Page Comparison

• Convener: Kristin Martin 
• Minute-taker: Owen Stephens 
• Re-architecting FOLIO Tri-Council Subgroup: Jen to initially convene to get things moving; looking for a meeting time. Application and Platform Formalisation Working Group:
  • Application and Platform Formalization Proposals WG wiki page
  • Initial meeting w/c 6th November
• Bulk Edit and Lists application are going to have a single group and Slack channel. Name TBA
• Bugfest ongoing and still tests to complete, so any time that people can give to testing would be very welcome
• Swedish Folio Days in Linköping, with 60 participating libraries.
  • Day 1: Information sharing
  • Day 2: Discussing forming a regional Folio community (inspired/similar to the German community)

Kristin Martin 
all

At the 10/19/23 PC meeting, we got an overview from Yogesh Kumar and Oleksii Petrenko on the current state of QA testing. There is some pre-bugfest UAT being done by SMEs, such as MM and Acquisitions, but could other SIGs do this as well? What dev environments are being used to do the UAT?

Current UAT Testing FAQs

You can add comments here too: Ongoing Discussion: Current state of FOLIO Quality Assurance testing

Jennifer Eustis:

• Involved in Bulk Edit UAT (PO = Magda Zacharska). UAT takes the form of a Google form to work through and answer questions, and usually also go through the UAT together as a group. UAT testing usually done before release. 
• Also done UAT for Lists app. Worked in a similar way
• UAT with data import (Ann-Marie Breaux) using a Rancher environment 
• UAT of data import "slicing technique" (splitting large files into smaller files to make large imports work more effectively) on 5 Colleges sandbox environment

• UAT for the Purchase Order "History" feature (PO = Dennis Bridges)
  • Worked in a dedicated test environment
  • Worked through a Google form
  • Significant time commitment
  • Posted results in Google form
  • Then reviewed in SIG

Martina Schildt

• UAT for Financial Year Roll Over (FYRO) (PO = Dennis Bridges)
  • Similar experience to Kristin
  • In both cases the need to setup realistic data was time consuming
  • Lots of note taking during testing
  • But worth it as allows for the influencing of development 
  • Worked locally as teams to do the testing and then were able to share experiences with the SIG

Is there anything that the PC can do to help with UAT

Owen Stephens: 

• Could we help coordinate announcement of testing - e.g. a dedicated slack channel (as well as the usual SIG channels etc.)?
  • A dedicated channel (suggested by Owen Stephens, supported by Martina Schildt as better than the existing channels. Ian Walls commented "encourage looking at REDUCING the number of channels in FOLIO Slack")
  • folio-general (suggested by several)
  • folio-implementers (suggested by Charlotte Whitt)
  • Also a wiki page to collate together? (suggested by Maura Byrne)
• Would there be a way of getting data set up in advance of the UAT commencing so that volunteers don't have the overhead of setting up data for testing

Maura Byrne:

• Does UAT happen at particular interval?
  • Owen response: Not really, as they tend to be at the point when the feature is in development/developed enough to put in front of users
  • Charlotte: UAT done at the point where the developers are able to respond immediately to feedback

Thomas Trutt:

• Can we use Slack Groups?

Kristin Martin:

• Does it matter to the POs? Is it easier for one channel?

Bad / Cons to UAT

Jennifer Eustis

• UAT is time consuming
• Should be clear about the purpose of UAT
• Is it for a purpose or just bug testing
• Are there other approaches that could be used instead (e.g. designs presented up front)

Thomas Trutt (in chat):

• I feel for new apps, and large changes UAT makes sense. For smaller changes it doesn’t.

Martina Schildt

• UAT important where there is very varied practice and data across institutions - where more testing and more involvement from across the community is important

Ian Walls:

• how I think it should work:

1. PC identifies functionality FOLIO should have, and delegates to a SIG
2. SIG defines features, and hands to PO + their dev team
3. Devs dev
4. PO presents code produced to SIG for UAT
5. repeat steps 3-4 as necessary
6. SIG signs off on features
7. Features undergo Quality Assurance (currently bugfest)
8. if passing, then PC approves release (later suggested this could be dropped)

• Clarifications from Ian:
  • The PC should have control over the definition of what features go into the product - isn't that the point?
  • Need to close the loop - so where ever decision made of a new feature, PC should sign off that it's been delivered and can be released

Kristin Martin

• PC haven't been involved in this level of detail previously

Owen