Information Technology Reference
In-Depth Information
types or data formats that are used by the available services would clearly
leverage the usefulness of workflow management systems. Just like strong
data typing in programming languages allows for the detection of erroneous
function calls, the thorough semantic characterization of service interfaces can
support the construction of type-correct, executable workflows [301, 128].
Fig. 1.7
Exemplary semantic type classification
Figure 1.7 shows an exemplary classification scheme for graphics format
types, which can be distinguished into
raster formats
(i.e. bitmaps such as
jpg
,
png
and
bmp
),
vector formats
(such as
cgm
and
svg
),
stereo formats
(such as
imo
and
pns
), and
compound formats
that can contain both pixel
and vector data (such as
ps
and
pdf
).Basedonsuchtypeclassifications,it
is possible to assess the compatibility/incompatibility of services, given that
they are properly annotated in terms of the domain-specific type vocabulary.
Fig. 1.8
Semantics-based compatibility assessment
As Figure 1.8 illustrates, this does not only allow to compare data types
directly via concrete instances (e.g.,
jpg
is compatible with
jpg
, but not with
svg
), but furthermore with awareness of their semantic categorizations. For
example, the ImageMagick services can deal with all kinds of bitmaps and
may thus be annotated as accepting
raster format
files as input. A service
that loads a picture in
raw
format from a camera can then be combined
with the
polaroid
ImageMagic service, as
raw
is a
raster format
. In contrast,
Search WWH ::
Custom Search