4pm UK time using MegaMeeting at RAC


JohnGarrett NASA
DavidGiaretta BNSC/CCLRC
BruceAmbacher NARA/ U Maryland
KatiaThomaz INPE, Brazil
HelenTibbo UNC

Record of the meeting

David Giaretta >> (All): I'm going to turn off the video frames to improve audio quality
HelenTibbo >> (All): David, Hello! Hope you are above water. I am drowning in the metaphorical type of water here.
RobertDowns >> (All): Hi, I am going to borrow headphones and will return in a few minutes.
David Giaretta >> (All): Try right-clicking on your name on the left and look at the settings
RobertDowns >> (All): I am back with headphones and can hear you, David
Katia Thomaz >> (All): hi
RobertDowns >> (All): Hi Katia
David Giaretta >> (All): Hi Katia
NancyMcGovern >> (All): I'm just listening today (and typing) but I'm here.
RobertDowns >> (All): As I had promised during our meeting two weeks ago, I entered a proposed revision for 2.9 to reflect the consensus in the chat discussion.
Katia Thomaz >> (All): I read it and agree
RobertDowns >> (All): That was B2.9
NancyMcGovern >> (All): Looks good to me.
David Giaretta >> (All): http://wiki.digitalrepositoryauditandcertification.org/bin/view/Main/DigitalObjectManagement
NancyMcGovern >> (All): It looks from the June 27 notes that the discussion got to 2.9 and was looking at 2.10
David Giaretta >> (All): Any thoughts on B2.10?
Katia Thomaz >> (All): We have comments at B2.11
NancyMcGovern >> (All): The evidence for 2.10 does not seem to address the requirements to document the process - more exampels might be good.
David Giaretta >> (All): I'll take an action to look at adding text to B2.10 - evidence etc
David Giaretta >> (All): Yes
RobertDowns >> (All): There was a chat discussion about 2.11 on July 11, but I do not see the chat notes online
Bruce Ambacher >> (All): Are the chatlines available still?
David Giaretta >> (All): B2.11 raise the issue of proof
David Giaretta >> (All): Bruce - I'll see if I can make the chat more persistent automatically
Bruce Ambacher >> (All): David, Are you raising concerns re: B2.11 when you mention "proof"?
Helen Tibbo >> (All): I dropped out for a while - trying to switch between the docs and the meeting but I am back
David Giaretta >> (All): B2.11 proof about completeness and correctness of the SIP (or AIP)
David Giaretta >> (All): Is is adequate to simply say that "Documentation should describe how completeness and correctness of SIPs and AIPs are ensured,"
David Giaretta >> (All): Can we say anything more specific? How can the docuemntation be judged?
Katia Thomaz >> (All): but AIPs come from SIPs
Bruce Ambacher >> (All): Yes. Auditors can check the records maintained by the repository, as suggested
Katia Thomaz >> (All): and if yo testify SIPs are complete you won´t have problems
Katia Thomaz >> (All): with AIPs
David Giaretta >> (All): Maybe if we focus on AIP - because at least we can say some definite things about them but SIP's are undefined
Helen Tibbo >> (All): I'm dropping out and trying to come back in. Not sure why. Can't hear anymore.
JohnGarrett >> (All): You have to correctly translate SIPs in AIPs, perhaps with other information already in the Archive
Bruce Ambacher >> (All): AIPs come from SIPs but may not be the same.  You may combine, divide depending on your storage commitments
Katia Thomaz >> (All): if SIPs are not correct you have a problem in the process
David Giaretta >> (All): The completeness and correctness of an AIP is defined in terms of whether it has enough information for long term preservation of its content object
David Giaretta >> (All): That at least can be  tested
Bruce Ambacher >> (All): SIPS-AIPs-DIPs a linkage but not necesarily a complete duplication.  And YES you must be able to verify you received what you were supposed to receive.
Bruce Ambacher >> (All): Does anyone have any problems with what is written or with the evidence?
David Giaretta >> Helen Tibbo: Helen - we are just using CHAT because it works for everyone and is easier to paste into the minutes
David Giaretta >> (All): Its not clear to me how the evidence is judged
Bruce Ambacher >> (All): Isn't that up to the auditors?
RobertDowns >> (All): The requirement should give guidance to the auditors and the repositories
David Giaretta >> (All): Sure - but should we provide more guidance? At least AIPs are defined.
JohnGarrett >> (All): Is 2.11 mostly supposed to be ensuring that SIP was correctly converted into AIP?  Or is it more a question of the AIP fulfilling the expectations of Designated Community?
Katia Thomaz >> (All): take a look at B1.4
Bruce Ambacher >> (All): Title would assume the former
JohnGarrett >> (All): Most of the discussion seems to focus on SIP being converted perhaps without many changes (Checksums don't change) in an AIP.
David Giaretta >> (All): B2.11 is looking at the AIP in particular - something for long term preservation. I'm not sure it is captured
