On Jun 12, 2007, at 3:53 PM, [EMAIL PROTECTED] wrote:
Hi Waclaw,
Matthias, if you look carefully at BFO, you'll see that roles are
entities. This means that evidences, as roles, are entities.
Of course. I just wanted to differentiate that an experiment is not
an instance of any class called 'evidence' (in other words, an
experiment 'is not' evidence). Instead, it should be associated
with an 'evidence-role'.
The only problem with this is that roles inhere in continuants rather
than in occurrents. One way around this is not to say that evidence
is an experiment, but rather the results of an experiment.
In the protocol application branch we've been discussing the idea of
"proxy". A common pattern in biological experiments is that you want
to measure one thing but in fact you measure another. In one of our
use cases, we for instance look at chromium release cytoxicity assays
(https://wiki.cbil.upenn.edu/obiwiki/index.php/Chromium_Release_assay)
In this case the desired measurement is the fraction of cells in a
cell culture that have died.
The cells are treated with a radioactive substance that is absorbed
into their cytoplasm, only to be released when they die and the
membrane is broken. So if you start with fresh medium and these
cells, you can estimate the amount that have died by measuring the
radioactivity of the medium.
In this case we plan to say that radioactivity of the medium is a
proxy for the the number of cells that die.
Now if you ask what the evidence for the cytoxicity estimate the path
leads via inverse proxy relations eventually to physical properties
that are measured by some instrument. If we were doing an analysis of
what could undermine the conclusion, each of these steps could be
questioned.
In this case the inference isn't quite as mysterious - all of it is
bottled up beforehand in the statement of the proxy relations.
-Alan
cheers,
Matthias
cheers,
Matthias Samwald
----------
Yale Center for Medical Informatics, New Haven /
Section on Medical Expert and Knowledge-Based Systems, Vienna /
http://neuroscientific.net
--
Psssst! Schon vom neuen GMX MultiMessenger gehört?
Der kanns mit allen: http://www.gmx.net/de/go/multimessenger
