The Minciu Sodas laboratory.  Materials for the draft of MindSet, an exploratory import/export standard for aggregates of thoughts.

Letters regarding MindSet

The materials below are in the public domain.  I invite you to send me your comments at ms@ms.lt.  Andrius Kulikauskas

Structural subspecification "simple as a spreadsheet", 7/18/00
Structural constraints are extensions of equality, 7/30/00
Which is "from", and which is "to"? 9/6/00
Thinking states, or why is movement a stable state? 9/12/00
Style guide for structural links 9/13/00
Exploratory IrDAKiss, Excerpts from Directional Proposal 3/27/00
TM Structure Template; and Request for Association Types, 7/5/00


Structural subspecification "simple as a spreadsheet"

 Tue Jul 18, 2000 4:53pm

              Hi,
                  I would like to gauge what interest there might be in an extremely
              simple format for aggregates of thoughts (a thought might typically be a
              note, but also an image, code, data, file, etc.)  The purpose of the
              format is to facilitate the purposeful import/export of aggregates of
              notes between tools for organizing thoughts, such as www.thebrain.com,
              www.mindmanager.com, http://thoughtstream.org, www.memes.net,
              www.multicentric.com, but also Lotus Notes, Microsoft Outlook, etc.
                  The format is a single six-column table that could be manipulated
              with a spreadsheet program like Excel.  It is a very simplified form of
              TopicMaps, perhaps - a bastardized form, so I am curious for whom else
              it might be useful.
                  Each record in the table is understood to be a thought.  The order
              of the records, in general, does not matter.  The six fields are:  ID,
              FromID, ToID, Intent, Prompt, Content.
                  The purpose of the format is for conceptual modeling, so the fields
              can be defined quite loosely, and well formedness is not an issue.  For
              example, any or all of the fields might be null, including the
              identifiers, and the identifiers need not be unique.  The model helps an
              author and assisting programmer transfer records (thoughts) from one
              software environment to another, by making explicit the following
              constraints:
                  1) The author distinguishes, for each thought, between the prompt
              and the content.  The prompt is what for the author evokes the content,
              but is not the content.  The content is that which holds the attention
              of the author.  Tools for organizing thoughts typically use the prompt
              to facilitate visualization.  [In Topic Maps, a "thought" is a topic, a
              "prompt" is a name, and a "content" is an occurrence.  Here there is the
              restriction that there can be only one name and one occurrence per
              topic.  Also, the occurrence is kept with (or "in") the topic, because
              the author by definition has authority to write.]
                  2) A relationship (or link) between thoughts consists of three
              thoughts: modeling (correctly or not) that our mental state moves from A
              to B by means of C.  Also: every thought "ID" is taken to be a
             relationship from "FromID" to "ToID" where the latter may both be
              possibly null, (in which case we have a node).  This is in the spirit of
              the mathematician Von Neumann, who suggested that instructions are data
              and should not be stored separately within computers.  More importantly,
              a thought that we take to be only a node, may always turn out to be a
              relationship between two other thoughts, and still be a thought.  [Here
              a "relationship (or link)" is an association.  This violates the spirit
              and letter of topic maps, where it is important that associations are
              not topics, so as to distinguish the annotator and the author.  However,
              if an author is looking at their own web of thoughts, it seems strange
              to force such a dichotomy upon them.]
                  3) The intent of the relationship is recorded by a structural link
              type, such as: S = step in a sequence, H = branch in a hierarchy, DN =
              link in a directed network, NN = link in a nondirected network, and so
              on, for all of the structural link types that we are able to find in
              use.  These are simply the local intents of the author, and need not
              hold globally, so that an author may intend to link some of her thoughts
              into a sequence, but she may not be finished, so that there are only
              segments.  Or the links may have inadvertently formed a circle.
              Validity is not an issue (in fact, invalidity is interesting!)  These
              structural link types are the most interesting bits of information
              because they are what we keep in mind when we map from our "mental
              interface" from one environment to another.  [In Topic Maps, the various
              allowed values for intent would be association types, and in general,
              you could abuse this column for your own purposes with your own
              additional association types.]
                  Why such a format?  Because it is the simplest format that addresses
              our problem: we would like to be able to do thoughtful conversion of the
              thoughts we accumulate from one software environment to another.  My
              goal is that the standard be simple, clear, and elegant enough so that
             users and assisting programmers understand the relevant issues, and are
              not confused by irrelevant issues.  We are currently designing a
              converter (in the public domain) between Excel and TheBrain, which means
              that if you could put your data in this tabular form, then you could see
              it in TheBrain.  There is a Software Development Kit for TheBrain which
              gives you much greater control.  But I am curious if you might see uses
              for such converters.
                  We need to have a community of users for our standard to be
              supported with a variety of converters.  We are working in the spirit of
              HTML (rather than SGML), trying to design a framework conceptually clear
              and simple enough that it could be used by anybody who can manipulate a
              spreadsheet, download and run macros, or even tweak a bit of code.  In
              other words, so that it would be as easy (or hard) as writing a home
              page was back in 1995.  The end result would be a ThinkerNet - a
              "slow-motion" kind of connectivity where somebody may go to Africa for
              three months, come back with a floppy disk full of ideas, or mail it to
              their friend.
                  I think this vision has a lot in common with TopicMaps, and in
              particular TopicMaps.Org  The purpose of using XML is to popularize
              TopicMaps.  What about using a spreadsheet?  Could such a format help
              popularize TopicMaps?
                  The format I've described can be written up as a subspecification of
              TopicMaps, a "child-like" version where each topic has only one name and
              one occurrence.  On the other hand, the very same format can, I think,
              capture the full generality of TopicMaps.  This is because we can extend
              the format in many ways.  We can code within the Content many
              occurrences, and within the Prompt many names, and within the Intent
              many types.  We can use markup languages for each of these fields.  We
              can use duplicate IDs to identify two records as being the same record.
              We can extend the number of columns.  We can use the Content as a
              pointer to read-only content located elsewhere.
                  The format is useful because it encourages the design of a network
              of converters that are simple but badly needed (at least I wish I had
              them so I could make use of many different software tools).  I
              appreciate your thoughts on whether such a format, and such converters,
              might be useful to you.

              Andrius Kulikauskas
 


Structural constraints are extensions of equality, 7/30

I have a new idea for how we can identify and define, for practical
purposes, the basic kinds of structural constraints that we rely on to
organize our thoughts.  The idea is that each structural constraint
works by defining, in its own way, what it means for thoughts to be
"similar" or "different".  These structural constraints are crucial for
doing meaningful transfers of aggregates of thoughts from one software
tool to another.

