It was the first time I met a researcher from SAP Research Germany. Of course I knew from their collegues from the United States, Australia, France and other countries because we see them on Demo Jam stage and they write lots of blogs about their activities. It seems to me that their work is quite transparent. I didn’t do comprehensive investigation but it seems to me that SAP Research Germany is not that visible within the community. But I believe that this getting better. In fact SAP Research Dresden put articles on SCN like Semantic Product Memory and in fact they invited me to discuss another project called “Aletheia” in SAP TechEd Berlin.
I really appreciate that Frithjof Dau offered an expert community session about Aletheia: http://www.aletheia-projekt.de/ Aletheia means “Truth” and is a product information system that offers access to product information using semantic technologies. It was very interesting to see Aletheia at work.
Every company offers products to customers and the information are spread on different documents, databases, forums and so on. For me the most important question was how semantic technology can offer more better ways to integrate the data sources compared to traditional mashups.
Getting Inspiration: From Customizing to Business Semantics
Another interesting thing I learned that an ontology can have the role as a central connector in a software system. For this leads to the question, whether old fashioned way of software parametrization known as customizing could be replaced by ontologies. For me this has some remarkable advantages:
- Ontologies support documentation – think of comments for example.
- Ontologies are extensible for customers who can link their own properties like links to specifications and other artefacts.
- We can prove consistency of customizing using inference techniques.
- Within ABAP Application Server parametrization is represented using customizing activities which can be maintained in implementation guideline (transaction SPRO).
The most important thing is that we should define try to connect isolated customizing activities in different SAP modules (FI, CO and so on) to get an overall view.
Semantic Interoperability is a Timeless Software Property
You may ask yourself why this is important. For me there are to obvious reasons:
- Any kind of software parametrization raises the complexity of a system and increases TCO. Getting more information (even comments) into this activity makes it more transparent and explains the wanted system behaviour.
- Another reason is that customizing in customizing tables has its own semantics using value helps or more general: dependencies between database tables. In my opinion this is much too technical and often lacks business semantics.
The second aspects gets more clear if you see it in the light of Vishal Sikka’s concept of timeless software: even after profound redesign of an application there must be possibility of tayloring the system and orchestration of its interfaces to other systems. So in my opinion we should ask ourselves whether customizable properties of a system should be expressed in a timeless manner.
As I can see the research strategy on SCN the strategy on semantic interoperability focuses on BPM and automated composition techniques of services but I think there much more applications for semantic technologies to reduce TCO and achieve more agility.
SAP Research and the Ecosystem
I have to admit that my idea of expressing customizing using ontologies is not completely thought through and I have no working prototype. But in my opinion it is an example of an idea that could be evaluated in the future.
And now I come to the last point of this blog entry: If research should serve a business enabler then business and research must come together. In fact SAP Research is well prepared:
- SAP Research has its place on SCN.
- There’s a Wiki space where you can download prototypes.
- There is a list of publications.
But if we want to achieve synergy SAP Research should try to get in contact to people who are both down to earth and not afraid of innovation. In my opinion you can find those people on SCN Community events like SAP Inside Track. So my suggestion is that more researchers should visit those events and especially get in contact with SAP Mentor Initiative. They are a friendly crowd that loves to discuss new ideas and I think within the US they already evaluated some prototypes and gave feedback.
In my opinion both sides can benefit from this approach: people from SAP Research get in contact with consultants, developers, enterprise architects and learn about their problems and get feedback. And I’m convinced that many IT experts would appreciate the contact to R&D and this could be a starting point for cooperation, too.