George Blood: The Hidden Value of Media Archives

November 19, 2026 • 00:55:18
George Blood: The Hidden Value of Media Archives
Broadcast2Post by Key Code Media
George Blood: The Hidden Value of Media Archives

Nov 19 2026 | 00:55:18

/

Show Notes

Media archives hold more value than most organizations realize. In this episode, George Blood joins Broadcast2Post to discuss how digitization and preservation are opening new opportunities to make archived content searchable, accessible, and profitable. From recovering historic recordings and government archives to repurposing legacy media for YouTube, social media, and new audiences, we explore how organizations can turn their media libraries into valuable long-term assets.

View Full Transcript

Episode Transcript

[00:00:00] Speaker A: Foreign. [00:00:11] Speaker B: Organizations across entertainment, education, government, and media are sitting on decades of valuable content that's stored on aging tapes or film or discs or other legacy formats. And that content may include historic performances, lectures, interviews, sporting events, public meetings, or institutional history, or even material that could support an entirely new generation of programming. But before organizations can make that content searchable, accessible, and profitable, they first need to recover it from formats and playback equipment that are quickly disappearing. Today, we're joined by the team at georgeblood LP to discuss how organizations can turn physical media archives into accessible data digital assets, and how those assets can create new cultural, educational, and, of course, commercial value. Thanks so much for joining us today. Let's jump right into the questions because we have a lot of stuff to cover. So for people who are unfamiliar with George Blood, which boggles my mind, can you introduce the company and explain the types of archive preservation and digitization projects that your team handles? [00:01:18] Speaker C: We work with a wide variety of institutions. Historically, a lot of the work that we've done was with glam government, libraries, archives and museums, everybody from the Juilliard School to the Library of Congress. We also have a lot of Fortune 500 companies. We work with IBM and Pfizer, McKinsey, network broadcasters from WGBH to Fox, host houses on both coasts and in between. And we've been seeing a lot of content for productions coming through from Netflix and Audible and Ken Burns. [00:01:55] Speaker B: That's a really wide swath there. So who typically is contacting you first? Is it the archivist, the media department? Or it maybe an executive team? Or is it someone who just started and stumbled into a room full of old media? [00:02:11] Speaker C: It's really all the above. You know, it could be a consultant or someone working in the C suite who's, you know, trying to do the entire archives. Could be a producer working on a documentary who has very specific needs, an archivist with a patron request for just a few items or, you know, somebody working on our 50th anniversary is coming up. This is how we started working with IBM, was for their hundredth anniversary. We pulled a whole bunch of stuff out of the closets and was sent to US for the 50th anniversary of Apollo, the landing on the moon. Just, you know, somebody wins a Nobel Prize. We recently did some analog instrumentation tapes that were early experiments in trying to discover or demonstrate gravitational waves. Or you end up with somebody, you know, in it who's staring at 15,000 LT05 tapes and realizes, I think we need to move these forward to a more recent generation or investigate the cloud. So really could be anybody. Curiously, because we try to collect this information when people call is how did you find us? And we're hearing more and more, just like I asked ChatGPT. [00:03:25] Speaker B: Of course there's the formats that I think anyone from this generation and a few back of course we get right. But it's the Betacam, vhs, U Matic, dvd, Blu Ray. But what are some of the more unusual legacy formats? You have already mentioned a few, but I'd love to hear some of the more oddball ones that you have. [00:03:42] Speaker C: Oh well, we could do a whole podcast on those. The bulk of what we're doing is, is what you would expect. So a lot of audio cassettes and quarter inch reels, a lot of VHS and a lot of U Matic. We support over 210 AV formats and about 50 data formats. We have two rooms that are set up just to handle audio multi tracks from quarter inch, eight track, half inch, three track, eight tracks, even 48 track digital tape machines. We do a lot of work migrating, logging tapes and it's whether you know, from back in the day, this call may be recorded for quality assurance or more likely cold cases for 911. So we have these machines that'll that are running at 15, 30 seconds, a quarter of the speed of audio cassettes. Single Tenant Trail holds 25 hours. We can do 10, 20, 30, 40 or 60 track formats. We do every format of of 2 inch quad. And you know, we have a couple thousand machines so we really run a whole lot range of things. We did some work recently for the National Archives. They came in with these tiny little recording discs and it was obviously analog and they were brown, but we looked at them and they had spiral on them, but they were flat, there was no modulation. But because they were brown, they were impregnated with iron oxide. So instead of having a stylus, had a magnetic head. And so it sounded like a really bad tape recording. And they folded in the middle like a modern phone so you could put them in an envelope. So that was a fun thing that we did recently. We did some wire recordings. This is a small format speaking. It's like the about the size of the new folding iPhone. That was a recording of Fidel Castro on stainless steel wire. And we see about three or four new formats every year of all these really strange and exotic things. We had some tape come