1
00:00:09,000 --> 00:00:13,400
The National Archives is at a pivotal moment 
in it's history.

2
00:00:13,500 --> 00:00:20,100
We've been a physical archive for 180 years 
and we're becoming a digital archive.

3
00:00:20,200 --> 00:00:28,800
Now many of the kinds of challenges, questions 
that we're running into are things that we 
don't yet know how to do.

4
00:00:28,900 --> 00:00:37,200
They're research questions, new ways in which 
we need to figure out how to preserve records, 
how to make them available.

5
00:00:42,000 --> 00:00:46,800
Almost everything that we're doing as a digital 
archive gives rise to a research question.

6
00:00:46,900 --> 00:00:54,500
Whether its how we preserve novel or more 
complex forms of digital object, particularly 
in the age of machine learning.

7
00:00:54,600 --> 00:00:59,400
How do we use machine learning technologies? 
How do we preserve large-scale neural networks?

8
00:00:59,500 --> 00:01:12,400
We're continually confronted by questions 
that mean we need to research our practice 
and our approach to delivering our work as 
an archive.

9
00:01:16,800 --> 00:01:22,400
So to deliver our research programme, The 
National Archives needs to work with our 
established partners.

10
00:01:22,500 --> 00:01:28,600
People particularly looking at things around 
digital humanities and how we make use of 
our digital collection.

11
00:01:28,700 --> 00:01:33,500
But also reach out to people who archives 
perhaps haven't worked with so closely in 
the past.

12
00:01:33,600 --> 00:01:38,400
People who are in computer science, but also 
in disciplines like mathematics.

13
00:01:38,500 --> 00:01:45,400
Especially when we're looking to explore 
questions like uncertainty, how we manage 
probability in our data.

14
00:01:45,500 --> 00:01:49,800
The maybes that are suddenly informing our 
world as a digital archive.

15
00:01:57,100 --> 00:02:07,100
One of my great privileges as a civil servant 
was to have the opportunity to be the principal 
investigator for the Big Data for Law project.

16
00:02:07,200 --> 00:02:14,300
And we were exploring really for the first 
time how we might apply large-scale data 
analytics techniques

17
00:02:14,400 --> 00:02:19,200
To understanding how our system of law, the 
statute book as a whole operates and is evolving.

18
00:02:19,300 --> 00:02:29,100
It gave us real insight, both in that particular 
sphere but also more generally about how 
we can apply big data approaches to a part 
of our collection.

19
00:02:29,200 --> 00:02:34,700
We were very fortunate to win funding from 
the Arts and Humanities Research Council 
for that project.

20
00:02:35,200 --> 00:02:41,500
And it really gave us the opportunity to 
do something that otherwise we wouldn't have 
had the chance to do.

21
00:02:41,600 --> 00:02:48,200
The Traces through Time project was the first 
opportunity that we've had in the archive 
to explore the concept of probability.

22
00:02:48,300 --> 00:02:54,900
And we were looking to try and identify the 
connections between people named in our catalogue.

23
00:02:55,000 --> 00:03:03,900
But rather than be sure that this person 
is that person, we're just trying to get 
a likelihood. What's the percentage chance?

24
00:03:04,000 --> 00:03:12,400
Now probability is one of our big research 
areas for the future and Traces through Time 
was a leg up into that world.

25
00:03:12,500 --> 00:03:19,200
And has given us some really vital lessons. 
It's also allowed us to improve the service 
that we deliver to the public.

26
00:03:19,300 --> 00:03:28,800
Through our Discovery service on our website, 
where if you find a record that lists a name 
you can go and find other records that relate 
to that person

27
00:03:28,900 --> 00:03:35,400
Based on how likely it is that the name in 
the first record is the name listed in some 
of the other descriptions.

28
00:03:40,900 --> 00:03:46,900
The ARCHANGEL research project is really 
breaking new ground for The National Archives.

29
00:03:47,000 --> 00:03:59,800
We're looking at a very hot technology, blockchain, 
distributed ledger technology and how we 
might use it in the context of the digital 
archive.

30
00:03:59,900 --> 00:04:09,200
We're working with people that we don't normally 
work with and exploring the cutting edge 
of digital archiving practice.

31
00:04:09,300 --> 00:04:17,100
It's exactly the kind of research that we 
want to do more of in the years ahead and 
that we're going to need to if we're going 
to be successful as a digital archive.

32
00:04:17,200 --> 00:04:27,900
ARCHANGEL is lead by the University of Surrey 
and its being funded by EPSRC, and its the 
first time we've been involved in a project 
that's been funded by EPSRC.

33
00:04:28,000 --> 00:04:36,400
And we're really pleased to be working with 
academics who are specialising at the forefront 
of computer science.

34
00:04:36,500 --> 00:04:43,300
And also looking at some of the economic 
questions, the sustainability of digital 
archiving, as part of that research project.

35
00:04:48,000 --> 00:04:57,800
Research is really key to our work as a digital 
archive. Our Digital Strategy sets out the 
ambition for us to be a disruptive digital 
archive.

36
00:04:57,900 --> 00:05:05,600
That means developing new ways of preserving, 
contextualising and presenting the digital 
record of government.

37
00:05:05,700 --> 00:05:15,700
Research is absolutely crucial to our work 
as a digital archive, as we're breaking new 
ground and building our capability for now 
and the future.