We have produced an initial model for how software may be described. There are several properties of software we are, at least initially, interested in capturing.
A model of software
Data and data formats – The description of inputs and output to software programmes and the formats in which they are encoded.
Software organizations – here we differentiate between two subtypes directly; the developers (responsible for authoring software) and the publishers (responsible for making software available to a particular consumer group).
Algorithm – description of the inputs, outputs and method for achieving a particular objective.
Programming language – the language in which source code is written.
Information processing – tying the algorithm and data inputs to the software process of execution.
Objective – the objective, task or purpose for which the software is being used
Version is also of particular interest although not shown on the diagram. Versions present an interesting modeling problem – how should a version be represented? How do you compare two versions? How do we say one version is more recent than another?
SWO 
This is interesting.
Are there any plans to characterize software such that people will be able to tell what kind of computer architectures it can run on?
Programming models such as many-task computing vs. fine-grained parallelism that requires low latency between nodes vs. purely serial programs, for instance.
An ontology of hardware, CI, etc. that could interface constructively with your ontology of software would be great.
Hi Miriam,
This is an interesting comment as we have also received similar requests via email. The answer is that hardware descriptions are out of scope for this project, however, we will leave placeholders that would make integration with hardware descriptions possible. I hope that is useful.
James
Pingback: Considering Versions « SWORD