I should first put this in context of what we've done so far.  I started
with the idea of Kestas Augutis (which also Roy Roebuck had) that our
minds rely on three different structures for organizing thoughts:
sequences, hierarchies, networks.  Computers allow us to use these three
structures in a balanced way, so that we can choose these different ways
of organizing our thoughts so as to promote different kinds of
thinking.  In September 1999, I wrote a paper "Developing Import/Export
Standards for Aggregates of Notes" summarizing work at the Minciu Sodas
laboratory on how to define these three structures from a human point of
view.  An important result was that visualizations involve restructuring
one of these three structures (S, H, N) with another, yielding six
visualization types: chronicle (S to H), evolution (H to S), catalog (H
to N), atlas (N to H), canon (S to N), tour (N to S).  This paper is at:

http://www.ms.lt/ms/projects/formatkinds/990917shn.html

In order to come up with a usable import/export format as soon as
possible, in response to the pragmatic perspective of Saulius
Maskeliunas, we then started an investigation, "Linking Locally is
Thinking Globally", sponsored by TheBrain, to see what kinds of
structural links are used by existing software and standards.  A cursory
view has suggested structures such as: Ordered Hierarchy, Unordered
Hierarchy, Radial Hierarchy, and Directed Network and Nondirected
Network.  However, we need to make a more concerted effort to find more
structural link types.  Also, some of these structural constraints
actually require pairs of links, for example, to code an ordered
hierarchy we need to know, given a branch, not only who its parent
branch is, but also which sibling does it come after in the ordering.
We need to have a way of reasoning, whether an ordered hierarchy is just
one link type (ordered hierarchy), or two link types (ordered sequence)
and (hierarchy).

So I am glad to have come up with a new way of thinking about these
structural link types.  The idea is that the purpose of the structure is
to give us a way of thinking about the sense in which two thoughts are
the same.  An example is the method by which I put together results for
our Thoughtful Wishing usage matrix
http://www.ms.lt/thoughtfulwishing.html of tools wished for by people
who enjoy thinking.  I had two or three dozen categories that I wanted
to relate, collapse, refine.  So I placed them on a sheet (using
Microsoft Paint) and worked to arrange them so that related ideas were
next to each other.  This gave me a diagram that I could think about how
to structure and purify.  What was I doing here?  Was the
two-dimensional constraint relevant?  Not really, not at this point.
Most relevant was the concept that categories adjacent to each other
were very closely related, whereas categories far away from each other
were related only by way of intermediary categories.  In other words, a
nondirected network (a network with bidirectional links, which TheBrain
calls "jumps") is what I was really using, even though I did not draw
them.  Thinking about this, I concluded that, in general, this is
arguably the whole purpose of nondirected networks, that thoughts are
equal to the degree that they are adjacent. So thoughts are equal to
themselves, thoughts that are adjacent are almost equal, and the more
thoughts in between, the less they are equal.  A strange, but I think,
very useful way of looking at "equality" - that structural link types
extend the meaning of equality in very different ways.  In other words,
equality is a structural concept, not a semantic concept, and perhaps
the only structural concepts are extensions of equality.

With this in mind, I give a preliminary list of structural constraints
that I know from practice.  This can serve as a "style guide" for using
TheBrain and other tools for thinking.  I invite your critique and,
especially, additions.
 

TREE = UNORDERED HIERARCHY
Sometimes I want to accumulate my thoughts over time regarding a topic
(an idea, category, problem...), so that I could later go over those
thoughts and see how they relate, and what kinds of issues come up, and
sort those thoughts into these issues, making for new subtopics, and so
on.  Here thoughts are different to the extent that they do or may
belong to different topics.

Comment:
Here and elsewhere, what is important is our global "intent" as authors,
not the actual structure, which may end up clashing with our intent.  So
our intent to structure our thoughts into a tree will naturally lead as
well to "forests", and possibly to "multiple hierarchies", or even
network type structures where a topic may be the subtopic of several
different topics.  What matters here is whether (or not) the author
intends the links to form a tree.

Comment:
There is no structural constraint "ordered hierarchy" or "radial
hierarchy".  A radial hierarchy (such as the branches extending from the
root of the tree in MindManager, www.mindmanager.com) involves two very
different kinds of relationships.  One is that thoughts are organized in
a tree, as described above.  Another is that we can think of the
branches as being adjacent to each other, or even to branches within
other branches, depending on how the tree is layed out.  This adjacency
has the same purpose as the nondirected network described above.  So
arranging the thoughts with respect to these two different bases (by
topic, and by adjacency-relatedness) makes for a creative tension, part
of what the MindManager website refers to as using both the "left-brain"
(by topic) and the "right-brain" (by relatedness).  Similarly, I think
that when we have an ordered hierarchy, then we are using both a
hierarchy and a sequence, as described below.

OPEN SEQUENCE
I think there are at least two kinds of sequences.
One I call an open sequence, which is used for listing priorities.  I
call it open because priorities can be reshuffled, and new priorities
can be inserted, and we can still think of it as the same sequence.
Maybe more important is that there is no concept of the end of the
sequence.  Each priority can be thought of as a filter, so that we have
a stack of filters.  We may think of the first priority as a null
filter, which says that we do not have to do anything.  Subsequent
priorities are interesting only to the extent that they are different
from all of the preceding priorities, for otherwise they do not have any
effect, do not require any response.  Thoughts are equal to the extent
that they extend the preceding thoughts in the same way.

CLOSED SEQUENCE
I call a closed sequence of thoughts one which has a definite beginning
and a definite end.  This is used for documenting reasoning, where each
thought follows from the preceding thoughts.  Presumably there must be a
single chain of thoughts connecting the beginning and the end, because
otherwise we would have to keep a separate record of what they are.
This means that there can be no "gaps" between thoughts.  Instead,
thoughts can be refined, over time, into subsequences of thoughts.  As
this happens, the tree structure can be used to identify the
subsequences of narrower thoughts with the broader thoughts that they
broke down.  (This gives rise to a chronicle: a sequence restructured
with a hierarchy).  I imagine that here subsequences of thoughts are
equal to the extent that they are interchangeable.

NONDIRECTED NETWORK
As described above, in a non-directional network (where the links are
bi-directional), thoughts are different to the degree that they are not
adjacent.

Comment:
The nondirected network should, I imagine, be implemented with pairs of
directed links.  There may be - unintended - cases where the link ends
up one-directional.

DIRECTED NETWORK, NO CYCLES ALLOWED
Sometimes thoughts are organized so that there is a link from A to B
whenever B depends on A (or B requires A).  This lets us see which
concepts are more fundamental, for example, the concept of "divorce"
requires the concept of "marriage", but not the other way around, so
"marriage" is more fundamental.  Or more complex molecules require
simpler ones.  Or, as in architect Christopher Alexander's theory of
patterns, certain patterns are layed down first, and they are required
by other patterns that refine them.  We have a lattice in which cycles
are not allowed, or at least, not intended to be allowed.  This
structure is for evolving complexity, and cycles would disrupt the
evolution of this complexity.  We use this structure to figure out the
underlying vocabulary of concepts or words or patterns, from which
"sentences" are generated.  Thoughts are equal to the extent that the
thoughts they require are the same.

