Glossary - sakura-team/sakura GitHub Wiki

Objects

Protocol: A series of precise and detailed instructions specifying all the operations to be carried out during an experiment. In Sakura, the protocols concern the operations to be carried out for the implementation of experiments in order to construct and evaluate the tools produced through research in computer science. (Definition of TLF: "Precise and detailed instruction specifying all the operations to be carried out in a certain order as well as the basic principles to be followed in order to carry out an operation, to implement an experiment.")

Datasets: These are of two types:

1. The projects projet/project described with metadata. They contain one or more jeux de données/dataset presented under different tabs on the sakura interface (a tab for each dataset) stored on the sakura database. There are two ways to specify metadata and import data from these datasets:

   • Using the web interface (metadata + importing CSV file)
   • By using the API (metadata + sending data in real time or not)

2. Existing datasets, which were created independently of the Sakura system (eg. geotweets, OSM). The Sakura system allows access to these datasets, without knowing their structure. The metadata associated with these datasets are indicated in the config of the daemon.

Operators: Algorithms that will allow data processing (management, join, stat, visu, ...). Some operators will be created by the members of the project, others will be created by the users according to their needs. An operator called "blank operator" enables development of a python operator during the construction of an analysis workflow.

Analysis: A user-driven process that links data sets and operators. In the Sakura workflow the user can add an operator made from a blank operator. The list of analyses gives access to the list of workflows. The analyses are described by the metadata during registration.

Results: These are the semantics brought to the user's analysis. On the Sakura platform, the analysis producer will have a tool to bring semantics to her results. She will be able to document its analysis process and annotate it. The list of results gives access to the annotated narratives. The narratives are described by the metadata during registration.

daemon: Process running on the Sakura server or on an external server to access or analyse the localised data at that place, through operators.

**TO BE DELETED ?? **

LINKS OBJECTS and TASKS according to the APPLICATIONS

	Protocol	Data Sakura	External Data (geotweets, OSM)	Operator	Analyses	Results
List	web	web or API	web	web	web	web
Create	web (thedre)	web or API		web (worflow)	web (worflow)	web (hubble)
Modify	web (thedre)	web or API		web (worflow)	web (worflow)	web (hubble)
Delete	web (thedre)	web or API		web (worflow)	web (worflow)	web (hubble)
Register			Config. Daemon