This document details the consolidated architecture of the WeGovNow platform prototype. In doing so, it builds upon the conceptual and methodological framework for the WeGovNow project presented in a dedicated deliverable (D1.2).
The document presents the technical architecture of WeGovNow in details and, in particular, how the different WeGovNow components are integrated. It is a good document but the relationships with use cases presented Del 2.1 are overlooked. Putting more emphasis on them would clarify better the connection between the functionalities of the platform and the use cases and improve readability of the deliverable.
Description of tools, datasets, patterns, libraries, and documentation for further platform extensions is highly appreciated.
Broad and promising interconnection of existing tools. It would be a challenge to properly set up them and continuously update/curate them, when very large amount of data are collected. Value added (for management purposes) by automatically analysing/learning from data is not very clear.
The document is very thorough in its coverage of intended system architecture. The challenge of integrating mature applications is acknowledged but perhaps needs more consideration of mechanisms for data exchange, aligned with authority views on interchange.
The specific applications appear well related (CommunityMaps built on GeoKey, OntoMap, FirstLife, LiquidFeedback and ImproveMyCity). However, the extensions are stated without very clear relationship to user needs (just kind of - we will do this - intuitive but not well explaiend). This is also true of their integration (how they relate as a meaningful service set responding to user needs).
The idea of "citizens as partners in delivering public services", stated to be central to the project, could be made evident in relation to architecture - how the architecture better supports that aim - the vision of architecture in operation.
Using INSPIRE is interesting, but INSPIRE doesn't support public safety very well and OSGeo could be taken into account (http://wiki.osgeo.org/wiki/INSPIRE) if, as suggested elsewhere in the project, an aim is to help citizens as reporters of concens.
Realistic approach, facing the classical problem of integrating existing applications; extremely developed on the user management side (authentication, preferences, etc.) would probably merit some more substantial reference framework for data and process integration (even i must admit that this light data and process integration is what make the approach realistic).
The main point from the software architecture point of view is that the integration of the different applications from the point of view of data exchange and process interaction, seems to be based on a ad-hoc approach, that as such would not be as applicable to other applications (i.e. without developing a new ad hoc solution for the new application).
Interesting the High level design; interesting the integration with Inspire, in term of Linked Open Geo Data; some Design element should have a deeper view (i.e. why is it used and Ontology to collect logs? What kind of technological framework are you thinking about ? what is the deployment architecture? what is the scalability model?)
Are there any Non functional requirement (I.e. Security) you are thinking about?
