Present:
| BNSC | David Giaretta | DG |
| Christiane Nill | CN | |
| Wyn Cudlip | WC | |
| CNES | Patrick Mazal | PM |
| Denis Minguillon | DM | |
| Claude Huc* | CH | |
| ESA | Nestor Peccia | NP |
| Gerhard Triebnig | GT | |
| DLR | Martin Pilgram | MP |
| Manfred Drexler | MD | |
| NASA | Don Sawyer | DS |
| Lou Reich | LR | |
| John Garrett | JGG | |
| Elise Blaese* | EB | |
| Stephen Voels | SV | |
| Alan Wood* | AW | |
| NASDA | Yasunori Iwana | YI |
| Naoyuki Toyoda | NT | |
| Ken Nakamichi | KN |
DS presented the results of the archiving task to date. This included:
It was agreed that an special distribution list for Archives will be created, such that everybody can follow its progress.
The major activities to date, including the various workshops at each agency, the dates of issue of the Reference Model paper and the forecast schedule are shown in Appendix G ( WP 700 ) . An ISO Committee draft is foreseen by May 97 and an ISO Draft International Standard by May 1998.
This document was produced by a Task Force on Archiving of Digital Information, which has the purpose to investigate means of ensuring continued access indefinitely into the future of records stored in digital electronic form. Members were from industry, museums, archives and libraries, publishers, scholarly societies, government.
The Executive Summary is as follows :
In the digital information the features that determine information integrity and deserve special attention for archival purposes include the following :
The Task Force gave a set of Recommendations ( See N12 ), which are at high level but identified a lot of concepts that the Archive Reference Model Group is addressing. Some of them are discussed below.
The main points on the operating environment are as follows :
The topics on migration strategy are as follows :
Additionally the Task Force is assessing how to model the costs, activity that is essential but very complex. Some limited studies have been done, but much work is needed on this area. The main conclusions were as follows :
DS presented the new outline of Version 7.0 OAIS Document ( Open Archival Information System ). This document is registered as N11. The name was modified ( now OAIS ) to distinguish from the many different functional entities that are called archives today.
The main purpose ( Section 1.1 ) is as follows :
The OAIS scope ( Section 1.1 ) is :
The OAIS applicability ( Section 1.2 ) is :
The section 1.5 on Definitions lags others due to timing of contributors. It is incomplete and some terms are not used. There are many known inconsistencies. Some definitions should include examples for clarification.
The OAIS functions ( Section 2.1 ) establishes the context, defines OAIS and references "problem statement" given by the paper ( Preserving Digital Information ).
The OAIS Responsibilities ( section 2.4 ) are as follows :
Sections 2.2 and 3.2 contain the Information Model based on OMT Object Model Diagrams. Section 2.2 presents a high level view, while section 3.2 contains a more detailed view. The Information Model is compatible with Z39.50 Digital Collections family of profiles.
Within AIPs the designated information object is the Content Information, which is the primary object to be preserved. It is the only place in the information model where physical and digital objects are distinguished. The content information contains data object and representation information. Examples of data objects include moon rocks, documents, and a file containing magnetometer readings.
Section 2.2.2 analyses how to categorise the information necessary to preserve the content information ( e.g. provenance, reference, context and fixity information, catalogue metadata, etc. )
Section 2.3 addresses the Environment Model with four external roles ( consumer as subscriber, consumer as searcher, producer, high level management ), two high level agreements ( producers - archives , consumers - archive ) and three types of sessions ( data delivery from producer, data dissemination to a subscriber and search session ).
There was a good progress on functional areas description, being Ingest and Archival Storage completed on version 7.
The section on archive domain and interface views need to be reworked. Only the initial OMT diagrams of the environmental domains have been included in Version 7.
Section 3.2.2 addresses the representation Information Model.
Some new sections are being worked :
The current issues are as follows :
The goals for this meeting are :
It was decided to discuss the Ref Model V 0.7 in two rounds, addressing first the general comments and then the detailed ones. Each person around the room gave their individual comments on the document:
The main topics were discussed as follows :
DS mentioned that the initial idea was to have the reference model as a framework and then start additional efforts to identify relevant existing standards and develop the missing ones. Some ISO technical reports are being used as reference. There are 3 types of ISO reports. A Reference model can understand the domain, the services implied and then go ahead with the standards ( as the IEEE model on mass storage ). CH mentioned that the archive domain is so wide that what we need is a common view giving an explanation and not a real specification. DG is more concern on the purpose and scope than the colour. After a length discussion we decided to go around the table to get the panel members opinions.
