Nice, very nice.
I am completely aligned with what you
say here.
Also with regard to bridging the gap.
--
Dennis
-----Original Message-----
From: Owen_Ambur@fws.gov [mailto:Owen_Ambur@fws.gov]
Sent:
Wednesday, July 18, 2001 13:20
To: dennis.hamilton@acm.org
Subject: RE:
Federal Open Source Conference - Records Management &
XML
Dennis, the short answer to your question is that I'm fed up
with excuses
that are tantamount to endorsing perfection as the enemy of that
which is
merely good. I'm also sick and tired of seeing "legacy"
systems being used
to justify continuing to live in the past, if not to
re-live the mistakes
of the past. (When paper was invented, we
didn't insist that information
continue to be recorded on stone.) Of
course, I want to allow for all
potential eventualities -- to the degree that
it is possible to do so
without, as you say, being "paralyzed".
Indeed, that is one of the
benefits that XML brings to the table -- the fact
that it is "extensible".
In his presentation at the XML WG meeting this
morning, Ken Thibodeau of
NARA pointed out that it is possible to have our
cake and eat it too (my
words) in the sense that all renditions of an
electronic record, including
the original one, can be maintained, if desired,
under the ERA model. In
other words, this is not an argument we need to
have. Instead, we should
move out forthrightly to *implement* those
more *basic* functions that we
all agree not only are needed but also
possible to implement in the
near-term. Once we've done that, then we
can move on together as well as
separately to add the next increments of
functionality that are common or
unique to each of
us.
Owen
BTW, with respect to IT architecture planning, for the
most part it has
been an ivory tower exercise and thus a failure.
However, in my white
paper for the Department of the Interior, I noted that
XML shrinks the gap
between the theory and practice of data standards and IT
architectures.
http://iosapps2.ios.doi.gov/eiastuff.nsf/5732a0cd0788c5e7852568e90075ffea/$FILE/eXtensible%20Markup%20Language%20(XML)%20Greek,%20Esperanto,%20Panacea%20or%20Snake%20Oil.htm
"Dennis
E.
Hamilton"
To:
<Owen_Ambur@fws.gov>
<dennis.hamilto
cc:
<ted@xanadu.net>
n@acm.org>
Subject: RE: Federal Open Source Conference
-
Records Management &
XML
07/18/01
01:26
PM
Please
respond
to
dennis.hamilton
Thanks
for your follow-up to Ted.
Funny, I don't think he is being
defeatist. I thought he was reminding us
of the reality of what happens
with documents and identifying a challenge
for us to consider. That the
result may be unruly and surprisingly
disorganized at times, in contrast with
what we might prefer to see, is,
for
me, something to be realistic
about. I would hope that systems allow
enough
ad hoc introduction of
connections that people can track what they are
willing to track in order to
provide a "complete" record consistent with
their purpose in having and
preserving the record in the first place.
I also thought Ted's
observation about being able to provide "overlays"
that
carry the
additional materials that come to be
associated-with/imposed-on
an
electronic document was interesting.
Ted is credited, in a different
discussion I'm in, with promoting the
idea that we need a way to carry
associations and structural information of
many abstracted kinds separate
from the raw content, which for many reasons
may not be alterable. That
resonates with me.
I will leave it
for Ted to speak for himself. But I didn't want to
overlook
the
value of considering that there is a messiness that needs to
be
appreciated.
A SMALL SOAPBOX AND SOME PONTIFICATING:
My
exposure to digital library systems and efforts to
preserve
deteriorating
19th century materials has been instructive in this
regard. When we were
scanning deteriorating mathematics books, it was
often important to capture
marginalia that provided corrections, either from
a reader or from the
author of the book, that were part of that particular
artifact. The idea
was to have an image of the pages of the book
that was reliable for
reproducing the book on acid-free paper for replacement
of the unusable
shelf copy and for reprinting for others who might want it
for some
scholarly pursuit. For this purpose, it is desirable to
capture everything
of the artifact, including notes and material that was
added to it. Even
the check-out history of a library book can be
important.
