2016-11-15 Tuesday Team Meeting
Dundee: Simon, Dominik, Mark, Simone, Will, Ola, Jean-Marie, Helen, Roger Remote: Sebastien, Josh, Gus, Melissa, Liza, Eleanor, David, Emil, Chris, Kenny (14:23)
Agenda - 2:00pm Start
Accepting minutes from last meeting
- accepted
Project Timelines (2-3 minutes each)
Spaces - 14:05 UK
-
Files/Formats/OME Files (Sebastien)
-
push for BF m2: JPEG-XR, Options API, Tiling API
- will advise if need to push m2 to early next week
-
Jean-Marie: when for more JPEG-XR testing?
- Sebastien: tomorrow
- Dominik: can help with Windows testing
-
push for BF m2: JPEG-XR, Options API, Tiling API
-
Metadata (Josh)
-
Demo3 (“banana”) up and running at EBI.
- interesting issue with siRNA making OMERO structured annotations inadequate for representing
- will do some smoketesting
- Getting annotations ready for swap.
-
Demo3 (“banana”) up and running at EBI.
-
OMERO 5.3.0 (J-M)
-
5.3.0-m5 tagged today
- open PR if version history is missing anything interesting
-
Simon: when for 5.3.0?
- Jean-Marie: February. m6 end of November; dependency inversion question may impact.
-
Jean-Marie for version/platform decisions need to first take into account impact on CI and avoid blocking others' work
- also need to consult community
- comment on Trello card this week
-
Question about Insight?
- moving to maintenance mode, refocusing on next-generation viewer
- Dependencies/version
-
5.3.0-m5 tagged today
-
Outreach
-
cmso
- register this month
-
frontiers in imaging
- Josh: about to submit abstract
-
cmso
-
Files/Formats/OME Files (Sebastien)
Other releases/upgrades - 14:17 UK
- Figure
- Learning
-
Sysadmin (Josh)
- upcoming discussion on requirements
- discussing cards tomorrow so ping on any greatly vexatious ones
- Website redevelopment/redesign (Liza)
- Smuggler (Montpellier)
Glencoe Update (Chris) - 14:20 UK
- working on scalability, also czi changes
- PathViewer 1.1 to be announced imminently
Demo of new grid viewer by Will Moore - 14:22 UK
-
Roger: can construct overall image from stage labels
- Will: need sample data with units not as "reference frame"
-
Josh: difficult to represent the complexity as need to get a lot of context into the display on the screen
-
Roger: other software like MetaMorph may provide good UI ideas
- e.g., collapse some panes to give up screen for thumbnails
-
Chris: different software has different ideas about these issues and perhaps little consensus on what is best
- Will's work has some similarities with Columbus
-
it can be helpful to see multiple plates at once, depends on type of assay: flexibility is good to cover different use cases
-
Roger: multi-plate is increasingly supported in other software
- Chris: one thing at a time, beyond scope of Will's current work
-
Roger: multi-plate is increasingly supported in other software
- Jean-Marie: please add other-software screenshots (or other useful information) to the SPW card
-
Roger: other software like MetaMorph may provide good UI ideas
-
Will: we probably now have data from Damir
- Chris: Damir's many-field well data is actually uncommon
-
Jean-Marie: of Will's pending SPW grid work what is higher priority?
-
Chris: with OMERO used to shepherd data through analysis workflow, users don't really touch the images until after analysis so initial rendering settings are less important, often only set once
- can set rendering settings on all fields based on baseline from a couple of control wells, can take a long time over realistically sized plates
-
Chris: for public data like LINCS, HMS, initial visual representation is much more important, metrics are there on import
- again, data size means don't want to regenerate thumbnails again
-
Will: so browser refresh after change of rendering settings would be okay?
- Chris: sure; also want to discuss knowing when thumbnails are "dirty"
- Chris: normal to apply the same rendering settings plate-wide
- Roger: generate thumbnail without rendering settings (like a shrunken image) and apply rendering settings on demand during viewing
-
Chris: with OMERO used to shepherd data through analysis workflow, users don't really touch the images until after analysis so initial rendering settings are less important, often only set once
-
Roger: can construct overall image from stage labels
AOB (5 mins max - technical discussions should be highlighted to relevant people and rescheduled) - 15:07 UK
- none
Done 15:07 UK