Thursday, March 10, 2011

How #Toolserver can do much more good

The Toolserver is a cluster of servers hosting tools written by many people. Some of these tools provide essential services like the traffic statistics of the collections provided to us by GLAMs.

Every individual tool is experimental and essentially provides a proof of concept. The implication is that once a tool has proven itself, it could go to a next level and gain all the trappings of a production service.

Criteria for the tools that deserve to be upgraded to production quality are:
  • the service it provides is being relied upon
  • the service can be of value to an international public
  • it stimulates the use of the Wikimedia projects by new demographies
Currently none of the Toolserver tools are of a production quality. The tools are in many code repositories and they do not comply with best practices. They are for instance not internationalised and when they are, they are not localised at Screens are not designed to take additional data into account.

As these tools are freely licensed, it must be possible to either fork the code or preferably work with the primary developer to provide the service that the concept hints at.

When you consider the size of the Toolserver server park, it is not a project that can be dismissed as "Spielerei".


No comments: