Skip to end of metadata
Go to start of metadata


We describe different stereotypes of Research Objects as a catalogue of patterns. A software pattern is defined as the abstraction from a concrete form which keeps recurring in specific non-arbitrary context 10; and in our case we focus in scientific-related resource aggregation situations. Patterns are convenient as they provide descriptions independent from the underneath implementation technology, and independent from the requirements of the domain that they are attempting to address. They become thus a catalogue of common situations that can be used by a knowledge expert in order to improve the knowledge acquisition process.

For each pattern we include a phrase that summarizes the depicted situation; the description of the pattern and its context; its graphical representation, where we identify the main elements and their relationships; and synonyms and related situations, which will point to the situations and similar approaches where the patterns were extracted.

Live Objects

Research Objects that aggregate, describe and encapsulate scientific work in progress.


Live Objets represent a work in progress, and as such, they are mutable. Their key characteristic are:

  • They have a temporal dimension. They evolve as they are modified by their original creators, and these modifications might have different nature (e.g. aggregation of new resources, data and metadata modification, etc.). Therefore Live Objects contain information about their lifecycle and to keep track of their current state; and on the other hand, all its constituent elements should be under version control.
  • They might have different contributors and authorships. Live objects are often the result of collaborative work and thus they may fall under mixed stewardship. To handle such situations credit attribution, accountability and licensing information become vital and are therefore taken into account for each of its constituent elements.

Graphical representation

Illustration 1: Live Object elements.

We presume that both the metadata and (in occasions) data that define and constitute every Live Object are exposed, published or referenced using the Linked Data principles [7] (otherwise see the Exposing Object). All the aggregated elements (both data and metadata) vary in time, this is why we depict these elements in Illustration 1 as a stack of elements, representing that they evolve across time; and we include a label with their version to stress that they are under version control. (Note that we write v U,v V,v W, v X, v Y, v Z,. to indicate that each element evolves on its own and therefore has probably its own independent version) .

We also include some high-level relationships, the aggregation and references properties. We denote the aggregation relationships as aggregatet to indicate that these relationships change over time.

Synonyms and related situations

Live Objects are similar to the concept of liquid state Scientific Knowledge Objects. This type of Scientific Knowledge Objects are those that are still evolving and being modified by their original creators[12] NanoPublications also take into account the possibility of evolution by means of richly annotated statements.

Live Objets might be used to represent different types of entities. A Live Object might represent a study that might go thorough a number of steps, an experiment with different phases, a PhD thesis, and taking into account the reusability property of Research Objects, it might even represent a whole project. In general any investigation that we want to capture in the e-laboratory that posses a temporal dimension.

Publication Objects

Publication Objects are the immutable presentation of a piece of scientific work.


Publication Objects are those that more closely resemble the traditional notion of publication. They are intended as a record of activity, and should thus be immutable [5]. Different versions of a Publication Object necessary imply different and distinct objects.

Graphical representation

llustration 2: Publication Objects elements.

We represent a Publication Object in Illustration 2. The figure is quite similar to the one proposed for Living Objects. The difference is that Publication Objects represent an immutable version, a snapshot in a concrete moment of time. Thus we just represent an unique version for each aggregated entity (we use the version labels v Ut,v Vt,v Wt, v Xt, v Yt, v Zt to denote the version of the element at the considered time t). The relationships that are stablished (aggregation and reference) are also immutable.

Synonyms and related situations

The paradigm of a Publication Object is the notion of traditional paper based scientific publication.
The characteristics of a Publication Object and their relationship with a Live Object resembles the notion of Solid Scientific Knowledge Objects and their transition from the liquid to the solid state (see [12]).
Publication Objects might represent static entities such as traditional papers, the report of a given experiment, the final version of a deliverable, etc.

Exposing Objects

Research Objects used to wrap and expose pre-existing scientific data.


Exposing Objects provide a wrapper for existing data, providing a standardized metadata container [6]; and an common access and management interface using the service layer around a Research Objet. As part of this pattern we include a prior process of publishing, exposing or referencing pre-existing scientific data resources into the Web of Linked Data.

Graphical representation

Illustration 3: Exposing Object elements.

In the Illustration 3 we represent the high-level elements and relationships that we identify in this pattern. We consider a set of resources (resource0 ... resourceN) that represent scientific data (e.g. data spreadsheets, workflows, workflow enactment data, databases, etc.). This resources are firstly exposed or published following the Linked Data principles (or annotated and referenced). Then they are packed and aggregated as part of the Exposing Object.

Synonyms and related situations

This Exposing Object stereotype represents situations similar to the approach taken in the SysMO research initiative ( Users have their own data format, most of them spreadsheets that store data from each experiment. The aggregation of of these spreadsheets along with their metadata and the description of the method used to obtain them is similar to a Research Object. 

Archived Objects

Archived Objects encapsulate an aggregation of data that is in some way finished, deprecated or no longer active [6].


Archived Objects are immutable Research Objets that represent scientific data that though is either no longer valid or active, there is an interest in preserving it. As such:

  • They are immutable, and they must be preserved in the exact state.
  • They can have two connotations:
    • A bad connotation, being abandoned or deprecated work that might be needed to be archived for some reasons..
    • A good connotation, meaning that the Archived Object represent curated research content and henceforth the archived object can be used as building blocks to new Research Objects.

Graphical representation

There is no specific graphical representation associated with this pattern.

Synonyms and related situations

OAIS model [14] offers mechanisms for describing archival information packages, providing a framework for the understanding and increased awareness of archival
concepts needed for Long Term digital information preservation and access. Solid Scientific Knowledge Objects [12] are also similar to Archived Object in their sense of a no longer active scientific object as they are considered as being at the highest grade of maturity. Also related with this connotation of Archived Object we find similarities with the notion of curated NanoPublications [13], considered immutable facts that conform current scientific insight [15].


[2] Mons B. and Velterop J. , Nano-publication in the e-science era, in: Workshop on Semantic Web Applications in Scienti?c Discourse (SWASD 2009), Washington, DC, USA, 200
[4] Barend A. M. Which gene did you mean? (2005) BMC Bioinformatics, 1, doi:10.1186/1471-2105-6-142,p 142,, Vol 6

[5] Bechhofer, S., De Roure, D., Gamble, M., Goble, C. and Buchan, I. (2010) Research Objects: Towards Exchange and Reuse of Digital Knowledge. In: The Future of the Web for Collaborative Science (FWCS 2010), April 2010, Raleigh, NC, USA.

[6] Bechhofer, S., Ainsworth, J., Bhagat, J., Buchan, I., Couch, P., Cruickshank, D., Delderfield, M., Dunlop, I., Gamble, M., Goble, C., Michaelides, D., Missier, P., Owen, S., Newman, D., De Roure, D. and Sufi, S. (2010) Why Linked Data is Not Enough for Scientists. In: Sixth IEEE e--Science conference (e-Science 2010), December 2010, Brisbane, Australia.



[9] Star S. L., Griesemer JR (1989). Institutional Ecology, 'Translations' and Boundary Objects: Amateurs and Professionals in Berkeley's Museum of Vertebrate Zoology, 1907-39". Social Studies of Science 19 (4): 387--420

[10] Gamma E., Helm R., Johnson R., and Vlissides J. (1995) Design Patterns: Elements of Reusable Object-Oriented Software. Addison-Wesley, Reading, Massachusetts.
[11] Heath, T., Hepp, M., and Bizer, C. (eds.). Special Issue on Linked Data, International Journal on Semantic Web and Information Systems (IJSWIS).

[12] Giunchiglia F., ChenuAbente R., SCIENTIFIC KNOWLEDGE OBJECTS V.1 January 2009 Technical Report # DISI-09-006

[13] Groth, P., Gibson, A., Velterop, J. (2010). The anatomy of a nanopublication. Information Services and Use, 30(1), 51-56.

[14] Consultative Committee for Space Data Systems, "Reference Model for an Open Archival Information System (OAIS)," Open Archives Initiative, Blue Book CCDS 650.0-B-1, 2002.

  • No labels

1 Comment

  1. Unknown User (paolo)

    I like the patterns catalogue idea.

    It seems to me, however, that these patterns differ mainly, or exclusively, in their state rather than in their structure. I.e., publication objects appear to be structurally identical to Live Objects, the difference being that prior versions of some of the resources are not accessible.

    I that is the case, then couldn't you express the same by saying that ROs are stateful, then enumerate these as states, and associate properties to each state (for instance, accessibility of resource history).