Although librarians, catalogers, and archivists are edgy about
the
prospect,
I also saw that much work on classifying and describing
books captured this
way could be left to future users of the book if we had a
way to introduce
annotations, index entries, and notes of scholars who were
willing to
contribute entries and notes as part of accessing the material
on-line.
There was no way someone was going to do a mass retrospective
indexing. It
would be valuable if all such additions had no impact on
the "original" of
record, but could be provided in conjunction with it in
some way.
Finally, once we move to a level of extracting the text and
providing
textual analysis, automated indexing, and content-based retrieval,
the
scanned images always remain as the authoritative source, especially if
the
automated process is lossy or not entirely reliable (e.g., it can't pick
up
the annotations unaided). Again, there might be opportunistic ways
to use
the energies of people who access the book to clean up the extracted
text
as
well.
I have a philosopher friend who is an expert on an
Anglican minister whose
sermons were influential in the era when
utilitarianism was being explored
as an ethical and moral principle.
This was at the time of a British
reform
movement. My friend had the
privilege of visiting the British Library
about
10 years ago and perusing
their volumes of the sermons. In this library,
there is a record of who
has checked out these books and there are notes
that can be identified with
particular individuals, including ministers of
government. It was a
delight for him to be able to peruse and examine
those
traces of the use
of the material as much as it was to have access to the
original works.
I thought it pretty amazing that he could take down a book
and see that
Benjamin Disraeli had left comments in it.
MY POINT:
It might be
thought that the requirements for governmental records
management and
accountability for electronic records is simpler than this.
Good. At
the same time, I would favor a model that does allow for
unanticipated
associations and usages over time. And I would not want to
foreclose
the integration of digitally-sourced and paper (scanned)
materials.
A
QUESTION:
I take the position that there is always a legacy system to
deal with, and
that introducing/upgrading the system is going to require an
accommodation
of the legacy rather than a replacement of it. That mixed
as well as ad
hoc
systems are always before us seems like a simple
practical observation.
Especially because there are imperatives for acting
now and doing what can
be done without having perfect understanding of how
requirements will
evolve
over time. Isn't this something we should
be appreciative of, and
attentive
to, without being paralyzed by
it?
In a job interview a long time ago (1972), the IT executive of a
major
corporation asked me why he needed a systems architect for his
largest
project, a major on-line processing system to be deployed in all
locations
of the enterprise. I told him that what system architecture
would provide
is a way to keep all of his technical options open while
evolving into the
future while understanding and preserving the essential
requirements on the
system design. I don't know whether that's the
answer he had in mind. I
did get the job.
Isn't this the kind of
approach you want to see in the creation of
structures for providing and
preserving accountability in electronic
records
over time?
--
Dennis
Dennis E. Hamilton
AIIM DMware Technical
Coordinator
------------------
mailto:dennis.hamilton@acm.org
tel. +1-425-793-0283
http://www.dmware.org
fax. +1-425-430-8189
ODMA Support http://ODMA.info/
-----Original
Message-----
From: Owen_Ambur@fws.gov [mailto:Owen_Ambur@fws.gov]
Sent:
Tuesday, July 17, 2001 14:21
To: Ted Nelson
Cc: dennis.hamilton@acm.org;
eharter@din.or.jp; ejw@cse.ucsc.edu;
Kathy_Moran@idg.com;
lewis.bellardo@nara.gov; marlene@xanadu.net;
Michael.Todd@osd.mil;
ted@xanadu.net; xanni@xanadu.net
Subject: RE: Federal Open Source Conference
- Records Management & XML
Ted, your response not only surprises
but also disappoints me. Of course,
perfection will never be reached on
this earth. However, we can do much
better than your defeatist response
suggests. Indeed, for we Feds, it's
the law! Reference the
Paperwork Reduction Act, Government Paperwork
Elimination Act, Sec. 508 of
the Rehabilitation Act, etc.
We need not be blind to see the truth.
