Henrik de Gyor: [0:00] This is Another DAM podcast about Digital Asset Management. I’m Henrik de Gyor. Today I’m speaking with Douglas Hegley. Douglas, how are you?
Douglas Hegley: [0:10] I’m good, thank you.
Henrik: [0:11] Douglas, how are you involved with Digital Asset Management?
Douglas: [0:14] Currently, I sit at the executive leadership level in a major fine art museum in the Twin Cities. I would be the ultimate decision maker. The Digital Asset Management systems would be operated underneath my responsibility.
Henrik: [0:27] Douglas, how does a fine art museum use Digital Asset Management?
Douglas: [0:31] What’s interesting, I think, what might be a misnomer for some people, the Digital Asset Management in an art museum is actually a business driver like it is in any business. Art museums have art objects. Those objects themselves have data records for them, and those are kept in a different system.
[0:47] But we do need a Digital Asset Management system for keeping photographs of those objects, and often there will be many of those. Various angles, raking lights. Sometimes x‑ray, other spectrometer those kinds of things, as well as images of people and parties and the history of the institution. It goes on and on and on.
[1:06] I would say at this point that museums are still sticking mostly with still images in terms of Digital Asset Management. We haven’t fully embraced media asset management. We’re producing videos and that production is accelerating. I don’t think we’ve really faced some of the struggles we’re going to have, similar to the ones we had with digital photography 5 or 10 years ago.
Henrik: [1:27] What are the biggest challenges and successes with Digital Asset Management?
Douglas: [1:31] There are many. One of the biggest challenges for us as an industry is that our metadata models are not mature. There are many different standards for the way that you would record what is in that picture. The built‑in metadata is easy enough ‑‑ date and file sizes and everything else.
[1:53] For us, since it’s often object centered photography so we’ve taken that three‑dimensional sculpture, we’ve taken it to the photo studio, lighting it, shooting it.
[2:02] How do you attach that asset to the record that’s in a different system that describes that object? We struggle with moving data back and forth, mirroring data, coming up with better methods of attaching the digital assets themselves to all of the other kinds of content that we have about an art object.
[2:22] Then, I think for us, being non‑profits, being small, being very tight funding models, affording a fancy Digital Asset Management system is a bit of a struggle. Then the first foray into digital photography that museums took beginning about 10 years ago, we had a tendency to over buy. We would be sold very fancy Digital Asset Management systems that could do lots and lots of wonderful things.
[2:48] None of which we ever took advantage of.
[2:49] We kept paying the fee every year, and throwing the assets in, and struggling with metadata models. Not really making much progress. The success is that when I worked at the Metropolitan Museum of Art the photo studio went digital. Within about a year, they had amassed an enormous closetful of CDs.
[3:09] How do we go back and find those images that we shot a year ago? If there’s a success that’s clear it’s in the capacity to locate, download, relocate, reshoot when necessary the assets that are actually needed. It’s not a manual process anymore. We can have multiple users log into a system, find the image you’re looking for.
Henrik: [3:25] You can more rapidly search, find, use, reuse, repurpose.
Douglas: [3:30] I think that was a clear business win. I also think it’s aged a little bit. That win really took place…at the Met it probably took place about 2003, 2004. I’m currently at the MIA. They had a system that’s about the same age. The systems are, in essence, aging because they’re becoming full of assets, and because the metadata model, as I mentioned before, is really not mature or specific enough.
[3:54] Really not mature or specific enough. We have issues with overflow of result set. People go in and they search on something like “Rembrandt.” They’ll get thousands of returns. Many of which are place‑holder records. They are old black and white study photographs. It’s not clear which one I’m supposed to use for my marketing campaign.
[4:15] I go and start asking my friends. Now we’ve blown it out of the water. The reason they have an asset management system is so that anybody, even with a cursory knowledge of what they’re looking for, should be able to come in and get what they need.
Henrik: [4:28] True. Let’s use that example of searching for Rembrandt and you get documents and records, and then maybe some photos of the Rembrandts that you may have. Can’t you filter down to, say, “paintings of” from the thousand records for the sake of argument?
Douglas: [4:41] Again, when you over buy a system of course that functionality is there. Users need a lot of training to understand how to use it.
Henrik: [4:49] Add that information in all fairness.
