Cargospotter technology

= BIA = Jeg klare ikke å lage en søknad som er 120% fram til deadline 13.2. Jeg vet at en 90% søknad er ikke noe problem, men da har vi ikke sjanse til å komme gjennom nåløret. Hadde en samtale med NOFAS som synes at ideen er meget spennende, og at de vil kan tenke seg med hjelp til å skrive, men da til en søknad til høsten.

= Teknologisk vei videre= Jeg er veldig fornøyd med de samtaler vi hadde, og ser en veldig spennede tid videre. Håper jeg at vi på sikt kan utdanne en PhD innen de områdene som vi identifisere. Enten som en NæringsPhD (50% støtte fra NFR) eller som en frittstående PhD (midler til UiO eller UNIK). Mitt utgangspunkt er at vi skal sikte på framtiden og "de beste løsninger", særlig når CargoSpotter har som devise å levere "100% korrekt".

I samtalen med Dagfinn identifiserte vi tre områder, som jeg kort beskriver på engelsk:

Semantic data analysis through rules
The goal of the work is to help your people to let things which can be done automatically done by the computer, and rather concentrate on the critical items. That includes the transfer of structured data into the database, then develop rules for checking the "validity", e.g. is the reported position of a ship reasonable as compared to the last position (speed, distance).

My suggestion is to go for a Semantic Model, either being a semantic CMS (Webnodes.com), Semantic MediaWiki (Skaperhuset, CWI: http://cwi.unik.no) or Microsoft OData (MVC4 development). This is a choice you have to take. Having decided, we can then start discussing algorithms which can help us to analyse the data which come in.

We could also consider to hire Susana on a project base or as a part time position (my study assistant in UNIK4710 "Mobile Semantics"). She came to Norway one year ago, with an education as Telecom engineer. http://cwi.unik.no/wiki/Susana_Rodriguez_de_Novoa My former PhD student Zahid Iqbal, http://cwi.unik.no/wiki/Zahid_Iqbal, is also without work,

GUI for graphical BPM on top of semantic data
The idea is here to provide the end-user with a graphical interface, where he can "drag and drop" symbols on a dashboard, and thus create processes. Though you (Dagfinn and Fridtjof) know well about processes, your customers might even have better ideas for finding out good ways for "enhanced knowledge" through processes. In the "management world" Trello.com has developed a good interfaces for "moving ToDo cards". I would like to think about solutions to develop that further.

A possible contact is Mathias Weske from the Hasso Platner Institute (P from SAP) in Potsdam (Berlin).

Enhanced knowledge through linked open data
We also discussed using available "open" data to generate knowledge which are of importance for freight. This may include: war, economic crisis, oil prices, meteorology (ice in the north east passage)...

Here I would discuss with the Logic Working Group (Alex Waaler++ at UiO) to get ideas and help

Think about, and then we'll talk further :-) Josef