Here you can have the presentation of the people presenting at the Smart Data Models session of the FIWARE summit.
Tomorrow at day 0 of the FIWARE summit there will be a training session about Smart Data Models.
Slides are freely shared here
This training will be delivered face to face.
- Board and what is Smart Data Models Program
- Current Status
- Contributos and dissemintions
- Agile standardization
- Current standardization status
- M.A.S.: Manifesto for agile standardizationç
- Standardization in a digital market
- Data model contents
- Contents of a data model
- Schema: review
- Additional documents: examples, contributors, adopters, notes*
- Generated documents: model.yaml, examples, README, context
- Data Model creation
- Incubated repository.
- From csv, json or open data portals
- Data Model moderation
- New services. Integration
- New data models. Mapping of standardizations
- Other services for the contributors
The Manifesto for Agile Standardization (MAS) describes the 7 principles that we apply to the Smart Data Models program.
0. Don't just standardize, be agile and standardize
1. Do not reinvent the wheel
2. Normalize real cases
3. Be open
4. Don't be overly specific
5. Flat not Deep
6. Sustainability is key
If you want to read the agile standardization manifesto’s complete explanation a one-page document is located at the root of our data models repository.
Not all changes in the data models are reported with a blog post, Twitter dissemination o LinkedIn post. But if you notice there is a widget in the right part of the front page (scroll down till Announcements) where these tiny changes are reported (see the image or browse the front page).
Most of them are issues, suggestions or PR solved thanks to the collaboration of the community. They make the data models better for all.
We cannot say thanks individually to all of them but please feel our gratitude.
You can download all of the tiny annoncements from this simple text file.
There are 3 new data models coming from the collaboration with the GreenMov project.
They are located at the Environment subject.
- NoisePollutionForecast. Noise Pollution forecast stores the expectation about noise pollution based on some input elements and the noise elements present.
TrafficEnvironmentImpact. Environmental Impact of traffic based on the vehicles traffic and their emission characteristics
TrafficEnvironmentImpactForecast. Environmental Impact of traffic based on the vehicles’ traffic expectations and their emission characteristics
Many of the datasets published in open data portals are extensively used elsewhere. Well-maintained portals have managers that document the data structure and provide definitions of the types and contents of every field in these datasets. These are some of the requirements for the successful publication of a new Smart Data Model.
With this simple python script (and others to come) at the utils folder it can be drafted a JSON schema compliant with the Smart Data Models Program contribution manual. It is an early version (not everything is updated) but you can check it out.
– base url of the ODS portal
– dataset_id of the dataset
A draft json schema compliant with Smart Data Models Program,
some limitations: it does not translate descriptions (required in English)
some data types
It prints the schema and also returns (if possible a file named schema.json)
test it with this command
SDM_OpenDataSoft_schema_converter.py https://data.ameli.fr/ effectifs
The Smart Data Models Program does not define canonically and uniquely the terms used in the data models. There are many ontologies and vocabularies providing solutions to this issue.
For those users of linked data solutions, every subject includes a context.jsonld file (see example) with long IRI for the terms used in the data models. Besides this, the IRI provided are in fact URL to pages with additional information about the term (see example).
not only this but also two services are available on the tools menu on the front page.
- Merging several contexts and detecting conflicts. (for merging several context from different subjects)
- Mapping a context with external ontologies. (mapping a local context from SDM to any external ontology). These are the ones available but more could be easily created on demand.
But when mapping an existing open and adopted standard now it is possible to customize the context generated by using a new file notes_context.jsonld (see this empty example) at the root of the subject. It will replace the automatic IRI for a term with the customized one.
The Noise Pollution data model and the AirQualityForecast have been published on the Environment subject. The first one merges specific and punctual noise measurements (coming, e.g. from NoiseLevelObservation entities) into average parameters referred to city areas, providing more city-related data about noise pollution status and evolution. The second one helps to store the forecast about the quality of air for a specific period.
If have approached the Smart Data Models Program (SDM) for the first time and you want to become a contributor there are some technical concepts that you need to know about the elements compiled at SDM.
Once checked this presentation maybe you want to review the contribution manual
The data models AirQualityMonitoring, AirQualityObserved, and NoiseLevelObserved have been extended with additional attributes. These data models have been tested in three use cases in the cities/regions of Murcia / Molina de Segura, (Spain), Nice Cote D’azur (France), and Flanders (Belgium).
Soon other data models will be released thanks to this collaboration.