DIRECTED NETWORK, CYCLES ALLOWED
Allowing for cycles makes for a qualitatively different kind of
structure, which by which we follow the movement of our attention.  For
example, we may have a network of questions (and possibly answers), each
question leading to other questions, possibly resulting in cycles.
Hyperlinks are another example where our attention moves from thought
(or document) to thought.  We can think of each link as a
transformation, and a cycle as consisting of "energy-conserving"
transformations.  Thoughts are equal to the extent that they occur on
the same cycles, that is, they can be readily transformed into each
other.

Comment:
Given the structure defined above, I doubt that there is any need for a
separate structure for cycles - which are quite rare for organizing
thoughts, and many cases where they do appear (like the water cycle -
"clouds-rivers-ocean", or Socratic questioning) are in the spirit of the
structure above.

THOUGHTS & FINITE STRUCTURAL TEMPLATES
Aside from the structures above, which can grow (like crystals), there
are many finite structures that one can impose on thoughts.  Or we may
simply have isolated thoughts.  In these cases the thoughts are all
different, but equal to the extent that they participate within the
template.

*************************************************
SUMMARY OF WAYS OF EXTENDING EQUALITY
I appreciate your help in identifying more types of structure that you
use to organize your thoughts.  This line of thinking has yielded the
following (which I associate with earlier work on visualizations from
the paper I mentioned):

CLOSED SEQUENCE
for: reasoning
Thoughts are equal to the extent that:
they belong to subsequences that are interchangeable.
visualization: chronicle (sequence restructured with hierarchy)

OPEN SEQUENCE
for: priorities
Thoughts are equal to the extent that:
they extend the preceding thoughts in the same way.
visualization: canon (sequence restructured with network)

TREE, HIERARCHY
for: relevance
Thoughts are equal to the extent that:
they belong to the same, but not different topics.
visualization: catalog (hierarchy restructured with network)

DIRECTED NETWORK, NO CYCLES ALLOWED
for: requirements
Thoughts are equal to the extent that:
the thoughts they require are the same.
visualization: evolution (hierarchy restructured with sequence)

DIRECTED NETWORK, CYCLES ALLOWED
for: attention
Thoughts are equal to the extent that:
they occur on the same cycles.
visualization: tour (network restructured with sequence)

NONDIRECTED NETWORK
for: adjacency, relatedness
Thoughts are equal to the extent that:
they are adjacent to each other.
visualization: atlas (network restructured with hierarchy)
*************************************************

CONCLUSIONS
One conclusion is that the notion of "extending equality" is helpful to
distinguish between what aspects are structural (formal, syntactic...)
and what are semantic.  The idea suggested is that equality (and the
extension of equality to various kinds of similarity) is a purely
structural concept (and the only structural concept).

Another conclusion is that a tool for thinking is successful (as an aid
for thinking) to the extent that it allows us to switch back and forth
between different kinds of structures.  For example, MindManager,
www.mindmanager.com, lets us switch back and forth between thinking
about the TREE structure, and thinking about the NONDIRECTED NETWORK
given by the layout of the branches, which is adjacent to which.
Likewise, TheBrain, www.thebrain.com, is - I believe - helpful as an aid
for thinking, to the extent that we switch back and forth between
thinking about the DIRECTED NETWORK (WITH CYCLES) given by the
parent-child relationship, and the NONDIRECTED NETWORK given by the jump
relationship.

I should temper the above statement to say that both tools can actually
be used to think the other structure as well.  For example, branches in
MindManager can be ordered by priority (OPEN SEQUENCE), or the
parent-child relationship in TheBrain may be thought of as a TREE.
This, however, just emphasizes the point that in doing import/export
between tools, the person performing the transformation should be aware
of the structure that the author of the thoughts had in mind.  Such
transformations are, first of all, a problem of modeling, and only then,
converting.

The associations with the visualizations suggest to me that we will not
find more than the six kinds of structures above.  So I very much
welcome your evidence to show me wrong, or ideas to point me right, both
large and small.  This will make for an elegant and useful modeling
standard for the import/export of aggregates of thoughts between tools
for organizing thoughts.

Andrius Kulikauskas 7/00


Which is "from", and which is "to"? 9/6/00

From: Andrius Kulikauskas <ms@ms.lt>
To: ourownthoughts@egroups.com

Hi,
     I'm working on my draft of the MindSet standard.  It's at:

http://www.ms.lt/ms/projects/formatkinds/000811specification.html

I've completed the Mathematical Formalism.  Now I'm writing up the
Modeling Language.  I've come across three unresolved issues:

1) What way do the "arrows" point?  What is the difference between
"from" and "to"?  This is important because the MindSet standard
implicitly compares different kinds of structural link types, so one
would hope that "from" and "to" have meaning for a step in a sequence, a
branch in a tree, a link in a web.  What is that meaning?
2) How do we deal with links in a "nondirected network"?  Do we use
pairs of links?  Do we use only one link?  If we use only one link,
which node do we list first?  If we use a pair of links, which link
carries the thought, the comment, associated with the nondirected link?
3) What structure explains why we have six link types (Open Sequence,
Closed Sequence, Unordered Hierarchy, Acyclic Network, Directed Network,
Nondirected Network)?  These link types arose from showing how
"extending equality" in various ways yields the different ways that I
know of working with thoughts.  These link types also relate intuitively
with the six visualizations (Canon, Chronicle, Catalog, Evolution, Tour,
Atlas) that are restructurings of sequences, hierarchies, and networks.
So there is reasonable evidence that the six link types form a complete
set.  However, it would be good to explain the six link types in terms
of their uses, respectively (priorities, reasonings, relevance,
requirements, attention, relatednes).  It would be good to have a
structure that would sharpen and explain these last six qualities.

I want to add, that it would also be good to get more of your experience
about how you use structural link types. For example, Saulius listed,
and it would be good to read more about:

> A. Gomez, A. Moreno, J. Pazos, A. Sierra-Alonso
> "Knowledge maps: an essential technique for conceptualisation"
> Data & Knowledge Engineering [Elsevier], vol. 30, pp. 169-190.
It has Table 1 "Some of the most common relations in a problem"
with indicated relation types:
> Equivalence, Taxonomic, Structural, Dependence, Topological,
> Causal, Functional, Chronological, Similarity, Conditional,
> Purpose.