through. Sony had prepared, you think of it like a 0.9 version of HD before the standard was set. Prepared an entire production chain, cameras, switchers, recorders, runs at 1035i, there was a half inch format in the cartridge and a reel to reel format in one inch. They didn't have any working machines so they brought half a dozen reels for us. It was a live concert in Tokyo in 1996 and it looked spectacular. It's where every engineer from Sony was there. [00:06:38] Speaker B: I'd like to hear from both of you as to what does it take to maintain enough functioning machines. I mean obviously you'd have a graveyard of devices you can cannibalize to repair devices. But are you sometimes building players for this? You mentioned the wire transmission earlier and does a reproduction device exist or is it something you have to engineer? [00:07:02] Speaker C: We have had to engineer solutions occasionally. Usually we were able to modify a machine that's close. We have a very large collection of machines both working and for parts. I'm an avid collector of parts and tooling and we'll go into a facility that's shutting down and buy the by the shop and they're usually ecstatic because it's all junk or it's like who wants that? But that's how, how we, we, we gather parts. But we have gotten to the point. We have a high precision machine shop as well. Milling machine, the high precision lathe. We have four people on staff who do the maintenance work. They are now all on Medicare. Seriously. We have over a thousand linear feet of service manuals just so we have access to this. Stefan, you walk around, what do you see what's going on here keeping our machines running? [00:08:03] Speaker D: I mean it is non stop. You hear George describe all of that and you start to understand just how much infrastructure sits behind what sounds like a very simple idea. Right? Let's digitize our archive. George mentioned that I tag along on every tour that I can, that I can join. And until you've done that, it's hard to grasp just the breadth of machines that are being maintained, maintained and in operation every single day. George came up with a fantastic. We had a project digitizing 78s, right? And in a single pass you're limited to, you know, one tone arm and the scribe. And George said, well, how can we develop something that we can do more passes or more recordings in a single pass? And George, what are the numbers on that device that you designed and engineered? [00:08:59] Speaker C: So 78s aren't standardized for speed, equalization or stylus size. And you can change the speed after the fact. You can change the equalization after the fact, but the physical interaction of the stylus and the groove is unique to each Stylus size and shape. And historically while transferring them, you'd put one size in, listen to it, put another size in, listen to it, and go back and forth until you find the one that sounds best, whatever that means. And we a large project for the Library of Congress to do 10,000 disks. And I wanted to find a way to be more efficient, more competitive about it. We engineered a turntable that has four tone arms on it, each with a different size, size and shape of stylus so we can compare them very quickly. And we have a project digitizing a large number of 78 RPM discs where they decided instead of the engineer deciding what was the best one, we keep them all. So we record left and right groove wall, four different stylus sizes with and without EQ. We're doing 16 different versions in a single pass and we do more than a hundred sides per shift. [00:10:20] Speaker B: That is unbelievable. And you mentioned that you're doing it without any equalization. So the RIA curve, you don't have to introduce that. You can then tweak that later and you're so you're getting [00:10:30] Speaker C: oh, we do it both with and without. [00:10:33] Speaker B: That's unreal. [00:10:33] Speaker C: Before. And these are 78s. This is before the RIAA curve where things are in standardized. And so the phono pre's in that era frequently used established curves. But you have adjustments for the, for the treble and bass, which you do separately for, for each record. [00:10:51] Speaker B: I, I think you're right. We could have a, another episode of this podcast where we just talk about different formats and, and, but we need to get down to some, some, some business here. So we deal here with, here at Kiko Media, with organizations that, that often think about digitization is, is kind of a preservation expense. How should they begin thinking about their archives as assets that, that they could generate new revenue with maybe aid in audience engagement or just institutional value. [00:11:22] Speaker C: This is, I think, something that could be applied more broadly. Generally, the archives of, you know, paper or AV are an expense until an anniversary comes along or somebody retires or there's some, somebody wins a Nobel, whatever that is. And then of course, you know, it's very time sensitive. And what do you mean it's not digitized? Was what you give us an expense. But there are lots of use cases that I think are being overlooked. You know, we've been digitizing since 1990, working with a lot of libraries and archives and museums. And if we hadn't been doing it for that long when the pandemic hit, every educational institution went remote and A lot of their collections were available and the classes kept going. So that investment that had been made over the course of 30 years suddenly allowed them to continue to hold classes and frankly stay afloat by being able to charge tuition. So even in the, in the glam world, there are use cases for that. What we're seeing a lot of interest in right now are collections that are large enough or specific enough to train LLMs. Or working on a large project for a university known for its science programs. And had dinner with the person who does the fundraising they're looking for. How can we do