes over some of this. This is one of the most important pieces of metadata you can collect for digital objects, truthfully, because it allows you to deal with them over time. And so tomorrow we'll go into this in quite a bit more detail. And then there's generally technical metadata. This is information about the digital object. And it's often created when the object is created, embedded in the object's header. I mentioned exif as an example of this. It's recorded by digital campus. So it records date and time. It's only accurate if you've already set this in your camera. It may include information about the make and model of the camera. Sometimes even the lens that's on the camera if it's an interchangeable lens camera. And other settings. So here I have a snapshot from Flickr, which lets you see the exif of metadata for objects in their system. This is a picture of the Library of Congress. I believe. And there is information. So this picture was shot with a Canon PowerShot camera. It was in 0.077 second exposure at f2.8 focal length, 5.8 millimeters, et cetera, et cetera. There's a lot of information in here in the exif. And that just is automatically bundled. And that's technical metadata that is managed in the object. You can pull it out of the object. There is a guide to technical metadata for digital still images. It's called a Data Dictionary. So it's just basically a list of terms with definitions and explanations of what they mean. You could implement this in your own asset management systems as custom fields and rely on them whenever you possible. And that makes it more likely that you'll be able to trade into your own asset management systems and more likely that you'll be able to trade information about your objects with other institutions or with other systems going into the future. Controlled vocabularies. Again, these are very useful when it comes to interoperability. And there are some places where they're really, highly, highly recommended. We recommend them always for subjects, for people's names, for all of that. Whenever you can find them, I talked about nomenclature as a type of controlled vocabulary. But there are four technical metadata fields where I really want you to consider using a controlled vocabulary. The first is format. The format of the digital object. We think of format often as being the file extension. So PDF files, PDF stands for Portable Document Format, have an extension of .pdf. But it's not always the same. That might be because someone accidentally or on purpose changes the file extension. If you've ever tried to open a document and your computer reported that it can't, it might be that the file extension was changed on you. But it might also be that there are generic file extensions and then more specific file formats. So multiple formats might all use the same file extension. Because the file extension is really just an instruction to the computer saying, use this program to open this type of file. So the list of formats we use is called MIME types sometimes, or media types, and there's a link here. Your homework for today will ask you to look at this list and use it. Another controlled vocabulary is language. And these are codes for languages around the world because increasingly we have multilingual materials in our collections. And so it's useful to record language and so you use a controlled vocabulary for that. Type is another piece of information that uses a controlled vocabulary. Doubling core, the same people who put out that 15 element set, also put out a vocabulary for types of materials. Again, you're going to see that more in your homework today. And date. Date is one of the hardest for all of us. We recommend using a defined standard and I've linked to one of them. This is the extended time and date format that the Library of Congress is developing. And the reason I recommend it is that it gives you the flexibility to include information such as circa and about and date ranges and all of that, which are very important to us in the cultural heritage community. We can't always pin a specific date and certainly not a date and time on our objects. So that's a standard that you can use and it may help you with your digital projects. Saving your technical metadata. Talks about saving it in your digital object headers and I'll show you a way to do it in Microsoft Windows in just a bit. I've linked to one that will let you do it in broadcast waves, but most image editing software, such as the products from Adobe or the Aputure product from Apple and many of the open source ones, will let you do it as well for static images. Not necessarily for AV, but at least for static images. You can take the information back out of your headers by using products like Jove or Jove 2 I've linked to the original Jove project or the New Zealand metadata extractor and those will look at your digital objects and then create XML files from the header information that you can then save with the object. There's a special type of XML information that is becoming more common, that's XMP. Adobe products use this a lot. There have been historic problems with XMP not being able to be re-imported back in to some of the image editing software though. This is a slightly problematic way of storing metadata. I much prefer that you store the metadata directly in the object in the header and then extract it for other types of uses. There are some guidelines from the federal digitization guidelines initiative. Minimum descriptive embedded data in digital still images in broadcast wave files and specifically in TIFF images. I've linked to these. You can get to them all through the website at digitizationguidelines.gov which is emerging as a tremendous resource for us in this field. This is a screenshot just showing you what the image properties window looks like in Microsoft Windows. If you right-click on an object it's a thing that says properties and it brings up the list of things you can add. You can add title, subject, rating, tags, comments, et cetera. It has information on authors, some of the technical metadata. And that's just built into your operating system. There are similar tools in Macintosh products and even in Linux types of operating systems. So don't be afraid to make use of those things that are just built into