Skip to content
This repository has been archived by the owner on Nov 29, 2017. It is now read-only.

Latest commit

 

History

History
92 lines (66 loc) · 4.05 KB

project_docs.md

File metadata and controls

92 lines (66 loc) · 4.05 KB

LibCrowds Project Documentation

The following document describes how to create and manage LibCrowds projects. The focus here is on details specific to the LibCrowds platform, more general information can be found within the PyBossa Docs.

Project ownership

Each project is assigned an owner who will be responsible for its creation and maintenance. Project owners have access to a dashboard, accessible via the Open Project menu, which provides access to various settings and additional information regarding the project, including:

  • Analysis: A link to the results analysis application (see below).
  • Results: To view the current results (useful for spot checking).
  • Statistics: Contribution statistics for the project.

Anatomy of a project

Apart from establishing the general requirements of the project (it's input and the desired output), when developing a new type of project the following components should to be provided by the project owner:

  • A thumbnail image, such as those displayed on https://www.libcrowds.com/project
  • A larger splash image used on the project introduction page, including any required attribution for that image.
  • A description of the project to be used on the project introduction page.
  • A tutorial page describing how the project works.
  • An optional tutorial video to be embedded in the project introduction page.

General workflow

For specific details of the workflow established for your project, please see the documentation available for each type of project on the LibCrowds GitHub page (e.g. project-convert-a-card).

Following is a general outline for all project types.

1. Creating a new project

The easiest way to create a LibCrowds project is probably to use pybossa-github-builder, which is a plugin that automates much of the setup process. It also provides a way to import templates for the new results and tutorial pages, via the web interface. The basic steps to create a new project using this plugin are as follows:

  1. Create a new project.
  2. Via the project dashboard click 'Sync with GitHub'.
  3. Search for a GitHub repository.
  4. Configure the import and update the project.
  5. Import some tasks.
  6. Set the task redundancy.
  7. Publish.

Note that there are various other ways to create new projects, all of which are described in the PyBossa documentation.

2. Analysing results

A separate application has been created to handle the analysis of results. The analysis process is automated, as far as possible, according to the rules defined for each category of project. Human intervention is only required for any results that cannot be analysed automatically (e.g. different answers were given for the task). A template is available for each category of project that aims to make the process of checking these answers as straightforward as possible.

The Analysis menu option, available via the project dashboard, provides a link to this application. A notification will appear next to this link showing when there are results that need to be checked. A password is required to access the application, which can be requested from a LibCrowds administrator.

Results can also be viewed via the Results menu option, also available via the project dashboard. While nothing can be modified here, this view can be useful for spot checking.

3. Project completion

Once a project has been completed and all results analysed a CSV containing the final results can be download from the LibCrowds Data page. The way in which this spreadsheet will be used depends on the workflow established for your particular project. For example, it might be checked and edited before being sent off to metadata services for the creation of catalogue records.