Skip to Content

The last days have been very interesting:

A glimpse on Standards and Open Source Tools

Why is this important? Let’s have a look at semantic standards and tool support. One basic principle of semantic techniques is the use of standardized vocabulary like RDF and OWL which are standardized by W3C. This techniques allow to code knowledge in a very sophisticated way so that it can be analyzed with tools from artificial intelligence.

This sounds really so why does not everybody work this way? The most important reasons is the complexity of this approach which I explain in three examples:

  • RDF is hard to produce. RDF consists of triples that build small sentences in the form subject, predicate and object. Because the object can be the subject of another sentence the result are complicated “nets” of statements. Unfortunately these statements are hard to produce manually even if you choose editors and a neat syntax like N3 or Turtle. One problem comes due to the fact that knowledge has aspects like time or place of an certain statement. In these cases it’s common pratice to introduce additional nodes .
  • Querying RDF you need additional ressources like RDF datastores for persistence and query engine linke SPARQL.
  • Ontologies are very complicated artefacts and should be created using tools like Protégé.  But nevertheless these are tools from academia which are only useful for key users. Protégé is extensible so you build you own plugins but this will take additional time.

So there is a variety and standards and Open Source tools which you can use for Semantic Applications but if you want to use them in the enterprise you’ll need some effort to create applications that are user friendly and easy to use.

From Academical Tools to Enterprise Software

There are companies who fill that gap. On the hand they offer solutions that can ran stand alone but they offer services & consulting to customize these tools to your needs and integrate them into you enterprise architecture. In another blog I’ll provide a concise overview about the economical potential but here I’ll only mention some companies whose tools I’m using frequently: Open Link Software, Clark & Parsia and ontoprise.  Even cloud based solutions are coming like Dydra which offers RDF as a Service.

In FindGrid SAP added tools of intelligent views to their solution integrated them with Enterprise Search and used NetWeaver Composition Environment as platform which shows the potential of CE especially for Cutting Edge solutions.

From Business Experts to Knowledge Managers

Unfortunately I could’t get my hand on FindGrid but I have some basic knowledge about K-Infinity of  intelligent views. It has a component called knowledge builder that allows to define semantic nets that are similar to ontologies. These allow you to define domain models, entities and relationship between them. So FindGrid is more than a Case Management because it allows collaborative research even which are more than bunch of documents & links.

I could imagine that this would be exactly the right tool for solving criminal cases, doing market research, enterprise architecting, knowledge stores for consultants and much more.

Let’s achieve Synergy!

Elements of IT Systems are data, functions and knowledge. For the first two aspects SAP is very strong: SAP Business Suite has a data model for exterprise applications and there many functions for all processes of an enterprise. These functions consists of ABAP transactions, workflows and Enterprise Services as SOA layer. And of course business knowledge is within the system as customizing for standardized processes and business rules within rules engines like BRFplus.

On the other hand we can integrate document management systems to SAP Business Suite and link business objects and cases (think of cases ad hoc workflows) to external documents. But these solutions are limited in the case of search and complex knowledge models.

In my opinion FindGrid has the possibility to close this gap and even more: SAP Business Suite become more and open using tools like Embedded Search and Project Gateway, so there is the chance to link business objects to concept models. In my opinion in the end this could lead very smart and and flexible software solutions.

To report this post you need to login first.

1 Comment

You must be Logged on to comment or reply to a post.

  1. Rui Nogueira
    Hi Tobias,, thanks for your few on this topic. We are trying to make it more transparent what SAP is doing around semantics and how SAP’s Technology Strategy overall looks like.

    One question: In your blog you connect the topic “SAP communicated a strategy for introducing semantics to their products” to a blog called “The role of Open Source in SAP’s Technology Strategy”. Is that intended?
    I have blog about the open source strategy called “The role of Open Source in SAP’s Technology Strategy” which can be found here: The role of Open Source in SAP’s Technology Strategy

    There is also another blog called “From apples to products: semantic technologies in SAP products” which can be found here: From apples to products: semantic technologies in SAP products

    Just wondering.

    But again thanks for your feedback. This is highly appreciated!

    Best,
    Rui

    (0) 

Leave a Reply