your system. Okay, so pausing again for questions. Anything that's specifically come up, Kristen? Yeah, actually some of them are kind of specific. So we're running short on time. So I might just ask one here. How do you embed technical metadata in derivative tag files? That was a question that was a little more specific. Yeah, so if you have your master file as a TIFF and you put all of the metadata in it, the theory is that when you create the derivative file if you use, say, an Adobe product to do that, that that metadata should transfer into the new file. Unfortunately, it doesn't always happen. And so it gets into some fairly specific and sometimes tricky things that you have to do to make that happen. It is the best practice, however. I'm just going to leave it at that for right now. Okay, and Robert had some specific questions, but I think we'll try to get that offline. And Cheryl also had a longer term project she was looking for some guidance on. So we'll handle that offline. Is it a quick answer to say if you're accepting born digital materials, should you recommend saving the file in a certain way before adding anything to that? That is a institution-specific question. The general best practice for digital preservation is to keep one copy of the item exactly the way it came to you and then keep other copies with all the changes that you make over time. That may not be practical for all institutions. It is, however, kind of the canonical way of doing this. So that's the short answer. Your mileage may vary. I think I just would like to go on and wrap this up, and we'll deal with a lot of your specific questions and follow-up responses later. Great, thanks. Okay. So preservation metadata. We've talked a little bit about provenance, authenticity, preservation activity. I wanted to mention technical environment. This is something that came up on Thursday where people were asking about technological obsolescence. What if you have a program that only runs in a particular version of a piece of software? Well, you need to record that information in your preservation metadata. If it's a program that only runs on Windows XP, well, you need to make a note of that as well. So that in the future, you can still get access to that content. And we've talked a little bit about rights management. I'll go over a little bit more. One resource I wanted to point you to, just as with the technical metadata, there's a data dictionary that has all the terms and their definitions. There is a dictionary for preservation metadata. It's available from the Library of Congress. It's version 2.0 currently, although they continually do maintenance on it. So this is a very useful resource. It doesn't tell you, okay, you use this specific field for this specific type of information. It talks about the types of information, and you can create custom fields in your own systems to do it. Premise does not cover descriptive metadata at all, and it only covers a limited amount of technical metadata. It only focuses on rights management insofar as you need it for preservation. So think of Premise as your framework of guidance, something that can help you create your own local implementation, your own application profile or cataloging manual that tells other people in your institution how to use specific fields in your system. Premise is sometimes talked about as if it is out-of-the-box preservation solution, but it's not. Again, it's just metadata, and it has to be implemented in particular systems to be of any use whatsoever. So I'm going to skip over some of the specific Premise entities. If you're interested, there are other webinars on Premise, or you can look at the dictionary itself, and all of these are explained. In terms of rights, I just wanted to make one minor note here that in terms of preservation, under U.S. copyright law, libraries and archives have certain rights that museums unfortunately do not. This was an oversight when the law was written. So even if a library does not own the copyright to a digital object, it can preserve that content over time. It can change it from version to version. It can keep it in its files. It can't necessarily provide access to it. It certainly can't put it online for the whole world to see. But there are things it can do. Unfortunately, right now, museums don't have that right. So if you're in a museum and you have material under copyright that you're taking in that you want to preserve over the long term, you may need to work on getting your deeds of gift or donor agreements to give you those rights. It's very important that you have the right to preserve the digital objects over the long term. I've linked to some copyright resources. This came up a lot in the first session. And I wanted to make sure that you had access to some of them. But this is one way to go. And one specific right schema I wanted to mention is Creative Commons, which allows creators to choose a license for their work. You're seeing this a lot more in digital objects, but people say it has a Creative Commons license. So it behooves you to know a little bit about what they are. And there are basically ways of saying I own the copyright on this, but I'm going to give you permissions to do something with it. There's a specific Creative Commons license called CC0, which means that you're putting the information in the public domain. You're no longer claiming ownership of it. The Digital Public Library of America is asking institutions to put a CC0 license on their metadata. And that doesn't necessarily mean that the objects are in the public domain, but they're just saying that the institution doesn't feel any particular ownership of its metadata. And since metadata is mostly about collecting facts, facts aren't copyrightable anyway. So if you see anything about CC0 and talk about the Digital Public Library of America, know that they're not asking you to surrender all of your copyrights to all of your digital objects. They're just asking you to put your metadata about the digital objects in the public domain. So I was going to basically wrap this all up with just a question about preservation metadata, but I don't think we have time, so we'll skip that. And just close with some final questions. Thanks, Danielle. There's some questions about bringing