Notes from Megameeting 18th May 2009

Attendees

BarbaraSierman Koninklijke Bibliotheek, Netherlands
BruceAmbacher UM
DavidGiaretta STFC
JohnGarrett GSFC
KatiaThomaz INPE
MarieWaltz CRL
MarkConrad NARA
SimonLambert STFC

Summary

There was a discussion of the structuring of the metrics, and particularly the distinction between mandatory and supporting text and where they should appear.

It was agreed to complete the actions from last time and meanwhile David will circulate a clean version.

Because 1 June is a holiday in some countries, the next MegaMeeting will be on 26th May at the same time.

Actions (see transcript for details):

  • DavidGiaretta to produce cleaned up version of the current working document.
  • All to amend the metrics in the sections for which they are responsible, so that supporting text is informative, and that all the normative material from the supporting text is in the metrics.

Transcript of chat

JOhnGarrett >> (All): I think Bruce's section is fine, but I think that is the 
only section that is organized that way.
David Giaretta >> (All): Yes
David Giaretta >> (All): Yes - assuming you meant what was on the WIki already - 
end of April
BruceAmbacher >> (All): Virtually all of the nesting of metrics is in B2
David Giaretta >> (All): Yes - that caused some difficulty with how much 
additional text to add
David Giaretta >> (All): SO what about the other sections?
David Giaretta >> (All): If people agree with keepng the "This is necessary..." 
in the supporting text then I can complete my action of producing a clean 
version
BruceAmbacher >> (All): I thought last week's action was for each section author 
to create the new netrics.  Correct?
David Giaretta >> (All): Bruce - yes - but only you and Simon responded
Barbara Sierman >> (All): Sorry, I was away
Barbara Sierman >> (All): Mark, you reviewed part of it, do you intend to review 
the rest also?
David Giaretta >> (All): But when I looked  over things there thought that in 
fact there was not so much to do
David Giaretta >> (All): Yes, there are also Mark's comments
JOhnGarrett >> (All): Yes, I agree it is OK to keep the "This is necessary ..." 
in supporting text.   And any "The repository must ... " from supporting text is 
combined into the metric or made a sub metric.
Mark Conrad >> (All): When I have a minute to come up for air, I will review the 
rest of the document.
David Giaretta >> (All): Mark - If you have not done so then perhaps you can 
wait for the clean version - Wednesday
Barbara Sierman >> (All): I did not intend to force you to something...
Mark Conrad >> (All): Ok.
David Giaretta >> (All): SO I'll go ahead and produce a clean version on that 
understanding
BruceAmbacher >> (All): David, how are we doing in terms of the schedule?
JOhnGarrett >> (All): Good.  One more question.  Should there be 1 supporting 
text, example and discussion for a metric or should each submetric have those 3 
sections
David Giaretta >> (All): Bruce - Behind - as normal - but I think that as long 
as we can do some test audits in June we should be OK
David Giaretta >> (All): John - we decided to be flexible I think - so if we had 
something to say about a submetric then we should say it otherwise not
JOhnGarrett >> (All): I think Bruce's section where there is only 1 per metric 
works better, but the other sections usually have those sections for each sub-
metric.
David Giaretta >> (All): ..oops - got garbled - I'll try again
JOhnGarrett >> (All): OK, let's go with what is there and see if we get 
comments.
David Giaretta >> (All): OK
David Giaretta >> (All): So the action is to complete our actions form last time 
and in particular I'll get the clean version out
BruceAmbacher >> (All): Mark, you are our purist.  Are you satisfied with the 
current text style?
Mark Conrad >> (All): What are you calling the current text style?
BruceAmbacher >> (All): The direction in which we are headed with more complete 
metrics and "this is necessary" as the beginning of the supporting text.
Mark Conrad >> (All): If you mean like what you sent around last week, yes. I 
could understand what you had written and it was easy to get the big picture.
David Giaretta >> (All): Except I was saying that what Bruce sent around 
