Introduction

The SIG considered two topics, namely developing a design for the Implementation of a Validation API and a feasibility study of Open Source rules engines. This document summarizes the activity, along with potential next steps.

Develop a Technical Design for the Implementation of a Validation API

The SIG reviewed, discussed and provided feedback on the core documents, i.e. Ed-Fi Data Validation Architecture and Ed-Fi Validation API Design. Feedback and open issues were discussed and noted within in the documents, and in the notes for  Meeting 2. 

There continues to be interest and desire for providing this functionality - particularly from mature Ed-Fi implementations and a subset of SIS vendors, and the SIG discussed options for moving forward, which include:

1. Doing a proof of concept with a mature Ed-Fi implementation

2. Adding the feature to the Ed-Fi core roadmap

Option 1 seems to be recommended the approach, as there are a number of aspects to understand in more detail, including:

• Adapting the existing Ed-Fi authorization approach to ensuring a robust approach to protecting PII information in validation results 

• Confirming that source systems (such as SIS) can consume and surface the information in a useful wa

• Determining what level of "smarts" the validation API needs to provide to avoid overwhelming API consumers with a flood of validation results

The Alliance will check back on the status and desire for field implementations to drive this forward during the Tech Town hall at the 2019 Ed-Fi Summit.

Open Source Rules Engine - Feasibility Study

The SIG reviewed and discussed the Open Source Rules Engine Survey and Recommendation. In addition to the main document, additional information and feedback is captured in the notes for Meeting 3 and Survey of Rule Engines.

There was general consensus that there is a growing need for LEA-based Ed-Fi implementations to get a handle on the state of data landing in an Ed-Fi ODS, and that a simple SQL-based approach to defining and executing validation rules (option 3 in the document) could address core needs, especially if there was a "community-source" approach to creating and sharing common validation rules.

The SIG discussed options for moving forward, and the ideal approach would be for one or more LEA's to take the lead in creating an initial framework and "starter" set of rules, and prove out if there is potential for this "bottoms" up approach to basic data quality to take hold in the ecosystem.

The Alliance will check back on the status and progress for field implementations to drive this forward during the Tech Town hall at the 2019 Ed-Fi Summit.