JohnGarrett >> (All): Personally just reading the title I didn't get that it was just the former.  I got that from the text.
David Giaretta >> (All): Maybe we need to say something specific about the long term preservability - the defining quality of the AIP
RobertDowns >> (All): And perhaps mention checking consistency of the AIP with the SIP
David Giaretta >> (All): But to create an SIP there may be more than one SIP
RobertDowns >> (All): I meant the SIP received from the provider
Bruce Ambacher >> (All): There can be a one to many or a many to one relationship
David Giaretta >> (All): OK - but the text looks as if it is more 1-to-1 conversions
Bruce Ambacher >> (All): The import is in a repository having a documented process for creating AIP from SIP (1 to 1 is probably the vast majority)
David Giaretta >> (All): Not sure I believe that
Katia Thomaz >> (All): take a look at B2.1, B2.2 and B2.3
JohnGarrett >> (All): I think in general it won't be 1-1.  On the other hand, perhaps the data objects with an SIP can be followed into an AIP and checked.
Helen Tibbo >> (All): Each time I switch out to look at the document I seem to be dropped even from chat so I'll just stay here.
Bruce Ambacher >> (All): Helen that is the advantage of having a paper version at your side, or two computers
David Giaretta >> (All): I think Katia is right - I'm losing the continuity of the metrics. 
RobertDowns >> (All): We should ensure that we limit redundancy
Bruce Ambacher >> (All): That is the problem wehn you focus on just one metric and try to load it with all you want done
David Giaretta >> (All): We define the AIP and B2.11 says that we check the AIPs we create are complete (according to those definitions)
Bruce Ambacher >> (All): So, does anyone want anything changed/added?
David Giaretta >> (All): I may have argued the oppoite in the past but now I think we need some more cross-referencing between metrics
David Giaretta >> (All): IN other words we add some text to B2.11 referring back to the B2.2 and B2.3
RobertDowns >> (All): Yes, the reference might help
NancyMcGovern >> (All): that sounds right - looking at them.
Katia Thomaz >> (All): i agree too
Bruce Ambacher >> (All): Could a introductory section explaining the build in the metrics accomplish that?  I'm not sure we want to begin wholesale 
                                     cross-referencing on an ad hoc basis.  That should be an offline exercise.
David Giaretta >> (All): We can collect the cross references as we go and then see how to make them read better.
Katia Thomaz >> (All): this is very important: prepare to the meeting
David Giaretta >> (All): ABsolutely right! Mea Culpa
Katia Thomaz >> (All): sorry. it was not my intention...
David Giaretta >> (All): I'll take an action to add some cross-ref text to B2.11
Bruce Ambacher >> (All): This type of activity may be good at the BoF session.
David Giaretta >> (All): Katia - no problem - but you were right!
JohnGarrett >> (All): Currenty text in 2.11 is discussing verifying correctness/completeness of SIPs as well as AIPs.  
                                I think 2.11 should just be about AIPs (as the title implies).  If SIPs are not handled elsewhere, it should be added as a separate metric.
David Giaretta >> (All): Yes - I was thinking that
David Giaretta >> (All): Will you take a look John?
JohnGarrett >> (All): Sure.  Is there another place where SIPs are addressed?
Bruce Ambacher >> (All): I don't see that way John.  SIPs are only mentioned as they relate to creating/verifying AIPs
RobertDowns >> (All): At the end of the 2.11 label, we could add "in accordance with documented procedures", which implicitly references the related requirements
NancyMcGovern >> (All): B1.2 refers to SIPs explicitly
JohnGarrett >> (All): For example the second paragraph starts off with a sentence about verifying SIP as well as AIP.
Bruce Ambacher >> (All): would the procedures differ?
David Giaretta >> (All): I would think so
JohnGarrett >> (All): I think so.  Procedures for SIP would entail verifying you got what was expected from Producer.  
Katia Thomaz >> (All): i am searching sometihinf "ensures transformation of SIPs into AIPs, but i don´t know how to do it mechanically.
JohnGarrett >> (All): Conversion into AIP is simpler in some senses, just need to ensure that what came in from Producer gets moved correctly to AIP.
NancyMcGovern >> (All): B1.4 talks about completeness and correctness of SIPs.  B2.11 should require making the link between SIPs and AIPs but focus on AIPs
JohnGarrett >> (All): Then also need to add extra stuff to AIP to ensure long-term preservability.
Bruce Ambacher >> (All): Is that B2.13?
David Giaretta >> (All): There is something here about an audit trail that perhaps we should capture explicitly