However, the following "Action
Recommendation" from the American Federation
for the Blind demonstrates
that it may help:
"Given the historic
opportunity now before us, all stakeholders should: ...
consciously disabuse
ourselves of the historic notion that regards a
printed publication as the
true source document and all other media as
variants. In truth, the
true source already is -- and must remain -- an
electronic file. Our
task is to understand that it must be an accessible
electronic file, and that
all media representations of this file -- whether
hardbound standard print
books, large print, braille, recorded audio, or
synthesized audio -- are all
alternative media to a single electronic
source ..." <http://www.afb.org/ebook.html>
See also the
characteristics of a record as set forth in ISO 15489.
Owen Ambur,
Co-Chair, XML Working Group <http://xml.gov/>
Vice Chair, FIRM <http://pages.zdnet.com/firmweb/firmcouncil/id1.html>
Ted
Nelson
<ted@xanadu.n
To:
Owen_Ambur@fws.gov,
Michael.Todd@osd.mil,
et>
lewis.bellardo@nara.gov,
dennis.hamilton@acm.org,
Kathy_Moran@idg.com,
ejw@cse.ucsc.edu
06/27/01
cc:
ted@xanadu.net,
marlene@xanadu.net,
03:59 AM
xanni@xanadu.net,
eharter@din.or.jp
Subject: RE: Federal Open
Source Conference
-
Records Management & XML
This is kind of a surprise
to me.
One thing that has to be mentioned is that when a
document
is printed out, frequently corrections are made by
hand
that never make it into the electronic record. To say
nothing
of electronic followups that chase after, but never quite
catch,
the versions they are correcting.
Even if we could index
it decently, the electronic record is and
can never be anything but a
terrible mess. Unless such corrections
in a variety of formats
can be overlaid.
Best, T.Nelson
At 04:39 PM 6/25/01 -0400, you
wrote:
>
>Dennis, obviously, Ted has had some wild dreams of his
own. What I'll be
>interested to see is the practical contributions
that he and his Xanadu
>colleagues might be able to make toward enhanced
management, access to,
and
>preservation of "We the People's"
records. Your analysis below provides a
>useful framework within
which to consider the issues.
>
>One observation I'd make
regarding your level 3 is that unlike
paper-based
>and other
non-digital activities, activities conducted in electronic
>information
systems *automatically* create records and those records are
>readily
subjected to various forms of enhanced (automated) management.
>Thus,
while the keystrokes and mouse clicks are not in and of themselves
>"the
record," the instantaneous results of those "activities" are indeed
>one
and the same as the "record". The only issue is whether we have
the
>will and wisdom to apply the appropriate elements of metadata to
those
>records, together with the programmatic logic to leverage those
metadata
>toward more efficient and effective management of the records
they
>describe. How we deal with that issue will determine the
*quality* of the
>records we create and maintain.
>
>With
respect your level 2, it seems to me that the number and
specific
>elements of metadata to be applied to any particular class of
records is
an
>issue that readily lends itself to incremental
benefit/cost analysis --
>perhaps even in a largely automated fashion on
the Internet. However, for
>anyone to take such matters seriously,
performance metrics must be applied
>and they must embody consequences for
someone. See
>http://users.erols.com/ambur/RMmetrics.htm
>
>Regardless
of what the records management metadata elements and schemas
may
>be
and how they may evolve over time, I hope and anticipate they
will
>eventually be registered and readily available (for
automated
>implementation) in the repository of "inherently governmental"
XML data
>elements, DTDs, and schemas that we plan to pilot at
xml.gov. See
><http://irm.fws.gov/xml43age.htm>, <http://irm.fws.gov/xmlops.htm>,
><http://xml.gov/registries.htm>, and recommendation #5 (on
PDF p. 2) at
><http://www.nclis.gov/govt/assess/assess.appen24.pdf>.
>
>Owen
Ambur, Co-Chair, XML Working Group <http://xml.gov/>
>Vice Chair, FIRM <http://pages.zdnet.com/firmweb/firmcouncil/index.html>
>
>
>
>
>
>
"Dennis
E.