something more with this? And whether they could take all of this classroom footage or. And train an LLM or better use it for the feedback loop in LLM. So the LLM is spitting out answers and you have this body that you can go and check whether the answers are correct and then feed that back into the LLM to improve its quality. And this is a project of 17,000 tapes, about 30,000 hours of content. And they said we have more. So that's an opportunity that they're exploring. We have another opportunity. It's 35,000 recordings of interviews with sports figures. Again, really high quality content that can be used to inform the training of LLMs. So this is an area we've been seeing a lot of traction in lately. Remember when we talked about how many cable television channels there could be as we've hit streaming, a lot of back catalog is coming online. We digitized the entire run of the McNeil there hours, the entire run of Masterpiece Theater, entire runs of television, 50 Years of Arthur so that this content that becomes available for new audiences. [00:14:25] Speaker B: You bring up a really good point and we've discussed this at Keycode quite a bit, which is there will come a point in time where there will be a cottage industry for this specialty data that can be used to train LLMs. LLMs obviously are great at a broad set, but when it gets into some of the nuances, it doesn't have enough. The corpus of information it has isn't enough to get those minute ANSW answers and, and by your clients and our clients having this unique set of media that may not have been utilized elsewhere for training, that becomes a great licensing opportunity. And that brings up some other questions. If it's not just licensing, where does the value also come from? Is it things like subscriptions? You mentioned streaming documentaries or documentaries, Pardon me, education, fundraising, Are there, are there something we've missed? [00:15:16] Speaker C: We have do a lot of work with documentarians, you know, and whether it's research Footage or footage that goes into the, into the materials. We've. We've literally had, you know, because I come from a post background and there's always these deadlines. We've literally had people wait in the parking lot while. While we digitized something. The. There's a lot of awareness about the long tail, about this niche footage that there's audiences out there for. And it's also nostalgia. We. We have a number of. Of artists that, you know, we're under NDA, but who come in who are just looking to stay engaged with their. With their audience. You know, it's so hard for artists to break through today. There's so much noise. But we have all of this, you know, we'll call it historically curated content. It's like, I'm sure that, you know, there was lots of dreck on the radio in the 50s and 60s, but what has survived the younger audiences, the new people can come and hear this content and how easy it is to access that. So there's engagement with existing fans as well as new fans looking for just great experiences. If it sounds good, it is good. [00:16:54] Speaker B: So over the years, I should say in recent history, the concept of detailed cataloging has actually become more of a topic. It used to be a wax pencil or it became a card that was taped to the outside of a tape case. Right. Or a Sharpie on the spine of a tape. So now that we're looking to reuse these assets, how important is cataloging and metadata that makes this monetization possible, and what kind of methods are you using to aid in that when you are restoring some of this content? [00:17:27] Speaker C: This has created a great deal of professional angst in our friends who work in libraries and archives and museums, where there is a strong professional background and rules and training about how to do really good cataloging so that things can be found that, you know that this name is always spelled correctly and the dates that this person lived, all of this cataloging metadata and authority control. And now it's just something that AI does particularly well. And that it's not that you're doing a free text search, but you can do a free form inquiry that it'll go and look at the data in whatever form it's in. So it's not like it needs to be Z3950 searchable, or that it's RDA or EAD, whether it's a Library of Congress standard or encoded archival description that AI benefits from that organization but doesn't require it. So the meticulousness, the Level of professional care nuance that goes into creating those very detailed, meticulous records. It's falling down the value curve. We have a system where you can put a tape on the glass and there are six cameras, take a picture of it, transcribe all the handwriting or whatever text on it, and just essentially does a text dump into a file that we could then ask some AI to do a preliminary organization of or just leave it that way. You can just put a whole catalog through, taking pictures of the box, and faster than somebody could walk down the hall or walk down the aisle looking at the shelving can find all of the information that's on that box and do searches on it. [00:19:38] Speaker B: You mentioned records earlier and I know recently you've worked on some projects with A Grateful Debt, including the concert recordings and decades of Farm Aid performances as well. If we shift towards music a little bit and creatives, what can music labels, artists, maybe sports organizations, broadcasters and other content owners learn from projects like these that are bringing older material back to market? [00:20:03] Speaker C: The Grateful Dead project is, I think, a really interesting example. It is a well established fan base. What we're digitizing are the originals from which many of those copies were struck. Two extraordinarily high levels. You have your 96K recordings and 192. These are being done at 384. And the nugs.net subscription has been a real success, not just for Nugs, but also for