Issue 1): What ways do the arrows point?
A helpful idea here is that we are modeling our experience of thinking.
With that in mind, the arrows are supposed to model our movement from
one thought to another.  In other words, we only deal with the
"relationships" between thoughts as a model of the movement of our minds
from one thought to another, and in particular, to reexperience such
movement.

We relate thoughts with a Directed Network in order to describe how our
attention moves from one thought to another.  So here "from" and "to"
are clear.

We relate thoughts with an Acyclic Network in order to describe
requirements.  For example, molecules may have atoms as components (the
molecules may require the atoms).  Or Christopher Alexander's
architectural patterns A and D may require that there are in place B and
C.  Or the concept of divorce may require the concept of marriage.
Which way is our mind moving?  Intuitively, I think that the purpose of
these structures is to have our mind work towards what is required.  If
our mind is moving from A to B, then as far as requirements are
concerned, it is because A requires B.  Sure, our mind can move from B
to the A that it allows for, and maybe a viewer may do so, and maybe the
"arrows" point that way typically, as in a flow chart.  But I think,
that from an author's point of view, the arrows and the mind go in the
opposite direction.  (You see, these are not trivial matters, but we
will have very profound results if we get it right).  So I claim going
in the author's direction is considering requirements, whereas going in
the opposite direction is to consider something else, for example,
priorities or attention.  It will be important to discuss and debate
these, especially if you have experience in requirements engineering,
that would be great.

In a Nondirected Network we are linking thoughts because they are
related.  How is our mind moving?  It moves in either direction.  So we
should have links in both directions.  Furthermore, there do not exists
"nondirected links" because they cannot model the movement of our mind.
So this answers Issue #2 pretty elegantly, I think.  Note, however, that
it is perfectly acceptable to have the links of a Nondirected Network go
in only one direction.  This is because Nondirected Network (as with the
other link types) is only the author's global intent, and doesn't have
to be enforced, which is to say, can be implemented as the author needs
to, or wants to, or is able to.  In general, A is related to B if and
only if B is related to A.  But the author may have a different view of
what "related" means, that's OK.  The intent is that the ideas are
related.  Actually, in an atlas it might be common to have
one-directional links, where a tiny idea may be related to a giant idea,
but not the other way around (which explains why the tiny idea does not
appear on the global view).  If A is related to B, then our mind moves
from B to the related idea A.  (I take "A is related to B" to mean that
B may perhaps be unrelated to A.)

In an Unordered Hierarchy we are grouping together ideas into a category
because they are relevant to that category.  How does our mind move?
When dealing with relevance, as authors, in creating links, I may say -
this new idea, what category should I put it in?  Or I may say, I would
like to work on this idea, which is a category (a folder) for a bunch of
relevant ideas I have been collecting.  Among those ideas I would like
to create new categories, new folders, to refine my thinking.  That's
where I use the unordered hierarchy for my thinking.  Typically, I'll
take two sibling ideas, A and B, which both belong to C, and then I'll
realize that A is just a particular aspect of B.  So my mind goes from
the aspect A to this new category B, and in this way, as an author, I am
developing my categories.  So I think our mind moves from the fact to
the topic.

I define a Closed Sequence to be one where the beginning and the end are
intended to be fixed and connected.  In a Closed Sequence of reasonings
I think our mind is checking to see how things are developing so we are
moving in the usual order.  From thought A to thought B means that A is
the reason for B, and B is right after A in the sequence.

I define an Open Sequence to be one where there is no intention to fix
an end.  An Open Sequence is for structuring priorities.  When do we
enter a new priority?  How does our mind move?  I think it moves from
the weaker priority to the stronger priority.  If A is stronger than B,
and we have a new priority C, we may say - hey, B is weaker than C, we
should make a note of that.  But C is weaker than A, and we should make
a note of that, too.  Or we may say, true - C is weaker than A, and we
got to focus on A first, but lets write this down so we don't forget
it.  So here I think the mind goes opposite to the expected direction of
the sequence, that is, From thought A to thought B means that A comes
after B, A is a lower priority than B.

So that might handle Issue #1.  What do you think?  In summary, "from A
to B" means:

Attention moves from A to B.
A requires B.
B is related to A.
A is in the category B.
A is the reason for B.
A is lower priority than B.

Note, in particular, that "from" and "to" defined this way let us decide
that, in an unordered hierarchy, "from-to" pointing away from the root,
in an Acyclic Network "from-to" go counter to the "flow chart"
direction, and in an Open Sequence "from-to" lead us to the front of the
sequence, whereas in a Closed Sequence "from-to" lead us to the end of
the sequence.  So it's not trivial!

As for Issue #3, the solution above may yield some new ideas.  As I
mentioned above, there is a nice intuitive relationship between:

The structural link types  (Open Sequence, Closed Sequence, Unordered
Hierarchy, Acyclic Network, Directed Network, Nondirected Network),
which are homogeneous (they involve only one kind of link).  They
involve one kind of link type, which makes them simple to record.

The visualizations (Canon, Chronicle, Catalog, Evolution, Tour, Atlas)
which are heterogeneous because they are restructurings of sequences,
hierarchies, and networks.  That is, they involve two kinds of link
types, and they are the basis for mental pictures.

So how could these two different kinds of structures be related?  That
should give some hints.  In particular, we can note that the reasons for
using the structural link types, discussed above, can be paired as
opposites: "attention" to the back and "priorities" to the front,
"reasonings" forward and "requirements" backward, "relevance" outward
and "relatedness" inward.  These pairs happen to coincide with pairs of
restructurings, as follows:

For attention: Directed Network, visualized as Tour, which is Network
restructured with Sequence, (NS).
For priorities: Open Sequence, visualized as Canon, which is Sequence
restructured with Network, (SN).

For reasonings: Closed Sequence, visualized as Chronicle, which is
Sequence restructured with Hierarchy, (SH).
For requirements: Acyclic Network, visualized as Evolution, which is
Hierarchy restructured with Sequence, (HS).

For relevance: Unordered Hierarchy, visualized as Catalog, which is
Hierarchy restructured with Network, (HN).
For relatedness: Nondirected Network, visualized as Atlas, which is
Network restructured with Hierarchy, (NH).

In each case, the structural link type is of the same nature as the
primary structure in the visualization (whether Sequence, Hierarchy,
Network) except in the case of Acyclic Network, where it may still be
the case that the hope is to create a tree of requirements.  I imagine
that the structural link types are homogeneous structures that arise
because we have the intent to assemble structure, and that intent may
have to be inferred.  (For example, there may be many trees, but the
mental wish is that they all come together into one tree).

