Knowledge preservation and PLM: a cultural perspective Bruno - - PowerPoint PPT Presentation

knowledge preservation and plm a cultural perspective
SMART_READER_LITE
LIVE PREVIEW

Knowledge preservation and PLM: a cultural perspective Bruno - - PowerPoint PPT Presentation

Knowledge preservation and PLM: a cultural perspective Bruno Bachimont Universit de Technologie de Universit de Technologie de Compigne Prologue Prologue PLM deals with information related to product life : The issue is to collect


slide-1
SLIDE 1

Knowledge preservation and PLM: a cultural perspective

Bruno Bachimont Université de Technologie de Université de Technologie de Compiègne

slide-2
SLIDE 2

Prologue Prologue

  • PLM deals with information related to product life :

– The issue is to collect knowledge to manage a temporal process. – The product is in time. Information helps managing this property.

  • Bad news :

Bad news : – Information is also in time and has a lifecycle.

1. A piece of knowledge elaborated and mastered by some specialists. 2 A piece of knowledge shared by society or designated communities 2. A piece of knowledge shared by society or designated communities 3. A forgotten piece of knowledge, whose medium is corrupted and content unintelligible.

  • The issue is knowledge preservation

– Digital information is much more fragile than ordinary one.

  • Digital information requires new approaches

PLM should deal with the digital nightmare of knowledge preservation : welcome in hell !

slide-3
SLIDE 3

Contents Contents

  • Memory gaps

y g p

– From knowledge management to memory management

  • Memory Conflicts
  • Memory Conflicts

– Two models : the static model, the dynamic model.

  • Digital memory

Digital memory

– Currently a deadlock, the dynamic model helps dealing with it.

M t

  • Memory management

– Need for new tools and approaches :

  • Criticizing contents, maintaining interpretation.

g , g p

slide-4
SLIDE 4

MEMORY GAPS MEMORY GAPS THE WILD LIFE OF KNOWLEDGE

slide-5
SLIDE 5

An usual perspective on KM An usual perspective on KM

  • KM deals with knowledge formats : from tacit to

li it d l explicit, and conversely.

  • No attention paid to time, memory and

intelligibility. intelligibility.

slide-6
SLIDE 6

Three challenges Three challenges

content Perceptible medium Trace Interpretation Technological

Physical corruption

Coded resource Published view Technological Medium

Physical corruption

Obsolescence gap Technical readability

p

Intelligibility gap Cultural readability Intelligibility gap Cultural readability

6

slide-7
SLIDE 7

Two complementary problematics Two complementary problematics

Knowledge and hypomnemata

Knowledge and memory

Knowledge and hypomnemata

  • Knowledge is not (only) cognition.
  • Knowledge as such should be

materialized in people mind, social

Knowledge and memory

  • Knowledge management is

a problem of :

p p , codes and correlated with material artefacts (documents, books, tools, etc.) : the hypomnemata

  • Knowledge is not a material object but

– Memory management – Knowledge preservation

  • Knowing is something like

g j the interpretative process enabled by the hypomnemata.

  • Knowledge relies on :

– the preservation of the material

g g remembering (Plato). What does that mean to be

the preservation of the material counterpart of the hypomnemata – the interpretative capacity of understanding them.

What are the relevant artefacts for knowing

What does that mean to be able to remember/know something ?

something ? How to preserve the interpretative capacity ?

slide-8
SLIDE 8

ADOPTING A KNOWLEDGE / ADOPTING A KNOWLEDGE / MEMORY UNDERSTANDING

slide-9
SLIDE 9

Two approaches Two approaches

  • “classical” approach

– One has memories:

  • Objective traces (documents, vestiges, indices): enables History;
  • Subjective traces (memories) : enables remembrance.

– If one keeps memories intact, one gets a trusted memory.

  • The issue is the preservation of intact items.

– Remembrance relies on intact memories

h h

  • An other approach

– Memories cannot remain intact:

  • Physical Corruption of medium
  • Unintelligibility and unreadability if contents remain unaccessed

– Through remembering as a process memories are reactivated and technically and intellectually actualized. Memories rely on remembrance – Memories rely on remembrance.

9

slide-10
SLIDE 10

Etruscans: intact memories, no b remembrance

10

slide-11
SLIDE 11

Aristotle: no memories, full b remembrance

  • Why is it still possible to read Aristotle

t d ? today?

  • No original traces: the oldest come from the IXth

century AD (Aristotle died in 322 BC).

  • Many copies done because Aristotle has always

been read and commented been read and commented.

  • Each succeeding generation has made his works

intelligible for its time, according to its own interests.

  • The preservation of the works of Aristotle

p is the process of explaining what these works mean using language and concepts intelligible by today’s public.

