Digital Asset Management: An Outsourcing Opportunity?

Co-chairs:
Alexander Pasik, CIO, Guggenheim Museum. (Formerly of wall street IT)
, and Howard Goldstein, VP, Center for Digital Imaging

Alexander - Back in 70s, most people used outsourced IT systems, because you couldn't afford your own computer. Then the PC revolution happened, and things changed. You couldn't push all that interactivity through a wire, so processing moved to the client. Enabled new capabilities. But then about 10 years ago, HTTP/HTML made it easy for disparate systems all over the world to easily present data to users in all sorts of format. So suddenly we're again looking at IT outsourcing again with fresh eye.

How many already have major systems entirely outsourced? How many don't? You people with your hands up? Do you really do all your payroll internally? (lots of hands go down, mine included.) My prediction, within 5 years 90% of your apps will be hosted offsite. Because you're a museum. You should be thinking strategically and not trying to run a data center.

A problem - many outsourced solutions need to be relatively generic. And organizations don't think of themselves as generic. BUT -- certain verticals (industries) have enough in common that they can outsource == particularly where technology is not used to give them a competitive advantage. We have an industry that has similar problems from one place to the next that could tremendously benefit from standardized solutions. I would put up a fight if anyone here tried to say that Museum A has totally different DAM requirements from Museum B.

So is there viability for organization(s) to provide outsourced DAM for museums? That's the question we're here to talk about.

Howard Goldstein, Center for Digital Imaging. This panel came out of a fun discussion Alexander and I had one day over trying to find solutions for DAM for small museums. Those who sat in on a workshop yesterday, sorry, many left feeling depressed. Left them with the feeling that there weren't lots of alternatives. The idea of this panel is to throw lots of darts up in the air and seeing what sticks.

Our Panel:

Douglass ____

Ari Davidow, Director of ONline Strategy, JWA

Nathan Bedd, Photog. for Nat Geo, first internet portals for Stock photog. Here representing OrangeLogic, which is a company that does hosting of images for corporate and some cultural institutions. Nathan is also a trustee for George Eastman House.


Doglas - we have one of the longest running photography studios in the world. Had to consider going to a digital workflow about 15 years ago. Eventually we did that. Then what happened? Tons of images, all over the place, and before you know it, you have a quarter million beautiful images, and god knows where they are, god knows if they'll open when you need them, etc. So what we need is some sort of asset management. So we started something called the Met Images Initiative. How could we solve this problem and create a persistent archive?

When we dove into this, we looked at places like Corbis and Getty and thought we could make money. Perhaps we were abit eager, this was overly optimistic about revenue. BUt we realized this was something we needed to do anyway.Problem that 8-15% of CDs were failing right off the bat, we were seeing multiple versions of photos, etc. We presented bck to the board and said three important things: preservation, access, and licensing/revenue. (although less agressively than we originally thought.)

Goal of good DAM - one good master image, then render any derivatives you need. (I'm talking about object photography -- of course, also need photos of people, the building, events, etc.) Couldn't find good options for outsourcing or sharing with others - art museums have different needs than other types of organizations. We needed to find software and purchase it. We went with MediaBin from Interwoven. What we liked was that we were able to customize and configure it in a way that worked for us. We pressed them pretty hard, I think, with our reuqirements. But they were responsive. They also provided capacity and APIs to let it work with other systems like collections management.

There were issues - any customizations we made complicates an upgrade path. There's also a need or us to have technical knowhow and staffing onsite. That's something we had to subsume. Costs. We had to buy the software and hardware. We now have 30TB and are using more than half of it. And not just storage -- how do you back it up?

SO what would be the issues if we were to outsource this effort? We're not biased against it. We're in a major effort to move all of our HR systems out now. We don't want to spend time running financial systems - we're not good at that. So if we're going to outsource DAM, what would we care about?

First, security. We've embraced openess, for example, images for intellectual publishing program. But probably wouldn't want to release all of our high quality images to the public. So we'd want it to be secure. But it begs the question, if we're willing to put all our financial stuff out on external services, why would we not do it for images? We even outsource art storage! So why is there fear about this?

We would want to make sure we had appropriate access levels and roles.

Bandwidth is always an issue. Images get bigger, people want to get them faster. So that's always an issue.

We would need to integrate with our collections system. That might be sort of an issue for outsourcing. [My q: need for standardization of this type of interface!]

Standardization: we want to move toward CDWA, OAI, etc. Would an outsourced vendor let us keep up with this sort of thing.

Workflow tools? Image ordering, workflow tracking, etc. We would need to feel confident about these things.

Q: Backups? A: We do lots of tapes. Differentials help - not much changes. But that's the best we can do right now. We do check them. I have a full time person who just does backups, and does testing. I don't think this is the best strategy. I think using the cloud and backing things up on live disks is probably a better solution.

