LEELELETT4II 
nding 
1 and 
ordan 
Metric 
netric 
aphy. 
small 
cord, 
)iece. 
, A, 
agery 
ch to 
ping. 
1sing, 
1996. 
high 
The 
digital 
Irvey, 
^ the 
3-D 
In 
n, H. 
sonal 
mote 
  
DTM System SCOP in a New Technological Generation 
L. Molnar, J. Wintner, B. Wóhrer 
Institute of Photogrammetry and Remote Sensing, The Vienna University of Technology, Vienna, Austria 
ISPRS Commission IV, WG 4 
KEYWORDS: DEM/DTM, Software Development, System Integration, GIS Integration 
ABSTRACT: 
According to the "Principles for a New Edition of the Digital Elevation Modeling System SCOP" [Mo92] presented at the 
Washington congress of ISPRS, a cross-platform, object oriented application frame has been created including GUI, 
graphics, and client-server components. Some major SCOP functionality has been transferred to this frame already, and 
new functionality has been added (in the first line Topographic Data Management [Ho96], and subdividing the model 
surface into regions). An introductory overview of the system is given, intended in the first line for qualified and 
experienced DTM/DEM specialists. 
1. INTRODUCTION 
At the Washington Congress of ISPRS, "Principles for a 
New Edition of the Digital Elevation Modeling System 
SCOP" have been presented [M092]. At the very least full 
ten man-years of R&D have been invested into realizing 
those principles. An application frame ('XX', see chapter 2) 
has been developed to dissolve "the contrast between 
merits and disadvantages of [university-based] research 
programs" as compared to commercial products - 
"concerning, in the first line, object oriented design, 
graphical user interfaces, interactive graphics, etc.". XX 
provides for client-server applications fully object oriented 
in their design; in practice, however, these applications 
realize an architecture that can be best classified as 
"componentWare" (see chapter 2). 
Created under XX, the component "Topographic Data 
Selector and Editor" of SCOP.TDM (Topographic Data 
Management, the successor of TOPIAS) is now in beta- 
test [Ho96]. A pre-alpha version of SCOP.DTM integrating 
under XX some of the major components of the SCOP 
system, is demonstrated at the workshop "Advanced DTM 
Technology" at this congress. 
The paper cited above [Mo92] is concentrating on R&D 
issues. This paper is one step closer to application: it is 
intended as a first introductory overview of SCOP under 
XX. 
2. ARCHITECTURE 
2.1 Versions 
Beside the stand-alone versions, versions to be integrated 
into/with host environments are developed. Languages 
Supported by the first edition(s) to become available are 
going to be English and German. 
All versions support three modes of operation: 
GUl-mode (including also command-line entries), 
- batch-mode to run unattended jobs, and 
slave-mode when controlled by host systems. 
2.2 Object Orientation 
A prediction that proved to be far too optimistic [Mo92]: 
“On a fine-grained level, with object oriented operating 
systems .. and with widespread processor support, 
distributed intelligence, and parallel processing - on this 
level, object oriented programming is not yet with us. For 
this, | think, even developers have to wait for maybe five 
years". - One cannot create any CAD class to inherit from 
AUTOCAD so to customize its functionality, and there is 
no Arcinfo consisting of distributed objects with virtual 
member functions - e.g. for digital terrain modeling that 
could be augmented by SCOP DTM class objects with 
members adapted correspondingly. Taligent Inc. was set 
up by IBM and Apple to create such a world - it seems to 
have failed, however, for technical and for commercial 
reasons [Co95]. "Object-oriented computing has failed. But 
component software ... is succeeding" - a headline of Byte 
Magazin (May 95). Parallel to this ISPRS congress, the 
first Component Users Conference is taking place in 
Munich, organized by the ComponentWare Consortium, 
supported by, and with speakers from IBM, Siemens, and 
other major companies; the issues are such as 
"Components: Another end to the software Crisis ?", 
"Evaluating Concurrency Models for CORBA Servers", 
“Survey of Data Management using CORBA", or "IBM 
OpenClass Component Framework". 
Components in this sense are (large) modules with 
standard interfaces allowing to compose them into 
systems. Inside, components may be (and increasingly 
are) object oriented, but they may also be written in any 
other way, using, e.g. procedural languages and 
structured programming techniques - or they may be 
hybrid in their design. These components are in most 
cases controlled by means of "object broker techniques" - 
i.e. as entities showing object oriented features on their 
surface but being not necessarily object oriented in their 
internal architecture. - Beyond doubt, this is exactly what 
'we termed "autonomous classes" [Mo92]. 
569 
We suspect, however, that the situtation declared 
aggressively as "failure" of large-scale object orientation 
above is just due to impatience, to commercial and 
journalistic ^ hectics. We tend to consider the 
"‘compnentWare contraRevolution" to be a transitory stage 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996