Notes from Megameeting 26th November 2007

Attendees:

CandidaFenton HATII, Univ Glasgow
DavidGiaretta STFC
DonaldSawyer NASA GSFC
HelenTibbo UNC
JohnGarrett NASA/GSFC
KatiaThomaz INPE, Brazil
NancyMcGovern ICPSR
RobertDowns Center for International Earth Science Information Network (CIESIN), U Columbia
SimonLambert STFC

All the discussion at this meeting was conducted by chat, so the following transcript of the meeting (with a few typos corrected) is complete.

Summary: All to try and work through the matrix of mandatory requirements marking Y, N or ?. SimonLambert to amend the statements of B1.3 and B1.6 as agreed at the previous meetings, and to add the definitions of "mandatory" and corresponding responses agreed at this meeting.

David Giaretta >> (All): Did I mention that the PARSE.Insight bid is being funded?
David Giaretta >> (All): PARSE.Insight is an EU bid which has some money in it for a meeting on certification
Helen Tibbo >> (All): Ah, that would be very useful!!
David Giaretta >> (All): The larger part of the work is a Roadmap for infrastructure to support preservation as part of the EU e-infrastructure
David Giaretta >> (All): Following the Roadmap is a bid survey, Gap analysis and then an "impact analysis tool" to help the EU focus its money.
David Giaretta >> SimonLambert: Simon - can you lead this please
SimonLambert >> (All): Yes sure
KatiaThomaz >> (All): hi
SimonLambert >> (All): Basically we were working through section B looking for mandatory reqts.
SimonLambert >> (All): Candida just sent a summary of the state of "voting".
SimonLambert >> (All): Last week we reached agreement on B1.6.
JohnGarrett >> (All): Hi, Just here for a few minutes today, then have to head out to another meeting.
SimonLambert >> (All): I think we also need to clarify B1.5 and B1.7.
candida fenton >> (All): Hi All. Form the 'voting' it looks a little like people started at the top, then didn't get all the way to the bottom. Is this the case? 
Helen Tibbo >> (All): It is with me
RobertDowns >> (All): Me too
NancyMcGovern >> (All): I started - I'm still struggling generally with the need for mandatory items.  I'll try to keep going.
Helen Tibbo >> (All): I keep struggling with the inclusiveness of some items and the sparcity of others.
Donald Sawyer >> (All): I saw the new 1.6 text, but I wasn't sure everyone agreed.  Does everyone now agree it is mandatory?
Donald Sawyer >> (All): I'm sorry I was behind, but I've now responded to matrix for section B1 only
JohnGarrett >> (All): Also the case with me.  I figured at the speed we are going, I could do a subsection a week and stay ahead of our discussions.  But I'll try to get more done.
David Giaretta >> (All): Before I forget to tell you - my PARSE.Insight project is going to be funded by the EU. I think I can use it to fund a face to face meeting. That might help us reach conclusions.
SimonLambert >> (All): I think everyone present last week agreed B1.6 should be mandatory in its revised form.
Helen Tibbo >> (All): Where is that new text?
JohnGarrett >> (All): It was sent out to the mail list
SimonLambert >> (All): The new text was only sent by email (by me).
SimonLambert >> (All): I suppose when we have agreed the revisions should go back into the working doc.
KatiaThomaz >> (All): i don´t consider B1.6 mandatory
KatiaThomaz >> (All): look at the matrix
Donald Sawyer >> (All): Helen had raised the issue as to whether it was required for preservation.  This brings up, again, the scope of what we're trying to cover.
David Giaretta >> (All): I suspose that if a valid response is "no response" then there is little difference between mandatory and non-mandatory
Donald Sawyer >> (All): What does 'mandatory' mean?  Perhaps it means a 'significant' deficiency, that then impacts the eventual score?  Or does it mean you 'fail' the certification?
NancyMcGovern >> (All): This discussion loop is why I think we should finish reviewing the requirements then think about a recommended set or something - leaving each repository to document and justify what it's doing - how and why
David Giaretta >> (All): I think we hoped that we could go through fairly quickly and see if there was an obvious set which we all agreed were absolutely necessary 
JohnGarrett >> (All): I think mandatory means you fail the certification if you don't meet a mandatory criteria.
Donald Sawyer >> (All): Nancy, I'm not sure I understand your point.  Would you elaborate?
RobertDowns >> (All): We could consider the possibility of either assigning weights to items or recognize that not mandatory does not mean failing certification
Donald Sawyer >> (All): Seems that, at a minimum, failing a mandatory must be a significant deficiency, however it is scored.
candida fenton >> (All): I understood mandatory to mean that you do not meet if you fail.
David Giaretta >> (All): It may be easier to say that we are first of all looking for these pass/fail metrics
SimonLambert >> (All): One thing the matrix is showing is a number of reqts that need further clarification e.g. B1.5.
NancyMcGovern >> (All): There are several strands here - what requirements are relevant for assessing trustworthy repositories - the basis for self-assessment by a repository.  How will the audit process work then further beyond that what levels or types or requirements for certification do we have.  we seem to be conflating those things and repeating bits of the discussion.  That's how I see it.  
Donald Sawyer >> (All): If it means you fail, if you don't respond to the Producer as previously agreed in B1.6, should this be enough to cause you to fail?  It is not essential for preservation, but doesn't  help with overall trust.
David Giaretta >> (All): Nancy - I  agree - we should be doing something clear and in a sense simple - what is realy really mandatory. And if we have time then we add comments to others
David Giaretta >> (All): Simon - yes, each time we look at this in a different way we will see something different
Donald Sawyer >> (All): I agree with simple, but we need to have the same understanding of what we mean or we are just fooling outselves in the short run and will have to come back to it later.