included explanatory text in the metrics
BruceAmbacher >> (All): Perhaps we should dissect one or two of mine or Simon's
Mark Conrad >> (All): Can we use the document that Bruce e-mailed so we can be 
sure we are all looking at the same document?
BruceAmbacher >> (All): Here is the first.
BruceAmbacher >> (All): 1.1.1   B2.1 THE REPOSITORY SHALL HAVE AN ASSOCIATED, 
PRINTABLE DEFINITION FOR EACH AIP OR CLASS OF AIPS PRESERVED BY THE REPOSITORY 
THAT IS ADEQUATE TO FIT LONG-TERM PRESERVATION NEEDS. TO ENSURE THAT THE AIP AND 
ITS ASSOCIATED DEFINITION CAN ALWAYS BE FOUND AND MANAGED WITHIN THE ARCHIVE. 
Mark Conrad >> (All): David, What do you see as explanatory in that text?
KatiaThomaz >> (All): This is explanatory text: TO ENSURE THAT THE AIP AND ITS 
ASSOCIATED DEFINITION CAN ALWAYS BE FOUND AND MANAGED WITHIN THE ARCHIVE. 
BruceAmbacher >> (All): David, If I understand you correctly, you would not 
include from "to ensure" on.  Correct?
BruceAmbacher >> (All): That part would go back to supporting text.
JOhnGarrett >> (All): I agree.  That's it for me today.  Talk to you again in 2 
weeks.
David Giaretta >> (All): Bruce - yes - the "To ensure.." is explanatory and 
would go back to the supporting text
BruceAmbacher >> (All): I agree on B2.1 but what about this submetric?
BruceAmbacher >> (All): 1.1.1.1   B2.1.1 The repository shall be able to identify 
which definition applies to which AIP to ensure each AIP can be properly 
parsed/interpreted. 
David Giaretta >> (All): The same applies...
David Giaretta >> (All): ...we just have to check that the implication about 
"parsed/interpreted" is clear from the metrics...
Mark Conrad >> (All): I think that most of the "to ensure" statements are a 
significant part of the metric - not explanation.
David Giaretta >> (All): ...i.e. that we might consider whether B2.1 needs some 
extra words about the Description - e.g. a description which allows us to 
parse/interpret
David Giaretta >> (All): Mark - I think I know what you mean - there do seem to 
be some issues which are caught in that explanation which is not explicit in the 
metric - as with the example I gave
BruceAmbacher >> (All): Obviously I lean toward Mark's view as that is how I 
rewrote my metrics.
David Giaretta >> (All): I'm happy to go with the consensus - my concern is that 
it becomes hard to see where the distinction is between normative and 
explanatory text
Mark Conrad >> (All): The to ensure statements lay out the absolute minimum that 
the auditors should be checking for for each metric or sub-metric. That does not 
seem explanatory to me.
BruceAmbacher >> (All): What is the consensus of the rest of you?
KatiaThomaz >> (All): I think normative is what and explanatory is how, why, 
when, where, who, how much
David Giaretta >> (All): Mark - I think the issue is that our metrics - as with 
the example I gave - missed some of the essence of what had to be done - and 
that is the underlying problem
David Giaretta >> (All): Katia - I guess I am saying that the "to ensure..." is 
about "why" 
David Giaretta >> (All): ..and so is Supporting
KatiaThomaz >> (All): yes, i agree
BruceAmbacher >> (All): Yes but we must show why a metric is mandatory and why 
it will prove what is is supposed to determine.
Barbara Sierman >> (All): so that is more then supporting
KatiaThomaz >> (All): so I think the metric should tell us only what and the 
supporting text the other issues
David Giaretta >> (All): Bruce - Isn't that explanatory text?
BruceAmbacher >> (All): But the supporting text is NOT mandatory, therefore it 
can be ignored
Barbara Sierman >> (All): but the why in the metrics might be a statement that 
is beyond discussion
David Giaretta >> (All): Bruce - surely the supporting text guides the auditor
Barbara Sierman >> (All): bruce, i think you are right
BruceAmbacher >> (All): The supporting text guides the repository and the 
auditor but it is not mandatory.  It does not even have to be considered.
BruceAmbacher >> (All): Any repository that prepares for an audit using only the 
metrics is shortsighted and could fail.
David Giaretta >> (All): But we cannot specify everything - it has to be down to 
the auditor's judgement - and if a reposoitory disregards the guidance they 
probably deserve to fail!
