Crowdsourcing at Crowdcrafting: we’re doing antimatter research!

[Wordpress is playing up so some images may be missing]

I am a strong believer in Crowdcrafting (volunteer “crowdsourcing” ay http://crowdcrafting.org):

CrowdCrafting is a free, open-source crowd-sourcing and micro-tasking platform powered by the PyBossa software. This platform enables people to create and run projects that utilise online assistance in performing tasks that require human cognition such as image classification, transcription, geocoding and more. CrowdCrafting is there to help researchers, civic hackers and developers to create projects where anyone around the world with some time, interest and an internet connection can contribute.

CrowdCrafting is different to existing efforts:

  • It’s a 100%% open-source
  • Unlike, say, “mechanical turk” style projects, CrowdCrafting is not designed to handle payment or money — it is designed to support volunteer-driven projects.
  • It’s designed as a platform and framework for developing deploying crowd-sourcing and microtasking apps rather than being a crowd-sourcing application itself. Individual crowd-sourcing apps are written as simple snippets of Javascript and HTML which are then deployed on the server. This way one can easily develop custom apps while using the platform to store your data, manage users, and handle workflow.

It’s the brainchild of Daniel Lombraña González
and uses the OKFN’s PyBossa (http://dev.pybossa.com/ ) softwareand supported by Citizen Cyberscience Centre . Crowdcrafting is not a single app or activity, it’s a framework that can be used for anyone to mount and run their crowdsourcing app. Daniel presented at OKFN about 2 months ago and among the projects were:

  • Do Yellowhammers (Emberiza Citrinella) have different dialects in UK and New Zealand?
  • Can we measure gender ratios in science from photographs on Flickr?
  • How does human time scale with graph isomorphism?
  • Which areas in cities are public parks?

I’m interested in using crowdcrafting in the biodiversity project that Ross Mounce, Matt Wills and I are starting in October – to extract phylogenetic trees and species from the literature. We aren’t quite sure what the details will be, but it may be creating a checked corpus, or disambiguating.

So I need practice and experience and so I was delighted to see a new project launched on antimatter http://crowdcrafting.org/app/antimatter/ . The simple goal, I believe, it to verify that antimatter has normal gravitational mass – i.e. that a proton and antiproton will attract each other gravitationally. (I won’t speculate more as I’m intending to find out in the project itself, and I’ll report back.). The way this is done is to capture annihilation tracks (I think in emulsion), and find out where they originated (“the vertex”). The tracks are faint and humans are better at recognising them than machines (I can believe this, having written image recognition software).

See http://home.web.cern.ch/about/updates/2013/08/join-dots-measure-antimatter for an introduction

Crowdcrafting is not Mechanical Turk. The participants are part of a community and may bring complementary skills. Some could be retired scientists like me, mathematicians developing financial software, web site designers, UI experts, neuroscientists looking at brain sections, electrical engineers, and even fully-paid-up HE physicists who want to be involved. Many will be students, the scientists of the future. And many will be taxpaying citizens who are interested in what their funding goes towards. Some will have social skills in developing online communities – the list is endless. It’s really exciting.

So for me the most important thing is to develop a sense of community. They/we are being asked to contribute time and in return they must have reward. This can range from a leaderboard, to a deeper understanding of the science, and even to discovering new effects (this has happened repeatedly in other projects). So here’s a rough set of what I think are important.

  • There should be a clear community infrastructure
  • We should know what the overall goal is
  • We should know how our contribution will help
  • We should understand what we are actually doing and be able to help refine it
  • There should be regular updates – stats are critical, but also summaries of more general progress. And measures of how far we have come and how far to go.

So I suggested that we should have a communal discussion group and DanielLG set this up. I posted some initial thoughts and questions and you can see these here https://groups.google.com/forum/#!forum/cern-antimatter-crowdcrafting . I’ve already got useful answers from Michael Doser (who runs the project) who writes:

first of all, thank [everyone] for your interest! As an avid follower of galaxyzoo and related projects, I have been thinking about how to open high energy physics research to public involvement in a similar manner. With the AEgIS experiment, we may have a first possibility, but I am sure that we will find more. AEgIS is built around the attempt to measure the gravitational interaction between matter and antimatter using antihydrogen atoms. [… lots more]

And I was about to blog about the projects and … But 5 minutes ago I got this incredible mail from Daniel:


Today all the available tasks have been analyzed by the volunteers. Amazing. Yesterday we had a huge [s]pike in the answers, and today all the data have been analyzed.
Michael and his team now have to download the results, study them, and if they have more data add it to the system 🙂
Meanwhile you can check the stats of the project. They are really impressive!

Cheers and thanks for your contributions,

Daniel

 

The stats show this:

  • Every dot on the map represents someone who understands Crowdcrafting and will be getting ideas! They are all scientists because they have all participated in science! And the roll of honour:

This is massive. 100+ users join in 1 day and completely finish the tasks. I think it’s taken Michael by surprise. It has certainly surprised me and it’s massively exciting. Here’s the results (they are updated in real-time and actually rotate). 4 of the tracks are from me!

 

 

 

This entry was posted in Uncategorized. Bookmark the permalink.

3 Responses to Crowdsourcing at Crowdcrafting: we’re doing antimatter research!

  1. Hi Peter,
    Thanks for this kind blog post! Hopefully more and more people will be using CrowdCrafting in the future for other projects!
    Cheers,
    Daniel

  2. “This is massive. 100+ users join in 1 day and completely finish the tasks. I think it’s taken Michael by surprise.”
    It definitely did. Right now, the tasks are a bit artificial, and the user interface is not even in beta, so it’s quite amazing how helpful people can be! Right now, we need to work on polishing the user interface, see what works best, listen to feedback, play around with the same sample data, but I hope that we’ll be able to provide lots more data (which we already have in hand, but haven’t really processed yet), and a more interesting set of tasks, soon.

Leave a Reply

Your email address will not be published. Required fields are marked *