One of the new experiences I’ve had here at LPB, among the many new experiences that have accompanied moving to Baton Rouge, Louisiana (I have way more opinions about gumbo now), has been working for an archive in a production environment.
The primary purpose of a TV station is, believe it or not, to broadcast television programs. Not, necessarily, to preserve and provide access to their archive. That being said, it can be a kind of two birds/one stone situation. However, given that archiving is not the first thing that springs to mind when one mentions a public broadcasting station, prioritizing archival functions in a production environment requires some justification.
Now, this isn’t all that different from a traditional archive. Whether one works at NARA or a local library, new preservation initiatives or procedures aren’t just rubber stamped because the archivist remembered to ask nicely (although I think saying “please” and “thank you” can’t hurt). Archivists have to advocate for archiving.
After spending several months at LPB learning the archival workflow, I had some recommendations I wanted to make to (hopefully) improve digital preservation practices. My host mentor, Leslie Bourgeois, suggested we call a meeting with several department heads and discuss the feasibility of my recommendations.
I think it’s important to acknowledge that I had several advantages coming into this meeting. LPB had to apply to host me in my residency, and a part of the stated purpose of my position at LPB is to make digital preservation recommendations. In other words, there was an understanding that I would be suggesting some changes. Leslie also helped pave the way for the meeting by reminding the participants ahead of time that this was why they had asked me to come work at LPB (thanks, Leslie). The choice to apply for and host a National Digital Stewardship Residency also acknowledges the value of archiving and information science. I can imagine a situation in which pointing towards “best practices” in the field would have no weight at all. While LPB is certainly more concerned with how other public television stations are archiving their collections than with how a university library preserves its collection, there is interest at LPB in the best practices of audiovisual preservation. This was a big help for me, because I didn’t need to justify the significance of complying with standards or archival best practices.
So, what did I say, and how did I say it?
I decided not to use a powerpoint. I wanted people to feel comfortable jumping in with questions, or sharing their expertise, and I thought looking at a screen in a dark room could inhibit that. I did use a handout, though (sorry trees). You can find that handout here.
I began by outlining our current practices and archival processes. Everyone in the meeting was more or less familiar with these, but since most folks are very familiar with their own part of the workflow, and less aware of the other elements, it seemed to serve as a helpful reminder. It was also a good opportunity to demonstrate that I knew the lay of the land and was making these recommendations from an informed perspective.
After a quick review of what our existing digital preservation procedures were, I pointed out some of the risks inherent in these current practices. I tried to boil down the risk to some key concepts, ideas that would be relatable even if you weren’t familiar with preservation. Below are some of those key concepts, and how they applied to our workflow.
Simplicity as Risk-Averse
The more things you do, the more chance you will run of one of them going wrong. Our current workflow involves moving media files across many different devices, that send signals to many different places, for different purposes. I suggested that the simpler the workflow was, the better.
Provenance as Troubleshooting
In order to encourage more fixity checks, I pointed out our current inability to diagnose the point of file corruption or loss of data integrity, in the event that a file does not perform successfully. If an error is encountered, there is no way to trace when the error took place, or if a previous version of the file would contain the same error. If we had checksums from several points along the way, we could trace an error back to it’s origin.
Planning for Updates as Planning for Obsolescence
Television stations are accustomed to the growing pains of technological evolution. Whether it is the shift from analog to digital or standard definition to high definition, changes in the media landscape are obligatory in television, but not necessarily easy. Given this reality, pointing towards the obsolescence of a particular storage media, or the necessity of more digital storage in the future, has more traction in a room full of professionals whom have had to manage these types of changes in the past. The station’s current dependency on the XDCAM format, and the risks involved in that, was not a surprise to anyone.
Once I had explained the risks inherent to the station’s current practices I outlined a series of potential solutions to those problems.
Lossless or Uncompressed File Formats
Best practices in the field of video preservation call for analog video to be migrated to digital formats through uncompressed or lossless digital video encoding. A fair question could be: “So what?” We currently migrate analog video to a high quality video file format – IMX50. (For the digital video nerds, IMX50 is a proprietary Sony codec, but it’s basically MPEG-2 encoding, standardized in SMPTE 365M as the D-10 codec.) Why bother with uncompressed or lossless when IMX50 looks really good? A couple reasons:
- You’re still losing information. Analog closed captions are encoded into a video signal on the horizontal “line 21” of the video signal. Compressing that signal can destroy captions! Michael Grant of NYU Libraries gave a cool talk about this at the last AMIA conference.
- IMX50 looks good now, but… As we hear more about the introduction of 4k broadcasting, it’s painfully clear that standard definition video is going to look worse and worse – by comparison – over time.
- Storage will always be cheaper. In the meeting, it was nice to have two IT professionals in the room to back me up on this assertion. LTO-6 tape is already relatively inexpensive in terms of digital storage. When LPB migrates to LTO-8, ideally following the release of LTO-9, the station will be able to store 5 times as much data on one tape, for likely the same price or even less. Besides, mathematically lossless file formats like FFV1 allow for the same quality as an Uncompressed file, while taking up about half the space. The effort to standardize FFV1 through the IETF’s CELLAR working group and other standards organizations and communities makes FFV1 even more appealing.
File-based submission to the Archive
We currently have a lot of material submitted to the archive on XDCAM disks, and we will also receive documentation of productions (contracts, releases, footage logs, etc) on paper. I argued that we should be moving towards file-based submission of material to the archive. Former NDSR resident Dinah Handel summarizes some of the advantages CUNY TV found in moving to a file-based workflow in the 2nd half of this blog post (the 1st half, by Mary Kidd, is great too). My primary argument for moving towards file-based submission centers around the ease of automating archival functions in a file based environment. When files are locked on an XDCAM disk, an archivist cannot document the video file’s technical metadata by creating a MediaInfo report, or validate the file’s fixity, without an expensive and clunky interface between an XDCAM deck and a computer (which our archive does not have). Even if with access to an XDCAM deck, this process could not be automated.
I suggested that LPB produced documentaries, like the recently released Deeply Rooted, be submitted to the archive on a hard drive. This submission would include the editing software’s project file, graphics produced for the documentary, and all accompanying documentation, such as scripts, supers lists, and contracts. I pointed to the advantages of digital information, and that keeping digital records in their original form – such as the connections between raw footage and a finished program – saved time. The same is true for a word processing document over a printed out version of a script.
So, how did it go?
Pretty well I think! There seemed to be a lot of support for some of the ideas I put forward. We had a follow up meeting a few weeks later to discuss which solutions we might explore, and how we might get there. I used a similar formula to prepare and present my ideas for that meeting. You can find the handout I made for that meeting here.
I would be surprised if LPB ends up following every single one of my recommendations, but steps have been taken to implement some of them already! It is certainly a move in the right direction, and I think the meetings we had helped raise the profile of some of our archival processes and procedures at the station. In hindsight, just getting everybody together in the same room and discussing what could be better about the workflow was one of the biggest advantages to the meeting. From there, things progressed pretty organically through teamwork. My suggestion of finding a higher quality file for the analog workflow inspired the team to start discussing how to collect a higher quality version of our recently producer programs as well! Don’t underestimate your colleagues’ interest in what you do, and their ability to help you accomplish your goals.
This post was written by Eddy Colloton, resident at Louisiana Public Broadcasting.