uva-dsi / open-data-lab Goto Github PK

an initiative to provide infrastructure for reproducible workflows around open data

License: GNU General Public License v3.0

Jupyter Notebook 97.98% Shell 0.15% Python 0.62% HCL 0.06% TeX 1.20%

opendata open-data open-research open-research-data open-research-experiments workflows openscience open-science datascience data-science

open-data-lab's Introduction

Open-Data-Lab

Contact: [email protected]

Introduction

The Open Data Lab is an initiative to provide infrastructure for reproducible workflows around open data. It is spearheaded by the Data Science Institute at the University of Virginia. Partner organizations include: the UVA library, the Center for Open Science, and the Public Library of Science.

Getting Started

For any and all questions please email [email protected]
To be added to our mailing list please fill out this form (< 30 seconds).
Currently we are in closed Beta and serve the Data Science Institute. To express interest in working with the Open Data Lab please fill out this form (< 5 minutes).
For help accessing local HPC see this link.

Goals:

Reference Data - Provision of a large corpus of open research and other data (CC0) and associated analytics (OSI-approved licence) across disciplines to be an exemplar for the representation of open structured and unstructured reference data.
Training - Use the reference data and analytics to train the next generation of data scientists and raise literacy around data at UVA and elsewhere.
Reproducibility - Facilitate scientific reproducibility through execution of workflows associated with the data and analytics.
Productivity - Science is currently a highly inefficient process. Provision of public executable workflows can accelerate the process.
Public engagement - Open data exploration pipelines based on open data are a powerful means to engage with the broader public, both around open data and data science in general and the issues that they can help address.
Broad Recognition - as a valuable community resource.

Currently we are in closed beta testing. If you are interested in the project please watch us on github and email [email protected]. Goals for closed beta:

Establish storage platform
Establish computation platform
Establish discovery platform
Accumulating a corpus of reference data

open-data-lab's People

Contributors

Stargazers

Watchers

open-data-lab's Issues

How should data in/ from the ODL be cited?

Of course, we should follow the data citation principles. Pinging @timclark .

Define criteria for reference datasets to be included in the Open Data Lab

some potential dimensions:

relevance
- to other DSI activities
- to communities around DSI
openness
- licensing
- formats
- standards

Set up a public ODL mailing list

Set up some suitable labels

as per https://github.com/UVA-DSI/Open-Data-Lab/labels

Document Continuous Integration

as per #3 and #4.

Define color code for labels

Probably not necessary for the small number of labels we have right now, but potentially very useful if that number grows.

Adding #12 for tracking.

Create Documentation

Set up suitable milestones

as per https://github.com/UVA-DSI/Open-Data-Lab/milestones .

Create CI

Identify datasets for potential inclusion in the ODL

One way to start looking into this would be to check open resources like

https://github.com/awesomedata/awesome-public-datasets
and see how sustainable/ usable the data are there.

On that basis, we could then decide (see also the inclusion criteria in ODL, as per #18 ) as to whether we'd like to go for datasets scoring high and/or low / average on those scales.

What kinds of stats/ metrics would we like to gather?

Would be useful to start thinking about that now, so that we can demonstrate impact right from the start of the public beta.

Launch VM protocol

send sms to AWS with project ID and email address
lambda function stands up vm
lambda function puts .bash_profile on vm with mounting commands
lambda function emails user with .pem file and command to connect