[Patina] Re: Patina Scenario

[Tom Frankland]

Hi everyone,

As promised, here is my scenario. As Graeme has already mentioned,
there is a lot in common with his modelling scenario:

1.       An archaeologist wishes to reconstruct a specific structure.
2.       They upload a floorplan of the structure they wish to
reconstruct into the procedural modelling software.
3.       They then choose an appropriate procedural rule-set based on
the estimated age and style of the structure they are reconstructing.
4.       The procedural modelling software then generates several
thousand possibilities for how the structure may have looked. It
outputs the values for each of the various parameters modified in each
reconstruction as metadata which is linked to each individual
structure.
5.       The archaeologist can then use an appropriate interface (such
as gestural/projection system, or a tablet or laptop) to begin to sort
through the structures and associated metadata. There are numerous
ways for the archaeologist to sort the data. They could choose to
focus in on a desired structural feature, for example, ‘only show me
structures with this style of roof’, or eliminate certain structures,
for example, ‘I don’t like this style of door’. Both actions help to
reduce the dataset in size towards a more focused interpretation.
Alternatively, they can sort through the data based on more abstract
values, such as the feasibility of a structure. In this way they can
rule out structures which appear to only be tenuously linked to the
evidence, or may be overly complex, such as suggestions for grandiose,
decorative buildings when the structure in question is in an
industrial or residential context.
6.       As they sort through the various structures and associated
data, each of the actions they take to reduce or sort the dataset is
recorded. This can then be consulted afterwards to see how the
archaeologist approached and understood the data.
7.       Should they wish to, an archaeologist can return to an
earlier stage in the interpretive process, to examine what could
happen if they had reduced or sorted the data in a different way.
8.       The final point for the process is a selection of suitable
structures. The process would never produce just one possibility, as
this would contradict the fundamental idea of the process, that
archaeological data can be interpreted in numerous ways. Another
suitable output for this process would be the process itself, showing
the decisions made by the archaeologist rather than a final image.

As I was writing this I was very aware that many aspects of this
approach have been chosen because I can imagine or can conceive a
potential way of achieving them. One concept both myself and Graeme
would love to see would be a procedural system that, based on user
input, could generate new structural possibilities in real-time, but
this is beyond my knowledge of what is currently possible. I thought I
would mention it here though so you have an idea of how the scenario
could possibly be extended. Equally, it would be great if this
scenario were to involve collaboration, so for example, two
archaeologists could refine and sort the data at the same time.

See you tomorrow,

Tom.