Technical Consortia

Although the SE team can assure quality institute-wise it cannot by itself enforce the same quality or standards across institutes or to the industry. It's necessary then a technical consortium to define the standards across different institutes especially if they're using the same technology or trying to achieve the same goal.

Each matter is different. Normally what works out for gene matching algorithms might not work out for proteins and what works for both might not work for multi-dimension data sets or more complex relationship between data, so it's clear that one technical consortium will not be able to embrace all queries of all teams in all institutes. Also, you have all the associated data with your main dataset such as journal listings, ontologies, annotation tools. All of that must be unified as well and each one has its own consortium.

So, we do have lots of consortia (UniProt consortium, GO consortium, etc) but they're much more interested in the data and not in how the data is produced or how well does that code cope with future changes or staff relocation. There must be one or more technical consortia to define such parameters, to standardise the way things are done and to unify all unique identifiers once and for all.

The easiest way would be if all institutes had their own SE teams and the SE teams would for a consortium on how to perform specific tasks and how to control the quality of their code. Other institutes would them be encouraged to follow the same standards (if it works in the current institutes, of course) and the community would start helping and being part of the definitions and discussions. More or less how the W3C (World Wide Web Consortium) was formed and today they define the standards of data formats and access for all the web.