Time: 10:00am ET
Audio Conference:
Join from PC, Mac, Linux, iOS or Android: https://duraspace.zoom.us/j/8128353771
Or iPhone one-tap :
US: +16699006833,,8128353771# or +16468769923,,8128353771#
Or Telephone:
Dial(for higher quality, dial a number based on your current location):
US: +1 669 900 6833 or +1 646 876 9923
Canada: +1 647 558 0588
Australia: +61 (0) 2 8015 2088
United Kingdom: +44 (0) 20 3695 0088
Meeting ID: 812 835 3771
International numbers available: https://zoom.us/u/MO73B
Designing a Migration Path - collection of project resources
|
Sayeed - nothing terribly surprising. one thing that comes to my mind - use cases can they be categorized? The use cases are different if we're using Islandora, Samvera, custom. In some cases, the front end application can help. Want to make sure the message is clear per community.
Linked data support - it doesn't need to be used. So can help people understand that they don't need to transform their data to linked data formats.
Mark J - per community messaging would also help us measure the outcomes and gaps within those groups.
Scott - what percentage of the users are using Islanodroa or Samvera? Are there early wins if we get users from those communities on Fedora 6?
David - in the survey we didn't ask for them to say. But some mention it in their responses.
Tim - When people said in the answers that they loved LOD - did they say how?
David - there was skepticism about those responses.
Erin - LInked open data is too broad a topic. We can't trust the responses from the survey around that.
Sayeed - libraries are more interested in LOD for bibliographic data - not for collections data.
David - we should call that out.
Tim _ this is why I keep hammering this I don't think we have reliable evidence of the need for LOD in Fedora 6
Scott - it could be potentially useful in the future. But I haven't seen anyone leveraging that in the Fedora community.
Tim - Migrations are hard. So what can we do to make it easier - other than everything? What are our obligations? What are the obligations of community members? Could this be another grant?
Erin - I was left with the impression that there isn't a whole lot we could do to get them to move. A major security vulnerability would make them move. They did mention wanting a pilot to look at.
David - We have pilot commitments once Fedora 6.0 is out.
Scott - and once I see there are others on Fedora 6 I would feel better about moving because I wouldn't be alone.
Tim - could the next grant be LYRASIS running a pilot and see what happens. It's not what we can do to get you to Fedora 6 but how can we get you onto supported software
Erin - This Fedora scenario will provide important information to the DSpace and Isalndora communities.
Este - justifying the value of the labour to migrate through a pilot is important - we need that to show people they should do it too. People are asking why don't we use Preservica? We are dedicated to Islandora and want to continue to use it but we're also acknowledging that things are going to change. We need to focus on value
David - Value was a major theme - that's why we're trying to do the preservation and oclf work.
Mark - I didn't see in the report the role of the service provider community. The decision to migrate might be reliant on service provider support.
David - a lot of institutions that don't have the staff to migrate also don't have the cash to pay a service provider
Erin - I would like to pitch a pilot project via the DTS team for the upgrades. It makes sense in the Islandora context because LYR has a hosted Islnaodra service
Scott - the AWS option is storage - whats' that about?
David - all of the options were checked at least once.
Erin - FSU said their goal was reducing complexity and for that reason, they are using AWS. But that doesn't make sense for organizations outside the US
Mark - What about Archepelagio>
Erin - I don't think they have the support to be the main competitor to a community product. The Fedora software didn't solve their problem so they created something that does and that's something to expect.
Este - we're part of the Islandroa collaboration group and it supports ISLE. We talked about Archepeligo and said the community isn't there and we want to go with the community. We are looking at how to work with AWSS to simplify things. Our org is saying "we want you in the cloud" and it's a good reason to move. I also think the service provider question is important. The expertise we have with Born Digital is part of how we are able to do this. We need to keep the eggs in the Fedora 6.0 basket and we need there to be a community of users in that community.
Erin - Value and effort should be the focus of our recommendations.
Sayeed - we're being asked to move to the cloud too. But using the cloud and be dependent on AWS and Microsoft infrastructure and it would be hard to take them out. If you use Fedora 6 in the cloud it gives you some insulation to those hooks. Is that part of the value or values? It can be framed as insurance
Do these users know about the terms and conditions of AWS or even figshare?
What will your community do if you need to change directions?
Scott - some schools might not have options to not move to the cloud.
Sayeed - one of my colleagues has gone to town on researching the terms and conditions of cloud services. ACTION - Sayeed will ask his colleagues to share the info he is compiling in a shareable format.
Erin - this is another area where we can categorize messaging - If you use AWS, Fedora 6 can help with X. If you use the European Open Science Cloud, Fedora 6 can help you with X
Tim - if you're not worried about lock-in, AWS will save you money. But if you protect yourself, if will not save you money. Sayeed, you should write a brief statement we can elaborate on ACTION -
Mark - executive roundtable at recent CNI (spring) was about this. We could cite that. Lots of similar themes. ACTION - get those.
Tim - this is similar to the publishing use case. It's easy for people to see themselves in. The customer is the product. We should draw these parallels. Even provosts see this and understand it.
Erin - protection is a counterbalance to reducing complexity in our messaging.
Scott - storage people are worried about where things should be stored. Storage and repository people can structure it differently.
Sayeed - we shouldn't cast this as complexity. It's insurance. It's protection from lock-in
Tim - Privacy and autonomy taxes - it is worth the resources for the organization to have these things? But tax is negative. I like insurance or protection better. What is our portability worth to us?
Recommendations - what is missing and what do we do with them.
Tim - some of the recommendations seem more like observations and not recommendations.
David - these are basically rough notes at this point. But we don't need to limit the recommendations to what's in the current notes.
How we frame Fedora (based on community, use of LOD, and if you use cloud storage) is useful, offering a pilot or service,
Sayeed - getting a grant for a pilot at LYRASIS would make sense. I'm not speaking as a bod member of IMLS but as an org
Tim - could we have Fedora and general recommendations. YES
The audience is broad.
Scott - part of this is marketing and addressing the things that people are concerned about.
David - make sure Fedora 6 is solving a problem, communicate that problem and how it solves it (for different constituents) and get people onto the new platform an communicate the outcomes.
The Fedora product position should be reviewed in light of the discussions in this project. We could add something about protecting and isolating from vendor lock-in (from cloud storage)
Tim - Awe to grind = I would like to recommend an explicit scale/performance test for Fedora 6.0 before its release. I would like to see that written down somewhere.
David - it's part of the development plan for Fedora 6.0. There are Czech institutions that have hardware and they can test it for us. They haven't been able to ingest because it's too slow.
Erin - should we pay to have these tests done? Could we get this grant-funded?
Erin - recommend a second grant project.