Interview with Charles Ivey (January 1998)

In an exclusive interview, Charles Ivey, CEOof the newly formed PI/Dwights/Petroconsultants group told PDM of the way the new combinedcompanies would be working in the future.

PDM - How will you approach the future, what changes are in store for the two companies, will they merge?

Ivey - We will continue to have two national flags - Swiss and the USA. In the longer term we are seeking to have a world-wide perspective. Already our customer base overlaps to a considerable extent with perhaps up to two thirds of our present customers having subscriptions to both Petroconsultants and PI/Dwights data services. We therefore need to be particularly aware of the way in which our market is segmented between domestic and international customers. These different needs must be addressed with appropriate technology and data services. Because of our position we will be extremely focussed on providing our clients with standardized solutions, with a view to improving data availability and reducing cost.

PDM - will the IRIS21database from Petroconsultants be subsumed into the Dwights/PI P2000 database in the medium term?

Ivey - I would not say that at all. In North America, it is the market that has driven PI/Dwights to the adoption of the PPDM data model. Elsewhere in the world where the basin and license paradigm is more prevalent, Petroconsultants' IRIS21 data model has been very successful. IRIS21 has "boxes" for just about everything, and with the PROBE subset is a very comprehensive solution. We will be very open with all our clients and we will attempt to find the best solution to deliver all their data - whatever that solution might be.

PDM - You mentioned standards, how do you see POSC's Epicentre in this context?

Ivey - One thing is for sure, we do not believe that one solution can fit all needs. We are very keen on the concept of late binding, whereby objects are composite and created "just in time" rather than "just in case". So in a real-world solution, one could imagine both PPDM and POSC based data models at a client location - which would be integrated through object technology much in the same way as the web does this.

PDM - Both PI/Dwights and Petroconsultants have done a lot more than just deliver data for quite a time now. You could be said to be encroaching on the "data management" field of the GeoQuest and Landmark Graphics of this world. Seen from the client view point, this situation could lead to a multiplicity of browsers and data access engines being necessary to build a project. Whose data browser and data delivery tools will prevail?

Ivey - We have no desire to compete with Landmark or GeoQuest. Our business is in data delivery and management, browsing and selecting. For a company with another system, there are two ways to move your data around - by push, or by pull technology.

Ivey - Pull technology is the way things generally work today. If you fire up your data management system and it goes looking for data in someone else's system, then that is pull technology. If however you fire up and then it is the server system that provides the data - on its own terms as it were - then that is pull technology. The difference is fundamental to us - especially in terms of interoperability with other vendor's tools. We want to be able to push our data into anyone's system and we can do this. On the other hand, a request to pull data from our system can quickly turn into a nightmare for us, as there may soon be hundreds of different ways of accessing our data all of which we must develop and support. We are currently working with Landmark on pushing data into Open Works projects. Ultimately this will be achieved through a middleware layer using business object type technology, but I believe that is unlikely that there will ever be a single way of achieving this. Site specificities and the market mean that we have to and will provide the data transfer mechanism that is right for the job in hand.

PDM - Both Petroconsultants and PI/Dwights have a long history of actually using data, how do you intend to build on that in the new group?

We have been working for some time on some new software - Powertools - which would be categorized as intelligent data retrieval with processing on the fly. So that for instance an engineer selects a group of wells and decline curves and perhaps net revenue are computed using company defaults - so that we are delivering data plus its meaning. This is a similar approach to that used in Microsoft PowerPoint where the image selection tool displays a thumbnail of images as they are being browsed in the file locator. In this context we are working on an intelligent tool that performs statistical analysis on the fly upon selected data. Most software vendors have underestimated the data problem. I firmly believe that it is data that is the heart beating inside the exploration and production business.