>> Narrator: Live from Las Vegas, it's theCube, covering NAB 2017, brought to you by HGST (lively music) >> Hey welcome back everybody, Jeff Frick here at theCube. We're here at NAB 2017 again with 100,000 of our friends. It's a crazy, busy conference. I think it's got three halls, two levels on each hall, more stuff than you could ever take in in four days, but we're going to do our best to give you a little bit of the inside, and we're going to go down a completely different path here with our next guest. We're really excited to have Linda Tadic on, she's the founder and CEO of Digital Bedrock. Linda, welcome. >> Thank you Jeff, happy to be here. >> Jeff: Absolutely. So for those that aren't familiar with your company, give us a little bit of an overview. >> Well what we do at Digital Bedrock is we provide the managed digital preservation services that are required to keep digital content alive. [Jeff] Okay, so managed digital preservation. [Linda] Yes. >> Okay, so what does that mean? >> Managed, meaning we do the work for you. You just have to give us the files and we take care of it, so you don't have to license software, you don't have to train people, you don't have purchase all the infrastructure, no big CAPEX, we just do the work for you with our staff and infrastructure. >> Jeff: Okay. >> Digital, meaning its all digital content. Any format, any kind of content, we don't care. And then preservation. And so what that means is keeping the content alive so it can be used in a hundred years. And that's not just storing it, because that means you have to know everything about how that file was created so that you can monitor obsolescence, because digital files will become obsolete over time. >> So it's a really different kind of spin because we're here in the HGST booth, and a lot of talk about storage or storage people all around us. But when you talk about archiving and preservation, how do you delineate that from just, it's a backup copy, I know I have a backup copy on a server someplace? >> Yeah, so the preservation part of it is it has to live somewhere. I mean the bits have to live on something, and so it can be spinning disk, it can be solid state, it can be tape, and so storing it is the easy part actually, but then the hard part is the managing it. So you want to make sure those bits are okay, that the bits are healthy, so you will be doing fixity checks over time, according to a schedule, and then you want to also make sure that the file formats themselves, so everybody's concerned about migrating the data onto other storage media in the future 'cause you just have to do that, end of life, you have to move things along, but it's those formats that can become obsolete over time, which means let's say you have a format, a specific format, which requires a software to render it, which requires an operating system for it to run, which requires a chip or a piece of hardware or a file system to run. So what you have to do is you have to monitor all those vulnerabilities in order to keep that format alive. So you have to either migrate it or you can emulate it, or use another software, or you can do nothing and just keep the bits alive until you can do something with it. >> So you'll do those things, so you'll, if there's a new file format that comes out next year to NAB that's the new preferred, the format, you'll take some of those assets you have in your protection, and go ahead and recreate them in whatever feels like a viable format going forward? >> Actually we don't do that. We don't do the transcoding work. What we do is we monitor it. We have a separate database that's tapped into our support database. It's called the Digital Object Obsolescence Database, or the DUDE is what we call it >> That's a good thing. >> So in the DUDE it's monitoring all those, what version of a software can be used to be able to render a file. So if something in our database suddenly is flagged as being, uh -oh, this is not, it's endangered now, because one of those vulnerability factors has now been deprecated, we'll notify the client and we'll say you have all these files you've given us to preserve that are now endangered. But we can't just do the media transcoding because you know that those digital objects also then have perhaps these underlying files that feed up into that object. If you change one of those subsidiary files, you can't then render that final object. And so you have to be very careful not to just suddenly flip something and change it. So we tell the client here's the files and here's all the relations between all the files, and here's what you can do to migrate it or to keep it alive. But we won't do that work for them because they probably can either do it themselves, they have to choose first of all what they want to do, or they might have a preferred vendor themselves who will do that work for them. >> Jeff: And the other piece you talk about a lot, in doing some research before we sat down, is the metadata, and how important the metadata is. There's a lot of conversation about metadata, especially in media entertainment because there's the asset itself that you need all this other information, so I wonder if you can give us kind of the 101 on metadata and why it's so important and maybe not necessarily just the 101, but something a little bit more advanced that people don't think about when they think about metada. >> Right. I would say that most of the folks here at the event, at NAB, they're thinking about metadata in two ways. One is the description, which is describing the content, so what is the nature of this content, what is it about, what's in it, do you want to search