Ari -- I feel like the odd person out because I work for a virtual organization, not a museum. We don't have a place you can come to and see things. It makes us significantly more paranoid, because we don't always have an original to rephotograph if we lose something.

We are very small. (I now have a web developer and an archivst. They're special people, but that isn't much.)

We use hosted services a lot. DonorPerfect for donor info, for example. Our nonexistent IT staff no longer had to spend time configuring computers for new people, planning strategy, etc. We now think about the actual info and strategy, not nuts and bolts. And we have a programming API that lets us do stuff with the info. I can assure you that the people who run DonorPerfect have far more backup and hosting resources than I do. If people freak out because it's not in my office -- but the people we're paying have far more resources and expertise than I do.

Get it off our plate. We're in the business of documenting and disseminating stories of people. We're not in the business of writing donor software.

We use Drupal for content management. We're using a host of open source software for a lot of things. Every time we think about something new, we think about whether it's something we want to spend resources on when there's someone else who already does it.

We're not outsourcing DAM because we're doing a lot of customization. We have disagreements with people locally we're working with on how to describe oral histories. Until we reach a point where there's a reasonably accepted standard for the things we're putting in, we can't host it.

I disagree with whoever said workflow doesn't matter. If your hosted app doesn't fit you, it's going to be much worse trying to fit your org into something that doesn't think the way you think. It's a morale hit.

Nathan -- My primary hat is from the company OrangeLogic. Based in France, provides digital asset services.

I want to speak more broadly than that and communicate two things. First, outsourced DAM solutions are widely used by photographers and photo agencies, and have been for years. Secondly, DAM is only one part of producing, managing, and storing media assets.

Today the global DAM industry is a billion dollar industry. 85% of which is for systems that are installed in enterprises. Most are not cultural institutions, especially those involved in marketing and communication. Average installation is 300k$. Early adopters of DAM outsourcing for marketing/communications were Miller Beer, Discovery Network, Nike, etc. Now hundreds use outsourced DAM serivce.

LArgest photo services Getty/Corbis have developed proprietary systems and spend tons on them. The third larges is xxx from Britain. They have 110 people on IT team (in India) managing 13M pics.

Most of us are engaged in museums where physical object is primary. But most photo businesses started that way -- the object might have been the print or the transparency. TOday we think of the photography industry managing digital objects. But the primacy of the physical object is at the root of IT in the photo industry.

There's kind of a tiered menu of software as service solutions for photo industry. At the top end, there are users like photo industries, museums, etc. 25k photos up to millions of photos and TBs of info. Costs start at 3k/month for smallest users and goes up to 10k's/month. Depends on scale.

There are in the low hundreds of companies in this tier. The leading vendors include Twinsoft, Aldoba (sp) in Paris, and our company, OrangeLogic.

Then there's low-customization solutions targeted at individual photographers and small agencies. Under 25k images, and up to maybe 100GB. Vendors: IPN, PhotoShelter, until last week, Digital Railroad, which just closed.

Then there are zero configuration systems for a scale of a few thousand, which cost zero. Like Flickr and Snapfish. It's Guerilla DAM.

Then at the desktop there's stuff like iPhoto, Lightroom, Aperture.

Installed DAM systems: average cost is 300k, and maintenace is 1 few thousands. But with IT staff, real operating cost is over 100k/year.

We were given an RFP by a museum this year. The issue is 25k images, and want a public facing front with transactional capability and maybe even self service license. They wanted it to be an installed solution inside their firewall maintained by their staff. But when we costed meeting their RFP and could operatte withint the ruls of the museum IT concept, it would have costed us over 400k. We offered them a disccount, but still in the hundreds of thousands of bucks.

To get to the same perspective from a user perspective, we could do the same job for 40k as an outsourced, hosted solution. BUt the museum doesn't want to go that way. We bid it that way, but it's not the right way to go for many museums who don't have the IT staff or budget.

The second part of my talk: DAM is only one piece of a total solution. Buying a DAM because you need to manage digital assets is like buying a jet engine because you need to fly somewhere.

To break it down, there are really two major modules. The first group of modules fit under DAM, and include prod. workflow, collection management, content management, public facing front, different interfaces for different users, etc. Tools to enable the internal and external users to find and share images. Ways to output images.

Then there's a whole other component: DRM. Tools to streamline rights management. Ordering, processing, licensing. Tracking what users and assets are doing on your system. The invoicing and accounting tools. Even if you give them away, still want to know where they went and who has them. And you need a system to track who the customers are. THat's a different thing, different even than Raiser's Edge.

The museum currently views all these things as seprarate systems, which creates a big ineffeciency in system specification and maintenance. Rather than trying to add anouther module one step at a time to their legacy, they should look at outsourcing.

Q: Trusted Archives? Is there an equivalent for trusted digital vendors?
A: We are a society that tends to be very careful of letting go of perceived security. But most of you still have money in banks.

(Didn't record most of the discussion.)