ESA does not have a problem between GB and BB, but we need to discuss the contents of the document, because it should be clear to every reader. PM said that it could be either. CH thought that GB is the best solution, but should be at the level of global understanding. DG is happy with a GB but adding some more specifics in the purpose and scope. Some material could be used in BB type material. At the end of the book we need a description of potential areas to be standardised. WC thought that GB is the best solution. For AW and EB is just GB. YI thought that BB is the best solution but only informative ( ISO Technical Report Type 2 ). LR thinks that BB is the best solution because it has more prestige. JGG is for GB. CN thought that it s important that ISO issues it as a Technical Report. SV is for ISO Technical Report Type 2, and after the 3 years review process can get BB status. MP preferred GB if it has a road map identifying the way to produce the necessary BBs.
Action Item
| DG will ask the CCSDS Secretariat at the MC about the procedure to follow with the future Archive Reference Model GB to achieve simultaneously ISO Report and CCSDS GB. | DG/141196 |
Conclusion:
P2 can go ahead with a GB or as a BB to become an ISO Technical Report type 2 or 3 as long as contains a section identifying the road map. Are the P3 Reference Model and terminology an accepted ISO standard or not? Action Item
| DG to ask to CCSDS Secretariat to follow this issue to ISO Secretariat | DG/141196 |
DG produced a new formulation of the "Purpose and Scope".
"The purpose of this document is to define the ISO Reference Model for Open Archive Information Systems. This will :
Archives will take in information from various sources, including other archives, disseminate information and some, perhaps all, will be responsible for the long term preservation of that information. Therefore this document will focus on the ingestion and dissemination processes and the standards designed to facilitate these. Long term preservation is a broader subject but this document will guide the production of Standards which will make that process easier.
The importance of long term preservation of data has several implications arising from the longer timescales implied.
In particular if we view archives of digital data over a long time scale, we expect to see a dynamic process with many transformations of the representation of the data, for example from one medium to another, following technology changes. Therefore the Standards presaged will, in addition to demanding data descriptions adequate for dissemination of data, will also specify metadata specifically needed for data preservation. These considerations should also be applicable to the digital component of archives of non-digital, e.g. physical sample, data.
The Standards deriving from this Reference Model will be expected to rely heavily upon existing Standards where applicable. Since many independent Standards do exist and will be developed in the future, the Reference Model must contain within it mechanisms for the use of multiple protocols. In particular the labelling methods, data description techniques and Control Authority system defined in other CCSDS documents, will be built upon.
Beyond this the model will also draw out the major functions of an OAIS in order that the interfaces may be well defined so that such systems may evolve in response to technological changes by replacement of these major components one at a time so that total rebuilding of such Open Archive Information Systems may be avoided. This component analysis should also allow the easier interoperation of such systems."
DS and LR will take this into account for a reformulation of the "Purpose and Scope". The concern is long term preservation and this may be applicable to short term archive as well. This concept will be added to the section. Definition of what "long-term preservation" is, should be explicitly stated in "Purpose and Scope".
DS and LR think that the level of detail present in the current version 7 of the document is the right one.
The issues are how the WB will evolve and the CCSDS identity. The decision will be taken at the end of the archive session
Schedule
Technical Issues
CH is more concerned on the information model and its objects. We need to describe very well these objects. The main issues are :
Action Items
| LR will put out some alternatives for the collection archive view | LR/151296 |
| Update the outline of scenarios including the new information model terms | DS/151296 |
The following comments were raised on the document structure :
There was a strong position within the US workshop and P2 to return to both digital and non-digital archives but to focus on digital area. Alan argued that the document should not break at a distinction between digital and physical. DS thought the best that might be acceptable would be to arrange the document in such a way that all the generic items are referenced first and then go into the digital archives domain.
Action Item
| Produce a concept paper to remove the digital / non-digital distinction at the highest level | AW/311296 |
Use of OMT
OMT can be used at different levels and its main objective is information hiding. We don't use event tracers. The main disadvantage is that some end users are not familiar with the technique.
Archives Classification
Additional topics to consider:
Standards Areas
The main interfaces shown in Figure 3.1 of the Archive Reference Model were discussed. Every P2 member gave their view on what are the useful areas for standardisation.