for a particular scene or a particular clip, and that's based on the content. They also may be thinking about technical metadata but technical metadata in the sense of interoperability with machines. And so you want to know that the software can work with this or with this system or whatever, and that's why this camera can then work with a certain system, and that's all because of the technical metadata behind the scenes. What they're not thinking about is the metadata that is required to keep that content alive. And that's all those obsolescence factors, and in order to monitor all that obsolescence as we do in the DUDE, is where you need to be able to validate a particular format. And you know immediately, yeah, this was shot with this camera, and it's a certain kind of raw format, it's this version of it, which can only be used in this particular system. >> A lot of complex variables that are moving very very quickly. >> A lot of metadata, yeah. >> I mean in the typical bit of technical metadata we extract off a file, we'll get over 400 bits of metadata and that's not even the descriptive metadata. >> 400 bits, 400 different classifications >> 400 different elements of metadata. And we just pull it off the file. >> Jeff: Wow. >> And if that's not complicated enough, we were talking a little bit before we turned the cameras on about virtual reality and a whole different way of really describing that experience. Probably experience is a better word than asset because there is no asset until you engage with what the software is feeding into your experience. >> It's kind of virtual metadata when you kind of think about it because it's like, so there's a code that creates the software for the virtual reality to all work, it's all required, but the actual experience that is what the human, the person who's using the software and how they're interacting with it, and so that metadata about your experience in the content is in your head. Unless you're recording it as you're going, your experience, and so then there's an output of it, but otherwise it's all in your head, in your experience. >> It's fascinating. The other piece we've heard a number of times here is, especially now with all the different content distribution methods, there's many many flavors of the same file. So are you keeping track of all the different variants as well? >> Yeah. And so in fact in the research for the DUDE, 'cause it's humans who are doing the research to add the data to the DUDE, they'll say okay, great, this one software works with all these different operating systems except for this one package that went out, so it's somewhere in the middle, so we can't even say this range from here to here, and we'll work with it, oh no, but there's always an exception in between. So it's very complicated. >> So it's complicated and expensive in a lot of versions, and storage is getting cheaper every day, but it's not free >> right >> and managing is not free, and so it begs a value question, and I'm sure you can bring up all kinds of sad tales of phenomenal assets that were lost in the past. But how are people thinking about the value of these assets so that they feel comfortable making the investment in this preservation and archiving. >> Yeah. Two different mindsets I think that people have to just start adjusting to. One is they're just creating so much data they need to start doing appraisal and retention policies on them. You can't save everything, you shouldn't have to save everything. So that means you should really in reality set those policies at the point of when you're shooting, when you're creating it, so that it's automated, so that it's not at the end of a huge project when you have a petabyte of data there. That's not the time to choose what you want to keep. You need to set that policy in advance and try to automate it. >> So are there best practices? What are some of the best practices? Or are there some reference points that people should kind of start from I guess? >> I think the bottom line that they should be thinking about is let's say that in a hundred years, so thinking about Paramount. Paramount just had it's 100-year anniversary. And they were able to go back to their original nitrates and digitization and they're showing films that were made a hundred years ago. So what about the content being created now? What if in a hundred years you want to be able to have your own one-hundred-year retrospective? What would you need in order to be able to render the file that you're creating now in order to show it then? So what elements do you need to keep in case you need to restore it or recreate it? So that's one thing you have to think about. >> That feels like it could be a complete rabbit hole though. >> It could be. >> So that's why you have to think about the bottom line, the hundred years. Now of course in a hundred years who knows, 'cause of all of this artificial intelligence and all of this automated capture, then there could be systems that will just recreate it for you. So you might, you know, I'll be out of business, as I call it, the virtual Linda. I'll be out of a gig in a hundred years. >> So this is a fascinating area. How did you get involved in this area? I started out as a creator, so I was a composer and a filmmaker way back when, but then I got into the archival community, the archival field. So I've been working in audiovisual, film video, auto and then digital. Really starting in 2000 all my work's been in digital format and doing that preservation because all of this content is important to me and whether it's your own personal home videos or images, of your kids when they were born, it's