Upcoming calls on next-gen bioimaging data tools (Starting Oct. 29)

Happy October, everyone. :jack_o_lantern:

We’re following up, admittedly a bit late, on one of the discussion themes of the 2020 OME community meeting that has been echoed elsewhere since. There was a strong message that we should continue having “regular” global meetings in the same or similar format to keep these conversations going. Various discussions have followed with different ideas on how to run such calls. There have been office hours, coffee hours, … happy hours. We’re all trying to make the most of these interesting times.

Here’s our starting proposal:

  • Frequency: once per month
  • Speakers: 1-2 per meeting, centered on some theme
  • Talks: 20 min max. From our OME2020 experience, we’d suggest these be pre-recorded
  • Length: Run two 60-90 min discussion sessions
  • Time: At least 1x coverage of the American, African/European, and Asian/Pacific timezones
  • Attendance: Free & open with public notes and recordings

The idea would be to set a schedule for the first 3-4 months, starting toward the end of October*, and then see what everyone thinks. The focus would broadly be on next gen bioimaging data tools. The biggest unknown is the name! We’ve been thinking about:

  • Nothing Special at All
  • BioImaging Data Network
  • World of Bioimaging
  • Next Gen BioImaging Fun
  • Or … ?

Most importantly, we want to keep in touch and up-to-date while moving all of our various activities ahead as in sync as is possible. If we you’d like to present or cross-organize, do get in touch. (Here or via a message to @ome)

As always, very much looking forward to seeing everyone again. Here’s to better times.
The @ome team.

P.S. The only major conflict noted so far is SciPy Japan starting Oct. 30, so currently holding October 29th. Please save the date!


Update: Connection information for those who have “registered” is now available in https://forum.image.sc/t/connection-information-for-next-gen-call-on-oct-29th/44210

21 Likes

CC’ing interested parties from recent calls:


* The few that had registered interest but aren’t on image.sc I’ve invited separately.

9 Likes

Certainly nothing crazy like “Multichannel Brightfield Time-lapse Human Networking”

Or “BioImaging Analysis Community Happy Hour”

There are good reasons I almost never name anything…

1 Like

To keep the focus on NGFF and to lure some trekkies: “Bio Formats: The Next Generation” - to boldly go where no format has gone before
Damir

4 Likes

Hi
Thank you @joshmoore for starting this thread.

I am definitely interested in the plan you delineate.

In particular, as you all know, considerable interest is in metadata guidelines and how the metadata should be represented in the NGFF.
In particular, one topic might be the need to develop community BEST PRACTICES on how a shared METADATA MODEL such as the 4DN-BINA-OME extension https://arxiv.org/abs/1910.11370 we are working on at BINA (https://www.bioimagingna.org/qc-dm-wg) and QUAREP-LiMi (quarep.org) (https://arxiv.org/abs/1910.11370), would be REPRESENTED in addition to the current XSD format (which might be abandoned soon).

Currently, at the model (https://github.com/WU-BIMAC/MicroscopyMetadata4DNGuidelines) we are working on is represented as:

  • XSD
  • xlsx spreadsheet
  • an Entity Relationship graphical representation of the schema (using graphml)

In addition, @norio.kobayashi at RIKEN and @joshmoore have proposed using RDF/OWL and JSON-LD as the recommended format.

The problem is that all of these formats have to be kept in alignment and there is for example the need for shared converters.

Is there any interest in discussing these topics?

Thanks for your attention.

Caterina

3 Likes

Thanks @joshmoore!

It’s clear to me that people are working on many aspects of ‘NGFF’, from formats & metadata to pipelines & visualization, but mostly in the shadows. GitHub issue stalking has been my primary method of attempting to keep tabs.

I’m not sure how to fit it in, but I like the idea of adding a few lightning talks somewhere along the way. At minimum, I’d love a show & tell of how people are using n5/Zarr today.

4 Likes

Very exciting @joshmoore, count me in!

1 Like

Thanks @joshmoore I am in :slight_smile:

1 Like

I also would love to join @joshmoore :slightly_smiling_face:

1 Like

I am in(terested)as well.

1 Like

Of course I am in :grin: as well.

1 Like

Wow. Ok! Loving all the title suggestions. Keep them coming. We’ll have to work in as many of them as possible…

In general, I’d say let’s take all this good strong pre-weekend enthusiasm as sign that we’re going to to run with this, so:

:point_right: For interested participants: Please save the 29th of October as the likely first event. Again, discussion sessions will be repeated.

:point_right: For interested presenters, consider the topic “Next Gen Data Access” roughly following @perlman’s idea of showing what people are already doing or being forced to do survive in these times. If someone wants to go ahead and record a talk, well, the content will be put to good use.

At the same time, @perlman’s other and @Caterina’s ideas sound complementary to this one. For the former, if lots of lightning talks show up we’ll figure out what to do with them. Otherwise, we can discuss on the 29th. For the latter, it sounds like this is more a working call that we should try to have earlier in October to add to the discussion. Does that sound right, @caterina?

One way or the other, a pleasant weekend, one and all.

4 Likes

@joshmoore I am interested as well. I am particularly interested in cloud-related topics and in particular in bioimaging tools deployment and use in the context of the European Open Science Cloud.

1 Like

@joshmoore
Thank you for your message.
I will reserve October 29th for the next meeting.

RE: my proposal, yes I think having an earlier working meeting, for example in early October, would be a good idea.

How do you want to go about organizing it?

Have a great week end.

Cheers

Caterina

1 Like

My calendar is marked with a link to this thread — so please post details here when they do arise! =)

1 Like

@joshmoore great idea I’d be interested as well.

A possible topic for a future call I’d be especially interested in is ROI/geometric object storage. I saw the specification for ‘labels’ associated to images, I guess storing surfaces/traces/positions/? as well would be doable and certainly useful. Common ROIs specifications, finally? I don’t know if that’s something that was discussed already…

Thanks again

4 Likes

Well, hello again! Love the idea, and I will definitely be in!

1 Like

Sounds great I’d love to join. Thanks @joshmoore!

1 Like

Thanks @joshmoore , I’d love to join!

1 Like

Josh, thank you for setting this collaboration opportunity up!
(Stephan Wagner-Conrad = swg08 (https://forum.image.sc/u/swg08), So, I’m on the platform already actually. ;))

1 Like