Notes from Megameeting 12th May 2008

Attendees:

BruceAmbacher UM
JohnGarrett GSFC
KatiaThomaz INPE, Brazil
MarieWaltz Center for Research Libraries
MarkConrad NARA
DavidGiaretta STFC
CandidaFenton UGlasgow
RobertDowns CIESIN, Columbia University

The discussion was a combination of audio and typed chat. For this reason the chat transcript (given below) gives an incomplete representation and may be hard to follow.

Progress made: Section B2 of the working document was discussed. Section B2.11 and B2.12 were completed. B2.4 was reviewed and the outstanding comment deleted. There was extensive discussion of B2.10. It was agreed that it should be make more specific to the Ingest process and and similar metric be added to section B3 in order to ensure continued understandability.

Actions:

  • All to check edits made to section B2.10.
.

Next meeting: Section B3.

KatiaThomaz >> (All): hi Candida
KatiaThomaz >> (All): hi Robert
RobertDowns >> (All): Hi Katia and Candida
BruceAmbacher >> (All): hello
RobertDowns >> (All): Hi Bruce
candida fenton >> (All): Hi
KatiaThomaz >> (All): Hi
BruceAmbacher >> (All): It is quite soggy here 7+ inches of rain in the last several days
RobertDowns >> (All): That is a lot of rain, Bruce.
BruceAmbacher >> (All): Are we starting in B2.10 or B2.11?
KatiaThomaz >> (All): i think b2.12
Mark Conrad >> (All): Bruce,
BruceAmbacher >> (All): yes, Mark?
Mark Conrad >> (All): I think we finished B.2.11 unless you had additional comments.
BruceAmbacher >> (All): No, I saw some unmarked comments in B2.10. that I guess we set aside for now.  Lets start with b2.12
Mark Conrad >> (All): Ok.
Mark Conrad >> (All): Candida can you hear the audio?
Mark Conrad >> (All): Bruce, do you have a suggestion for a word to use in place of audit?
RobertDowns >> (All): How about inventory?
BruceAmbacher >> (All): Mark, would you accept a substitute sentence:  The repository must provide a way to  independently demonstrate the completeness and orrectness of its collections and their contents." 
candida fenton >> (All): I can hear now
BruceAmbacher >> (All): inventorying would be ok in Requirement.
BruceAmbacher >> (All): replace "audit  of" with inventorying
Mark Conrad >> (All): Ok with me. 
RobertDowns >> (All): Ok with me, too.
KatiaThomaz >> (All): ok with me
Mark Conrad >> (All): Any objection to this change to the Requirement?
candida fenton >> (All): No
Mark Conrad >> (All): Any objections to Bruce's proporsed substitution for the first sentence in the Supporting Text?
RobertDowns >> (All): No objection.
Mark Conrad >> (All): Katia? Candida?
KatiaThomaz >> (All): no objection
candida fenton >> (All): No objection
Mark Conrad >> (All): Ok. I will make those changes. What about Bruce's proposed change for the second sentence in the Supporting Text?
Mark Conrad >> (All): Is anyone there?
KatiaThomaz >> (All): many negative words made me confused
RobertDowns >> (All): I agree with Bruce's proposed change: "This is necessary to show that any incomplete data objects are not the results of actions taken by the repository."
Mark Conrad >> (All): I agree with Katia that the use of so many negative words makes it difficult to understand. Could this be re-phrased as a positive statement?
Mark Conrad >> (All): Good Morning, John
JOhnGarrett >> (All): Good Morning
BruceAmbacher >> (All): Should it stress that the repository has ingested all SIP data that was available?
Mark Conrad >> (All): Can anyone hear Bruce?
cclrc >> (All): Hi guys - David here
Mark Conrad >> (All): Welcome back!
BruceAmbacher >> (All): I hear Mark but no one else.  Welcome back Davis.
RobertDowns >> (All): Hi David, it is good to see that you are participating again.
BruceAmbacher >> (All): Its been so long I forgot how to spell David's name
BruceAmbacher >> (All): Is this better? "This is necessary to show that the repository ingested everything that was in the relevant SIP(s)."
KatiaThomaz >> (All): hi David. Nice to see you again
cclrc >> (All): I'm not quite up to speed but I wanted to pass on the news that CCSDS voted to make the BoF and Work Group. The only thing left is for me to submit a Concept Paper - which is either TRAC or else a snapshot of the Wiki, and an updated Charter.
candida fenton >> (All): Good news
JOhnGarrett >> (All): Hi David!  Good to see you back!
RobertDowns >> (All): That is great news, David!
BruceAmbacher >> (All): David, can you submit both TRAC and the wiki snapshot to show our progress?
cclrc >> (All): In terms of the Charter I need to update the effort which people are committing - which for most people will be something very small like 5%. 