NancyMcGovern >> (All): About B1.6 I have taken B1.6 to be about notification mroe than about responding - producers may not send messages to respond to but the repository should notify the producer during the process.  We are establishing three points of notification here- at receipt, at acceptance (after ingest as content goes to archival storage) and at release.  A repository should not fail if it does not meet this one requirement but this requirements contributes to lifecycle management of digital content.
David Giaretta >> (All): Nancy - there seems to be good agreement that B1.6 is not mandatory. 
David Giaretta >> (All): Maybe the way to make progress here is to not try word-smithing - instead just Y/N and if the "Y" is based on some assumption then that should be put in the comments
Donald Sawyer >> (All): We could take the view that we are looking at the requirements, for now, as mandatory or not for doing preservation and being able to demonstrate that.  In such a case, B1.6 would not be mandatory in my opinion
RobertDowns >> (All): Yes, we seem to agree that B1.6 is not mandatory, but we should consider amending it as suggested and we also should consider the suggestion that John had offered, for the repository to post the status on the web as an alternative response
David Giaretta >> (All): ....and correspondingly if "N" then move on
David Giaretta >> (All): Shall we try focussing on the "Y"'s and try to get to the end by  next week?
NancyMcGovern >> (All): I guess but is it the most useful sequence to vote Y/N when it seems that a number of rhe requirements are candidates for revision?  Might adjusting the wording influence Y or N?
Donald Sawyer >> (All): David, I'm still not clear what 'Yes its mandatory' means'.  Until we agree on this, I can't proceed.
RobertDowns >> (All): Also, the comments accompanying a Y indicate interpretations or qualifications
David Giaretta >> (All): Don, I guess that in part was what I was aiming at - namely are there any that are so obvious that we all say "Y" it is mandatory i.e. if this is not done then fail. Maybe there are no such metrics. The next step would be to go through and say definitely "N" and that will leave a group in the middle which we can focus on. 
Donald Sawyer >> (All): How about 'mandatory means it is a critical requirement that must be done in order to either be doing preservation adequately or demonstrating the same'?
David Giaretta >> (All): Sounds OK
NancyMcGovern >> (All): There may be a fine line between essential and important - perhaps the binary Y/N is tricky.  By the way, I missed the very start of the mandatory discussion - why does the voting not include section A?  A number of those would also be showstoppers for audit and certification.  
David Giaretta >> (All): I'll try to go through and add any "Y"'s to the full list.
SimonLambert >> (All): The only reason was that at the time we were working on B.  It should be completed with A and C.
NancyMcGovern >> (All): Got it. 
Donald Sawyer >> (All): Nancy, 'important' could be given by use of the ? mark,
SimonLambert >> (All): If Y = yes as it now stands and N = no and ? = needs clarification, that would be sufficient for a complete pass quickly.
David Giaretta >> (All): Sounds good. 
RobertDowns >> (All): Sounds good to me, too.
Donald Sawyer >> (All): Are we also agreeing 'critical for preservation or demonstrating the same'?
RobertDowns >> (All): That seems reasonable. If we all agree, we should post the definition on the matrix
David Giaretta >> (All): I suspect we'll see some with all "Y"'s and some with a split vote - we can then discuss those in detail, and perhaps clarify what each of use means by "critical"
Donald Sawyer >> (All): yes, I would think that will be the case
NancyMcGovern >> (All): It's a fine process if we think this is a desirable outcome.
Donald Sawyer >> (All): Nancy, I think we are trying to take TRAC to the next level of agreement.
KatiaThomaz >> (All): ok for me
Donald Sawyer >> (All): I would change a few of my 'votes' based on the above definitions.  B1.6 would be 'n' for me now.
NancyMcGovern >> (All): I can see that's one way of looking at it and an important step to take.  It feels like we stoped doing one thing - completing a review of the requirements - to do another vote on mandatory requirements, a process that would beenfit from the completion of the review.
RobertDowns >> (All): With this definition, I also will need to review my votes
David Giaretta >> (All): nancy - I think this is desirable thing to do as long as it as done fairly quickly.  
David Giaretta >> (All): ...but we should finish going through the metrics - there were just a few left I think.
SimonLambert >> (All): Going back to the discussion that motivated mandatory reqts, the emphasis was on identifying what we also called "Core" reqts.
Donald Sawyer >> (All): My view is that use of the matrix is forcing us to take a closer look at the requirements, and is really doing the review.  However if there are 'agreed' updates to text, they should be reflected back into the document - not just sent via e-mail.
David Giaretta >> (All): Good point
RobertDowns >> (All): Yes, I agree that we need to update the text to reflect our discussions
David Giaretta >> (All): I need to go now but I'll try my utmost to mark my "Y"s by next week
KatiaThomaz >> (All): david, is it possible to include a feature on the wiki for sending messages directly to the group? as you know, i haven´t been able of sending you messages from my emails...
NancyMcGovern >> (All): Next week then.
David Giaretta >> (All): Katia - I'll see and let you know
KatiaThomaz >> (All): ok, thanks.
KatiaThomaz >> (All): bye for all and have a nice week
Donald Sawyer >> (All): Simon, would you put up the definitions of the Y, N, and ? with respect to our discussions today?  Maybe at the top of the matrix?
SimonLambert >> (All): Yes, OK
Donald Sawyer >> (All): Great!  Bye
SimonLambert >> (All): Also amend B1.3 and B1.6
Donald Sawyer >> (All): o'k
SimonLambert >> (All): Is that all for today?
JohnGarrett >> (All): OK, bye, type at you next week.

-- SimonLambert - 26 Nov 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