I'm still looking for an answer to Issue #3.  Although it's not crucial
for our standard, but it would put it in perspective.  I think they can
relate to the ideas on "Who is my neighbor?" from
minciu_sodas_en@egroups.com, if you've followed that, perhaps:

priorities: (importance goes beyond particular priorities)
relatedness: (responsibility goes beyond particular roles)
relevance: (joy goes beyond particular gifts)
requirement: (challenge goes beyond particular strengths)
reasoning: (belief goes beyond particular knowledge)
attention:  (rights go beyond particular environments)

That would be a good sign, if the link types related to aspects of human
life.  A little outside the scope of the standard :), but not as far as
knowing it hangs together.

I appreciate your thoughts, especially to double check my reasoning as
to how the mind moves in each case.

Andrius Kulikauskas, 9/00


Thinking states, or Why is movement a stable state? 9/12/00

From:  Andrius Kulikauskas <ms@ms.lt>
To:  ourownthoughts@egroups.com

Writing shapes thinking.  Writing up this draft certainly helps identify
the strong and the weak ideas.  I've got this point of view that I like:

What are we modeling?  We are modeling our own thinking.
The goal of our standard is that we may be able to reexperience our
thinking.  So we need to focus on stable states.
A thought is a stable state in our thinking.
Content is the information that sustains this stable state.
Prompt (=name) is the information that evokes this stable state.
Our minds can move from one state to another.

What I want to say now is that:

The cool thing about our minds is that:
Such a movement is also a stable state - a thought.

The model I'm drawing up assumes that there is no difference between
nodes and links.  Every thought is potentially a full-fledged link (from
one thought to another), and every link is potentially a full-fledged
thought (with prompt (=name) and content).  Our mind has this liquidity
by which it does seem to be something like that.

Who can make sense of this?

I wrote a couple of months ago that I think there are two basic
distinctions I would like our modeling language to be able to help us
reexperience.  One is the distinction between focusing on a thought, and
jumping from thought to thought.  Another is the distinction between
being immersed in our thinking, and being reflective on our thinking.

One solution is to skip around the word "thought" and talk about
thinking states.  A thinking state can be interpreted as a stable state
(a thought), or as a movement between stable states (a movement between
thoughts).  Also, a thinking state can be interpreted to be immersed
(the content engrosses us), or reflective (the prompt invites us).  So
when we interpret a thinking state, there are four possibilities,
depending on what we've loaded up in our minds:

Stable Immersed: We are engrossed in a thought.
Dynamic Immersed:  We are swept up by a motion from one thought to
another.
Dynamic Reflective:  We stay suspended in motion by relying on the
prompt for that motion.
Stable Reflective: We keep a safe distance from a prompt that invites
us.

So the ideas is that a thinking state can be experienced - or
interpreted - in any one of the four above ways.  Our standard would not
worry about what thinking state we are experiencing, that's a matter of
our personal freedom, how we want to interpret it.  We are free to
reexperience the thinking state any way we choose.  But, in practice,
our interpretation of thinking state is pressured by the data recorded
(whether Content or Prompt is null, for example) and by the media used.
As in my diagram of the IrDAKiss point of view of the Thoughtful Wishing
matrix http://www.ms.lt/thoughtfulwishing.html#connectivity  I think
media - if we allow it - has us experience our thinking states in the
following ways:

Video: Stable Immersed: We can be shaped.
Phone: Dynamic Immersed: We can be involved.
NotePad: Dynamic Reflective: We can be invited.
Printer: Stable Reflective: We can be included.

We can struggle against media by refusing the interpretation it offers,
I suppose, by blanking out the levers (Content, Prompt, etc.) for
controling us.  For example, we can be in the presence of a television
show (like Survivors) and just be included as a viewer, refuse to let it
shape us, by not getting engrossed, just acknowledging the mechanics of
the names.

Be shaped, involved, invited, included - these are four different ways
that we can be cared about, which came up in work I did at our
minciu_sodas_en@egroups.com discussion group regarding "Who is my
neighbor?" and the objectives of our laboratory.  I suppose here it says
that maybe we want to be careful as to who "cares about us" and how.
Our standard, as shown in the IrDAKiss diagram, will make it more easy
for us to switch media, switch from interpretation to interpretation,
especially for aggregates of thinking states.  For example, instead of
having to watch the content of Survivors, we could look at an equivalent
diagram of names for scenes,etc..

It would be helpful to hear your thoughts about these "thinking states",
and alternative models.  This model I've described here has the
advantages that it's rather simple, I think models important and real
phenomenon, and explains the value, and suggests the soundness, of a
very simple model.  In particular, I like the idea that every "thinking
state" is - at least potentially - both a stable "thought A" with prompt
and content, and a dynamic "movement from one thought B to another
thought C".

Andrius Kulikauskas


Style Guide for Structural Links, 9/13/00

From: Andrius Kulikauskas <ms@ms.lt>
To: ourownthoughts@egroups.com

Hi!  I'm happy to share new findings on the usefulness of various kinds
of structural links.  With the help of your analysis, and more examples,
I would like these to develop into a style guide for using structural
links.  I don't imagine such a style guide as defining a standard, it
would be so hard to write nonsubjectively.  But I think it will be
important both as an explanation for why we have the "reserved link
types" that we do, and what would be the expected use.  I appreciate
your feelings on whether they can be helpful or not.

I looked through the methods in "The Thinker's Toolkit: 14 Powerful
Techniques for Problem Solving" by Morgan D. Jones.  His fourteen
methods are: Causal Flow Diagram, Chronology (Time/Line), Decision/Event
Tree, Devil's Advocacy, Divergent/Convergent Thinking, Hypothesis
Testing, Matrix, Probability Tree, Problem Restatement,
Pros-Cons-and-Fixes, Sorting, Utility Tree, Utility Matrix, Weighted
Ranking.

I wanted to see how they relate structurally to the six ways of working
with thoughts that I've written about: Open Sequence, Closed Sequence,
Unordered Hierarchy, Acyclic Network, Directed Network, Nondirected
Network.

My conclusion is that his methods fit very nicely, and in fact, provide
a lot of evidence for the various uses of each structure, and also, the
direction that the mind moves in establishing each kind of link.  There
is also a seventh kind of structure, which I call Template and basically
is a table of records having fields, so that there are columns of
independent significance that can be analyzed in terms of there
interrelationship.  This structure, though, is not an external
structuring, but an internal structuring of thoughts, and therefore I
think falls outside of the scope of our standard, at least for our first
version.  His tool kit points to the following general uses:

Unordered Hierarchy (Catalog): for channeling convergence.
Nondirected Network (Atlas): for inspiring divergence.
Acyclic Network (Evolution): for appraising scenarios.
Closed Sequence (Chronicle): for selecting context.
Open Sequence (Canon): for ranking judgements.
Directed Network (Tour): for examining legitimacy.
Template: for analyzing independence.

I included above in parentheses the associated visualization types from
Saulius Maskeliunas and my papers. I also include them below where
S=Sequence, H=Hierarchy, N=Network and (S,H)=Chronicle means that a
Chronicle is a Sequence restructured as a Hierarchy.