Mark Conrad >> (All): Simon asked us to update that part of the wiki last week. Several of us have already done so.
cclrc >> (All): Can people update http://wiki.digitalrepositoryauditandcertification.org/bin/view/Main/EffortAvailable please
cclrc >> (All): Great!
cclrc >> (All): Bruce - yes, I can submit both TRAC as well as a snapshot of the Wiki.
cclrc >> (All): I'll submit things later this week.
Mark Conrad >> (All): So anyone who has not updated their Effort Available should do so ASAP?
cclrc >> (All): Yes please
KatiaThomaz >> (All): ok.
BruceAmbacher >> (All): will do after the webchat.
Mark Conrad >> (All): Great! Is there anything else that you need from the group before you submit things to the CCSDS?
cclrc >> (All): No - I'll circulate the reviswed charter - it will just be very minor changes for consistency with what we decided and the effort update
RobertDowns >> (All): I just added my effort available.
cclrc >> (All): For the Wiki snapshot I'll put  in the CCSDS boiler-plate
cclrc >> (All): OK - where are we now - I'll just lurk in the background for now
Mark Conrad >> (All): We are on B.2.12. We are talking about a replacement for the second sentence in the Supporting Text.
Mark Conrad >> (All): Bruce,
BruceAmbacher >> (All): Comments on my suggested replacement for my replacement?
Mark Conrad >> (All): I am not sure it gets at what this requirement intended.
KatiaThomaz >> (All): What if something like "This is necessary in order to make sure any external action could damage the repository"?
BruceAmbacher >> (All): What about: "This is necessary in order to demonstrate which data losses resulted prior to ingest and which may have occurred after ingest."
Mark Conrad >> (All): I changed my mind. I like your second proposal better.
JOhnGarrett >> (All): This last suggestion works for me.
Mark Conrad >> (All): I reread the original requirement in the TRAC document and your second proposal seems to fit.
Mark Conrad >> (All): We have one vote for proposal #2 and one for proposal #3. What do the rest of you have to say?
BruceAmbacher >> (All): For clarity we are considering: "This is necessary to show that the repository ingested everything that was in the relevant SIP(s)."
RobertDowns >> (All): I believe that I also liked the second proposal.
Mark Conrad >> (All): Bruce, yes. That is proposal #2.
JOhnGarrett >> (All): That is also OK with me
KatiaThomaz >> (All): ok with me too
candida fenton >> (All): OK with me
cclrc >> (All): OK by me
Mark Conrad >> (All): Ok. I will make that change.
Mark Conrad >> (All): The next comment is on the Examples section. This is really a housekeeping comment.
BruceAmbacher >> (All): I accept Mark's next comment to drop the reference to B2.6
KatiaThomaz >> (All): I agree
RobertDowns >> (All): I also agree
candida fenton >> (All): I agree
Mark Conrad >> (All): So it would read B.2.1. through B.2.5.?
JOhnGarrett >> (All): Yes
cclrc >> (All): OK
Mark Conrad >> (All): Ok. I will make the change. On to B.2.13?
BruceAmbacher >> (All): In B2.13 I accept Mark's comment to delete reference to PREMIS
JOhnGarrett >> (All): I agree with removal of the PREMIS reference
Mark Conrad >> (All): This comment is consistent with what we have done for previous references to Premis.
KatiaThomaz >> (All): I also agree
cclrc >> (All): Yes
RobertDowns >> (All): Yes
Mark Conrad >> (All): Ok. I will make the change. Do we want to go back to look at B.2.4. now?
KatiaThomaz >> (All): ok
Mark Conrad >> (All): The question here was whether or not we needed another requirement on the basis of John's comment under the Discussion session.
BruceAmbacher >> (All): Will a cross-reference to/from B2.12 that we just modified suffice?
Mark Conrad >> (All): I don't think so. John is asking what happens if the repository has not received all of the SIPs necessary to form an AIP. Is that right, John?
KatiaThomaz >> (All): I think it´s not necessary, because the AIP is not created
JOhnGarrett >> (All): I'm not really sure why it was decided that this is not part of this requirement.  I was talking about case where a SIP makes up part of an AIP.
JOhnGarrett >> (All): Yes, I imagine there would be a policy regarding only get part of data 'needed' for an AIP.
cclrc >> (All): The example for B2.12 looks as if it is about that
BruceAmbacher >> (All): John,  you are talking about a situation such as where a repository makes an annual AIP from 12 monthly SIPs?
cclrc >> (All): But I agree with John - looking at B2.4 it's not clear to me why it does not cover John's case. If would just need a piece of text to make it explicit
