Page tree
Skip to end of metadata
Go to start of metadata

Time/Place

  • Time: 11:30am Eastern Daylight Time US (UTC-4)
  • Dial-in Number: (712) 775-7035

Attendees

  • Julie Allinson
  • Chris Awre
  • Robert Cartolano
  • Aaron Choate
  • Stefano Cossu
  • Dan Coughlin 
  • Karen Estlund
  • Tom Cramer
  • Jon Dunn regrets
  • Declan Fleming 
  • Sarah Fredline
  • Mike Giarlo
  • Wolfram Horstmann 
  • Neil Jefferies
  • Debra Kurtz
  • Susan Lafferty
  • Mark Leggott 
  • Jonathan Markow
  • Steve Marks (star)
  • Tom Murphy
  • Sandy Payette
  • Matthias Razum
  • Robin Lindley Ruggaber
  • Dan Santamaria
  • Thorny Staples
  • Jim Tuttle
  • Keith Webster
  • Evviva Weinraub
  • David Wilcox
  • Andrew Woods
  • Maurice York
  • Patrick Yott

Agenda

Topic

LeadNotes

Membership update

David

 

Fedora Value Proposition BrainstormingDavid 
Import/Export FeatureAndrew 
API SpecificationAndrew 
OSF/Fedora Integration Use CasesDavid 
Fedora Performance at Scale IssuesAndrew 

Previous Actions

None mentioned.

Minutes

1) Membership update - summertime is slow, but MIT has shifted funding from Dspace exclusively to gold for Dspace and gold for Fedora.

They are diversifying their repository platforms and are attempting to allocate staff time to Fedora work. Exact use is not known to us.

Prospects research is slow over the summer, but David is going to pick it up in the next few weeks. As always, please update the spreadsheet with any prospects you might know/think of.


2) Value Proposition Convo

Tricider page is for voting on some one-sentence statements to determine what is considered "valuable" for fedora. Upvote the ones that resonate with you or add the ones you want.

Q: What is the level of pitch? It's a high-level one, so for administration and others who are making technology decisions. More management level than in-depth technical specs.

This will be folded into a working document that will lay out the value proposition for Fedora. Marketing material, of a sort.

Noting we also do want to reach out to technical folks too though.

Do we want to highlight the Hydra and Islandora communities or focus on Fedora itself? General consensus that we need to address them, but understand that choosing Fedora != choosing Hydra or Islandora. "Use one of these exciting front ends or create your own!"

Possible other selling points/things that need to be addressed:

  • Can use a standard library discovery layer overtop.
  • What is Fedora and what isn't it? List of adjacent/competing projects - one to use one or the other.

After a couple weeks, this list can be shopped around to technical group and other broader communities for inclusion into a proper statement.

Leaders, please go through the list!

 

Andrew W presents current top technical priorities for this calendar year.
3) Import/Export Feature

Have had a couple of meetings, last two Fridays + this Friday to pull together stakeholders/use cases for import/export. Sprint to occur later in August incl. Import/export in advance of upcoming Penn State sprint to further this effort. Goal: i/e functionality meets discussed use cases. Power statement: this will allow the "rebuild" capability that was present in Fedora 3. Usable for repair/migration/object transfer to other Fedoras or other preservation systems.

4) API Specification

Significant milestone for Fedora. This separates the versioning of the API from the versioning of the underlying implementation. This provides a significant level of stability for client interactions. Fedora API will continue to be more aligned with community and industry standards. Different pieces of this conversation are being led by different members of the community. Andrew highlights that although other things may seem more pressing, this is really damn important. NB - we do have people doing this!


6) Performance scaling

Work that has come out of Princeton re: the way Fedora is modeling relationships. If you have a collection with a lot of members, it takes a long time to load all those members. Ben A and Esme are looking at solutions for this long load time. This is a good example of the community converging around problems manifesting as systems move to production. This isn't related to Sufia, but rather the way Fedora handles objects w/ a lot of links. Exacerbated by PCDM's modeling of these objects.

Is this a Modeshape issue or Fedora issue? When you have an object that has relationship to a lot of children, those all get loaded by Modeshape. Is it worth pushing the Modeshape community to help get involved? Declan: it isn't a problem with modeshape, it's a problem with the way we use it.


5) OSF/Fedora Integration Use cases

Interest exists in integration between Fedora and OSF. Some folks already doing this: Centre for Open Science at JHU and Notre Dame. JHU has a fairly lightweight implementation that uses Fedora as a deposit location a la Dropbox. David & Andrew going to have a conversation with OSF about more fulsome ways to integrate the two.

Robin: Does OSF treat Fedora as a source or a destination? DW: not sure, uses "WaterButler" to connect and show the objects in repos.

Notre dame has interest in "one-touch archival deposit".

UVA interested in OSF as a research workflow framework. Possible conversation at the VIVO conference?

There is a general interest in people exploring this. Might be worth convening a call w/ Fedora/OSF stakeholders together to think as one. Andrew and David will take that idea forward.

 

Other Business

Q: request to get Nick to Penn State. It's all going to work out. We are most likely looking at a cost share arrangement.

Is there a way to get Nick/York on Leaders call? Probably, based on his contributions.

Actions