WZDx Specification Subgroup Meeting Notes, 2019 10 21

Virtual Attendees

• Volpe - Nate Deshmukh Towery
• USDOT BTS - Derald Dudley
• Booz Allen Hamilton - Ashley Spurlock
• Booz Allen Hamilton - Mahsa Ettefagh
• ICF - Amy Tram
• General Motors - David Craig
• Ver-mac - Serge Beaudry
• Utah DOT - Chuck Felice
• KCUS - Jacob Brady
• Kentucky Transportation Cabinet - Drew Clark
• ICF - Rachel Ostroff
• SharedStreets - Kevin Webb
• KCUS - Michelle Boucher
• Seattle DOT - Craig Moore
• Trillium Transit - Aaron Atrim
• Booz Allen Hamilton - Eric Ricciardi
• Noblis - Kellen Shain
• Los Angeles Metropolitan Transportation Authority - Kali Fogle
• RTC of Southern Nevada - Brian Hoeft
• Regional Transportation Commission of Southern Nevada - Julia Uravich
• Volpe Center - Mark Mockett
• TomTom - Rob Hoyler
• Spacial Front - Leslie Haig
• Michigan DOT - Jake VanZalen
• Uber ATG - Sammy Ghorbanian
• Waze - Dani Simons
• Google Maps - Sandra Tseng
• Oregon DOT - Joseph Kuenzi
• Massachusetts DOT - Neil Boudreau

Presentation Slides can be accessed [here]

Purpose and Intended Outcomes:

• Finalize the voting results of pull requests which are now closed on GitHub
• Discuss the pull requests which are open (or will be open soon) for voting on GitHub
• Remind all to vote for open pull requests
• Decide next steps on pull requests still not open for voting
• Discuss meeting timeline for broader WZDWG meeting

Agenda

• Sign in and Welcome
• Group Discussion - Closed for Voting Pull Requests
• Group Discussion – Future Pull Requests
• Group Discussion – Alternative Pull Requests
• Action Items and Next Steps

Discussion Summary

Pull Requests Closed for Voting: (#16)

• 12 yes’s, no, neutral
• We were hoping for 50-100 votes, so please go comment. (#33)
• Also looks like it will be passed, 11 yes’s
• Quorum is 12 votes, we expect additional input. (#38)
• 10 yes’s, still need 12 votes for that to pass
• Quorum means 12 votes overall, not 12 yes’s, just need 12 votes overall

Future Pull Requests for Voting: (#50 – Relational Data Model)

• Employing relational data model will allow geospatial visualization tool that v1 did not allow, and helps to organize the data and makes it more consistent, easier to retrieve, and more flexible/scalable.
• Also promotes standard-centric development of applications.
• The specifics of changing to a purely relational data model impacts the subsequent pull requests. (#40)
• The first of several requests contingent on closing out #50 first. (#44)
• Again, the content being asked for from data producers would be the same as we said in Derald’s request #50, but instead of an enumerated number structure, it would be a table. We’ll put up the content up for voting on this once we decide on the data structure, i.e., continue with existing or move to the relational structure. Then, after we decide that on (50), we can vote on it.
• This sort of mirrors Waze-compliant feeds, referring specifically to vehicular impacts
• In response to Kali - We would define content but we’d be wary of defining field types if that makes sense, initial thought was to model it something closer to GTFS which is all text-based strings
• Aaron - The field types are specified in, GBFS, for example
• There are certain fields for which a number isn’t provided a validator would be thrown in there
• Overall, we’ll take field types on a case-by-case basis
• Kali asked question about using LRMS so for example, if lane 2 is closed, people would still need to know how many lanes there are to determine which one is lane 2
• Responded that we’d still have the element detailing total lanes, and that it would be captured, so that would address that Alternative Pull Requests:
• If 44 doesn’t go through, we’ll move forward with the #45, #46, #48 pull requests on screen

Action Items and Next Steps

• Some pull requests got votes during the meeting, so now they’re at 12 votes and will pass. Please go vote.
• We will open Derald’s pull requests #50 ASAP. Any other comments on #50? None.
• After Subgroup has voted on these changes, they’ll be packaged and presented to the larger WZDWG on November 12th or so. That group has 10 days with their own quorum, and then based on those votes from the WZDWG, we’ll go back and package everything together to release the final v2 specification.
• We received questions about how that package will look like when presented to the WZDWG, whether it’ll be specific line items presented to them or just the package overall.
• Have a check-in with co-chairs later this week.