>
Hamilton"
To:
<Owen_Ambur@fws.gov>,
>
<dennis.hamilto
<Kathy_Moran@idg.com>
>
n@acm.org>
cc:
<ejw@cse.ucsc.edu>,
<ted@xanadu.net>,
>
<MORRISR1@LEAVENWORTH.ARMY.MIL>,
>
06/22/01 08:33
<Michael.Todd@osd.mil>,
<lewis.bellardo@nara.gov>
>
PM
Subject: RE: Federal Open
Source Conference
-
>
Please respond Records
Management &
XML
>
to
>
dennis.hamilton
>
>
>
>
>
>Owen,
>
>You
are a continual surprise to me. In my wildest dreams, I hadn't
placed
>myself on the same e-mail distribution as ted@xanadu.net and
these
>distinguished contributors.
>
>I have been looking at
your summary. I readily align with what I read
into
>it as an
underlying manifesto: that the peoples records
(including
>documents
>and data and other forms for capturing
coordinated human activity) are
>preserved for access and reuse without
impediments. In this vein, we
>desire
>that those who operate
in the public interest -- in government and
>elsewhere -- have
technologies that serve full accountability and that
>afford unparalleled
levels of visibility and usability across time
and
>space.
>
>At a grounded, practical level, I see three
areas:
>
>1. Demonstrating practical records management
with available and
emerging
>tools. Tools that preserve what is
familiar and used today in forms that
>are durable for access, faithful
presentation, with derivative usage
>afforded as much and as long as one
desires. I see two sub-themes here.
>First, having records in forms
that are publicly established and freely
>usable, independent of the
private, proprietary or public characteristics
>of
>the tools
employed to create, store, access, comprehend, and
manipulate
the
>content carried in those forms. Secondly, having
open, fully-disclosed
>implementations of essential tools for those
operations such that there is
>always a path onto new platforms and media
that preserves the value of
>legacy materials and ensures continuing
usability of material in
>established, public forms as long as
necessary.
> I say that what we are seeing with
XML, with WebDAV, and with
>open-source
>tools such as those
available through W3C, OASIS, and other sources.
>
This is very important, not only for giving us a grasp on
practical
>day-to-day concerns sometimes most-easily explored in the
small, but in
>creating a level of shared experience around what it is
like to seriously
>take on accountable preservation of the record of
enterprise.
>
>2. Looking at what must be kept and
preserved *about* records that
>supports
>accessibility,
interchange, convertibility and so on to the degree that
>time, space, and
the obsolescence of media, equipment, software and other
>platform
specifics force us to comprehend. It is not clear to me that
we
>have a grip on the subtleties of data about records and questions
of
>semantics (or intent) versus syntax (or form). I certainly thing
that
work
>on metadata interchange is leading us to where we are
beginning to
>understand the question. Again, the provision of open
mechanisms, public
>formats, and assured existence of some covering set of
tools that are
>themselves public property become
important.
>
>3. Finally there is the area that people
more visionary than I are
>continuing to extend. That has to do with
novel (still) arrangements that
>involve the intimate incorporation of
records, as we might think of them,
>activity itself (and probably vice
versa). Where, to put it badly, the
>activity is the record (and
probably vice versa). These may cast content
>in
>quite
different forms, and deal with questions about the many views
and
>structures of content that not only exist side by side, but that
are
>themselves dynamic, growing expanses of information about the topics
and
>activities that concern us. Xanadu struck me as an instrument
of this
>level
>of vision and I disqualify myself from having
anything more to say that
>hasn't already been said
better.
>
>My interest.
>
>The work that I support on
AIIM DMware is situated, for me, in (1) and,
>very
>slowly,
(2). I see one as essential for providing the particulars
that
>demonstrate why we are looking to (2) and (3), and what the barriers
are
>that have headway in these areas be non-trivial. Although once
we find an
>appropriate perspective on metadata and content structures,
perhaps it
will
>end up seeming obvious and almost trivial after
all. One would hope.
>
>I shall now blush and quietly step
away from the pulpit.
>
>Owen, is this what you are looking for from
us?