Rhino and Winery that owns the content. This ongoing engagement and the ease with which the fans, new and old, can access that content. And this is a really obvious, of course, the Grateful Dead, they're really big, but you could just go down any of the bands that we could rattle off and there'd be plenty of content to put into that kind of environment because you don't have the overhead of, you know, mastering a CD and pressing that disc, holding the inventory, all the graphics, the. The barrier to entry is much lower and you could make, you know, you know, best of album. Remember how novel those were? You can make the whole freaking catalog available. So there's still the capital up front, there's still the risk, but that, that is so much lower. And I think this extends into both academic collections. Who can do this collection of oral histories or lectures for which there may be a small audience, but you can then make it all really available and do pretty good description of it using AI tools. And whether that's full on transcription or just summaries makes the material much more accessible. [00:22:00] Speaker B: I'm curious how you decide whether a Recording needs straightforward preservation, maybe a more extensive restoration or maybe a specialized process before it can be released. I mean, obviously we could do some light cleanup. But then there's also folks who say, well, you know, we want to hear that era. We want to hear that this was live and it's not pristine. Sometimes you want to hear the pops and clicks in the vinyl. So is it seasoned to taste? Is it a judgment call? How are you making these decisions? [00:22:29] Speaker C: At its root, we're making digital surrogates of the original as it exists. We want that to be the highest quality, warts and all, and that we hand that off to the client to decide what to do about it. But there are different degrees of that. What does it take to get the best quality? So we talked about 78s and choosing the right stylus size. We want to be sure that when we're digitizing and whether it's audio or video or film, that we're within the legal range, that we're getting the maximum color range, the maximum dynamic range, the maximum frequency range, that we really are being sure that all the information that's on that carrier is sent forward. In film, you're doing a lot of color balancing for the fading channels. In audio, we generally aren't mucking around with the EQ because that doesn't really change in the way that the color channels fade. In film, analog, particularly SD videos, notorious for interoperability, interchange is always a problem, so there's always a tweaking going on there. But by and large, the enhancement that we're doing either becomes another project that's handled in our production studios, or we do some light enhancement for listenability or intelligibility for general access copies. But by and large, our role, particularly with regards to what we're discussing today, is in making the best possible playback. So we're relieved of that. And as someone who worked in post for many years, a lot of those decisions are best made with the creative, whether that's a producer or the artists sitting right there, because so much of it is subject to taste and you could do make a really lovely job. And you know, I know the number of times when I was mastering a recording and came back the next day and wondered what the hell I was thinking, or I forgot to save that and so you have to do something over again and it sounds great and then you find the original one. It's just like they're both good. So anything with a subjective component usually gets handed off to somebody else, unless that becomes another project. [00:25:01] Speaker B: It doesn't sound better? It sounds different. [00:25:04] Speaker C: Yep. [00:25:06] Speaker B: Let's talk a little about streaming platforms. And Stefan, I'd love to hear any input you may have. How have streaming platforms changed the business case case for digitizing large entertainment archives? Is this just hunger the need for content? [00:25:21] Speaker D: I think that, you know, it has. Streaming platforms have changed the, the delivery significantly, right? So distribution has changed. The material might not have had an obvious outlet 30 or 40 years ago, suddenly has dozens of potential outlets now. So that gives archives a relevance they simply couldn't have had when much of this material was created. So simply by providing access to this content, which may live on, on a shelf or on a reel that hasn't been digitized or transferred yet, your audience is much larger than it was originally intended for. If there was an intent. I think that as time has gone by, there are more and more use cases for the material. So you might have been delivering a program at the start or recording at the start. And now it can be used in a multitude of different ways. In a TikTok soundtrack, in an ad, in storytelling of a company's history, there's a multitude of options throughout, through the streaming avenues. Now, [00:26:35] Speaker C: you know, just a moment ago we were talking about what we're doing for post. It's like if we have a standard def video, okay, something really great off a 2 inch quad or a 1 inch type C, what are we going to do with it? Like I said, there's one use case to up res that to hd, which is widescreen, and then figure out what to do with the, with the pillar boxes and do we like AI that in or do we crop it or whatever. And then you've got another audience where it's this way, you know, the vertical formats that are so popular. So there are just so many different ways to repackage the material. And you know, there's a lot of, you know, footage you go through and there'll be like one gem, but the, the outlets now provide for that. [00:27:22] Speaker D: And it's not only the audience, right, but it's those different departments, right? So on a typical day, we digitize about a pallet of media