To indicate, please add your name here so I know all eight of us have access. This will be a good way to do issue tracking against our milestones, and for Dav Clark to look over our shoulders.
First, we should make a high level milestone for a near-term achievable, like initial ability to examine the data - summary stats or viz. We'll break that down into achievable issues that we'll assign to undergrad teams.
We can have this conversation below in the comments for this issue (and elsewhere), but please communicate with me here once you've got specific goals.
@jessicajji@reginacpp Will you both be able to review shiny (see link) and present its capabilities at our group meeting this Friday? This will help us to decide if it is the best tool for our purposes, and see how it compares to the templates Jessica presented to the class last Thursday. Please let me know, thanks!
The repository URL is meant to be a project info URL. For example, The twitter bootstrap repo links to their public facing docs site.
So, a link to google drive should go either in a text / markdown file (e.g., in the README) or in the wiki. It's confusing and presented in a "misleading" context now.
If you want less choices, move the URL to the README. The README also has way too much content.
A data management tool that enables our project to collect, measure, analyze and visualize mobility data for the purpose of evaluating our hypothesis - namely, that our common locations lead to common routes. The tool will enable anyone using Moves to export and upload their data in two easy steps to our project's hosted data management tool, a combination of RDS on AWS, R Studio and Shiny.
In addition to the survey, this is the other deliverable our team will present to the class on 12/3
We'll present the multiple factors that created the need for and influenced the design of this all important tool:
privacy - instead of offering to anonymize their data, we enable them to submit it anonymously - furthermore, instead of sharing all of their "quantified self" we upload a small subset of their collected data and parse it into a chronologically-sorted list of trips across all users, preventing us from doing pattern analyses on the mobility data of any single user
ease of use - a simple website for The Carpool Project enables users to do all the things we need them to do to participate in our project, so we envision a website with four buttons: 1) Download Moves; 2) Export Your Moves to Your Device; 3) Upload Trips to Our Server; and 4) View our Project Data
transparency - the fourth button above enables users to see the entire data set of anonymous, parsed data that's used by our data management tool.
@zhuwr0423@luyangyi Could you both discuss at our meeting Friday how we can adapt the spreadsheet on google drive (the one that has my Moves data for September) to our data model? Please let me know, thanks. Some of the issues you should cover are:
how can the anonymous data be parsed in order to further protect the user ID's privacy
how can the data be cleansed/restructured before applying proximity filter to entire data set of trips
how can we show the data once we've applied a proximity filter to it?
Our evolving marketing and advertising piece that will enable us to convert ordinary, mobility constrained citizens into mobility data collectors (see link). This will be our third deliverable on 12/3
Survey a sample of car owners by asking them a series of questions related to carpooling. The survey would ask them to select the circumstances under which they would be willing to use a mobile app that enables them to offer rides to others, provided they never have to change their routes or their timings. These circumstances would include questions on the following considerations and perhaps others:
driver safety
driver preference
driver cancel-ability
rider accountability
rider payment
rider review-ability
Any others?
Joseph, will you lead this exercise with help from me, others in our group, and Dilek? Dilek from the Air-Quality project has already created and run a survey (see link).
What user guide content might compel someone to become a location leader and 1) download Moves and 2) persuade their location peers to do the same. Four key content areas are:
convince them that their efforts will lead to a a new approach to carpooling in which they will be in the driver's seat and be able to...
save money
save time
save the environment
tell them what they have to do:
download the app
allow the app to access their location, even when they are not using it
create an account on the app (this will allow them to export their data to their laptop)
address their concerns:
assure their anonymity
assure their privacy
offer them some tangible benefits in exchange for their help
let them know how many carpools they might have participated in during the period
visualize their mobility for them (optional, and only if they request it)