[Moims-dai] OAIS and the OAIS Architecture

Mon Mar 12 20:03:09 UTC 2018

Mike,
I concede the DAI has been open and sharing.  However, I have not really that transparency regarding the review of OAIS comments and suggested revisions.  Or . .  I just needed to be smacked upside the head to realize the full implications of where the OAIS revision was going.

To address your questions below:

I am not opposed to any interoperable architecture or protocols effort in digital preservation.  I am not opposed to  any such data architecture compatible with OAIS processes.

I am for retaining the reference model as just that - a broad, high level schema sitting above any specific repository or community's application.  ISO 16363 illustrates how an interoperable architecture schema can be achieved.  ISO 16363 is "nested" below the RM-OAIS as a/the specific implementation of the call for certification.  ISO 16919 is similarly nested as the specific way to conduct the certification audit for digital preservation information systems.

I am not against the architecture concept as a concept and model. My vision would be a similar nesting for any architecture schema(s).  This is where any interoperable architecture or protocol should be - as a unique ISO standard.  It must clearly show non scientific communities, especially the cultural heritage digital preservation efforts, how their efforts would be included.

Perhaps what I am envisioning is a three tier approach.  Tier One is the RM-OAIS.  Tier Two is the nested efforts such as ISO 16363 and ISO 16919 and any interoperable protocol(s).  Tier Three is/are the community specific architecture/protocol implementations, which or may not be ISO standards.  I stress the plural because the needs of the diverse communities are just that, diverse.  I just do not see how an architecture for any one of the space sciences could be adopted, without change, by the airplane manufacturers, the drug makers or the financial communities.

Vint Cerf comes at this from a different perspective.  We have not been building a single, tightly bundled underlying structural protocol (such as TCP/IP) thatrequires every community to get "on board" the same way.  We have consciously been building a looser approach that allows each community to accept and adopt the core (RM-OAIS) and add its own embellishments or definitions, a blending of efforts enhancing digital preservation.  There are still many folks who dislike "ingest" and immediately say that is what we call "accessioning" or whatever specific term they use for bringing digital objects into their digital preservation environment.  As digital preservation sites proliferate and communities have even less need to interoperate, the need for a single architecture diminishes.  Some communities will be able to impose a single architecture; others never will be able to do so.

Hopefully this gives you a better idea of where I am on this.

-----Original Message-----
From: Mike Kearney <kearneysolutions at gmail.com>
To: 'MOIMS-Data Archive Ingestion' <moims-dai at mailman.ccsds.org>
Sent: Sun, Mar 11, 2018 9:09 pm
Subject: Re: [Moims-dai] OAIS and the OAIS Architecture

Bruce, I highly respect your experience in the community, and I appreciate that you took the time to provide your insight.  

I have to say that I’m surprised by the response, though, because I thought that even for folks who have not been in the meetings or telecons, we have explained the rationale and background for this effort in emails with presentation material, minutes, etc.  I thought we had pretty good consensus that developing architecture and protocols for interoperability was the next important step for developing a healthier and longer-lasting digital preservation ecosystem.  

Just to make sure I understand your message… you’re not only against the term “OAIS Architecture”, you’re against any interoperable architecture or protocols effort in digital preservation?  Against any such data architecture compatible with OAIS processes?  

When Vint Cerf brought the discussion to me and to the DAI WG about the need for interoperability in the digital preservation archive world, as a newcomer to the group, I was immediately struck by the “gap” that existed in the arena.  It’s totally beyond me how adding a capability for improved interoperability can “diminish” OAIS or the work that has gone before.  I’ll quickly concede that I don’t have as much experience in this arena as senior experts like you.  But I think that, in turn, does not diminish the obvious benefits that some new measure of interoperability between archival IT systems would bring.  It’s exactly the “forest of architectures” (I would say forests, plural) that is a contributing obstacle to long-term access.  A ubiquitous protocol (ubiquitous like TCP/IP) can only improve long-term access to preserved assets.  And while it won’t be imposed on the diverse communities, it will be available for those that recognize the benefits and begin a migration towards it.  

