Op/Gov Experiments

From crowdresearch
Jump to: navigation, search

Introduction

As we prepare a paper for the CHI conference (http://chi2016.acm.org/wp/), the time to start planning experiments for a September 25th deadline is now...With that, what do you feel are the organizational, leadership, communication, engagement, operational challenges we can measure, address and potentially solve. Sometimes the hardest part of research/experimentation is converting a gut feeling or a perception into a hypothesis, hence our first Google Hangout on Wednesday, Aug 5th at 12:00noon EST to help with brainstorming, spitballing and refining ideas. In no short order we'll need to determine a research agenda: which experiments we're gonna run for the next 1.5 months, when they are gonna be run, survey methodology, data collection, etc. The more contributors the better, so if any of these areas of engagement interest you, don't wait until Wednesday to share/declare! Coordination, transparency, caffeine and intellectual curiosity will carry the day! Let the fun begin...

Ideas/Hypothesis'

hy·poth·e·sis hīˈpäTHəsəs/ noun a supposition or proposed explanation made on the basis of limited evidence as a starting point for further investigation. "professional astronomers attacked him for popularizing an unconfirmed hypothesis"

When you post, please denote your name/date posted....

(Trygve 8/3)

This is some feedback MB received from an outside reviewer. A point for us to address: one of my alumni read our UIST submission and had a question that can help inform any revisions: "Thanks for the paper, that’s both interesting and intriguing indeed. I really like the idea of a prototype task to built a trust relationship between workers and requesters and make sure that everyone is on the right path. On the other hand, I didn’t quite understand the open-governance model thing: is there only one leadership board at the global scale of Daemo (is this what you refer to as “the platform” in the text?), or is there one for each project posted on Daemo (although that doesn’t sound feasible)? If we’re talking about the former solution, I still don’t understand what this leadership will be doing (is this like a trade union?). Nonetheless, this is a great demonstration of a crowdsourced paper, and that’s really cool!"


(Trygve 8/3)

In response to the well balanced critique of MB's reviewer. Just as we did in choosing participatory democracy as the vehicle/method to elect the LB; we should choose from three/four organizational models to determine how the LB should function. We can align a feature that will be prototyped using the mechanisms of that particular model. Using surveys and other methods we can assess and evaluate options. At the end of 8 weeks we will have a LB model.

  • A. What models should we explore (Oligarchy, Wikipedia approach, Executive within a Democracy, Open Allocation/no LB, etc) and test?
  • 2A. How do we test these models in a short time frame. It might be confusing but can we work in parallel (Using two approaches on two projects...at the same time!)
  • B. We are in the midst of a transition...should we look at a model to Build the platform and another to lead it once its built? Is there a mechanism to segue from build to lead?
  • C. How do we consider an option of having no LB at all as open allocation models?
  • D. As we're looking at organizational models: what are the impules for choosing the model (Control, contribution, Trust, etc..)


Jsilver (August 6)

I think, at this point in time, it's more appropriate and efficient to implement than do experiments.

I propose an LB version (a leaner version of the Custodian Group concept [1]) I've had in mind since March/April -- a hybrid approach that offers a good balance between democracy and efficiency. Both decentralization and centralization are in play. In this LB model, the community proposes suggestions/issues within the platform forum and upvotes the best suggestions or most critical issues. Voting allows the community's voice to be heard. The LB would then debate, sift, and execute. That is the basic idea (process loops are highly possible).

  • Rijul approached me via DM to discuss the Custodian Group concept and he presented the idea with a little twist: [2]
  • @Shirish.goyal: "A Idea/issue sourcing, B Prioritization, C Debate, D Decision, E Execution"

Possibly subject to further (minor) modifications, I firmly believe this is the best opgov model for our platform.

My favorite reference, which I believe bolsters my idea, is Prof. Thomas Malone's Making the Decision to Decentralize [3]

Daemo opgov decision flowchart1.png


M.Kambal ( Aug 6)

1- The Oligarchy approach is something that we should already left it behind us the moment we decided to go for open-gov, such model would no be supported by the workers and requestors in the long term , and it will sound that we are replicating the closed system adopted by Amazon and Upwork. BUT the positive side is that this approach is : FAST because the LB will take decisions directly and have kind of democracy since the LB will be selected by the crowd . 2- For the second scenario , that sounds really interesting , my only point to add is why not decision taken by LB directly been implemneted and it could be revisted if ONLY 2/3 or +50% of both requestors and workers object it. this will blance the openess and the need to move fast. in other words.

Reference Material