Guilds and computational compatibility...The system

From crowdresearch
Revision as of 17:58, 22 February 2016 by Trygvecossette (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search


Computational comparability. It scales and has expertise baked into the outcome because of the extent to which attributes are pulled from cv, github, prior work, social media, application, etc. We choose three vectors that apply to each task..for image processing: color, location recognition and data base query. Using graph data bases we take attributes and workers will literally gravitate to their expertise. We use this to intake workers and the system constantly feeds itself so we can have a built in leveling module, as well as a module that can see the skills missing for a worker to take on additional tasks. We can align with a training program. Leveling is vertical and horizontal.

The System

  • 1. when a worker decides to join a guild or is recruited into said guild, they will present a dossier. Links to CV, Github, Social Media, Current on line work, MOOC, etc.
  • 2. We pull from these data sets. These attributes will gravitate to one/if not all, of the vectors the guilds support. For example, if the guild supports image processing, DB queries, Color Identification and location recognition will be the vectors. Vectors will be pulled/refined/updated from the Task Authoring process.
  • 3. We can than create a levelling system based on proximity to the various vectors.
  • 4. For those that are not proximate to any of the vectors, the system can than recommend training/professional development to acquire the needed attributes
  • 5. With a realtime system, managers of the guild, can have an accounting of workforce when tasks arrive. Impact of price/time.



  • 1/ The way the system gets input appears to be practical regarding advanced skills. Not sure we are talking about micro-tasking anymore. *2/ I like the matching system, the gap analysis and the recommandation engine. I think you provide a very good framework for gate-keeping recruitment.
  • 3/ We probably should articulate that with initial mentoring, ranking in the guild and subsequent evolution.
  • 4/ On a personal note, I remain an advocate of human intervention in this later,
  • 5/ but computational tools will be necessary to support this: my first impulse is to expose data to guild elders (granting info access to them is part of the guild compact) and let them use plug-ins developed by the community (meaning we must expose an API) to manage alerts, performance issues, etc.