Mark Conrad >> (All): Using the B.2.1.1. example. If we replaced the text that 
begins, "To ensure", with " THE AIP AND ITS ASSOCIATED DEFINITION MUST ALWAYS BE 
Available. linked AND MANAGED WITHIN THE ARCHIVE. Then would you still say it 
was explanatory?
David Giaretta >> (All): Mark - that looks exactly what I meant!
David Giaretta >> (All): ...we include the concepts from the existing supporting 
text into the metrics and separate the "why" into the supporting text
Mark Conrad >> (All): David, It is the same meaning as the to ensure statement - 
except the to ensure statement is much easier to read and understand.
David Giaretta >> (All):  But it reads  like normative text - i.e. do this! 
rather than an explanation
Mark Conrad >> (All): I don't understand why we are getting wrapped around the 
axle about what is normative and what is explanatory. It seems to me that the 
important distinction is what is mandatory and what is optional.
David Giaretta >> (All): Yes - I had hoped this would be a fairly simple thing 
to resolve
KatiaThomaz >> (All): this another thing. you can have an optional metric or an 
mandatory metric
BruceAmbacher >> (All): I thought we had rewritten each metric to be mandatory
Mark Conrad >> (All): Katia, No! We agreed a long time ago to limit metrics to 
mandatory requirements.
KatiaThomaz >> (All): i know
David Giaretta >> (All): Katia - the auditor must decide how far the metric has 
to be satisfied by the particular repository since none is a simple binary
KatiaThomaz >> (All): i am only trying to explain thet they could be optional
Mark Conrad >> (All): If there is a metric in this document it is not supposed 
to be optional.
David Giaretta >> (All): Mark - I think it just comes down to  wording - it may 
be an English vs US use of words.....
KatiaThomaz >> (All): the words shall, should, may etc. will say this.
BruceAmbacher >> (All): The only optional is in what materials prove your 
compliance
KatiaThomaz >> (All): and we decided to let all the metrics mandatory
David Giaretta >> (All): ...so wording like "adequate to ensure..." would be OK 
for me i.e. it clarifies the meaning - what - rather than explains why
BruceAmbacher >> (All): The supporting text amplifies, explains the metric but 
it is optional
Mark Conrad >> (All): Bruce, When was that decision made? For the longest time 
we said that the metric and the supporting text were mandatory and everything 
else was optional.
KatiaThomaz >> (All): the metric is mandatory or optional, not the supporting 
text. the supporting text is explanatory.
Mark Conrad >> (All): Katia, The metric is mandatory.
KatiaThomaz >> (All): we decided this in this standard
BruceAmbacher >> (All): But haven't we virtually eliminated the mandatory 
("shall") part of the supoprting text, leaving only the explanatory?
BruceAmbacher >> (All): eliminated by moving it into the metric itself
David Giaretta >> (All): Bruce/Mark - yes I thought that was where we were 
heading
David Giaretta >> (All): ...so as a matter of tidiness we were trying to move 
text from under the heading of "Supporting Text" into the metric
Mark Conrad >> (All): Bruce, Certainly in the document that you e-mailed last 
week that is the case. I am not sure what is the current version for the other 
parts of the document.
BruceAmbacher >> (All): Simon was the only other author to attempt a rewrite
David Giaretta >> (All): .....and the issue was to do this in a consistent and 
non-repetative way.
David Giaretta >> (All): Bruce - that's right
David Giaretta >> (All): But I still think this is not a big issue. Can I try to 
do it in the clean version by mid-week?
Mark Conrad >> (All): It seems to me that we are allowing issues of style to 
trump issues of substance. Decide which way we are goiing to do it. Make sure
that it is abundantly clear what is mandatory and what is not. And be done with 
it.
David Giaretta >> (All): Yes!
David Giaretta >> (All): I don't think this is a big issue
BruceAmbacher >> (All): Next Monday is the federal holiday for Memorial Day.  
Should that affect the timetable?
Mark Conrad >> (All): I will not be here.
David Giaretta >> (All): It's a holiday in the UK also
BruceAmbacher >> (All): The following MOnday I will be on travel
