22 
ice. How- 
sified and 
;liing new 
>f software 
3. The re- 
mfortable 
iry scopes 
ositioning 
y order of 
for direct 
ry effects. 
is (GIS). 
presenting 
id remote 
revolutionary if compared with the situation of 25 years ago. The changes «are not only drastic 
by degree and magnitude, they also represent change of concepts and operations on new 
conceptual levels, with much wider scope and application, interfacing with other disciplines to a 
previously unknown degree. Such developments seem to follow a certain pattern which is typical 
for the general features of automation and of systems and which explains to some extent the 
revolutionary effects. 
The pattern implies that technical progress often goes through 3 steps, each one leading to a new 
dimension. We have seen that the developments in photogrammetry started first on a purely 
technical level, attempting to improve previous performance by making use of new technical 
tools and by developing their application. There may be further reaching anticipations already 
on that level, but usually the first phase is a struggle with the gradually evolving technical 
possibilities. 
As soon as the technical problems of the first step are solved, then there is a certain explosion, 
as tasks are solved or come within reach beyond the original scope. In other words, the power 
of the tools and the related methods push into new thematic expansion, either by increased 
complexity and increased economic performance, or by the sheer new accessibility of tasks pre 
viously out of reach and therefore often out of concept. 
ke GIS. It 
al plotter 
graphics, 
, starting 
ose range 
uutionary 
it the de- 
i way our 
iipiies not 
le field of 
laps, that 
ore auto- 
;h drastic 
«ore quite 
Many developments go through a third step by merging into complete and totally new systems, 
with a new paradigm. On that level the original starting point is not recognizable any more, as 
fusion with other techniques and methods has taken place, and the system may have outgrown 
even the original discipline entirely. 
3.2 A few examples may illustrate that scenario. We have reviewed that aerial triangulation, 
analytical plotters, digital terrain models, computer assisted plotting took quite a long time 
and extended efforts to accomplish the first step, i.e. to establish themselves on the first level 
of improved technical performance. During that struggle the extended concepts were gradually 
clarified and the breakthrough prepared. 
By successfully mastering the first step, and by solving the problems in a rather general approach, 
suddenly new qualities appear, leading to new more demanding applications. In this way the 
block adjustment for aerial triangulation spread out into high precision cadastral or close range 
photogrammetry, as well as into photogeodesy, apart from killing the classical first order analog 
instruments which were designed for strip-triangulation. Computer assisted plotting blossomed 
into digital mapping. Digital terrain models and orthophotos established themselves as new 
products in their own right, forming the base for multiple derived products. 
There are also examples in photogrammetry for developments which pass beyond the second 
step by producing or making part of larger systems which as such had not existed before. They 
imply a completely new look at tilings, with new demands, new requirements, new results. If 
it will be possible, for instance, to measure all external orientation parameters of photographs 
or sensors-accurately enough (with GPS we are partly on the way) then all indirect orienta 
tion methods in photogrammetry, aeri.al triangulation included, might become obsolete, which 
have dominated the working methods for more than 60 years. The real examples of change of 
paradigm are, however, the steps from digital mapping to geo-information systems, from mul- 
tispectral classification to complete object restitution, and in particular the step from present 
day digital image processing to «automatic image analysis and to computer vision, including 
knowledge based systems and artificial intelligence as most powerful disciplines.