Skype call on coordinating development between the Agrovoc manual indexing module and the AgroTagger client module

Adam Sanchez (Condesan)
Armon Chen (Sun-Yat-Sen University)
Burley Wang (Sun-Yat-Sen University)
Ahsan Morshed (FAO)
Valeria Pesce (GFAR)
Adam Sanchez from Condesan has developed the first version (ready for testing) of an Agrovoc manual indexing module for Drupal (the zipped module is available here). Sun-Yat-Sen University is going to implement an AgroTagger client module for Drupal for automatically indexing nodes against Agrovoc.
The architecture of the manual indexing module had already been agreed upon (see attached slide) and it was deemed important that before Sun-Yat-Sen starts developing the manual indexing module we all agree on that architecture and make sure that the two modules use the same approach in those parts of the architecture that are common to both. 
To agree on a common architecture for the two modules (same storing of keywords and URIs), on the best way to distribute them as modules (two independent modules, one module with sub-modules, or one module with different "widgets") and on future coordination between Adam and Armon for the further development of this solution.

  • Valeria introduced the background.
  • Adam demonstrated his module and explained how it follows the agreed architecture.
  • All participants agreed that the module's architecture and functioning is what was expected.
  • Burley and Armon said that their module can adopt the same architecture for the storing part.
  • The discussion then moved to how the two modules can work both together or alone and we all agreed that the difference between the two could just be in the fields that can be selected for content types. The Agrovoc manual indexing module creates a new CCK field that allows to select Agrovoc terms through an autocomplete textbox connecting to the Agrovoc web services 2.0. The Agrotagger module could just create a similar CCK field that automatically gets the Agrovoc terms by sending both node body and attachments (if any) to the Agrotagger web services. In this way, website administrators would have two CCK fields available for their nodes and could decide which one to use each time, even both together (Armon suggested that the automatic indexing CCK field might suggest best terms and the user might still decide to use the manual indexing field if he is not happy with the automatic indexing).

Follow up:
- Adam will send some documentation about his module to the others, in particular to Armon.
- Our team in FAO will test Adam's module.
- When the AgroTagger web services are ready, Armon will start developing the Agrotagger client module.
- Adam and Armon will coordinate and decide on the best way to present the options for manual/automatic indexing to the user, based on the ideas illustrated in this meeting.
- When the module(s) is/are ready, our team in FAO will incorporate it/them in AgriDrupal..
- Adam and Armon will publish the module on and post an article to the AgriDrupal group on AIMS.

Start date: 



People involved: 





Agrovoc manual indexing module for Drupal: architecture40 KB

© 2007 - 2020 Valeria Pesce
Twitter icon
Facebook icon
LinkedIn icon icon
StumbleUpon icon
Digg icon
Reddit icon
Technorati icon