>
>-- Dennis
[ ... ]
>
>-----Original
Message-----
>From: Owen_Ambur@fws.gov [mailto:Owen_Ambur@fws.gov]
>Sent:
Wednesday, June 20, 2001 14:26
>To: Kathy_Moran@idg.com
>Cc:
ejw@cse.ucsc.edu; ted@xanadu.net;
infonuovo@email.com;
>MORRISR1@LEAVENWORTH.ARMY.MIL;
Michael.Todd@osd.mil;
>lewis.bellardo@nara.gov
>Subject: Re: Federal
Open Source Conference - Records Management &
XML
>
>
>
>Kathy, per your request, here's a quick
first-draft synopsis of the
session
>I'd like to lead ... IF we can
find some folks who are willing and able to
>share some degree of
pertinent knowledge and wisdom with us at the Federal
>open-source
conference:
>
>"Simply finding records for internal business
purposes -- much less
>responding to FOIA requests and subpoenas in
litigation -- is far more
>cumbersome, time-consuming, and costly than it
should be. Moreover, at
the
>root of virtually every scandal as
well as every program management
>inefficiency criticized by GAO is an
ineffective records management
system.
>Indeed, "accountability" was by
far the top benefit of eGovenment cited by
>members of the public in a
poll conducted by Hart-Teeter for the Council
on
>Excellence in
Government. Former FBI Director Louis Freeh has observed
>that the
hype surrounding rapidly evolving technology has diverted
>attention from
the basic need to manage records effectively. ISO 15489
>highlights
the conceptual requirements for records management that are
>applicable to
all organizations, worldwide, and DoD Std. 5015.2 specifies
>the legal and
technical requirements that are applicable by law to all
U.S.
>federal
agencies. Various COTS products have been certified under
the
>5015.2 standard; however, the standard does not provide
for
>interoperability and all of the certified products are proprietary
in
>nature. This session will explore the needs and potentials for:
a) an
>XML-enhanced, open-source alternative certified for Federal (We
the
>People's) records management purposes, and b) internationalization of
the
>5015.2 standard as a logical extension of ISO 15489 to provide
testable
and
>implementable interoperability specifications for records
management on
the
>World Wide Web."
>
>I'm copying a few
folks who might be able to help us determine whether
>sufficient knowledge
and wisdom might be available for presentation at the
>conference so as to
justify scheduling of such a session.
>
[ ... ]
>
>
Kathy_Moran@i
>
dg.com
To:
Owen_Ambur@fws.gov
>
cc:
>
06/20/01
Subject: Re: Federal Open
>Source Conference
-
>
04:04 PM
Records Management &
XML
>
>
>
>
>
>
>
>
>Great;
thank you! Please send me an abstract for the session you
would
like
>to
>lead. There may be a slot for me to fit your
session in. I can't create a
>"Records Management/XML" track at this time,
but let's see if we can fit a
>session into the program and see how it
goes.
>
>Thank you for your interest in our
event!
>Kathy
>
>
>
>Owen_Ambur@fws.gov on
06/18/2001 09:48:19 AM
>
>
>
>
To: ejw@cse.ucsc.edu, Kathy Moran/WORLD
EXPO/IDG@IDG
>
> cc:
>
>
> Subject: Re: Federal
Open Source Conference -
Records
> Management
& XML
>
>
>Jim, FYI in the event there may be
value-additive opportunities to engage
>Ted and the Xanadu people in
pursuit of an open-source RM solution.
>
>Kathy, FYI in the event
there may be opportunities to turn Ted's interest
>into greater
participation in the Fed open source
conference.
>
>Owen
>
>
>----- Forwarded by Owen
Ambur/ARL/R9/FWS/DOI on 06/18/01 09:39 AM
-----
>
>
Ted
Nelson
>
<ted@xanadu.n
To:
Owen_Ambur@fws.gov
>
et>
cc:
ted@xanadu.net,
>marlene@xanadu.net,
>
xanni@xanadu.net,
>eharter@din.or.jp
>
06/18/01
Subject: Re: Federal
Open
>Source
>Conference
-
>
08:48 AM
Records Management & XML
>
>
>Fascinating! Wish I
could go. Best, Ted N.
