Last week, Digital Bedrock officially announced their new secure, managed digital preservation service in what they describe as a “hack-free architecture.”
NOTE: You can read their introductory press release here.
Their website states:
“Everyone has digital content they’d like (and are sometimes required) to keep forever: businesses, archives, artists, producers, consumers. But file formats become obsolete, bits die, and files get lost.
“We future-proof your files so they will be available and playable when you need them. By identifying long-term vulnerabilities, we create complex metadata about an asset’s characteristics and dependencies and monitor its health over time. We offer offline redundancy on LTO-7 tape in three geographically separated locations, and migrate your assets as new storage media becomes available.
“Through our digital file analysis, we run virus scans, create checksums, validate formats, and extract technical and add preservation metadata. Clients can search for their files through their secure, online portal. We’re not a cloud-based file access service, but we make it easy for our customers to find and track their assets as needed.
“We offer ongoing format obsolescence checks through our proprietary Digital Object Obsolescence Database. If a format is flagged with obsolescence vulnerabilities, we notify our clients and work with them towards the best decision in what to do: migrate the file, apply an emulation environment, or only perform bit-level preservation until a solution is identified.
“Digital Bedrock is in a high-security ISO 27001 certified data center located in the retrofitted historic US Post Office Terminal Annex building in downtown Los Angeles. Physical security is high: access to racks and our office requires passing through six levels of security, including biometric scanners.”
Last summer, I interviewed Linda Tadic, CEO of Digital Bedrock, in anticipation of their up-coming launch.
Linda began her career in media, but for the last 30 years she’s been involved with digital preservation, metadata, and digital production operations at organizations such as ARTstor, HBO, and the Media Archives and Peabody Awards Collection at the University of Georgia. She is currently an adjunct professor in UCLA’s Moving Image Archive Studies program teaching Digital Asset Management, while previously she was an adjunct professor in NYU’s Moving Image Archiving and Preservation program.
She, like all of us, is looking to find the best way to securely archive our media, guard against the continual obsolescence caused by the on-going rush of technology, and make sure that our assets are safely stored for the long-term – yet still readable many years into the future. This was the driving idea behind her starting Digital Bedrock.
What they’ve done is create a facility where media assets are stored on tape and stored in libraries that are not connected to the web, which prevents hacking. Prior to storage, they extract the metadata (the information about our files) and store that metadata on an accessible, cloud-based database.
Clients are then able to search the metadata contained in their files, but not modify it, through this web-based portal. And, because the metadata is read-only and separate from the actual archives themselves, this minimizes opportunity for a hacker to mess things up.
This metadata is more than just file names, it also includes obsolescence information like what program created the file, or what codec it uses. Their system then tracks this data. When a codec or application is marked as “end of life,” their system notifies us with options on how to migrate our data into a more current format.
To me, this is a huge benefit to filmmakers. We deal with so many assets, even on a single project, that after a few years, it is impossible for us to remember which programs and technology we used on an earlier project. But, if we fail to migrate our media, we will end up with perfectly archived assets which are impossible to play back. This is not just an issue of hardware – like SyQuest or Jazz drives – but an issue of software – how many PICT images created in MacPaint can you still open today?
Here’s their process, as Linda described it:
This process of active archive management is handled by the Digital Bedrock Digital Preservation Application. This performs on-going obsolescence checks, and, in general, keeps an eye on all your data. For example, files are tested every six months to make sure the data is still OK, the tapes don’t just sit on the shelf.
The benefit to using tape is that it is inherently stable, safely stores data when the power is off, is not subject to online hacking, and doesn’t require Internet access. If, due to drop-out, there’s an error on one tape, it is not likely to be on all three copies.
PRICING AND OTHER QUESTIONS
As part of this article, I spoke again with Linda Tadic, CEO of Digital Bedrock about pricing and she told me: “We’re offering early adopter pricing that’s less expensive than simply storing files on Amazon Glacier, and we do a lot more work for the client. I want to make this high level of preservation accessible to anybody, not just wealthy institutions. But the pricing matrix is complicated so I didn’t want to distribute it publicly and risk people misunderstanding it. It’s based on the number of files and the amount of storage; there’s a one-time only fee for processing, and then ongoing maintenance and storage. “
The Digital Bedrock website has a Contact Me page that you can use to get more information and pricing help.
Larry: Your curation tool allows clients to get accurate pricing information, but that requires them to send files to you first. Is there a way to determine ROUGHLY how much your service will cost before sending you their precious data?
Linda: Good question! They can contact us first and provide a rough estimate of the number of files and storage they think they’ll send. We can tell them over the phone or via email the cost for their estimated content. That’s also a good strategy to follow if they want to borrow a HDD to deliver their files.
Larry: How do you want customers to send you data?
Linda: Data is only being accepted on detachable media sent via UPS/FedEx. We don’t want anything sent over the cloud. Clients can send us data on their own HDDs or LTO tape, or we will loan a HDD so they don’t have that extra expense. If they send their own media, they have the option of having their media returned, stored with their LTO7 tapes, or destroyed.
Larry: How do you want metadata submitted that isn’t associated with the file itself. For example, data stored in media asset management systems or Excel spreadsheets that describes the file contents, not just its specs?
Linda: We love it when clients send us descriptive metadata about their files! It helps them search for their files on their own client portal; otherwise they’d be searching by directory or filename. When they send their package of files, our tool has a tab where they can add .CSV or Excel files with their own metadata. We map the client’s metadata into our system so clients can search by their own data, plus we’ll save the metadata file as well. They can also send any database output to be saved and preserved just like the content assets.
Archiving is essential to almost all media projects. However, long-term archiving is challenged by on-going obsolescence, security issues, and an inability to remember what was stored where.
Digital Bedrock seems to have thought carefully about all these concerns. I especially like their avoidance of The Cloud. It will be interesting to see if their system is embraced by the industry.