Transcript Document

Grand Designs:
reflections on
archaeology, the historic environment
and the E-science programme
E-science? Eh?
Collaboration?
Data processing
Can archaeology use these tools?
Dr William Kilbride
Is E-science different?
Co-ordinated problem solving
Resource sharing
Virtual organisations
Access to scientific instruments
Grid: data, access, computation
Hype?
Mystery?
Gartner’s Hype Cycle:
Helping William understand the
E-science programme
Jargon is power!
Helping William understand
the E-science programme
Exclusive language
Not always clear what projects are
doing
Computing as discourse
Standards should be controversial
Political / industrial drivers
Hidden (overt) expectations of what
arts and humanities are about?
In spite of all that …
Can we use E-science tools to
be better at archaeology?
Taken as read:
Focus on archaeology needs
Access grid for meetings etc
Moving large objects
Collaborative tools meets data grid
Computational tools meets grid
Need to articulate needs
Need to be understood
Collaboration in espace
Sharing data across multiple computers
Not just http
Distributed file stores
Federated databases
Multi-site use
Big (and small) research
Channel Tunnel Rail Link
Heathrow T5
Stansted
M74 …
OASIS as an example
OASIS:
what it used to be like …
Fieldwork Print out
unit
Post
National MonumentsBacklog
Record
Backlog
Local
Govt.
Post
Print out
In an ideal world the machines
should do the talking …
Fieldwork
Print out
Post
Backlog
Local Archive
govt
National
agency
Backlog
Post
Print out
But at each point where data is keyed in,
it is validated by experts …
Local govt: local knowledge
Is this what it claims to be?
Do we have monuments like that here?
That’s the wrong parish name
That field unit is no good
National Agency: national standards
It’s not MIDAS compliant
It’s not like other records
The terminology is different
That SMR is very good
So need to capture the validation
process but eliminate the drudgery
National Monuments Records
Fieldwork
OASIS record and report
Local SMR
ArchSearch
Strengths and
weaknesses …
Shared standards development
It works!
Inherently collaborative
Roles and responsibilities clear
More records in the pipeline
Backlog and the rest of the country
DC as big stick
Backlog bigger than front-log
Resolves duplicates
Quality of grey literature
Persistent and pervasive
Quality of DC archaeology
Only grey literature
Not weaknesses:
Geophysics? Survey?
Areas for growth!
Only DC archaeology
ac.uk? DNA? C14? Dendro?
Closed process
Import and export issues
Only UK
Sharing data not processes
Single data source
Computational
power?
CPU resources from different
machines are used to address a
single problem:
Desktop scavengers or server grids
Computational
archaeology
Geo-computation: viewshed (etc)
Geo-temporal computation?
Visualisation and recording
VR applications
Data mining and processing
Simulation?
Conservation and management
Others
Content-based image retrieval?
Video and audio?
William’s data mining example
How does it work?
Knowledge Map (classification)
Gatherers
Document Set (XML chunks)
Text Mining Tools
Creates …
Adaptive Concept Map
Every possible query already processed
Click and browse instead of type and hope
Visualise the distribution of data against concepts
What’s the problem?
Scalability …
Huge numbers of possible combinations
5000 concepts and 1,000,000 documents
= 50,000,000,000 possible combinations
Conceptually simple
Huge memory overheads
Space is infinitely refined
Grand Designs:
can archaeology,
use the tools of the E-science programme
Of course we can!
Only scratched the surface here…
Need to concentrate on archaeology
Hype and jargon are problems but …
Collaboration
Computer processing:
… both have a role in archaeology
Dr William Kilbride