[Larry Lannom]

I think retirement is the right option. The activity and discussions around types has moved on to other projects such as the FDO Forum and other RDA groups, e.g., Data Fabric. Ulrich Schwardmann, who is quite active in this area, and I have discussed various options involving him as the new co-chair but both he and I are now focused on these other groups.
Does retirement involve some sort of closing statement that could point to these other activities? If so perhaps Ulrich and I could draft such a statement.
Best,
Larry

[Larry Lannom]

I was about to reference these P11 DTR slides and could not find them, so adding them now.

Agenda & Introduction

ePIC/GWDG

RPID Overview

RPID PIDs

 

[Larry Lannom]

Thanks Tobias,
I made a few minor typo-level changes to the Future Trends document, which I sent just to you, and have no changes to the Virtual Layer document, which I see is a short summary of the longer Virtual Layer document that you and Peter edited last year. The main issue left over from the call, from my POV, is naming what it is, exactly, that we are pursuing now in the Data Fabric group, as you show below in the GDON comment.
Trying a little logic chopping on the topic I came up with:
The IG is still Data Fabric and that is a perfectly good name for summarizing our interests, although I note that if you search for Data Fabric you’ll find it in wide use, and not because RDA is suddenly famous in the commercial world. See this Forbes article
https://www.forbes.com/sites/johnwebster/2018/01/12/big-data-meets-data-…
for quotes like
Thanks Tobias,
I made a few minor typo-level changes to the Future Trends document, which I sent just to you, and have no changes to the Virtual Layer document, which I see is a short summary of the longer Virtual Layer document that you and Peter edited last year. The main issue left over from the call, from my POV, is naming what it is, exactly, that we are pursuing now in the Data Fabric group, as you show below in the GDON comment.
Trying a little logic chopping on the topic I came up with:
The IG is still Data Fabric and that is a perfectly good name for summarizing our interests, although I note that if you search for Data Fabric you’ll find it in wide use, and not because RDA is suddenly famous in the commercial world. See this Forbes article
https://www.forbes.com/sites/johnwebster/2018/01/12/big-data-meets-data-…
for quotes like
> Data Fabrics—essentially information networks implemented on a grand scale across physical and virtual boundaries—focus on the data aspect of cloud computing as the unifying factor. To conceive of distributed, multi-cloud computing only in terms of infrastructure would miss a fundamental aspect of all computing technology – data. Data is integral and must be woven into multi-cloud computing architectures. The concept that integrates data with distributed and cloud-based computing is the Data Fabric.
Did we steal the term or vice-versa? Anyway, not a bad description of what we are doing, although I would say that the author of that quote thinks of ‘infrastructure’ much too narrowly.
And now we are focused on a digital object model. PIDs are a given and certain other elements seem to be close to givens – among others we are looking at types, PID kernel information, and some level of uniformity across operations on objects, including collections, which we want to consider as objects. The PID piece seems to be pretty well established, both conceptually and operationally. The others are established conceptually, but not operationally. Can Data Fabric IG help with this? I think that’s a good question for our breakout session.
As to names – there are now a lot of names floating around. CNRI for years has been talking about Digital Object Architecture (with the unfortunate acronym DOA, which in English usually stands for Dead on Arrival), and we have used DOIP for Digital Object Interface Protocol (also known in the ITU x.1255 spec as DEIP, for digital entity interface protocol). We have C2CAMP for a project that is also looking to flesh out this DO virtual layer, we have GDOC or Global Digital Object Cloud, which came from a slide that I created a few years ago, and now perhaps GDON or even ViLaNDO (which I love saying – thanks Maggie). One question I have is why we need a new name. Aren’t we just talking about the Data Fabric IG talking about ways to implement the digital object model? Not at all sure, but I worry about too many names causing confusion.
Looking forward to Berlin,
Larry

[Larry Lannom]

Its good as is, but I have a few comments regardless:
Any way to sneak in the wonderfully recursive notion that a collection of digital objects is itself a digital object? Perhaps a second phrase after
(Research) data management beyond single objects
Manage a collection of objects as an object in and of itself
I find the small diagram in which a collection descends from a service to be a little obscure. I understand it but I’m not sure its helpful. I can imagine other, much more elaborate diagrams, that show collections, encompassing members, sharing some services universally and some services selectively, perhaps using colors or some sort of differently shaped portals into the collections, but nothing that would fit on this slide.
Larry

[Larry Lannom]

and I would like to be in that meeting but i will be in a dentist chair at that time.

[Larry Lannom]

We missed you. We had some struggling discussions re. the nature and purpose of road maps, but then we had a very productive two hour exercise at the very end that you will hear more of in Berlin.
Best,
Larry

[Larry Lannom]

Opps, hadn’t meant to send that to the list….

[Larry Lannom]

Its all your fault??

[Larry Lannom]

Opps. I had 1 pm my time. Will try to join shortly.

[Larry Lannom]

Given a lack of changes or objections I was going to push it to the two larger lists this coming week and see what happens then. If we get no (useful) comments, entirely possible, it could be sent on to Council exactly as it is now, but with the added comment that it excited no added conversation. But maybe it will spark a debate.
Larry

[Larry Lannom]

Thanks Stefan, Mark, et al
We have been accumulating comments and the question now is how to proceed. We agreed in the meeting itself that the next step was to distribute it to the entire Chairs group, but how to not lose what has been said to date? The comments have been well made and thoughtful, but I doubt that they would change fundamentally what the Bob/Rebecca/Keith group wrote or what Peter provided. Our end goal is to present something to council, but it probably should not be a collection of email threads. There is no Chair of the Chairs and as Mark pointed out this group is different from the TAB, while admitting that it does not represent all of the Chairs but only the group that tends to cluster around this series of meetings, with some coming and going. I feel at least somewhat responsible for taking the next step and would appreciate any advice.
Best,
Larry

[Larry Lannom]

Thanks. We’ve gotten a couple of comments now and I encourage anyone else who has view to comment as well. We will move on to the entire Chairs’ list at some point in the not too distant future.
Best,
Larry

[Larry Lannom]

The latest draft agenda is on the meeting site
https://www.rd-alliance.org/sites/default/files/7th%20Chairs%20Meeting%2…
We start at 13:00 on the 12th, end at 18:00 followed by a self-pay dinner.
We start at 09:00 on the 13th, end at 17:00 and are invited (registration was required) to a very nice dinner at 19:00
On the 14th, we start at 09:00, end with lunch at 12:30, and the side meeting scheduled for one hour follows lunch.
The side meeting schedule was up in the air until yesterday and so did not make it onto the agenda.
Best,
Larry

[Larry Lannom]

Correct, with the exception that lunch was pushed to 12:30. So, following what I assume will be an hour or so of lunch and chat, which I hope Phil can attend, we will have an hour or so of VRE4EIC/W3C/RDA meeting. As far as I know we have the room for the rest of the day and anyone who wants to hang around and continue the discussion is welcome to do so. But I assume that the meeting proper will close around 14:30.
Larry

[Larry Lannom]

My other comments aside, if we have a recommendation across the three options (my vote in the current circs would be #3) then I am ok with this approach.
Larry

[Author]

Replies