In the second group discussion we talked about specific examples of software preservation. Groups were asked to identify a real or hypothetical example, and then to answer the following questions:
- Consider the why, what, who, where, when, how…
- What are the preservation requirements?
- What is your preferred approach?
- Where are the skills, resources and funding?
- What are your next steps?
We had four great examples, and a lot of interesting issues:
- The research software involved with brain imaging can be from a variety of sources, including commercially bought packages and open-source based programs written by research students. This variability complicates decisions on how to preserve software in terms of overall strategy but the consensus is that globally there should be a push towards better software engineering. This includes better documentation, the sharing of expertise and a long term outlook. The group proposed an advocacy strategy where a small group of consumers could put pressure on the software designers in order for this to come about. It is also important to identify the benefits to the funders who would be putting investment into better software engineering.
- This hypothetical example involved a piece of data entry and search software that uses a number of different formats, across three institutions and where the original developers have moved on. The group explained that continual availability of the software was dependent on:
- operational funding;
- accurate and maintained rankings search to provide consistent precision (ie search terms give the same results each time);
- a preserved interface (due to a lack of training funding if the interface were to be changed).
Security and confidentiality were noted as important aspects as was consistency in authentication. It was decided that a positive effect of software preservation (apart from the preservation itself) for current users would be to improve the speed of the software while preservation is taking place – ie the performance of the underlying systems would increase with each new generation of hardware, so that the searching may become faster over time.
The group also highlighted the importance of cultivating a community for future users and that a testing group across the involved institutions would be valuable. If attempts to engage the original developers failed, the group suggested engaging the support of top level professors and research councils to raise funds, awareness and possibly corporate sponsorship.
- A library keeps archives of the works of famous scholars and writers, which sometimes includes entire PC’s with all their files. The aim is to have a complete environment to present to future scholars so they can explore the famous scholars file systems and have access to previous drafts and other documents. To do this the library has successfully used an emulation approach (using both Mac and Windows) which is greatly appreciated by the academics and students who use it. However, the emulator itself is now becoming obsolescent and they are now thinking about how to migrate the emulator to a new platform (eg away from XP to a new platform). A major migration issue is whether repeated emulation is possible, sustainable and leads to information degradation over time? Also, though current one machine jobs are relatively straight forward, the required software can sometimes be spread across more than one machine. Increasing the ‘standalone desktop’ model for personal computing is out-dated, to be replaced by a mix of laptops, mobile devices, and online services. This identified another huge problem – that of preserving data and software for research groups that use the “cloud”.
- A Museum holds a physical video collection where both content and context are important. The collection is also publicly accessible and so has to be easily available and understandable to non-experts. Preservation is relatively simple for a single particular platform or console (eg the Nintendo SNES), but for PC games preservation can become complex as the hardware and software are often very varied. The group explained how the numbers of users who remember the preserved systems are starting to decrease and there is a now race to preserve people’s knowledge. Generally there is a throw-away attitude to “old” software in the commercial sector and the group suggested raising the profile of organisations that preserve old defunct hardware/software as an example of good practice. Due to the large workload, software preservation has a low priority at the museum, and the group assessed that their first approach should be to raise awareness in this area. Secondly they proposed a change the museums collections policy so that it covered software, which would then increase the importance of software preservation to the museum.