Legacy In the Digital Age #32 #cong24 #legacy

Reading Time in Minutes

4

1. A legacy is really two parts. The legacy item itself and the record of your involvement in it.
2. Making the latter permanent is more tricky now in the digital age than it was in the paper age
3. We have no consensus yet about how to make digital artifacts long-lived
4. The problem is becoming more acute because “digital transformation” is mostly removing the old way. Namely, using paper versions as “ground truth” in the event that digital versions become obsolete or of suspect fidelity

Sean McGrath is a trainee digital curmudgeon with a 40 year career in information technology. He is an amateur musician and a keen arm-chair philosopher with a particular interest in language and epistemology. He likes to turn the latter into country songs in the style of Hank Williams.

You can connect with Sean via email

By Sean McGrath

I suspect most of us want to “leave something behind”. We want to make some sort of a difference to something, somewhere somehow.

In addition to making the difference itself – the actual legacy – most of us would also like to be remembered for it. In other words, having a legacy requires two things:

• the thing you want to leave behind for posterity, and
• the record that you were the person that actually did it

The latter is often referred to as attestation record. For example a written statement that this particular Sean McGrath played a part in the thing called X (for any value of X – an invention, a construction, a work of art, a body of knowledge, an endowment fund for the arts, an old oak tree etc.). This needs to be some sort of long-lived artifact that future observers can use to find out about your legacy or to convince themselves that you are responsible for some legacy item.

Ok. But what is a “long-lived artifact” in this context?

For most of the technological era (from steam engine to space travel) we mostly think of paper books as exemplars of long-lived attestation artifact. Examples include histories, chronologies and ledgers.

We can increase the trustability of these by adding in academic/social credibility and also by giving them legal status in some cases. We typically do this with, yes, you guessed it, more paper!

All these paper-centric methods have “gone digital” since the Seventies and therein lies the problem. Making any digital artifact “long lived” is actually a very tricky proposition compared to “old fashioned” paper.

A good example of the problem is the BBC’s Digital Domesday book project from the early 2000’s. This “digital book” was supposed to last 1000 years but actually lasted just 15. The original digital incarnation is now completely obsolete. Unless that is, you happen to have a 12 inch video disc player and a BBC Micro Computer with all the right ancient software loaded. The great irony here is that the original Domesday book from 1086 can still be read today. It is housed in a museum in the UK. It was written on animal skin parchment.

We are in the midst of what is generally referred to as “digital transformation”, aka Industry 4.0, aka Digital Twins. There are many facets to digital transformation but the one that concerns us here is that we are increasingly jettisoning the concept of paper-based records. This is mostly happening without much consideration of societies need for “long lived artifacts” in order to create trustworthy repositories of knowledge.

So far, we have mostly got away with it because we have always had good old fashioned paper records that can be referred to if any “copy” in digital form becomes obsolete (“bit rot” or has questionable provenance (“digital fakes”).

This is recognized as a growing problem but no silver bullet appears to be on the horizon. Meanwhile, the AI revolution continues apace and this has made the need for reliable “ground truth” records more pressing than ever before. It is not a stretch to say that we are reaching the point where we cannot trust any digital document, any audio, any video as being real any more. AI is becoming scarily good at deceiving us and scarily good at just making stuff up.

If that wasn’t enough, there was recently a hack on the internet Wayback Machine. In principle, an attacker could take control of, say The Wayback Machine or WikiPedia and continue to serve up pages but doctor those pages behind the scenes and we would never know. Put another way, there is now a clear and obvious digital attack vector for any bad actor wishing to essentially “rewrite history”. So much for legacy!

The good news is that there are signs at last that this problem is increasingly being taken seriously. The model law known as UELMA (Uniform Electronical Legal Materials Act) in the USA for example, is encouraging governments to think through how the electronic versions of their laws can be considered “authentic”. In other words digitally introduced as evidence in courts of criminal law, in an age where there is no “master version” on the shelves of the Law Library if counsel for the defense casts doubts about its authenticity.

While all this shakes out, if there are aspects of your intended legacy that are digital in nature – including the documents-of-record that attest to your legacy – I’d suggest printing them out on preservation grade paper with non-fugitive inks and keep them in a fire-proof safe or in a safe deposit box and give your estate the keys.