Project Open Data layout redesign

[rebeccawilliams]

This issue is to discuss how Project Open Data can be laid out as a more user-friendly playbook for all agencies, including those not currently mandated under the open data policy. A Project Open Data v 2.0, if you will, but let's not change the name.

Some inspirational options include:

opendata.guide by @mheadd (CC-BY)

The U.S. Digital Services Playbook

@melodykramer also mentioned an answering common questions approach.

What else would be helpful here?

cc @leahbannon @ErieMeyer @philipashlock

See also #450.

[leahbannon]

The upcoming gov style guide work might be really useful for this (maybe in a month-ish).

And some other ideas...

[rebeccawilliams]

• http://datasf.org/resources/

[rebeccawilliams]

Adding these as a reference as well:

• Government Design Challenges: http://healthdesignchallenge.com/
• The Budding Gov Wide Pattern Library: https://pages.18f.gov/govt-wide-patternlibrary

@mollieru @juliaelman, do you two have thoughts guidance on designing for guidance? (also dashboards? https://github.com/GSA/project-open-data-dashboard)

[rebeccawilliams]

Current sections on the website include:

• 1. Background
• 2. Definitions
• 3. Implementation Guidance
• 4. Tools
• 5. Resources
• 6. Case Studies
• 7. Open Data Engagement

Adding some notes about what I think the redesign should include:

• Be approached as a tool for all agencies, not just the CFO-Act agencies that must comply with the Federal Open Data Policy
• Bundle tools, resources, and case studies w specific tasks, for example
• Consider full data lifecycle
  • Plan Data Collection
  • Data Collection
    • procurement / grants (data should be supplied to gov in )
    • electronic / structured / validation checks
    • tagging records management / review issues upfront
  • Review
    • Common PII issues: HIPPA, FIRPA, SSNs Names, Addresses
    • FOIA protected? what parts, what exemption?
  • Store/Share - data.json POD metadata
  • Maintain - automated updates to data.json from collection/review space if needed
  • Engage
    • To prioritize data improvements / release / new data collection
    • Contribute to / validate USG data with non USG data sources (see Open Street Map)
    • Collect new data (see Smithsonian digitization project)
    • more things
  • Archive / Dispose / Iterate
• Answer common questions:
  • How to conduct a comprehensive data inventory
  • How to fulfill data.json / ODP requirements for data.gov harvesting
  • How to ETL (ft ETL -> data.json)
    • cleaning data
    • reformatting data (parsing PDFs, HTML -> CSV)
  • How to host big data
  • How to create APIs
  • How automation can be inc to your data management process
  • How to include open data in records management/FOIA process
  • How to protect PII with math

[philipashlock]

Here's a generic layout to start with to have something to talk around. Starting with the frontpage can also help us decide the items that should be prioritized and refine the organization of the different sections

[rebeccawilliams]

Thanks for this mock-up!

While Guidelines are mentioned in the NAP, they don't exist apart for the Implementation Guidance right now so I wouldn't lead with them in mind.

I think the high-level categories to include right now are:

• Federal Open Data Policy
• Implementation Guides - "How To"
• Resources & Tools
• Dashboard (include a screenshot of the Project Open Data Dashboard rather than an icon)
• ? Cross-Agency Data Initiatives (CAP Goal+)
• ? Data.gov
• ? Impact/Case Studies
• ? Training

[rebeccawilliams]

Can we see this mock up with the Handbook template? http://opendatahandbook.org/

[philipashlock]

Here are some analytics for the site over the past six months. The v1.1 schema page is by far the most accessed page, receiving over 40% of entrances compared to less than 30% for the frontpage.

And here's a drill down of traffic flow starting from those 12k (~30%) frontpage entrances

[rebeccawilliams]

Thanks @philipashlock for the web analytics. My takeaways:

• It looks like users use this site for fundamental open data concepts and research (principles, open-standards, glossary).
• It looks folks are going directly to the schema v1.1 page, which makes a lot of sense since we link to it for troubleshooting. Is it coming from any other pages that link to it online right now, like the Digital.Gov's How to Get Your Open Data on Data.gov, blogposts, and webinars? Also, how many of these Page Views are coming from inside GSA/USG/the US?
  • It looks like users are also navigating to the v1.1 schema page from the home page.
• I'm surprised (and discouraged) that the guides explicitly called out in the IDC are not getting more traffic:
  • https://project-opendata.cio.gov/redactions/
  • https://project-open-data.cio.gov/engagement/
  • https://project-open-data.cio.gov/open-licenses/
  • https://project-open-data.cio.gov/v1.1/api/
  • https://project-open-data.cio.gov/v1.1/collections/

[rebeccawilliams]

I think when refreshing the site with a homepage we should balance these goals:

• Limit the number of high-level buckets - (would 6 be the max? 3? -- I see some of the examples above have more, but I think the Implementation Guides bucket could list more and the homepage should be as straightforward as possible)
• Tell a coherent story about open data in the Federal government (beginning, middle, and end)
• Provide the most used content on the homepage
• Encourage more eyes on content that may have been overlooked: CAP Goal; API documentation guidance, etc