Instrumentation & Measurement Magazine 26-4 - 5

fundamentalsmeasurement
of
Luca Mari and Dario Petri
Measurement: The Spread of Social Knowledge
T
his paper is devoted to introducing measurement as
a process aimed at producing knowledge that is socially
shareable. Some subjects of the paper are the
metrological system, metrological traceability, and calibration
as tools to guarantee the intersubjectivity of measurement
information.
Introduction
As introduced in the previous paper of this
series [1], measurement aims at producing
data, and then information, and then
knowledge, about empirical properties
of objects (we are using the term " object "
in the generic sense of anything that has
properties, thus including physical bodies,
systems, phenomena, processes, and so
on - along the same line, ISO terminological
standards define 'property' as " feature
of an object " [2]). In a semiotic framework
[3], data may be accepted as something that
is given, thus according to the meaning
of the Latin term " datum " , what in English
is sometimes called " evidence " . But
data is only about the recognition of differences:
it is this, and it might have been that:
hence, a binary digit is the simplest case of
data. It becomes information if it is referred
to something (e.g., 0 stands for " no " and 1
for " yes " ), and then possibly knowledge, if
the information is embedded in a context and acknowledged
to have a usefulness. Given that measurement data originates
from and is about empirical sources, such a process of interpretation
and contextualization cannot be taken as evidence in
turn, and this generates a substantial issue of quality of the related
information and knowledge [3].
This happens even in the case of simple measuring instruments,
like an alcohol thermometer (a lexical note: we treat
here " measuring instrument " and " measuring system " as synonyms,
and use only the former). When operated according to
the expected procedure, the thermal contact of the thermometer
with the object under measurement makes the alcohol flow
in the tube and its upper surface reach a position, more or less
aligned to an etched mark. The instrument is designed to make
June 2023
the identification of the mark as unproblematic as possible, to allow
us to treat the identified mark as evidence, and therefore an
example of measurement data. For sure, some reading mistakes
could always be made, but that the alcohol reached a given position
is something that may be taken for granted. It should be clear
that there is nothing special in this example: were, as another
example, the thermometer based on a thermocouple, the instrument
output would be a voltage instead of a position, but the
structure of the arguments that follow would remain the same.
Such a physical transduction is supposed to causally connect
the temperature of the object to the position of the upper surface
of the alcohol in the tube. Hence, this setup allows us to interpret
alcohol positions as standing for object temperatures, and
therefore conveying information on them: accordingly, while
the observation that the alcohol reached, say, the twenty-third
mark is a measurement data, reporting such a mark as referred to
the temperature that triggered the transduction, thus as
temperature of a given body → twenty-third mark
of a given thermometer
(1)
is an example of conveyed information, where in fact the arrow
has to be meant as something like " is represented by " . This
process, called pre-measurement in [4], is structurally straightforward
but has the obvious drawback that its results are not
transferable, since they convey information that is correctly
interpretable.
◗ only in reference to the given instrument, thus implying
that the information about the temperature produced by
different instruments is not comparable, and
◗ only under the condition of stability of the instrument, so
that any change in its transduction behavior makes again
the produced information not comparable.
Of course, the usual strategy is instead to report measurement
results on a public scale of the measured property, as for
example obtained in reference to degrees Celsius or kelvin in
the case of temperature, and this offers at least two key benefits,
as it produces information
◗ that is interpretable independently of the instrument by
means of which it was generated, and
◗ that can be reported as a comparison, and in particular
as an equation, between the measurand and a measured
value.
IEEE Instrumentation & Measurement Magazine
1094-6969/23/$25.00©2023IEEE
5
Instrumentation & Measurement Magazine 26-4

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 26-4