ESA mentioned that the producer / consumer interfaces are the most important ones to be addressed first. LR agreed on this. JGG added also Data / metadata Management standards. SV added fixing standards. For EB the ingest standards are useful to help to get info in and archive it. AW mentioned Administration standards to monitor performance of archives. WC interest is interoperability of archives ( i.e. access and dissemination ). DG pointed out that since "long-term preservation" has been added as a main aim then this implies that we need to have preservation standards ( assessing the technology evolution ). CH mentioned that the first thing to address is the external interface to the archival system ( consumer / producer ), and the second topic is storage, data management and ingest standards to support internal archive management. PM supported the external interface standards ( first consumer and then producer ). MP supported PM's views.
A first conclusion is that we need a set of WPs on areas of standardisation, from which the external interfaces is the first candidates. DG / SV to produce a concept paper that assist the panel in agreeing on several archive standard WPs DG / SV 311296
White Book
As the US workshop meet more frequently than P2 the document evolves a lot without P2 members following its progress. NASA suggested to have telecons or e-mail after every US workshop to inform the rest of P2 members. NASA proposed to have it as WB version 1 with CCSDS identity incorporated by beginning of January 1997. It is doubtful to get all the recommended changes in the next issue. LR is proposed as editor. Mike Martin will help with some sections. RD will expand the scenarios. We need a road map on the Reference Model identifying how p2 moves from the Ref Model to the production of standards.
There is a general consensus to produce a WB as version 0.8 of the Archive Reference Model.
Actions
The following actions were identified on the Ref. Model Version 0.7 :
| Update descriptions of the functional areas | MM/311296 |
| Update purpose and scope | DS/311296 |
| Expand of scenarios on version 4 | RD/311296 |
| Expand classification, another round on migration | SV/311296 |
| Regenerate section 3.3 ( domain views ) | LR/311296 |
| Update life science scenario | AW/EB/JT 311296 |
| Clean-up of definition section | ES/311296 |
| Include Road Map section | LR/311296 |
| Issue Archive Reference Model WB Version 1 | LR/030197 |
A dedicated telecon on the Archive Ref. Model will be held on the 3rd or 4th week of January 1997 after having Ref Model WB version 1 and the output of the US workshop ( 8th / 9th January 1997 ).
The next Panel 2 workshop will be hosted by NASA at GSFC during May 1997. ESA is planning to organise the fall meeting at one of its sites. The next Panel 2 telecon shall be on December 1996 and ESOC will arrange the telecon.
At the end of the workshop, the action list was reviewed and agreed upon by all.
The active work package in Archiving is WP710.
WP710 Archiving Reference Model
A. Progress
NASA held two US archiving workshops since the second ISO/CCSDS archiving session in Pasadena. Results of the Pasadena workshop and US workshops have been put on the WEB.
Version 5 of the reference model paper was produced following the guidance received in Pasadena. No international comments were received. Review at the July US workshop led to substantial changes and version 6 was produced in time for the October US workshop. This version began to more clearly incorporate data modelling views from the paper "e Preserving Digital Information "e and from the Z39.50 papers on access to digital collection and access to digital objects. These data modelling views are quite similar to those given in Claude HucÕs paper reviewed in Pasadena. Comments from CNES on the version 6 paper have been registered for the Oberpfaffenhofen workshop.
Version 7 of the reference model paper has also been registered for this workshop. This paper is believed to be a significant advancement over the previous papers and has significant material in all currently proposed sections.
B. Changes
NASA proposes that the reference model should be published as a Green Book, rather than as a standard. This would make it an "e information "e ISO document rather than a "e normative "e document. The rationale is that the nature of the material is not something for which conformance can be checked, and therefore it is not a proper standard. If we obtain agreement on this approach, we will need to identify the ISO Type as I, II, or III.
C. Problems
No unusual problems are noted, other than the issue of "e informative "e vs. "e normative "e .
D. Forecast
NASA proposes that version 7 be adopted as a White Book. It is necessary to get a good review of this document at this workshop and obtain commitments for writing assignments. Completion of this document should still be seen as May, 1998.
We should also begin to identify areas for standardisation and attempt to prioritise them, as scheduled. Our Methodology document should assist in this process.