David Giaretta >> (All): Me too - on June 1st - Simon also
Barbara Sierman >> (All): me too
David Giaretta >> (All): How about Tuesday 19th May?
Barbara Sierman >> (All): tomorrow?
David Giaretta >> (All): Oops - I meant 26th May
BruceAmbacher >> (All): The schedule is really slipping.  Will we fall before we 
cross the finish line
Barbara Sierman >> (All): I'm not available
BruceAmbacher >> (All): I can be available without mic
David Giaretta >> (All): It would be nice to have a chance to see if we can 
agree on a clean document next week - I am also worried about schedules 
David Giaretta >> (All): Fine for me
Mark Conrad >> (All): I am available at this time on the 26th.
David Giaretta >> (All): Can we meet on the 26th - those that can just to see if 
there are major issues?
BruceAmbacher >> (All): David can you send a broadcast to all after the new 
version is posted and announce date
David Giaretta >> (All): ...Yes
KatiaThomaz >> (All): ok for me
Mark Conrad >> (All): Will we meet at this same time?
David Giaretta >> (All): That would be best for me
BruceAmbacher >> (All): Does anyone want to give David any further guidance?
KatiaThomaz >> (All): thanks and have a nice week. bye.
David Giaretta >> (All): All guidance welcome!!
Mark Conrad >> (All): Bruce, What would you suggest?
BruceAmbacher >> (All): I do not have anything but I do find the lack of 
consensus troubling this late in the work.
David Giaretta >> (All): I really believe this is not a big issue - we are just 
a little at cross-purposes - I think seeing a clean version will help immensely
Mark Conrad >> (All): As I said before, I think what you have written is good. 
It is easy to understand. I am not concerned about whether the to ensure clauses 
are explanatory. They are mandatory. As long as that is clear, I don't care if 
they are in the metric or the Supporting text.
BruceAmbacher >> (All): I would like to see the global "Preservation 
Planning/Preservation Plan re-examined.  It turned some metrics into 
preservation metrics when they were something else.
Mark Conrad >> (All): Yes. I believe the meaning of several of the metrics in 
Section A. were changed.
BruceAmbacher >> (All): I also am leary about creating an annex of just the 
metrics' text
David Giaretta >> (All): Mark - that is a more substantive issue that requires 
some checking
David Giaretta >> (All): Bruce - I tend to agree - a tool would be better
BruceAmbacher >> (All): Marie, did you get the discussion you needed to re-
examine your last draft?
Mark Conrad >> (All): Bruce, If we include the text of the metrics in the table 
of contents that would allow readers to see the big picture.
BruceAmbacher >> (All): That is a very long TOC
Mark Conrad >> (All): You need the text of the metrics somewhere so that the 
reader can see the forest.
David Giaretta >> (All): We can decide how far down in sub-sections to go - we 
have various groupings which help with the big picture
David Giaretta >> (All): ..I meant in the ToC
BruceAmbacher >> (All): Can a technical editor help steer us on this?
BruceAmbacher >> (All): or the CCSDS style, or the ISO style?
David Giaretta >> (All): Maybe - but I don't really think so
David Giaretta >> (All): We can re-read things like 27001 again
Marie Waltz >> (All): Sorry all, its a busy morning, I had to step out. I got 
feedback on the first metrics and am planning to make Mark's changes
David Giaretta >> (All): Great
BruceAmbacher >> (All): OK.  We can try an expanded TOC.  If we don't like it we 
do an annex
Mark Conrad >> (All): Bruce, We should be consistent with the CCSDS and/or ISO 
styles, but that should not trump making the mandatory and non-mandatory 
portions of the standard explicit.
BruceAmbacher >> (All): agreed.  Substance before style
David Giaretta >> (All): Substance > clarity > style
BruceAmbacher >> (All): OAIS has a brief TOC
David Giaretta >> (All): Must go - bye all - I'll do my action and include any 
contributions I receive by mid-week
BruceAmbacher >> (All): See/hear you all next Tuesday
Marie Waltz >> (All): David I'll send some by then
Mark Conrad >> (All): See you on the internet!

-- SimonLambert - 18 May 2009

Topic revision: r1 - 2009-05-18 - SimonLambert
 
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