all digital or whatever, to a studio product a station, government documents, it doesn't really matter. If that content is important to you, it should be preserved, because it documents your personal history, it documents our cultural history, it documents governments who are going forward for evidence, for law enforcement, all of that if it has to be preserved you have to really focus on that and how to keep it alive. And it's all important, and that's why I got into it. >> And as you spoke, you're involved in some really interesting cultural heritage preservation, which is a completely different kind of value chain than a movie or my home video of the kids. I wonder if you can kind of talk us through that use case that you described earlier, 'cause this is a very different way to think about virtual reality, preservation, and digital assets. >> Yeah. So I also do some consulting work, and I'm working with this organization in Dunhuong, China, which is on the Western part of China, so that's out in the Gobi Desert, far out. So what this organization is in charge of are these caves that were created by Buddhist monks starting in three sixty four A.D. going up to around eleven hundred A.D. Hundreds of caves out in the desert, carved out of sandstone and the monks would then paint murals, and beautiful, incredible murals showing Buddhist culture, history, and the culture of the time. You can see how people lived, how they farmed, 'cause they have that representation on the murals. So the Dunhuong Academy, they came to me and they said they're doing digital capture of the caves, high-res capture of the murals, and they said Linda, these caves are fifteen-hundred years old. We know they will not be around in fifteen-hundred years, so these digital assets must be around in fifteen-hundred years, 'cause those will be the only representations of these caves that are there. So I'm helping them build a digital repository to keep those digital images alive. Because if they are, they consider them to be the embodiment of the caves. So I've seen some great examples of virtual reality implementations in the cultural heritage environment, again thinking about some of these critical places around us, in the world and the environment. They won't be around in fifteen-hundred years, either because humans have destroyed them, through the environment, or just natural deterioration and destruction. So what virtual reality can do is go out and capture those environments, capture those sites, so that we can experience them, or people can experience them when those sites are no longer around. If the humans are still around in fifteen-hundred years. >> Fascinating. And what a great application of virtual reality. >> Yes, absolutely. It's my favorite. And entertainment is fun, to pretend you're somewhere, but it's not just to go to a different site, go to a different place. >> I want to shift gears just a little bit. As you've done all this archiving and you look at these old movies, 'cause we're here at NAB and it's all about media entertainment, I'm curious if you have any kind of historical perspective of how the storytelling has changed over time. Is there a consistent thread that you see or just reflection as you've spent so much time with this historical archive footage, that you could share with the audience, that maybe will get them to go look at the ... that aren't opening this weekend at your local cineplex. >> Okay, so think about film. So film in the early days was basically just a representation of theater. Because that was the moving art form of the time. And so it was really static, just one camera standing there and people would act in front of the camera. And then of course that changed what with D.W. Griffith and others to mold the intercutting into the show and then things happening at the same time in different locations, that was really radical in 1912, 1913, just over a hundred years ago. And then you go into the golden age of cinema in the '30s and the spectacle, and so it's more, and so now we're in the age of virtual reality where instead of we're being told a story, it's more like we are part of the story and going through that. And we'll see how if people still want to go back and return to "tell me a story," just like when we were little kids we all wanted "tell me a story daddy and mommy," kind of thing so when we're in the theater maybe we want to be told that and just be engrossed in somebody else's story and relax our brains instead of feeling like gosh I just want to rest and relax, do I have to interact with this thing? >> Right. Do I have to work? I'd rather have somebody who's really good at it, like Quentin Tarantino, tell me his interpretation of this story. >> So I'm really curious to see, it's still new with virtual reality and augmented reality to see how it's going to really expand. And people ... it might just be a fad, I know people who don't want to hear that, but it has all these other great uses as a cultural heritage or in gaming and that kind of thing it's totally fun, but for narrative, sometimes you just want a story. >> Well Linda, you're doing great work, so we have to let you get back to the booth so that more people can take advantage and keep track, and I think the word that you used a number of times, keep these things alive for future consumption, not just in cold storage in a vault someplace. >> Yes, absolutely. >> Alright, well thanks again Linda for stopping by. >> Thank you. Thanks so much Jeff. >> Alright. Linda Tadic. I'm Jeff Frick. We're at NAB 2017, you're watching theCube, and we'll be back after this short break. Thanks for watching. (lively music)