Douglas: [4:50] Right, exactly. The only keyword on the photograph is Rembrandt. I should say the photograph on maybe 700 photographs. There isn’t a really good mature metadata model. Now, maybe the photographers remember because they know that only Charles would be shooting the master image. He shot those paintings about in 2007.
Henrik: [5:12] At high resolution blah, blah, blah with the proper lighting.
Douglas: [5:14] They can go in the system and they can say, “I only need things shot by Charles. I want them 2007. I want them only the TIFFs.” They can get that for you.
Henrik: [5:22] To your point, you can search for the TIFF, or you can search for the file type, meaning, “I don’t want a .doc of Rembrandt’s about the insurance record, or the transfer record, or the purchase record or whatever. I want the TIFF or the raw file or the JPEG or whatever.”
Douglas: [5:39] Although, to be clear, we’re not currently in the DAMs that we have storing any .docs. They could, I suppose. We’re not doing that.
Henrik: [5:45] Or PDF, for that matter?
Douglas: [5:46] There may be a few PDFs. That’s not really the core business case right now. The core use is still images, high res, primarily objects. Secondarily, events, people, activities of a museum being recorded.
[6:02] We also have an archive dating back 130 years, but it’s a physical archive. A few of those things get digitized now and then because there’s some need for them in a publication or something, so some of those things in there. Right now, it’s mostly just still image.
Henrik: [6:19] What advice would you like to share with DAM professionals, and people aspiring to become DAM professionals?
Douglas: [6:23] It’s a really good question. First of all, I don’t consider myself the world’s expert on answering this question. I would say that in the museum arena, which is the arena I know best, museums are in need of people to come into our world and help us adapt best practices, help understand how businesses are running in this way.
[6:48] One of the core differences, in a way, is that we’re all looking at Digital Asset Management systems as if they are at their core set up to be persistent electronic archives. We’re not a for‑profit vendor who is creating products for which there are seasons and catalogues and websites to be made, and campaigns to be run, advertising, marketing, press, everything else, and then a year later it’s all new products. It doesn’t matter what happens to the photos of the shoes from last year.
[7:18] For us, every time we take a photo, there are a number of things. First of all, I’d say it’s a fine work of art. You’ve moved it from its safe storage space into a photo studio. Any time you move something that old and that fragile, you’re damaging it. Maybe it’s not obvious, but you have micro‑fractures, or you’re exposing it to different atmospheric conditions, or different lighting conditions, whatever it may be. You’re actually not doing good by the artwork.
[7:46] I don’t mean to belabor that point, because people are very professional and very careful. Accidents almost never happen, but it’s still a fact that it’s a risk. If we’re going to do this, we’re going to move this work of art into a studio, light it, shoot it, let’s do it at the absolute most professional, highest resolution that we can.
[8:07] Let’s get as many angles. Let’s get as many types of spectral photographs that we can manage right now so that we put that wonderful and rare and unique object back into its secure storage space and don’t touch it again for years.
[8:20] What we’re doing is we’re capturing these incredible photographs, but we’re amateur in then what we do next. We have a very professional production process, followed by a very amateur archival metadata process.
Henrik: [8:36] Does the workflow fall off? Is that your point?
Douglas: [8:39] Workflow falls off a little bit. The folks who are doing it are probably the photographers themselves, and/or relatively junior people, probably not a strong metadata library background. I don’t mean to single anyone out. There are certainly people there who are skilled. If any of them were to leave their positions, it’d be hard to replace them.
[8:58] It’s specialist knowledge. Even with that specialist knowledge, what’s missing then is some real world experience of having run this kind of system, where it’s a really rapid fire production environment.
Henrik: [9:10] You’re embedding the information, to your point. That may be missing because most photographers don’t like adding metadata to their files. There’s a lot of value to finding it again if they add a lot more than just the word Rembrandt, to your point earlier.
Douglas: [9:24] You’re right. There’s been talk here at the Henry Stewart DAM New York about having workflows that would capture data that would then automatically become metadata. That’s terrific. There you get subject and photographer assigned, and all these other kinds of things that can happen automatically.
[9:41] In the use cases that we’re seeing, though, whether it’s internal. In the internal, you would have content creators, writers, editors, people working with the press, marketing, whatever it may be. They don’t think in those more academic, scholarly ways. They want the hero image of “Lucretia” by Rembrandt, and they want to be able to get it right now because they’re on the phone with someone who wants to do a story.
[10:03] We need keywording in a very…
Henrik: [10:07] Consistent way?