Mark Conrad >> (All): Actually, I think the last paragraph of the Discussion for B.2.12. covers this case.
JOhnGarrett >> (All): I think the example in 2.12 covers what I was thinking
Mark Conrad >> (All): So can we strike the comment under B.2.4?
BruceAmbacher >> (All): ok
KatiaThomaz >> (All): ok
RobertDowns >> (All): ok
cclrc >> (All): OK
JOhnGarrett >> (All): Yes
Mark Conrad >> (All): Ok. I will make the change. Shall we move to B.2.10?
JOhnGarrett >> (All): yes
KatiaThomaz >> (All): yes
Mark Conrad >> (All): There are several issues to be addressed. One of them is whether or not this requirement should be moved to B.3.
cclrc >> (All): It does look odd in the Ingest section
Mark Conrad >> (All): B2.10. clearly extends beyond the ingest process. It is related to B.2.3.
BruceAmbacher >> (All): What would be lost if it is deleted?  What is conveyed that is not in B3.2?
Mark Conrad >> (All): Sorry B.3.2.
RobertDowns >> (All): Was there a reason why it was put in the Ingest section. Was this intended as an ingest activity?
Mark Conrad >> (All): Bruce, B2.10 is about understandibility of the content B3.2. is about the understandibility of the formats.
RobertDowns >> (All): Perhaps this was intended as part of the review during Ingest
cclrc >> (All): I'd have to look back at the various versions but it looks wrong now. I suspect that it was a metric which was edited and transmogrified but then we forgot to move it.
BruceAmbacher >> (All): I believe B2.4 was intended as a check on the ingest process - that everything was ingested or otherwise accounted for, that nothing just disappeared.
cclrc >> (All): Good point
JOhnGarrett >> (All): B3.2 doesn't talk about bringing content back up to a level of understanding.  Perhaps it should.
BruceAmbacher >> (All): The requirement in TRAC flows down a decision table from B2.3
Mark Conrad >> (All): John, I think B3.2. and B2.10. need to be separate requirements.
JOhnGarrett >> (All): I could see looking at B2.10 as a place to check that content is understandable when it is first ingested and B3.2 is monitoring of the situation.
cclrc >> (All): But it does look as if B2.10 was generalised to cover more than just ingest
Mark Conrad >> (All): John, I believe that B.2.10 and B.3.2. are monitoring two separate things.
JOhnGarrett >> (All): And as you have pointed out, B3.2 is only format oriented and B2.10 is a more general understandability.
Mark Conrad >> (All): I would suggest moving B.2.10 to a new requirement in B.3.
KatiaThomaz >> (All): i think "approaches obsolescence or is no longer viable" is very different from "understandability"
JOhnGarrett >> (All): OK with me.
cclrc >> (All): Perhaps B2.10 should be moved to before B3.2 and a cross reference is made to B2.3 (?) to say that the initial AIP has to have adequate RepINfo
BruceAmbacher >> (All): B2.10 deals with initial understandability and B3.2 deals with loss of understandability due to technology or other changes
KatiaThomaz >> (All): i agree with Mark
RobertDowns >> (All): If we look at the requirement only, it appears to be relevant to the review that should take place during ingest to ensure understandability
BruceAmbacher >> (All): We need to just read the requirements as they flow within a section.
BruceAmbacher >> (All): Robert is correct. and B3.2 deals with technology and preservation issues.  I recommend they stay where they are.
Mark Conrad >> (All): Bob, The Supporting Text for B.2.10. says "continue to be understandable" I.e., beyond ingest.
cclrc >> (All): Yes, clearly it applies beyond ingest
RobertDowns >> (All): Yes, I agree that it applies beyond ingest, but it appears to describe the review to be done during ingest.
cclrc >> (All): We need to make it clear that the initial AIP is understandable and that it continues into the future
BruceAmbacher >> (All): That continuing requirement was not in the original text
Mark Conrad >> (All): How about this? Restate B.2.10 where it is so that it applies at the time of ingest and have a new requirement in section three for subsequent testing of understandibiliity of the content in section B.3.?
JohnGarrett >> (All): Hi, back, just got dropped.  I think if we just drop the word 'continue' there it helps
cclrc >> (All): Mark - sounds perfect
KatiaThomaz >> (All): I agree with Mark
BruceAmbacher >> (All): But B3 deals with preservation planning, not object understandability directly.  I am hesitant about adding that without looking at other requirements.
cclrc >> (All): The only thing is that we would then have 2 very similar metrics about understandability