I also noticed that the uses above resonate with a structure which I
call the "doubts and counterquestions". I write some more about them
at:

http://www.ms.lt/ms/projects/reasonfeatures/991102counterquestions.html

When we have a doubt regarding our experience, so that we can't trust
our experience because we may be brainwashed, the corresponding
counterquestion lets us find our bearings without accepting or rejecting
our experience.  So a counterquestion opens up a fresh way of thinking,
and seems to drive the generation of a corresponding type of
structure.   Also, each structure seems to deal with two of the
following four outlooks on a solution:

Whether it is a solution - is it, or isn't it?
What is a solution - describe it!
How is it a solution - show how we got it.
Why is it a solution - how does it address the problem.

So there is a practical result.  If we want to know, about a solution:

Whether it is, use:
Unordered Hierarchy=Catalog (channel convergence) or
Nondirected Network=Atlas (inspire divergence) or
Directed Network=Tour (examine legitimacy)

What it is, use:
Unordered Hierarchy=Catalog (channel convergence) or
Acyclic Network=Evolution (appraise scenarios) or
Open Sequence=Canon (rank judgements)

How it is, use:
Nondirected Network=Atlas (inspire divergence) or
Acyclic Network=Evolution (appraise scenarios) or
Closed Sequence=Chronicle (select context)

Why it is, use:
Closed Sequence=Chronicle (select context) or
Open Sequence=Canon (rank judgements) or
Directed Network=Tour (examine legitimacy)

I'm very happy that because of Morgan D. Jones book I can rely less on
my own experience.  It's important for me to know how these results
accord with your own experience. What must we add, subtract, or change?
What are the core ideas?  What hangs together, and what does not?

I include below a list of his methods, describing especially their
uses.  In certain cases I have written up his method more than once,
based on how he variously describes and uses it.  He gives a nice
range of examples and exercises.

Yours,

Andrius Kulikauskas
Minciu Sodas
http://www.ms.lt/importexport.html
ms@ms.lt
Public Domain 2000

******************
Doubt raised: Do I truly like this?
Counterquestion: How does it seem to me?
This channels convergence.
Structure Generated: UNORDERED HIERARCHY  (H,N)=Catalog
We learn about a solution: Whether it is, and What it is.
******************

"Convergent Thinking"
Idea A belongs in Cluster B.
Cluster the ideas, select promising ones.  [start with "Divergent
Thinking"]

"Sorting"
Record A has value B.
Select a feature with a range of values.  Group records according to
their values.  Then within each group, sort the records according to the
values of the new feature, and so on.

"Devil's Advocacy"
Position A supports viewpoint B.
Focus on evidence that bolsters an opposing viewpoint, in order to
find out the truth.  Promotes objectivity by supporting all viewpoints
with equal and independent vigor.

******************
Doubt Raised: Do I truly need this?
Counterquestion: What else should I be doing?
This inspires divergence.
Structure Generated: NONDIRECTED NETWORK  (N,H)=Atlas
We learn about a solution: Whether it is, and How it is.
******************

"Divergent Thinking"
Idea A suggests idea B.
Generate a comprehensive selection of alternative ideas.  Come up with
many ideas, build on them, inspire wacky ideas, do not evaluate them.
[Then switch to "Convergent Thinking"].

"Problem Restatement"
Statement A is the basis for restatement B.
Express the same meaning in different words.

"Problem Restatement"
Aspect A focuses newly on problem B.
Take the opposite view of the problem. Take a leap to focus on a
different aspect.

"Hypothesis Testing"
Evidence A is relevant to hypothesis B.  [Note: the more important
part of Hypothesis Testing is considering inconsistencies].
Consider the relevance of evidence to determining the truth of
hypotheses.

******************
Doubt Raised: Is this truly real?
Counterquestion: Would it make any difference?
This appraises scenarios.
Structure generated: ACYCLIC NETWORK  (H,S)=Evolution
We learn about a solution: What it is, and How it is.
******************

Although Morgan D. Jones considers only trees, these structures become
acyclic networks whenever the same state can be arrived at in different
ways, which is especially important in calculating probability.

"Decision/Event Tree"
Determination A makes relevant possibility B.
Structure and distinguish alternative scenarios.  A scenario is a chain
of events.  Display choices and outcomes.  Branches are mutually
exclusive, and collectively exhaustive.  Brings out alternatives,
identifies sequence of determining events.  Helps in making choices.

"Probability Tree"
Condition A makes relevant possibility B.
Calculate the likelihood of a scenario.  See the impact of estimates.
Recognize sources of uncertainty.

"Utility Tree"
Outcome A makes relevant the Potential B.
Rank courses of action according to expected benefit.  Branch to show
options to gamble on, and attach the estimated probabilities and
expected utilities.  Separate the effects of probability and utility,
make both explicit.

******************
Doubt Raised: Is this truly problematic?
Counterquestion: What do I have control over?
This selects context.
Structure Generated: CLOSED SEQUENCE  (S,H)=Chronicle
We learn about a solution: How it is, and Why it is.
******************

Identify the weak link, the crucial link, the missing link, within the
middle of a chain of problems.

"Chronology (Time Line)"
Context A influences Event B.
Sort and organize events chronologically to appreciate the context for
events, and to understand the influence of prior events.

"Problem Restatement"
Reason A is context for problem B.
Continuously reframe a problem to select the expression that most helps
and interests us.  Focus the problem so that it is not too broad or
narrow, not too vague or clear, not driven by assumptions or solutions.
Consider the problem in a wider context.  In particular: start with the
concrete issue, and continuously shift the issue to a higher level by
asking why it is important, until the issue becomes completely general.
Then work with the level that is most crucial.

******************
Doubt Raised: Is this truly reasonable?
Counterquestion: Am I able to consider the question?
This ranks judgements.
Structure Generated: OPEN SEQUENCE  (S,N)=Canon
We learn about a solution: What it is, and Why it is.
******************

"Pros-Cons-and-Fixes"
Disadvantage A addressed by Fix B
Identify strengths and weaknesses of an idea.  Modify the idea to make
up for these deficiencies.  Consider the positive aspects first, then
the negative aspects, and then solutions to the negative aspects, to
find out the negatives that cannot be eliminated.

"Ranking"
Benefit A overshadowed by Benefit B.
Rank in a list according to preference.  Each item on the list should
feel overshadowed by the item of higher preference.

"Pair Ranking"
Benefit A overshadowed by Benefit B.
Compare each pair of items, as to which you prefer.  Make rationale
explicit for each pair.  Rank based on the total outcomes of the
pairings.  Check the ranking by going down the list again (emphasize
neighbors on the list, just as in "Ranking").