  • Preservation thus results from the
  • Preservation thus results from the

attempts of scholars to understand and interpret Aristotle

slide-12
SLIDE 12

Re‐inventing the past Re inventing the past

U d t di Understanding Reading

Readings reduce the

Intelligibility gap

Readings reduce the cultural distance between an archived

  • bject and its

interpretation

Reading Archive item

interpretation

Reading Archive item Preserving integrity Preserving readability g g y g y

slide-13
SLIDE 13

A success story : music A success story : music

  • Scores

– Keeping a user manual telling how to reproduce a given piece of music;

  • Organology:

– Coupled with scores, organology is an artisanal tradition preserving an artisanal tradition preserving and transmitting instrument elaboration ;

  • Music School / Conservatory

Music School / Conservatory

– Preserving know‐how by a perpetual practice merging generations. g

slide-14
SLIDE 14

First conclusions First conclusions

  • Our experience from the long term history
  • Our experience from the long term history

learns us that:

Contents are never completely preserved – Contents are never completely preserved

  • Libraries always burn one day…

We are nevertheless able to remember because – We are nevertheless able to remember, because remembrance is an invention and interpretation process based on critical tools and methods. p

  • Memory is a dynamic process inventing its
  • bjects and faithfully using them
  • bjects and faithfully using them.
slide-15
SLIDE 15

ESCAPING FROM THE DIGITAL ESCAPING FROM THE DIGITAL NIGHTMARE

slide-16
SLIDE 16

The original paradise before the Fall The original paradise before the Fall

Universal: every content may be digitized H thi b Copy : bitwise copy: perfect and and eternal Ubiquity: multiple access to the 0010111001110110001100100010010001111101000101 Homogneous: everything may be digitally processed same resource 0010111001110110001100100010010001111101000101

slide-17
SLIDE 17

But… But…

In fact…. 0010111001110110001100100010010001111101000101 0010111001110110001100100010010001111101000101

slide-18
SLIDE 18

The digital damnation: the four horsemen of the digital Apocalypse horsemen of the digital Apocalypse

Platforms: Windows, Linux, MacOS, etc. Formats (metadata) : XML, LaTeX, mpeg‐7, mxf, rdf, TEI,… Environments : Word, WMP, Realplayer, VLC, EMACS, VI….

0010111001110110001100100010010001111101000101

Formats (codes) : unicode, ascii, iso‐latin1, mpeg, jpeg,tiff,aiff, pdf…

18

slide-19
SLIDE 19

The digital schizophrenia The digital schizophrenia

  • From the theoretical stance:

– Everything should be digitized to be preserved – Digital memories are preserved intact Th i i t i thi i di it l – The main issue : turning everything in an digital counterpart and keep it intact.

  • From the practical stance:

p

– Digital contents should evolved to remain accessible:

  • Format obsolescence
  • Medium decay
  • Medium decay.

– The main issue : transforming contents to keep them alive without loosing their authenticity / fidelity and identity identity.

slide-20
SLIDE 20

Why such a schizophrenia? Why such a schizophrenia?

  • We do not want to abandon the theoretical

t b it t it i th l stance because it seems to us it is the only way to preserve knowledge :

we fear to loose our memories and hence our – we fear to loose our memories, and hence our capacity of remembrance.

  • But:

But:

– According to the dynamic model of memory, digital contents are nothing particular; g g p ; – One should remember (!) how to do from our past experience of knowledge preservation.

slide-21
SLIDE 21

TURNING THE “OTHER” MEMORY MODEL INTO METHODOLOGICAL PRINCIPLES PRINCIPLES

slide-22
SLIDE 22

Two problems Two problems

Dealing with contents Dealing with interpretations Dealing with contents

  • Contents should evolve and

cannot remain identical

– Physical and technical

Dealing with interpretations

  • Interpretation is a cognitive

and social process;

  • Interpretation should be

y transformations

  • Contents should be preserved

through their use

– Interpretations to keep them alive

  • Interpretation should be

preserved according to a social need and a given community

  • Question:

Interpretations to keep them alive and intelligible

  • Question:

– How to transform content and exploit it on a respectful and

Q

– How to ensure the possibility of interpretation ?

  • Answer :

i i l d

exploit it on a respectful and authentic manner?

  • Answer:

– Some old disciplines to consider with a closer look.

– Principles to manage and

  • rganize the memory process:

OAIS and its designated community.

with a closer look.

slide-23
SLIDE 23

Contents

Resources are unaccessible in themselves: a mediation

<course title="…"> <definition> … </definition>

is necessary How to assess that the published view is fair and i h ?

</definition> </course>

right ? How to compare and sort the different t ti ?

?

reconstructions?

?

?

slide-24
SLIDE 24

The problem of a reference bl h d establishing content identity

A R

?

