Our guides for the evening were LIKE stalwart Lena Roland and Adrian Brown from the Parliament Archives.
Lena opened with a general description of the problems relating to the retention of digital materials and some of the approaches than can be used to address these. Adrian then explained the specific issues that Parliament has and the work that they are now doing.
In the past most information was stored on physical media (paper, film, etc.) and while these have their own preservation problems they are generally understood. Whereas now most is created and stored digitally. This creates new problems that we need to solve.
Compounding the problem is the sheer volume of information. Once we just had a few newspapers and TV channels to worry about, now we have to add the likes of blogs, twitter, Google+, Facebook, etc. etc.
This means that a "save everything" solution is difficult to create and even harder to search. The alternative "save the important stuff" solution relies on good judgement on what is worth saving and that judgement often only comes with hindsight. For example, all my old family photos may become useful if one of my sons becomes famous but otherwise they have little external value.
Some organisations are wise to the problem with initiatives to, for example, archive Twitter.
Searching will become more of a problem. Books have ISBNs to identify and classify them but there are no equivalents for blogs, tweets, etc.
Customisation of web sites to individual users, e.g. the Amazon home page, complicates matters further. How do you archive Amazon if every person's view of it is different?
The Parliamentary Archives in some ways have a simpler problem to solve in that it is just concerned with a relatively small and discrete set of records but this is made harder by the historical records, some of which are on parchment, and the need to have a precise and accurate record.
The approach they are taking is to separate the content (the words) from the media (a PDF file on a CD-ROM). Adrian gave the example of an old movie that requires the preservation of both the acetate film and a projector to display it. Separating content from media allows the preservation of both to be addressed separately.
Approaches to be considered include Preservation (keep everything as it is today), Migration (move it periodically to new media, e.g. from WordStar to Word) and Emulation (a new machine pretending to be an old one).
There are some decisions to make about what constitutes "content". Clearly it is the words but it may matter how the pages are laid out, which font was used, etc.
Links are an increasing problem. For example, a debate may refer to a policy document that is held on a Departmental website that is not under the Archives' control, i.e. it could move or be deleted. Archives are working with the Departments on this and one approach is to get them to keep everything in its original place even if it is no longer directly addressable from the Department's website.
Archives are considering whether the only way that they can be sure of keeping an accurate record is to keep a copy of all the things that they link to, with the implications that has for storage.
My notes from this session broke my golden rule and went on to a second page, which only goes to show that I am more interested in this topic than I thought.
Overall the debate was encouraging in that bright minds are aware of and understand the problem. What was less convincing was whether anybody has yet got the answers.
Thanks for the writeup Matthew. I wasn't able to go to this was as I was at work. Sarah
ReplyDelete