*************************************************************************
		Report file for COMP2800 Session 7, 2001
	    Mr Simon Kravis, CSIRO's Genesis 2000 project, Aug 20
*************************************************************************
* name		: A. G. Student
* student number: 3399999

0. FEEDBACK  (after the`:', please enter an integer from 0 to 5;
	     `0' = strongly disagree, `5' = strongly agree)

* Q1. Generally, I understood the material presented.	: 4
* Q2. I found the Session interesting.			: 2
* Q3. I think the Session was potentially valuable.	: 3

* Q4. Please enter any other comments relevant for feedback below:


1. SUMMARY OF THE SESSION (200-300 words)

The CSIRO's Genesis project is a software system used for oil and gas
well planning and analysis. The software is categorised as dinosaur
software.

Dinosaur software generally...

   - Is large and complex (over 1 million lines of code).
   - Is specialised.
   - Has a long lifetime (over 10 years).
   - Is expensive (Genesis costs $10000 US per seat).

The Genesis project was conceived in 1989 by CSIRO. IT has come along
way since then. It has a large development team made of professional
programmers, scientists and some students (mainly engineering).

The system consists of 90% Visual Basic code, 5% Visual Basic
Application for Excel code and 5% Fortran. It uses about 20 3rd party
components. The system uses Microsoft Access has its database
application. It is currently made up of 75 compiled object and 150
tables.

The system is used to manage well data, analysing well data and for
planning new wells. Some of its distinguished features include a QC
process for data and the ability to identify trends.

CSIRO has sold 50 seats in 6 oil companies. CSIRO owns the
intellectual property rights to the Genesis project. Noble Drilling,
however, have bought the commercial rights. Maintenance is performed
by Spektl, a CSIRO spin off.

Some problems that the project is working on currently are migrating
to a client/server model. Dealing with the concentration of
expertise. This involves getting programmers to understand the science
involved and getting scientists to understand the programming
involved. The last problem is common to most software
developments. That is, dealing with the lack of testing and
documentation.