Anonymous Resources M Vi Many Views Putative Canonical Canonical reference document

slide-25
SLIDE 25

An old problem! An old problem!

slide-26
SLIDE 26

Manuscripts studies Manuscripts studies

?

(original)

slide-27
SLIDE 27

Token Identity A unique material

  • bject

Digital Codicology

… …

Formal Identity

A given formal presentation

Digital Diplomatics

Content Identity Content that should be invariant

Digital philology

Semantic Commentaries

Digital Hermeneutics

Semantic Identity and content explications

27

Digital Hermeneutics

slide-28
SLIDE 28

Strategies Strategies

Focused on materials Focused on meaning Focused on materials

  • Museal approach

– Keeping operational the whole set‐ up: contents, devices ;

Focused on meaning

  • Descriptive approach

– Scoring : Contents are too complex to be preserved,

up: contents, devices ;

  • Emulation approach

– Keeping contents intact ; emulating

  • bsolete rendering devices on

t t

complex to be preserved, emulated or migrated. A description is made and kept in

  • rder to re‐invent content with

the future current technologies

contemporary systems

  • Migration approach

– Migrating contents from obsolete formats to contemporary ones in

g – Modeling : contents or objects are modeled to extract the very

  • bjects meaning and express it

in a consistent and faithful way.

formats to contemporary ones in

  • rder to use the current rendering

devices (proposed by the market)

y

  • But :
  • producing new descriptions /

models that are documents

  • N

d f th i ti

  • Needs for their preservation.
slide-29
SLIDE 29

Interpretations Interpretations

concept

This is not a concept This is not a concept (piece of knowledge)

This is not a pipe p p

slide-30
SLIDE 30

An essential tension An essential tension

Formalised language “natural” language Formalised language

  • Syntax is set up to control

semantic interpretation:

natural language

  • Semantics depends on

context

– No ambiguity – Formal calculus – Formal reasoning

  • Context cannot be fully

explicit bl

  • Expressions are not contextual
  • Problems :

R d d i it

  • Problems :

– Rich expressivity – Semantics should be

– Reduced expressivity – Need for a convention to interpret syntactical expressions

negotiated according to context, actors, situations.

expressions.

slide-31
SLIDE 31

A necessary tradeoff A necessary tradeoff

Freezing the context: from context to formalization Making sense of the formal: from formalization to context context to formalization

  • Formalization as a means for

reducing ambiguity, variability d h i i t bilit

from formalization to context

  • Putting the formalization in

context to apply formal t t l it and enhancing interoperability. constructs on real items.

forall x, H(x) ‐> A(x) forall x, H(x) ‐> A(x) H, A: predicates, whose reference are defined through sets in a H A : interpret them in the effective g reference model, based on set theory: how to relate these sets to effective reality? H, A : interpret them in the effective contexts, according to actors’ knowledge.

slide-32
SLIDE 32

Context cannot be eliminated Context cannot be eliminated

Formalizing knowledge and context Tacit knowledge inherent to context and situations and context to context and situations The borderline can be deplaced The borderline can be deplaced, never suppressed.

slide-33
SLIDE 33

A traditional approach A traditional approach

  • Historical examples :

– 1537 : royal library of France (édit de Montpellier) 1530 llè d l – 1530 : collège des lecteurs royaux (former « Collège de France »)

  • Lesson :
  • Lesson :

– On one side, you keep materials – On an other side you

Guillaume Budé 1467 ‐ 1540

– On an other side, you preserve the reader, and maintain a continuous tradition of reading

François 1er 1494 ‐ 1547

slide-34
SLIDE 34

Relying on context and current state of k l d l f tacit knowledge : example of OAIS

  • Designated Community:

g y

– An identified group of potential Consumers who should be able to understand a particular set of information.

P

Preservation Planning Descriptive Info.

R O D U Designated Community

queries result sets

Ingest Access Data Management Descriptive Info.

  • rders

SIP AIP AIP DIP

U C E R Community

Archival Storage Administration

MANAGEMENT

slide-35
SLIDE 35

Conclusions Conclusions

  • PLM needs KLM
  • KLM relies on contents and interpretations:

– Preserving contents and their authenticity/identity

  • Conceptual and critical organization of knowledge

– Preserving interpretation by managing / monitoring designated communities

  • Social organization of knowledge
  • KLM needs tools to manage hypomnemata

(ontologies document engineering metadata (ontologies, document engineering, metadata, etc.), and methods to manage communities (OAIS, CSCW, etc.).

slide-36
SLIDE 36

Thanking audience for being here and quiet Thanking audience for being here and quiet

Thank you for your attention!

Thank you for your attention!

Time flies….

Thank you for your attention!

This was the excipit Usual formula in XXIst century in colloquium