We’ve been working on an experimental digital preservation project.
Sometimes, digital preservation means something besides preserving an exact bit-by-bit copy of an item. Computer hardware and software required to support digital documents becomes obsolete and falls away from common existence, or else evolves so significantly that documents created with original versions of the hardware and software can no longer be used.
As an example, let’s imagine that we’ve collected a set of video journals of undergraduate students that were kept as part of an introductory but innovative telecommunications course in 1997. These hypothetical video journals were given to us on 100 MB Zip disks, and were created with RealVideo. In order to preserve these videos, one of the first things we’d want to do is copy them to another location besides the Zip disks, because we don’t want to depend on the storage lifespan of a single Zip disk. This move would not only support preservation, but also access. In 2016, it’s unlikely that any person – whether a casual user or a professional – would have a Zip drive available to access these files. It’s better to place them online somewhere, so that they can be accessed worldwide. Additionally, the RealVideo format is no longer practical – few people have players capable of playing RealVideo installed in their web browsers. So we’d want to transcode these videos into other formats, for example mp4/H.264 for access, and Matroksa/FFV1 for preservation. Zip drives are obsolete for our purposes, and RealVideo is obsolete for our purposes, but the content of the video journals may yet be interesting to someone, so we preserve the content in a way that is usable.
Sometimes the digital artifact that we are most interested in preserving is not a digital document created by a piece of software, but rather the software itself.
There’s an iPhone app that was released several years ago, in April 2010, by an Icelandic company called Fancy Pants Global. Fancy Pants Global described the software as “a handy little app that will help its users make those tough everyday decisions.” This decision making app is very basic, but highly rated by users because of its simple and attractive interface.
To some of its adoring users, the app is no longer usable. The iOS operating system for iPhones has been updated several times since the final release of the app, until the app finally became unusable with the release of iOS 9. Resolver crashes and exits whenever an on-screen button is tapped.
One way for users to resolve problems with apps that no longer function is to contact the original developers and ask for bug fixes. This doesn’t always work; motivation is sometimes low for a developer to provide never-ending, free support to an application user whose original purchase netted approximately seventy cents.
My friend and colleague, Amanda, who favors the original app, made unsuccessful attempts to track down and contact the original developers.
The developers couldn’t be found, and the company appears to be out of business, but perhaps there was another avenue for resolution – the app source code. The source code is created in a format that is understood by software developers – a computer programming language – and used to build the package that is distributed and used to run the app. Many computer applications are open source, meaning that this source code is freely available to users, with a license that allows for modification. However, the source code for the Resolver app could not be located.
One possibility to provide access to the original app would be to preserve the entire operating system and make it available in an emulator that can run on newer equipment, as Emory does with the Salman Rushdie archive. However, there may be significant technical hurdles in setting this up for a new environment, and we prefer instead to allow users to access the app with equipment that they are likely to already possess – their phones – without requiring significant additional technical knowledge.
Upon confrontation with the obstacles of missing developers and missing source code, Re-Resolver was born. The app will be coded from scratch, with the goal of duplicating the functionality as well as the look and feel of the original Resolver app as closely as possible – at least until people request new features. We are modern day monks, meticulously copying a classic work that we will be inspired to embellish.
As a basis for the project, Amanda created a video that demonstrates the use of the original application.
Working from Amanda’s video, I coded an initial beta that attempts to duplicate the functionality of the original Resolver app. It missed substantially on the look and feel, as well as the emotional connection. The app has since been revised to look more similar to the original, and is in the beta testing phase. The code for our project is open source and available on GitHub.
This particular project is good for this type of experimentation. The app is small and doesn’t have any complicated functionality, so the project should be relatively quick to complete. It will be easier to get a sense of the time spent on the project.
The experiment gives us an opportunity to ask and to explore many philosophical and practical questions. Is this method cost-effective and ultimately worth it? Is an app still the same app when it is nearly identical, functionally and visually, but was re-written from the beginning in a programming language that didn’t exist when the original was created? What should we do when people request new features?
Is this preservation or intellectual property theft? Please weigh in using the blog comments.
The next blog post about this app will be about accessibility.