"Weighted Ranking"
Criteria A overshadowed by Criteria B.
Benefit A overshadowed by Benefit B.
List criteria.  Pair-rank criteria. Apply same criteria to all items.
Apply criteria with a weight appropriate to its importance.  Compare
each pair of items.  Rank based on the weighted outcomes of the
pairings.  Check the ranking by going down the list again (emphasize
neighbors on the list, just as in "Ranking").
 

******************
Doubt Raised: Is this truly wrong?
Counterquestion: Is this the way things should be?
This examines legitimacy.
Structure generated: DIRECTED NETWORK (N,S)=Tour
We learn about a solution: Whether it is, and Why it is.
******************

The truth of A explains the truth of B. In other words, the truth of A
has a wider context, a wider reality,
than the truth of B.

"Hypothesis Testing"
Evidence A is inconsistent with hypothesis B.
Rank explanations in terms of how inconsistent they are with the
evidence.

"Causal Flow Diagram"
Cause A leads to effect B.
Consider graphically how causal relationships between main components
generate the problem.  Characterize relationships as direct (increase
causes increase, decrease causes decrease) or inverse (increase causes
decrease, decrease causes increase).  Analyze cycles (feedback loops) as
inherently stable or unstable.  Clarify underlying assumptions different
analysts may have.
 

******************
Template
Counterquestion: Am I doing anything about this?
This analyzes independence.
******************

Involves internal structuring rather than external structuring.

"Matrix"
A and B are independent features of C.
Compare and correlate the features.  Identify possibilities, express
constraints.

"Sorting"
Various features are independent (fields).
Identify features.  Isolate and query on desired features.

"Utility Matrix"
Consider options (courses of action) as records having features
(fields).  Consider probabilities as features, and the total benefit as
a feature.
Separate considerations of probability and utility.  Consider entire
scenarios, analyze them from different perspectives, focus on outcomes.
Sum over outcomes, compare overall effects.

Andrius Kulikauskas, 9/00


Exploratory IrDAKiss, Excerpts from Directional Proposal

Submitted to the Infrared Data Association, 3/27/00
Andrius Kulikauskas

1. Introduction

Infrared is one of several contemporary technologies, including the Internet, that have
opened up wonderful cultural possibilities. Some of these have been realized, but
others have not. Authors can choose from a variety of software tools to accumulate,
organize, and reflect on their own thoughts. Such ongoing projects can, in the long run,
be very rewarding, but are typically cut short when software is discontinued and authors
are forced to abandon the thoughts they have tended. Exploratory Irdakiss is a
proposed standard for an infrastructure that establishes the concepts aThought and
aLove. These concepts facilitate experimentation with the transfer of structured
collections of notes and other expressions of thoughts. The accumulation of thoughtful
and personal structurings of ideas and experiences ultimately encourages their social
exchange. Exploratory Irdakiss will carry special significance across an Infrared
connection, which is arguably the most personal way of exchanging electronic
information.

1.1 Exploratory Irdakiss is for authors

Exploratory Irdakiss is proposed with authors in mind. Authors love their piles of notes.
They generate new notes from old notes. Writing down thoughts on a daily basis over
several years can yield a collection of ten thousand notes. Different kinds of writing
promote different kinds of thinking. The discipline of writing in a sequence helps an
author distinguish between strong ideas and weak ideas. Organizing thoughts in a
hierarchy distinguishes between the broad and the narrow. A private language of
personal concepts by thoughts, vague and clear, is easier to write down as a network of
associations. Computers allow authors to organize their thoughts in large sequences,
hierarchies, and networks.

Remarkably, there is no mainstream tool for independent thinkers to accumulate,
organize, and reflect on their thoughts. We live in a reader-driven, as opposed to an
author-driven world. We are presented with single documents rather than aggregates of
thoughts. We are invited to visualize and navigate through seamlessly integrated
media, meticulously edited, formatted, polished. We are supplied with tools to publish
such documents for others. Why are we organizing documents instead of thoughts?
Why is there no streamlined environment for working with masses of our own thoughts,
raw and unedited, juxtaposing old thoughts and creating new ones?

Software tool makers have created dozens of tools for organizing notes. Unfortunately,
notes get trapped when software products are discontinued, as has happened with even
the most popular of products. Competing products do not benefit, and in fact, suffer
because serious users know that the more thoughts they accumulate, the more they
may lose. This vicious circle stifles the market for these tools.

Exploratory Irdakiss creates a virtuous circle by enabling the import and export of
structured aggregates of notes, especially sequences, hierarchies and networks.
Knowledge workers are assured that they their notes will not get trapped in products,
but are also invited to explore working on them with a variety of tools. Use in the
corporate world encourages the development of user interfaces for working with
aggregates of thoughts. We all benefit from tools which invite us to approach our world
as authors.

1.2 Infrared is personal

Exploratory Irdakiss facilitates transfer over any connection, but has special
consequences for IrDA infrared communication protocols. The Infrared universe
accentuates our social universe. Infrared exchange is deliberate - often personally, and
negotiated - often socially. We are encouraged as authors to participate at each step,
with each exchange.

The Infrared world is populated by more than one hundred million devices with a
wonderful range of functionality, processing power, memory, bandwidth, user interface,
and output. Why are so many of them silent? The impact of HTML on the Internet
suggests that a technological standard is awakened by a conceptual standard.
Exploratory Irdakiss aspires to awaken a Sleeping Beauty. Exploratory Irdakiss
emphasizes that Infrared exchange is an act both personal and social, a kiss - an
irdakiss - with a wide range of meaning just as a kiss may. Such exchanges are
special, not that frequent, and so should pack a lot of meaning. They can be tedious if
we only send items singly. They can be thrilling when we send thoughtfully assembled
aggregates, which the receiver may examine at their leisure, and then transfer to a wide
variety of devices for the sake of viewing, changing, reassembling and exchanging.
Exploratory Irdakiss is a standard that facilitates experimentation with structuring and
exchanging aggregates of aThoughts, which are ideas and experiences. Exploratory
Irdakiss establishes a conceptual infrastructure for bundling together such aThoughts
into aLoves, which may be sequences, trees and webs of attachments. An aLove may
be put together in the spirit of a “care package”, which may bundle together images,
notes, voice, cash, and any kind of file imaginable. This may be delivered in many
ways, but an Infrared exchange - an irdakiss - is the most personal way.

2. Underlying concepts for Exploratory Irdakiss

Exploratory Irdakiss is, first and foremost, a conceptual standard, defining what is an
aThought, and what is an aLove of aThoughts. Exploratory Irdakiss is an intermediate
conceptual format which can be used with existing software tools for organizing
thoughts. This conceptual format is a structurally neutral means for representing
aggregates of information structured by any and all software tools and standards. An
author needs to agree to any structural interpretations that will be made regarding how
they have been structuring their notes. Exploratory Irdakiss is designed to be
conceptually clear both to the author affirming a structural interpretation, and to the
developer of a converter which maps the file format of a software tool onto an
implementation of the conceptual standard.

