ODK 1 TSC Call - 2019-02-20

(i'm taking over facilitator role from Tom so he can focus on the project on-boarding policy)

ODK 1 @TAB calls bring together the Technical Steering Committee to discuss roadmaps, working groups, and other issues of technical governance. Everyone is welcome to come to these calls, but only TSC members may talk.

The calls are held every two weeks in our UberConference room from 16-17 London time. We put the agenda, audio, and transcript of every call in this document.

Our next call will be Wed, Feb 20th, from 16-17 London time (see in your timezone).

The agenda is tentatively as follows:

  • Visioning session for 2019 (~30 minutes) looking at how we might evolve the engagement of the TSC beyond the calls.
    • Goal oriented (e.g., We set a goal of n high-quality contributors)?
    • Non-code oriented (e.g., Schedule topics on marketing, funding, etc).?
    • Code triage (e.g., we pick a repo, get it running, clean up issues)?
    • ???

The agenda can be seen in the agenda document

If there are topics you would like to add to the TSC's agenda, please comment below. :point_down:

I wanted to float the idea of perhaps making certain checks in Validate warnings, as opposed to fatal errors. Some things are obviously always fatal errors in a form, eg unsupported data types, unknown binding properties, etc. But others checks are arguably more best-guess, and sometime prone to throwing false positives; eg type checking XPath expressions, and dependency checking (ie apparent cycles in DAG graph which are not actually run-time recursive, eg index-repeat(), coalesce(.,), etc).

All Validate's checks are useful, but when it gets it wrong (false negatives) it prevents being able to deploy valid forms until the problem is fixed and any tool(s) using Validate are updated. So it might be time to take a closer looks at all Validate's checks and make a conscious decision which should actually be fatal errors vs warnings.

It might also be an opportunity to compare (align?) ODK's form validation strategy with Enketo's, since they differ (@martijnr?)

Is this sufficiently controversial to warrant TSC discussion? :slight_smile:

Since we have current members running, I think we should take some time during the meeting to discuss the process of selecting new members in a way that reduces conflicts of interest.

My proposed process is:

  • The votes will be the same as always: +1, +0, -0, -1
  • Consensus approval will require three binding +1 votes from different organizations and no binding -1 votes over a 72 hour period.
  • Current members of the TSC (including Alex and Hélène) will send their private votes via email to all three of the members who are not up for election (Guillermo, Dan, Martijn) with their votes for each candidate.
  • Guillermo, Dan, Martijn will tally votes and report back to the current members.

I think we should also have a quick non-recorded discussion about the new applicants.

@Xiphware I think we should have that discussion for March when Martijn will likely have some cycles to contribute to it.