18 
ms. 
le 
nd 
7. Will the computer output provide all the information wanted, 
and in a form that is acceptable? 
8. Are the drawing styles acceptable and will the drawings 
reproduce satisfactorily? 
9. How does the system cope with compound curves, junctions, 
curve interference, changes of alignment and similar 
complications ? 
10. How much manual intervention is necessary? 
11. What are the inbuilt safeguards against error and are they 
sufficient? 
12. Does the user retain effective control or is the system so 
complicated that his needs become subordinated to it? 
13. Is a rapid turn-round guaranteed? 
14. Isthe service charged on à cost-plus or fixed price basis? 
These questions and many more must be raised and inevitably some of 
the answers will be unsatisfactory because every system is developed as a 
compromise between the conflicting requirements of potential users. Ifa 
system accords in general with the user's principal requirements it is wiser 
to adapt the remaining requirements to the system, than to insist that the 
system be modified. Because of the interdependence of so many of the 
computational processes, what might appear to be à simple modification is 
often found to have unsuspected repercussions, putting in doubt both the 
reliability and the efficiency of modified programs. 
CON CLUSION 
Clearly it would be futile to embark on à computer -aided design 
project without most careful and detailed planning, requiring the co-operation 
of all concerned. In this the photogrammetrist as provider of one of the 
principal supporting services will play a major part, and the success or 
-14 -