Bruce Ambacher >> (All): John, B4 covers storage and preservation
NancyMcGovern >> (All): A lot can happen at the generate AIP point - in many cases, it doesn't seem like generate AIP would be simpler than creating a SIP?
David Giaretta >> (All): Nancy - yes I agree that an AIP would be more complex probably
RobertDowns >> (All): The creation of the preservation information could be contingent on the level of preservation that was decided for a particular AIP
Bruce Ambacher >> (All): Repositories receive SIP and generate AIP and DIP.  I agree each process van be equally complex
David Giaretta >> (All): SIP is trannsient. AIP is built tto last
NancyMcGovern >> (All): That's it - yes
JohnGarrett >> (All): Right, I can see it going either way, verifying that you have the right information in a SIP can be complex too.  
                        Just think the AIP could be easier because it is all under Archive control.  SIP verification is a two-person interaction involving both Producer and Archive.
David Giaretta >> (All): In terms of the audit trail we can check that the SIP is what was intended. Then we check that part of all of one or more SIPs 
                                     get put correctly iunto an AIP and then we ensure the AIP is built to last
JohnGarrett >> (All): I agree
NancyMcGovern >> (All): Yes
David Giaretta >> (All): Maybe we need something like that in an introduction (unless it's there already and I forgot!) to show how things hang together
Bruce Ambacher >> (All): Here is where a cross-reference to B1 will help and to B4
RobertDowns >> (All): So, should 2.11 state that Repository verifies each AIP for completeness and correctness with the SIP(s) received at the point it is 
                                generated and inclusion of preservation information for the determined level of preservation
David Giaretta >> (All): Looks like we need a volunteer to add some such text
Katia Thomaz >> (All): and the description of this is at B2.3
Katia Thomaz >> (All): so B2.11 check what was defined at B2.3, ok?
Bruce Ambacher >> (All): I am going to have real participation problems for the balance of the year.  I will be teaching on Wednesday during this time 
                               from late August through late December.  I also am away next week.
David Giaretta >> (All): and then adds the things needed for B2.2
David Giaretta >> (All): DO we need to re-consider the day of the week for the meeting?
NancyMcGovern >> (All): yes 
David Giaretta >> (All): Shall I put up another voting table for day of the week?
NancyMcGovern >> (All): that would be great
Helen Tibbo >> (All): Monday would be good for me in the fall. I teach on Tuesdays and Thursday at this time and maybe on Wed.
Helen Tibbo >> (All): Friday could also work for me.
David Giaretta >> (All): ACTION on me to put up a table for meeting day
Katia Thomaz >> (All): for me wednesdays have been ok
Helen Tibbo >> (All): I have been dropped out of chat each time I tried to send until just now. Not sure what the problem is.
David Giaretta >> (All): Can we allocate an action for adding the intoductory text we talked about?
Bruce Ambacher >> (All): yes
David Giaretta >> (All): ACTION Bruce to add introductory text to show how things hang togehter for an audit trail
Katia Thomaz >> (All): ok.
David Giaretta >> (All): ACTION David to make a small change to B2.11 as discussed (just a few words from Bob plus a cross ref)
Helen Tibbo >> (All): I think I am out again
David Giaretta >> (All): Is that OK?
Bruce Ambacher >> (All): I will not get to my action until after August 8.
Helen Tibbo >> (All): Can we have the chat text up on the wiki automatically? How is that captured?
David Giaretta >> (All): Can someone else take Bruce's action
Katia Thomaz >> (All): so i propose next meeting start at  B3.2, ok?
Helen Tibbo >> (All): Can we have the chat text up on the wiki automatically? How is that captured?
David Giaretta >> (All): Helen - we can just cut and paste from the chat
David Giaretta >> (All): Katia - yes - good to note that
JohnGarrett >> (All): I can't seem to cut and paste from the chat.
RobertDowns >> (All): David, I sent the July 11 chat text to you by email
David Giaretta >> (All): Robert - I'll check - mea culpa again! Thanks
RobertDowns >> (All): I just sent it about half an hour ago
David Giaretta >> (All): Shall we call it a day - my pain killers are wearing off 
Bruce Ambacher >> (All): yes
NancyMcGovern >> (All): I have to run too.  Next week then... hope you feel better, David!
Katia Thomaz >> (All): yes
Katia Thomaz >> (All): bye
RobertDowns >> (All): Bye, all!

-- DavidGiaretta - 25 Jul 2007

Edit | Attach | Watch | Print version | History: r2 < r1 | Backlinks | Raw View | Raw edit | More topic actions
Topic revision: r2 - 2008-02-13 - KatiaThomaz
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