that's 30 terabytes of data that doesn't fill a single LTO10 tape, right? It's the massive cost savings that George was talking about that always gets my attention because it makes me think of the economics of it, all right? A pallet of physical media becomes something that you can hold in your hand, albeit one box at a time, right? So even before you start thinking about like Licensing or the new uses for the content. There's the very practical question of what does it cost an organization every year simply to hold all of that physical media? Going back to like, thinking about digitization gets categorized as preservation, almost like insurance, if you will. Once material has been made accessible, it starts behaving much more like an asset. Marketing can use it, production can use it, researchers can find it, you could license it. Maybe it's used for new programming. There are use cases that nobody even anticipated, you know, when this stuff was made. One of, one of the, you know, the, the, the reasons that projects like this install is organizational, right? Storage falls under facilities or records management. But the value of that digitization shows up elsewhere, like George was saying, in marketing or in production and other parts of the organization, other departments. And nobody necessarily owns both sides of that equation. So at the most basic level, content that you can't access is very difficult to create value from. [00:28:58] Speaker B: We could spend all day on entertainment, but I do want to branch off into some, some other territory. And you had mentioned earlier that you work with Library of Congress, which is a client of Key Code Medias as well, but you're also working with colleges, libraries, museums, government organizations, and of course cultural institutions. What are those organizations discovering inside their archives and how are they creating value from material that may have been obviously inaccessible for decades? [00:29:25] Speaker C: I think the example that I given earlier about, you know, just doing general purpose reformatting, working through the collections, things that are requested frequently over the course of 30 years, provided a very large corpus that they were able to call upon when the pandemic hit. The large institutions, MIT Open Learning, repackaging a lot of their classroom footage into online courses, whether they're free or paid or for some nominal fee, and a way to maintain that engagement with the alumni. Auburn University had us digitize every motion picture film that had a game that Bo Jackson played. And they digitized these all at 4k. And it's been something of a bonanza for them that they're able to say they're taking care of their archives, that they have this famous alum and that the history is being supported. Won't you donate to the school? So that they're many different ways in which the archive can become a resource to be seen as part of the endowment, if you will, that then contributes to the core mission of the school and rethinking of that from an asset that we can monetize, which is not a vocabulary that they're comfortable with, but a resource that is part of the long Term value in the culture that has developed within the quadrangle that can then be transformed in a way to continue the educational missions of the institution. So these are some of the changes in the way that institutions are thinking about how they can access that material and make it useful. [00:31:34] Speaker B: I think that's a great thing to bring up because I think you hit the nail on the head, which is we're so conditioned to well, how can I monetize this as opposed and have instant revenue? Or how can this be used in, dare I say, more altruistic way? Right. And so if we translate that to kind of the government organizations, do you find that, that this, retrieve this retrieval of the archival materials is, is primary for. Is the primary goal for public access? Is it for historical accountability research or kind of all of those [00:32:11] Speaker C: huge project. On a crazy deadline we did, we digitized 26,000 audio cassettes for the Texas State Library and Archives Commission. And this was every bit of government work, as you could expect. It ranged from, oh my goodness, somebody was the low bidder on this brand of tape we'd never heard of and it was just dreadful. To the fact that normally when we digitize, say an audio cassette or a reel, we'll digitize the entire duration of the tape to be sure that there's not other content after it appears to end and then we'll trim it back. Well, in this case, because they were records of legislative events, the completeness of it was something that we needed to project into the future. So we didn't trim anything. So there wasn't something got cut off. All that completeness was was there by law they have to keep these records alive. We have a really interesting project now with the data tape. These are mostly IBM 34. Yeah. 3480 tapes, which is like way before DLT. Way before exabytes is really is like right after nine tracks first. One of the early cartridge based formats to recover that information to keep alive, you know, most of its prison records or mailing lists of donors that really play a role in the historical record that may set somebody free one day. And they are completely, absolutely, positively inaccessible on that cartridge until we pull the data off. And this particular example, they had been migrated to another format that we haven't been able to pull apart. We had to go back to the earlier formats. So there are lots of these really interesting cases that come out of government archives. We've been called upon to do some materials for the Warren Commission and jfk. Many years ago we digitized a live broadcast of The Boston Symphony from November 22, 1963, where in real time we're hearing the hometown, heard about, hear about the assassination of jfk, the memorial concert that was done by the Boston Symphony. These really important historical records that show up in government archives because somebody thought to hold onto them. [00:35:12] Speaker B: So when should an organization act? I mean in one camp we have organizations that say we digitized part of our collection five, ten years ago and they assume the work is complete. And then there's others who are postponing projects because the tapes appear to be okay. They're sitting on a shelf and it's somewhat temperature controlled. Right. So how should organizations evaluate the remaining timeline on their physical media and determine when action becomes urgent? [00:35:38] Speaker C: Looking at the tapes is looking at the wrong part of the system, right? So these are all information professionals called machine dependent records. And there's this entire life cycle we've spoken about. The metadata and the cataloging, the physical playback of the tape, the physical condition of the tape. All of these things are part of a system. And you can see either sitting on a shelf or on a spreadsheet for a budget line paying for storage, that there are tapes there and the tape is going to last a lot longer than there will be machines to play them. There's definitely some really serious issues with tape degradation and other formats shrinkage and embrittlement of motion picture film. So there's lots of what archivists call the inherent vice, the physicality of the object that's, that's just ailing. According to the natural progression of time and physics. There is now for the replacement of video heads. One supplier in the world where there is one person at a bench and one trainee to provide heads for the world. For audio there are two suppliers worldwide. A year ago there were three and a half. Literally the market is for that relatively small and the people who know how to do it are old. Father time always wins. And the best maker of audio heads, Greg Orton at Flux Magnetics, is not physically able to do it anymore. 84 years old. It just, it's this vision is up, isn't up to it. High precision work and it's a supplier we really miss. So we can build a machine shop and build some of the parts, other parts, no way, no how. We've had machines with really exotic bars, belts or unusual component values. Custom silicon and chips, unobtainable audio heads for digital multi tracks, unobtainable at any price. The latest generation of high definition digital video machines heads are not available at all. Last month we bought A facility out of the northeast that I really wasn't interested in. You know, it's. I, you know, if they went to auction, there might have been 10 machines that we would have bought. I'm trying not to buy, you know, more, more than a few models, you know, such as for maintenance. We ended up buying the entire operation, Packed it up, filled a tractor trailer, we had a 53 foot truck pull up to our building, packed to the gills, machines, parts, tooling spares. And I bought it for Head Life. It was the entire reason I found the funds to buy that facility is because I expect to live to see the day when formats go extinct. And so what should institutions be doing that have these libraries? I can tell you if you don't get to it, somebody else is going to get those head hours and you don't have enough money to recreate. Because we were trying. The suppliers are going away or the suppliers of the suppliers. So the one problem we're having right now is the ferrite that audio heads are made from. And video, rather, video heads are made from the cores that audio heads are made from. [00:39:37] Speaker B: For organizations that have just a massive, massive back catalog, you know, thousands or hundreds of thousands of items, do you have any rules of thumb on how to prioritize? Is it just what's older? Do that first, or is there some other logic baked in that you recommend for how to approach something that is this, this large? [00:39:58] Speaker C: There's some very specific form formats that have, that are shorter, lived, instantaneous recording discs, lacquer discs, mini discs, fail, motion picture film, that's not stored. Well, out of 210 formats, there's a dozen that would be physically highly endangered. And it was a really lively conversation at a conference, the association of Moving Image Archivists, where somebody got up and said, look, as the specialists who know these collections, if we don't know what's really important, we should be fired. So you can go into any archive anywhere, and whoever's sitting behind the desk catalog this. They know what the stuff is because when their friends come in, they give these really great tours of all the really sexy stuff that they. That's where you start. You know, the stuff that is most likely to have value, however you define that. So whether it's something you can monetize this or something that's historically valuable, or this is something that that researchers request frequently, you should be able to get through a quarter of the collection in the time it takes to pull it off the shelf. Beyond that, the harder you think about it, the More meetings, you have to try to decide. You're just driving up the cost and you think of everybody who's around the table or answering emails for an hour is costing you two or three takes. And we've often recommended just like start at the upper left hand corner and just start going even to decide whether or not something's a copy. So there's also, you can, you know, so we talked about the scanning machine that gets you a whole lot of, of descriptive metadata about the object and the text that's on it. You might be able to sort it, but the, the more time you spend trying to be the curate a collection, the, the clock is ticking against you. And even if you just like, you know, get a line item over the next five or 10 years, I think you should just go because you can't predict what people are going to want for all the time. We were doing really high production value materials. Who could have predicted TikTok? We did a project for the Museum of Glass for Corning and it was outtakes of commercials for the release of Corelware and they had this really great juggler who was juggling all these plates and of course part of it was that they would fall and not break. And it is hysterical, all these outtakes and how things would go off the rails. This is great marketing footage that we digitize a whole bunch of stuff and there's three weeks of, of social media feeds. So I think people should think creatively. Maybe that's the answer to the question is, you know, what's really important though, and then think creatively. If we had access to this, get a bunch of people around the table. Well, if I had this, I could do that. Since I'm always looking for content for Instagram. What kinds of things could you use and just spitball things? Because I can tell you in, in the May, May we all live long and prosper, but in our lifetimes, formats are going to fall like flies. We are simply not going to be able to digitize content. And in our lifetimes, [00:43:49] Speaker B: if it hasn't become already apparent, this is much more than just connecting an old tape deck to a computer and pressing record right at the scale you're doing it. It requires much more sophisticated production techniques and a data infrastructure. Can you walk us through a little bit what happens after a collection arrives at your facility? [00:44:07] Speaker C: Sure. And it's a really interesting mix of different skill sets and as the media moves through different stages in the life cycle of the digitization, the way you deal with things. So as things Come through the door. It's extremely high touch. The registrars are checking each object against the manifest. So we receive the manifest electronically, usually a spreadsheet. We ingest it into the project management software that we write in house and then they physically check in every object to be sure what it is. We want at the beginning of the funnel to reduce as much doubt about what this object is format, what it's called, that it's properly identified and we have a file name for it. From there it gets queued for production where it goes to our prep department. Motion picture film is physically touched from start to finish, looking for tears and broken sprocket holes, failed splices and then it gets cleaned. All videotape is cleaned. Audio tape is often cleaned. Anything polyester based is probably going to need to be baked. So we have this days of time that is spent preparing the object so that it is in its at the highest physical condition for the digitization goes off to somebody who is specializing in this formula or sorry, somebody specializing in this format. People who specialize in doing disks or handling cassettes or reels or quad video or obsolete format or oddball formats. The exotics we call them appropriate machine. Be sure the machine service the appropriate time based correctors, frame syncs A to D conversion. The files are universally digitized in uncompressed. So standard definition video is flying around at 100 gigs an hour. And we're typically running about 60 streams of either standard def or high def content uncompressed. When it's finished, the engineer checks the work. Yes, this is the content matches the record, top tails, everything appears stable, the file is complete. And then behind the scene there's a setup that was done calling for what flavor of checksum, what the folder nesting is going to be. What are the derivatives? Does this one get transcription or different sorts of metadata? This is all configured and automated behind the scenes and they click on a button. This goes off to a render farm where we render about 200 hours of video every day. And that's everything from a simple checksum to up resi to transcription that goes to another department for QC which runs in a pretty binary environment. Does the tape play? Does the derivatives match? Is the checksum correct? Does the metadata. All there, but along the way there are all these checkpoints. So when the tape arrives, this tape with this barcode on, it doesn't have the right file name or it doesn't have the right. The content doesn't match the barcode. When the engineer clicks on that button to send the file to Render. If it wasn't ingested according to the specifications of the project, the script will kick it back. So we have many control points to be sure that things are correct along the way. So QC isn't something that you do at the end. QC isn't about finding errors, it's confirming that all the work was done correct. [00:48:11] Speaker B: How are you delivering this? Is it something that you put in a cloud and here's a download link or is it here's an LTO tape or are you having some fun with it and giving them a FireWire drive? How are you normally giving them the that media? [00:48:25] Speaker C: We have some, some, some robots that burn CD ROMs. Most of the stuff leaves here on spinning disks on external hard drives, some on SSDs. Very little leaves as on LTO as the primary medium. But we'll make an additional copy usually to be split. [00:48:51] Speaker B: And can I ask why is that? Just because not everyone has an LTO reader and that version sitting around. Is that kind of the reason why LTO isn't kind of the go to for something like that? [00:49:01] Speaker C: Well, we're comfortable with all 10 generations of LTO. You know again key code supplied us with the, the libraries and the drives for our and the support infrastructure just for putting adding LTO 9 and 10 it's accessibility so that unless you're accessing that object regularly, you have no reason to spend 10 or 15 or $20,000 for a drive. But as an additional copy. So what we're often finding like the work we're doing for the American archive public broadcasting is once all the stations have approved the files, we then write the masters out to LTFs. So because changing a file on data tape is possible, but there's a lot of friction to that and the end users aren't usually familiar with that object and how it behaves. So for a lot of the institutions that we're working with, it becomes an additional copy that's off site on a, on a different format update to our Internet access and a firewall that will support that fire hose of data so that we can deliver either to an AWS bucket or even directly to somebody's cloud instance. We have a couple of clients where we're delivering to their on prem where they have either running S3 protocols Object storage within their facility and we get access to other institutions that have security issues, will pull from at will build an embargo space for them and they'll pull the data. Being able to have that bandwidth to support that kind of delivery is becoming more and More necessary. So the bulk of it's still on spinning disks, a little bit of ssd, but more and more data tape and to direct Internet delivery, where everything just lives. [00:50:58] Speaker B: So if we take a step back to the beginning, right, we just talked about delivery, but if we go back to the very beginning and someone has watched this and they're interested, but they're still kind of overwhelmed when they don't have a clear budget or an accurate idea of what their catalog is, or even a digitization strategy, what is kind of the first practical step that organizations in that space should take? [00:51:24] Speaker C: Count. So how many do you have of what? And don't get picky about it because there are so many other variables. How long is it, what the condition is, you know, what your storage strategy is, what other value added services you're looking for. You know, plus or minus 10% is the broadside of a barn. You know, open a box, count the tapes. It's, you know, whole, whole wall of count five boxes, average it, multiply the boxes close enough, I can get you a budget. And then we'll run down a bunch of questions. Tell me about the background of the collection. What are you trying to achieve? Who's your designated user Community File formats. Do you want to follow the federal agency's digital guideline initiatives, the best practices there, or is it more of a production environment? Figure that out. Figure out the storage. We can usually get you information for the, for the meeting that you have in an hour that somebody just asked you for. We can do that pretty quickly, just as a place to start. And then once you find out that it's going to be, oh well, 300,000 times, any number is going to be a big number versus oh, that's not as bad as I was thinking it was going to be. It's not $500 a tape. So quick counts. Background. What's your goal? And we can start asking those questions that will normally go back to the other stakeholders. And Stephen, we're talking about, we have this, what we're paying for storage that you take to the, you know, CFO and the, the budget people for handling facilities. And then what's on the tape that you take to the marketing and salespeople and who may be interested in accessing the content. So we can talk through all these use cases, put them in touch with either here's the best practices document or here are three other companies that we've worked with that have found their own solutions for this. Here's the person you need to talk to who's already worked through this question. So it's a lot of where do we start? Get you started with a budget and questions and then calling upon our network and whether it's who you need to talk to, who's already solved this problem, or oh, you're going to need some infrastructure, you know, call key code. [00:53:58] Speaker D: I think once you can start seeing the collection clearly, you can begin making rational decisions about it. And the biggest mistake may be simply assuming that you have to solve the whole problem before you can start. And that's just not the case. Take a take a pilot, take a chunk, start working on it, and the answers will present themselves before you. [00:54:20] Speaker B: George and Stefan, I could ask you questions for the rest of the day. This is fantastic. Thank you so much for being with us here today and answering both our creative but both process and technical questions as well. [00:54:34] Speaker C: Stop by for a tour someday. [00:54:37] Speaker B: I will be at NAB New York in the next month or so and may I have to hit up Walter and Rapp and take you up on that. [00:54:44] Speaker D: We'll see you there. Thanks for having us, Michael. [00:54:46] Speaker C: Thank you. Bye now. [00:54:48] Speaker B: Preserving an archive is not just about protecting the past. It's about making valuable content available to new audiences, new platforms, new research, and potentially new sources of revenue. Thank you so much to the team at georgeblood for helping us understand both the opportunity and the urgency surrounding archive digitization. Thanks for watching Broadcast to Post don't forget to follow Keycode Media on social and contact us about your projects at keycodemedia. Com.

Other Episodes

Episode

June 06, 2025 • 00:45:15
Episode Cover

Remote Post That Works: Reliable Workflows for Creative Teams

In this episode of Broadcast2Post, Key Code Media workflow experts Michael Kammes and Jeff Sengpiehl break down what it really takes to design, deploy,...

Listen

Episode

March 23, 2023 • 00:42:19
Episode Cover

Riedel Communications Discuss IP, 4K, and Cloud for Live Production

Riedel Communications Discuss IP, 4K, and Cloud for Live Production on #Broadcast2Post Podcast Interview + Live Audience Q&A https://www.keycodemedia.com/2023/03/17/riedel-communications-discuss-ip-4k-and-cloud-for-live-production/

Listen

Episode

January 24, 2025 • 00:35:56
Episode Cover

SMPTE Production Workflow Standards | The Vision for 2025

Join us on Broadcast2Post as we dive deep into the future of production standards with SMPTE President Renard Jenkins and upcoming SMPTE president Rich...

Listen