E. Milestone Table
| # | Description | Mgmt. Plan Completion Date | Status | Projected Completion Date |
| 710.2 | Draft Archiving Reference Model | 96.09.01 | Closed | 96.10.28 |
| 710.3 | Draft ISO Committee Reference Model | 97.05.31 | Open | |
| 710.4 | Draft ISO Standard Reference Model | 98.05.31 | Open |
Assigned number MUN/96/P2/N10
TITLE : CNES comments on Archival Reference Model version 6.0
AUTHOR:Claude HUC
AFFILIATION: CNES
ABSTRACT:
Copy available from:
and/or from esaccsds server as
/meetings/mun_p9611/arch_com.do6
Assigned number MUN/96/P2/N11
TITLE : Reference Model for an Open Archival Information System
AUTHOR:Don Sawyer and Lou Reich/editors
AFFILIATION: NASA and CSC
ABSTRACT:
The is a substantial improvement over version 6. It is still incomplete in some sections, but there is something in every section.
We believe this document should be a Report, or an Informative ISO document, as opposed to a standard.
Copy available from:
and/or from esaccsds server as
books/archive/refmodel/RefMod7.m51
Assigned number MUN/96/P2/N12
TITLE : Preserving Digital Information
AUTHOR:RLG Task Force (D. Sawyer-provider)
AFFILIATION: NASA
ABSTRACT:
This report is from the Research Libraries Group Task Force. It is a good summary of the problems associated with the long
term preservation of information in digital form, and directly addresses the archiving concerns that prompted the ISO archiving
effort. Everyone attending the ISO archiving session should have read this paper.
Copy available from:
and/or from esaccsds server as
meetings/mun_p9611/presdgtl.do6
Assigned number MUN/96/P2/N23
TITLE : Initial comments on OAIS Reference Model ver 7
AUTHOR:D Giaretta
AFFILIATION: BNSC
ABSTRACT:
These are a few initial comments on the OAIS reference model with particular emphasis on the Purpose and Scope.
Copy available from:
http://www.ssd.rl.ac.uk/ccsdsp2/mun96/bnscrefmodcmt.html
and/or from esaccsds server as
ACTIONS CLOSED
ACTIONS OPEN FROM PREVIOUS MEETINGS
ACTIONS OPENED AT THIS PLENARY MEETING
| AI # | WP # | Description | Actionee | Date | Status | Comments |
| P/9611/27 | 700 | Ask CCSDS Secretariat the procedure to follow with the future Archive Ref. Model GB to achieve simultaneously ISO report and CCSDS GB | DG | 961114 | Open | |
| P/9611/28 | 700 | Ask CCSDS Secretariat if P3 Reference Model and terminology are an accepted ISO standard | DG | 961114 | Open | |
| P/9611/29 | 700 | Put some alternatives for the collection archive view | LR | 961215 | Open | |
| P/9611/30 | 700 | Update the outline of scenarios including the new information model terms | DS | 961215 | Open | |
| P/9611/31 | 700 | Produce a concept paper to remove the digital / non-digital distinction at the highest level | AW | 961231 | Open | |
| P/9611/32 | 700 | Produce a concept paper that assist the panel in agreeing on several archive standard | DG SV | 961231 | Open Open | |
| P/9611/33 | 700 | Update descriptions of the functional areas of Ref Model | MM | 961231 | Open | |
| P/9611/34 | 700 | Update purpose and scope of Ref Model | DS | 961231 | Open | |
| P/9611/35 | 700 | Expand of scenarios on version 4 | RD | 961231 | Open | |
| P/9611/36 | 700 | Expand classification, another round on migration | SV | 961231 | Open | |
| P/9611/37 | 700 | Regenerate section 3.3 ( domain views ) | LR | 961231 | Open | |
| P/9611/38 | 700 | Update life science scenario | AW EB JT | 961231 | Open | |
| P/9611/39 | 700 | Clean-up of definition section | ES | 961231 | Open | |
| P/9611/40 | 700 | Include Road Map section | LR | 961231 | Open | |
| P/9611/41 | 700 | Issue Archive Reference Model WB Version 1 | LR | 970103 | Open |
URL: http://ssdoo.gsfc.nasa.gov/nost/isoas/int03/minutes.html
A service of NOST at NSSDC. Access statistics for this web are available. Comments and suggestion are always welcome.
Editor: Nestor Peccia
Curator: John Garrett (garrett@ncf.gsfc.nasa.gov) +1.301.441.4169
Responsible Official: Code 633.2 / Don Sawyer (sawyer@ncf.gsfc.nasa.gov) +1.301.286.2748
Last Revised: 15 November 1996, John Garrett (14 April 1198, John Garrett)