In terms of the feasibility or chance of success, I think we’ll just have to get farther downstream.  So far nobody else has raised that as a concern.  I believe we need interoperable protocols and interfaces, and I am convinced that it would be a mistake to start work on them without having an overall data architecture concept.  So that’s why we’re starting with architecture discussions.  

So, that’s my high-level perspective, expressed frequently in the DAI WG.  And the subject of meetings with other external entities (OMG, IIPC) and the subject of papers at several upcoming conferences this year.  So if there is a problem with this work being done in the DAI WG, we need to resolve that issue quickly.  

If I’m overreacting, or misinterpreting, I apologize.  If you objection is only with the name “OAIS Architecture”, that’s considerably easier to fix.  We’ll be discussing that at the next DAI WG telecon, I’m sure.  

   -=- Mike

Mike Kearney
Huntsville, Alabama, USA

From: MOIMS-DAI [mailto:moims-dai-bounces at mailman.ccsds.org] On Behalf Of Bruce Ambacher
Sent: Sunday, March 11, 2018 12:10 PM
To: moims-dai at mailman.ccsds.org
Subject: Re: [Moims-dai] OAIS and the OAIS Architecture

Mike, 

I have not been a part of the DAI weekly meetings as I am no longer directly involved in digital preservation systems.  But as an original OAIS team member, a retired digital preservation system manager, NARA's representative to several federal data communities, a team member on the development of NARA's Electronic records Archives, and an information studies visiting professor, I think i can make an overall comment.

Since its initial development, beginning in 1995, through adoption, and a ten year review, all involved were quite vocal in insisting that OAIS was NOT an architecture.  To turn a position held for nearly a quarter century 180 degrees is unwise and endangers its credibility.  The greatest value of OAIS was its simplicity, its plain text, its ability to be understood at every level of digital preservation, and its use of high level components to allow individual communities to be compliant and to utilize their own architecture below that framework.

I also believe that effort will diminish the OAIS. Beneath it is a forest of "architectures" developed by different communities to meet their community needs.  Too be successful you would have to eliminate/incorporate every such architecture into an over riding super architecture.  The end result would be an architecture that communities might pay lip service to but it would be an architecture so complex that most such communities would reject it and continue doing what they were doing.

I believe the proper place for such an architecture is in annexes to the standard or in a separate standard.  I also should point out that OAISIS is already in use in the cultural heritage digital preservation community and to utilize that would create both confusion and resentment - just one example of how changing something has unexpected consequences.

You note:  "The use of the term “OAIS Architecture” should not be considered intended to change anything about the OAIS Reference Model or certification thereof" If that is true why do it, especially when the consequences could be different than expected?  You also note: "  If use of the term “OAIS Architecture” make people think it will change OAIS, then maybe we need a new term."  I believe the answer is more than a "Maybe."  It is a definite.

-----Original Message-----
From: Mike Kearney <kearneysolutions at gmail.com>
To: 'MOIMS-Data Archive Ingestion' <moims-dai at mailman.ccsds.org>
Sent: Sat, Mar 10, 2018 10:29 pm
Subject: [Moims-dai] OAIS and the OAIS Architecture

DAI WG members:  After last Tuesday’s discussion, I realized that we have an issue at least with terminology for the new architecture effort.  So I’m attaching a short writeup for your reading pleasure and discussion during the next telecon.  

The main points you should take away from this discussion are:  

·         The use of the term “OAIS Architecture” should not be considered intended to change anything about the OAIS Reference Model or certification thereof, and;

·         If use of the term “OAIS Architecture” make people think it will change OAIS, then maybe we need a new term.  

I would prefer to stick with “OAIS Architecture,” but we can discuss other options.  

   -=- Mike

Mike Kearney

Huntsville, Alabama, USA

_______________________________________________ MOIMS-DAI mailing list MOIMS-DAI at mailman.ccsds.org https://mailman.ccsds.org/cgi-bin/mailman/listinfo/moims-dai 

_______________________________________________MOIMS-DAI mailing listMOIMS-DAI at mailman.ccsds.orghttps://mailman.ccsds.org/cgi-bin/mailman/listinfo/moims-dai
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ccsds.org/pipermail/moims-dai/attachments/20180312/2f540622/attachment.html>