cclrc >> (All): Bruce - part of preservation planning must have to do with continuing understandability
KatiaThomaz >> (All): but we must think if it's really necessary checking understandability at the moment of the AIP creation
Mark Conrad >> (All): Bruce, the continuing aspect is part of the original TRAC requirement.
BruceAmbacher >> (All): David, that is covered in B5.
Mark Conrad >> (All): Bruce, Are you saying that B.3.2. should be removed from B.3.?
cclrc >> (All): Bruce - no B5 has to do with discovery not understandability
JohnGarrett >> (All): I think it is probably a worthwhile check at ingest to test understandability.  
BruceAmbacher >> (All): No.  I am arguing against adding an equivalent to B2.10
BruceAmbacher >> (All): Mark, I do not see anything about continuing in TRAC B2.10
Mark Conrad >> (All): Bruce, B.3.2. and B.2.10 both have to do with aspects of the continuing understandibility of the AIP.
Mark Conrad >> (All): Bruce, See the Evidence section of the TRAC requirement.
cclrc >> (All): We need (1) understandability initially and (2) understandability into the future
BruceAmbacher >> (All): B2.10 is in the Ingest section; B3.2 is in the Preservation Planning section
KatiaThomaz >> (All): B3.2 more with obsolescence and B2.10 with understandability
cclrc >> (All): ...the question is whether we can reduce the duplication of text 
BruceAmbacher >> (All): Mark, does the evidence section apply to Ingest or also to continuing understandability?
Mark Conrad >> (All): Bruce,  It talks about "periodic assembly of designated..."
JohnGarrett >> (All): Should we check understandability of content at ingest?   If so, we should keep a B2.10 requirement and reword it to ensure it is talking just about ingest time.  Decision about adding more to B3 is another, but separate question.  So should we check understandability at ingest?
BruceAmbacher >> (All): Katia, I think continuing understandability is in B5
BruceAmbacher >> (All): John, yes we check understandability at Ingest.
RobertDowns >> (All): B2.10 is needed to ensure that we check for understandability at ingest
KatiaThomaz >> (All): Bruce, i think B5 is anothet thing
cclrc >> (All): Perhaps we need to look at B3 - add our comments - for the next meeting.
JohnGarrett >> (All): Ok, I propose then that we edit B2.10 to make it apply to only ingest and that we make a note in B3 or B5 to ensure that we check continued understanding.
cclrc >> (All): Katia I think B5 is about discovery rather than understandability
KatiaThomaz >> (All): David, I agree with you
cclrc >> (All): John - I'm OK with that - put a placeholder in B3
Mark Conrad >> (All): I agree with John to reword B.2.10 to apply at the time of ingest. I think we may need an additional requirement like B.2.10 in B.3. 
RobertDowns >> (All): I also agree
cclrc >> (All): But I suspect that we will have some duplication of wording and will need to review whether we might combine them!
cclrc >> (All): However right now I agree with going ahead with John's proposal
JohnGarrett >> (All): Suggest we change requirement from "... understandability of content information ..." to "understandabiliyt of content information AT INGEST.."
BruceAmbacher >> (All): ok
cclrc >> (All): OK
KatiaThomaz >> (All): checking understanbility at the moment you receive something sounds strange to me... but i accept...
BruceAmbacher >> (All): I have to go to get ready for another meeting.
KatiaThomaz >> (All): i must leave you now too
cclrc >> (All): Must go also. Bye all. Good to be back!
Mark Conrad >> (All): John, I think that it will take more than that to correct B.2.10. I will propose changes for next week.
Mark Conrad >> (All): David, Glad you are back. 
KatiaThomaz >> (All): bye. have a nice week!
JohnGarrett >> (All): For our case at NSSDC we check understandability at ingest normally.  Scientists often send in dataset that 'everybody' understands.  In reality only people on the project understand what they are.
JohnGarrett >> (All): Bye see you all next week
Mark Conrad >> (All): Comments on B.3. for next time?
JohnGarrett >> (All): In supporting text, just drop the word continue in the first paragraph.
JohnGarrett >> (All): OK, we'll talk more next time.
JohnGarrett >> (All): Bye
Mark Conrad >> (All): See you all next week. I will capture that chat and send it to Simon for posting.
JohnGarrett >> (All): Thanks Mark
RobertDowns >> (All): Bye

-- DavidGiaretta - 12 May 2008

Topic revision: r1 - 2008-05-12 - DavidGiaretta
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2018 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback