UCB deployment planning, July 2011

Owned by Chris Hoffman
Last updated: Jul 01, 2011Version comment
5 min read

Here are some planning notes for July 2011 (especially given some vacations and absences).

Please do refer to this!  Please do update this!  Please do understand that there are too many moving targets to consider this a well formed and definitive plan!

Major milestones

  • July 11 -- next version of UCJEPS for NSF seaweed project
  • July 18 -- next version of HAVRC for initial testing by HAVRC and CEDVRC
  • July 25 -- first customized version of PAHMA for initial testing and review

Goals: General

  • Get team members up to speed on customizing CollectionSpace, especially given the new tenancy model. (Rick and Aron are best experts for App and UI layers; Richard for Services.)
  • Patrick and Richard: Best model for where code lives, on branches or trunk, in UCB svn or CSpace svn.
  • Do some training or orientation for Leslie and Michael so they can do some customization work -- minimally changing field names, hiding fields, reordering fields; hopefully adding simple fields; hopefully adding field groups; hopefully adding repeating fields and field groups.
  • Set up development environments on Leslie's (new) Mac and Michael's laptop so they can do some work on PAHMA.  Patrick recommends that Michael and Leslie try to use the existing documentation to do this.  Chris will try to find the email thread where this was discussed recently, but do we first need to figure out how customizations work given the new tenancy model?  As of July 1, Rick was still reporting fundamental issues.
  • Place authority: John and Jan work on the default schema for Place authority records and the schema extensions for Art and Natural History (unless they can all live happily together).
  • Custom authorities: For now Patrick thinks we should assign these to his team to build once schemas are complete.  As long as we keep these semantically simple, they should be pretty easy to do.
  • Monitor bugs that were not fixed in 1.8 to allow custom authorities to work properly. (I can't remember the Jira tickets for those.)
  • Chris: Look for or create Jiras -- complete Media Handling "Link to external image"; complete other relationships (e.g., synonyms); pan and zoom; update display name on related records.
  • Patrick: Complex relationships between objects -- evaluate whether same design and approach as in wireframes for vocabulary terms would work (hierarchical plus other relationship types in drop down).
  • Patrick: Some additional tasks for Leslie.
  • ETL best practices and training
  • Lam: Learn how to do data import from Glen.
  • Glen: Learn how to do batch processing from Patrick.

Goals: UCJEPS

  • Main goal: Get a next version of UCJEPS ready for acceptance testing with Andrew. Available beginning of week of July 11 if not earlier.
  • Coordinator: Glen
  • Leslie: Wireframes for next version of 1.8-UCJEPS with minor changes
  • Where should code live and how do commits happen (based on approach defined by Patrick and Richard)
  • Rick: Make customizations to UI
  • Glen: Wire to app and service layers.
  • Leslie and Rick and/or Glen (whatever subset wants to): Test 1.8-UCJEPS with Andrew (with camera, specimen sheet)
  • Richard and Glen: Develop game plan for finding out what is slowing down media upload on Media Handling screen.
  • Later -- Chris: Request UCB VM for UCJEPS CSpace (but what do we call it?! ucjeps-cspace.berkeley.edu?)
  • Lam -- get up to speed with Talend and data: Chris and Lam to work on next steps for UCJEPS data though for this round, the goal is to have something to do next round of acceptance testing.  Minimal data needed and mainly for authorities.
  • Question for Patrick: Is there anyway to create a "Save this record and create a new empty one, maybe with the scientific name from the last one" in the Media Handling screen? Use Refresh for now, AFTER record is saved.  Leslie, we need to test this!
  • Leslie?: Develop specs for batch process that creates initial catalog record from Media Handling record.

Goals: HAVRC

  • Main goal: Have a next version of the HAVRC CollectionSpace prototype ready for initial testing (by Lynn and Jason).  Target beginning of July 18 week for availability and testing.
  • Coordinator: Jan (keeps an eye on the main goal, the other activities outlined below, knows where we are, and if someone is stuck, they contact coordinator. Coordinator can make a decision, even to defer, or call Chris on his cell phone)
  • Assumptions: Based on CSpace v1.8, running on havrccspace; with person, organization and cataloging (work) records in the 1.7 instance; with minimally same customizations to UI and schema as in 1.7 version though with all fields saving to database.
  • Where should code live and how do commits happen (based on approach defined by Patrick and Richard)
  • Chris: Clean up 1.7 jiras and set up v1.8-based project and initial tasks
  • Glen: Upgrade havrccspace server to CollectionSpace 1.8.
  • Jan: Based on review of UI and schema and comparison to SMK approach, update schema; work (with Leslie?) on new wireframes.
  • Jan: If needed, create some new high level Jiras (e.g., "Media Handling screen: add new fields and make some revisions per the wireframes at http://...." -- the wireframes can contain the details).
  • Rick: Make customizations to UI; redo earlier customizations if needed given new tenancy model.
  • Glen: Wire customizations to app and services.
  • Jan: Testing, coordination, and liaison to HAVRC and CEDVRC.
  • Later or if Jan is bored: Develop specifications for batch process that creates Media Handling records from information on Intake (Order) screen
  • Lam get up to speed with Talend and data: Susan has passed her Talend and Kettle files to Chris.  Jan is documenting some of the high level assumptions in current data (e.g., only one Creator Role in each Person record).  Probably get Lam involved in next steps for HAVRC data, though for this round, the goal is to have something to do next round of acceptance testing.  Minimal data needed and mainly for authorities.

Goals: PAHMA

  • Main goal: Have a solid 1.8-based prototype, demonstrating a significant set of customizations, available for testing by Michael, during the week of July 24.
  • Coordinator: Michael
  • Assumptions: Run on one of the existing slicehost slices.  Increase RAM if you need to.  Later, we will move this to a VM in the UCB data center.
  • Where should code live and how do commits happen (based on approach defined by Patrick and Richard)
  • Glen: Convert one of the PAHMA slices to 1.8 (work with Michael on which one and what to do with data)
  • Michael: Monitor our ability to do different kinds of customizations in order to decide for example, whether we can do 0, 1 or more custom authorities.  Probably assume we hold off on new NAGPRA Claim procedure until a later iteration.
  • Michael with assistance from Leslie. Wireframes and customization docs. This iteration of documentation basically done by July 8.  Enough work to do now ;-)
  • Rick, and possibly Michael or Leslie: Make customizations to the UI.
  • Glen, and possibly Michael or Leslie: Wire customizations to app and service layers.
  • Glen, or Glen with Lam? Data for this iteration.  Maybe just extract XML versions of Person and Organization authority data from existing slicehost, as XML using Nuxeo workbench, and transform as needed for loading into 1.8?  -- Susan has give Michael and Chris her Kettle and Talend files.  We probably have CSV files for Org and Person somewhere.
  • Later -- data for next rounds.  Michael might continue to use Kettle; then take output from there to XMLify using Talend?