information out of objects and tracking them. So Maggie said she used Windows to add photos, but couldn't find that information in Adobe Bridge. Elisha said when you add the metadata in the properties of the digital file, is there a way to export it all onto Excel? There are some specific ways to do this. One of the things I mentioned was Jove and Jove2, that these are ways to extract metadata out of your files, and they allow you, it gives you an XML file. It takes a little bit of technical know-how, but you can then take that and import it back into your asset management system. And this allows you to get access to all of it. Adobe is very attached to the XMP stamp that I mentioned. And so they don't always support metadata produced in other systems. They want everyone to do it their way. And that's an unfortunate limitation I found with some of the Adobe products. But once the object is in there, it's in there. Just maybe you have to investigate a little bit more to get it back out. Okay. Was there a site Jack wondered with regard to Excel and metadata? There's a site for finding aids in Excel, so creating a spreadsheet that you can then export it out in an encoded archival description. I think that's maybe what you're referring to. And that was a site from the Orbis Cascade Alliance. The link is in the handout. It might not be something you can read. I'll make sure it's in the resources. Okay, great. Yeah, we will double check to make sure all the URLs mentioned by our audience and Danielle mentioned in her PowerPoint are pulled out and put on the website. So Natalie was getting congratulations for her questions. So other people have it too. She said she's both a museum and an archive. Is the museum site under the same umbrella as the archive? And we get mixed content. And how can she sort of maybe coordinate the metadata they're collecting for each side of their institution? This is a challenging thing. I don't know if in your institution the collections get split up and some of them go to the archives and some of them go to the museum. I've seen that happen. This is where at the time of accession when a collection is accepted, whether it's digital or physical, some sort of collection description and metadata record for the collection should be created. And then from there all of the individual objects or groups of objects should be tracked with that original information. Unfortunately, that doesn't always happen. But that would be my ideal, is that we keep that together. I may be missing the point a little bit on some of it, but ask in the chat box and I'll try to get back to it. And there's sort of, I think there's similar questions that Bryce and Julie had. Is there any kind of way to automate the process of embedding metadata within files? Bryce is working with Windows Explorer files and Julie was thinking of her some digital files. Does that have to be done file by file? I mean you open each individual item and type in. And Windows Explorer is typically done item by item. And again, I don't really recommend doing that. If you're working with images, the Adobe products and other image editing programs do allow you to do bulk actions. So you can apply metadata to a whole lot of files all at the same time. Now they all have to have the same metadata. So that means you have to group things carefully when you do this. There's also a lot of tools out there that help you extract that information back out. And that can be done on a file by file basis or on a bulk basis. Some of the tools I've seen, I don't know that I would recommend because they're not really well supported. But I'll try to put together a list of some more of these tools and include that in your resources. Okay. Just see if we can get a couple more questions in. But could you talk about the homework briefly? I'm going to put up a link. And as I mentioned, when you go to the website, the course page, you'll also see the four websites. The Daniel wants you to look at. And I hyperlinked them there as well. Yeah. So, homework, we're going to do something a little bit different. What I want you to do is to look at a particular digital object. And I've provided the link to it. It's a film online at the Internet Archive. You don't have to watch the whole movie. Actually, it's pretty sad if you're a pet lover as I am to watch the whole movie. But I want you to look at it and to look at the metadata that they have for that file. And to think about how you would use that for preservation. And so there's some specific questions there. I've also asked you to use the Dublin type vocabulary for format. And the media type, sorry, Dublin type for the type element and the media type vocabulary for the format element. It's one of the most common mistakes people make with their digital objects. And as you see, even I make it. So I just wanted you to see how those two controls like capabilities work. Great. And if you have any questions or concerns as you're working on the homework, you can do the best you can. But if you want to ask anything further, just infoadheritagepreservation.org. And we can get that to Danielle. We don't create the homework, but it really helps us get a sense of how much we're presenting is sticking and if there are other points we need to clarify in our next webinar. OK. So people are filling out the group attendance form. That's great. A few things coming in. We'll post the question and answers to the website after, but we've had a chance to look through them. The question and answer log for the first webinar is online. But we haven't put the one from Thursday up yet. And we'll try to get that to you as soon as we can. And then I'll get to you today. That's great. And I just want to remind everyone our next webinar is actually tomorrow. We don't usually do them day-to-day back-to-back, but it's tomorrow at 2 o'clock Eastern time. And we'll be talking about backup copies. So with that, if there's something we didn't get to today, feel free to put it again into the Q&A box so that it'll be in our log for Danielle to take a look at. But with that, I want to thank you for your time and attention today. And thank, Danielle, for all this great information. And thank you for helping each other out in the chat as well. We look forward to speaking again with you tomorrow. Thanks.