>
>
>At 12:07 PM 6/5/01 -0400,
you wrote:
>>Ted, Andrew, and Katherine, I'll be surprised if any of
you'd be up for a
>>trip to Washington, D.C., to participate in this
conference. However, I
>do
>>know that Australia has been a
leader in specifying international
>standards
>>for records
management (e.g., ISO 15489) and it would be interesting to
>>know what
contributions the Xanadu People
>><http://xanadu.com.au/people.html> might be able to make to
the cause.
>>
>>Owen Ambur, Co-Chair
[ ... ]
>>
>>Additional
References:
>>http://xml.gov/recordstupid/index.html
>>http://xml.gov/firm/index.html
>>http://users.erols.com/ambur/
>>
>>
>>-----
Forwarded by Owen Ambur/ARL/R9/FWS/DOI on 06/05/01 11:57 AM
-----
>>
>
>>
Owen
Ambur
>
>>
To:
Kathy_Moran@idg.com
>
>>
06/05/01
cc:
Doug
>Newcomb/R4/FWS/DOI@FWS,
>>
11:27 AM
MORRISR1@LEAVENWORTH.ARMY.MIL,
>Ted
>>
Weir/WO/BLM/DOI@BLM,
>Michael.Todd@osd.mil,
>>
cbrock@opic.gov,
>Bette.Fugitt@usda.gov
>>
Subject: Federal Open Source
>Conference
-
>>
Records Management & XML
>>
>>
>>Kathy, this
morning I received a flyer on the Federal Open Source
>>Conference
scheduled for October 2 - 5 at the Reagan Building. I
see
that
>>the program has not yet been posted at <
>>http://www.fedosconference.com/confprogram.asp?> but that
the flyer makes
>>no reference to records management or XML. The
purpose of this message
is
>>to encourage you to seek speakers and
exhibitors who can relate the
>>potential of open-source applications
-- and XML in particular -- to the
>>requirements for records
management by government agencies.
>>
>>For U.S. federal
agencies, the basic requirements are specified in
DoD
>Std.
>>5015.2. Although their products are
"proprietary" in nature, the list of
>>vendors whose products have been
certified under the standard may be a
>>prospect list for
speakers/exhibitors at your conference -- at least to
>>help define the
boundaries between what can reasonably be expected to be
>>open-sourced
and what is likely to remain proprietary for the
foreseeable
>>future. (For example, it seems logical that the
software programs by
>which
>>E-records are created and processed
may always contain proprietary
>>features, and that customers should be
free to use whatever software they
>>choose, but that the *records*
themselves should be in non-proprietary
>>format -- for access by any
XML-enabled application immediately, as well
>as
>>long-term
preservation and access.) See the list of
DoD-certified
vendors
>>and contact information at <http://jitc.fhu.disa.mil/recmgt/#register>.
>>
>>How
the 5015.2 standard might be enhanced through the application of
XML
>>will be the focus of the agenda for the September 19 meeting of
the XML
>WG.
>><http://xml.gov/agenda_20010919.htm> See also the
agenda for our July 18
>>meeting, which features NARA's E-records
Archive project and XTM: <
>>http://xml.gov/agenda_20010718.htm>
>>
>>Carol,
perhaps this might warrant brief discussion at the June 13
FIRM
BOD
>>meeting.
>>
>>Finally, Kathy, if you do
decide to pursue a Records Management/XML track
>>at the open-source
conference, I'll be happy to help get the word out on
>>the XML WG
listserv and I trust that my FIRM BOD colleagues will be
>>supportive
of publicizing it on FIRM's listserv as well.
>>
[ ... ]
>
_________________________________________
Theodor
Holm Nelson
Project Professor, Keio University SFC Campus, Fujisawa,
Japan
Visiting Professor, University of Southampton, England
?
e-mail: ted@xanadu.net ? world-wide fax
1/415/332-0136
? http://www.sfc.keio.ac.jp/~ted/ ?
http://www.xanadu.net
?
Coordinates in USA Tel. 415/ 331-4422
Project Xanadu, 3020 Bridgeway #295, Sausalito CA
94965
_________________________________________