Douglas: [10:08] It’s consistent, but it’s also natural language. We have keywording that says things like in the acrylic on canvas.
Henrik: [10:18] Which you probably have a few.
Douglas: [10:19] Yeah. Oil on canvas, oil on copper, terracotta, these kinds of things, which are very important and they are the fact.
Henrik: [10:26] Yes, the medium.
Douglas: [10:27] When your press agent is on the phone with a reporter from the New York Times, they don’t go to the system and type in terracotta.
[10:34] They are on the phone, they’re talking, they’re trying to type to try to type things like clay, pot, Africa, bead work and you do desperately trying to find the image, like, “I am trying to find it for you right now, Mr. such and such.”
[10:47] Because we don’t have that piece in there, it makes the system of much less use to them. So instead what they’re doing is emailing somebody, like a photographer, their friend, saying, “What do you have that pot for Africa with the beads?” They’re like, “Oh yeah, sure” and so two people get involved in the work when it really should just be one.
Henrik: [11:03] It is really tied to, in part, institutional knowledge.
Douglas: [11:06] Here is what I want, because I am not a Digital Asset Management worker, expert, it is not my training, but if you had an organization that was constantly feeding stories to the press. So whatever that may be, there must be folks out there who do sports photography, something like that.
[11:28] They got to be uploading those things quickly, they got to be tagging them with the kinds of words that sportswriters are going to use, like “World Series Game 3″ and you better have it or no one is going to use your images.
[11:36] We don’t have that discipline, is a weird word for it because it is kind of lightweight, but it’s so absolutely necessary to make the asset findable across a much broader swath of people.
[11:49] If I were to tie it back to some of the strategies that we’ve been talking about in the art museum world anyways that we have been in an industry that for 150 years has been in the kind of, if you build it, they will come mode.
[12:04] We’re great, we’re fancy, everyone should come. If you’re not coming, there’s something wrong with you.
[12:10] This was the old model, the old elite model in having an art museum. What you’re seeing art museums do in the last, I don’t know, 10 years, maybe a little bit more and say, “No, no, no, no. Look, this is important. Art education being gutted in this country, creative thinking gutted, innovation gutted.”
[12:27] You can’t get this kind of stuff in school anymore. You should come to a museum. That makes museums need to be more engaging, more embracing of different points of view.
[12:38] Instead of saying to someone, “Welcome to the museum, you must be very proud to be here,” we have to say, “Welcome to the museum, we are really glad you are here. How can we help you have an experience that you will never forget?”
Henrik: [12:49] To your point, even with DAM, it needs to be more accessible. Once they are more accessible, then people can obain it. Hypothetically in the virtual museum sense, I’ve worked with some that are doing that piecemeal. That’s the future challenge, I assume, with some.
Douglas: [13:07] That’s the tactical implementation of a philosophical point. Let’s say our PR marketing hire a couple of young social media folks. They want to throw together a Tumblr site or a Pinterest or Instagram. They want to grab some photos from the collection.
[13:23] They go into the system and start saying like, “Give me an exciting photo that shows women having fun.” The system doesn’t have anything like that in there. It has like Matisse “Bathers,” but that is not what a 28‑year‑old social media manager is looking for.
[13:38] I know we’re a little bit, I’m beating the horse to death here, a little bit, but it is a metadata model is less about this sort of deep scholarly academic information and more about, “Hey, guys, what exactly do we have here?”
Henrik: [13:49] That could be controlled but that could be a taxonomy, because of those events, to your point earlier, happened regularly in the kinds of activities that happen in the museum, because there’s only so many things that will happen in a museum.
Douglas: [laughs] [14:00] There is a lot of things that will happen in the museum.
Henrik: [14:01] That are permissible in a museum.
Henrik: [14:05] If it’s a fundraiser, or…
Douglas: [14:08] It is funny, we’re beholden to our own approach when I worked at the Metropolitan. There were years when there were almost 20,000 events on the event calendar in 365 days.
[14:18] Now I am at a smaller regional museum, but the number of events still is in the thousands for year. Now that’s counting things like tours and school groups coming in, but each of these things happen and we are slaves to our own success in this way. We do not want to stop doing all of that.
[14:36] It is a little overwhelming, there’s photography of all a lot of it that nobody can find and that is the whole point of having a DAM in the first place.
Henrik: [14:44] Thank you, Douglas.
Douglas: [14:45] Thank you.