Several standards currently exist that can be used for the import and export of
sequences, hierarchies and networks of notes. Exploratory Irdakiss differs from them in
the conceptual clarity with which it approaches the authorship of notes, purposefully
imposing conceptual constraints and avoiding conceptual pitfalls.

2.1 Avoiding conceptual pitfalls

Exploratory Irdakiss differs from other standards in that it more faithfully represents the
structural relationships between capsules of information. In doing so, it avoids several
conceptual pitfalls for authors.

2.1.1 Conceptual pitfall: Not emphasizing the boundary of a thought

In order to assemble aggregates of thought, an author needs to be clear as to the
boundary of a thought, and especially the difference between internal structuring within
a thought, and external structuring among thoughts. When we are structuring thoughts
with regard to each other, then the boundary between what is inside and outside of a
thought is the only one that matters. Many standards, such as HTML, muddle this by
introducing the notion of a document, which in general has nothing to do with a thought.

On the other hand, standards such as XML often emphasize expressing the internal
structuring within a thought, which is semantically very rich, but overwhelmingly
complicated. This is typical with markup languages. There is a tendency to think of
every marked up item as a thought in its own right. Markup is handy for computers that
want to act on a well defined lexicon, but is primarily a nuisance for authors who should
not have to care about what they mean. An author working independently with their own
thoughts has no use for documents or markup, which only muddle their authority as an
author as to what they take to be a thought.

Documents and formatting are relevant when we want to present material only for the
sake of other people. Markup and semantics are relevant when we want to present
material to an information system. But they are all contrived and in the way when we
are working independently with our own thoughts, or inviting others to share our outlook
on life. Exploratory Irdakiss focuses exclusively on the external structuring among
thoughts and therefore imposes on authors the requirement that they be explicit as to
what is a thought. Exploratory Irdakiss places no requirements regarding the content of
a thought, and leaves this open for other standards.

2.1.2 Conceptual pitfall: Not preserving what gives thoughts their value

It is a misconception that an author treasures their thoughts and may therefore have the
desire to organize their thoughts very neatly. In fact, we all have more ideas than we
know what to do with, and we simply do not have the time or the need to be formatting,
editing, indexing or even reading most of them. Instead, our notes have value as an
organic heap which helps us generate new ideas, and which as we structure, brings out
the most important ideas. Ideas are valuable through their relationships with each
other. It is therefore crucial that import and export of aggregates of notes focus on the
structural relationships between thoughts. An author would rather have isolated ideas
lost - they can be reconstructed or replaced as needed - than have their entire system
structurally mangled beyond recognition. Authors always find time to create new
thoughts, but dread the drudgery of reformatting isolated thoughts, each of which by
itself is of questionable value. Exploratory Irdakiss is designed to respect a system of
thoughts in its entirety.

2.1.3 Conceptual pitfall: Preserving what the environment intended

We want to encourage the creative and unexpected use of software tools. We are
therefore concerned only with the structural constraints that they impose. Semantic
interpretation is left entirely to the user, and may not agree with the intended use of the
software tool. For example, someone working on a genealogical tree might choose to
import their thoughts into a graphic editor for components of a product. In doing so,
they will necessarily be subject to the structural constraints of their chosen software
tool. Exploratory Irdakiss does not place any additional semantic constraints that would
discourage them from using a tool as they see fit.

2.1.4 Conceptual pitfall: Preserving what we see rather than what we think

An author chooses to work within an environment as a way of affecting their thinking.
The structure they are thinking is often different than the one that they are actually
seeing. A list may look ordered, but an author may take it to be unordered. In fact,
unordered lists are typically laid out in some order. What is relevant for an author is that
import and export preserve the conceptual environment. The transfer of irrelevant
structural information can distort and cloud thinking.

HTML, XML, UML and TopicMaps are standards that are all able to express one or
more kinds of sequences, hierarchies and networks. None of them do this in a
structurally neutral way, but rather, each favors a certain visualization. In general,
visualizations favor one structuring over another. For example, XML list elements in
sequence, then groups them in a hierarchy, and then relates them in a network of links.
Exploratory Irdakiss handles sequences, hierarchies and networks in a uniform
structurally neutral way.

2.2 Imposing conceptual constraints

Exploratory Irdakiss lets authors know that if they follow certain constraints, either
explicitly or implicitly, and if converters are available, then they will be able to transfer
their thoughts along with their structural relationships from their chosen file format to an
implementation of the Exploratory Irdakiss conceptual format. These constraints have
the author affirm the representation of each thought as an aThought consisting of eight
components:
· Content
· PromptToThinkTheContent
· PromptToGuessTheKey
· PromptToReadTheCode
· IDforTheContent
· IDforTheKey
· IDforTheCode
· TypeOfStructuring
The descriptions below only serve to clarify the nature of the conceptual standard, and
are not a specification of an implementation.

2.2.1 Conceptual constraint: Identify every thought

Exploratory Irdakiss requires that the author make explicit each thought, identify it with
some capsule of information. An aThought is, conceptually, any such explicit expression
of "a thought" that an author construes. An aThought is intended to be self-contained,
inasmuch as it is to have some permanence independent of us. The author decides
what is a thought. Generally this means that a thought has an internal structure which
may be nontrivial but the author choose not to make the components explicit. For
example, an author working with a word processor might consider each paragraph to be
an aThought, although they may have an aThought be a word or two, a sentence, an
entire document, or even null.

2.2.2 Conceptual constraint: Distinguish content and prompt

Exploratory Irdakiss requires that the author of a thought divide the information within a
thought between the content and the prompt, either of which may be null. As authors
we create thoughts for ourselves so that we might think them again. A prompt is what
allows us as authors to return to our thought, and a content is what we are returning to.
An author uses a prompt for many of the same reasons that a computer uses an ID. A
prompt is a sign that evokes our thought, and so should attract attention to itself only to
pass it on to the content. A name, label, or icon may serve as a prompt, and they help
make visualizations more compact. A content is typically a note, or may be a
document or file. A content may have complicated internal structuring that draws the
author deeper.

2.2.3 Conceptual constraint: Understand relationships to involve three thoughts

Exploratory Irdakiss requires that each thought - each aThought - be considered to be a
relationship between two other thoughts, either of which may be null. Moreover, these
are the only kinds of relationships allowed. In this way, everything is reduced to
aThoughts. A set of aThoughts is defined to be an aLove, which is to be understood as
a system of interrelated thoughts for which there is a caring author.

An aThought which relates one thought to another, may correspond to a link in a
network of thoughts, or a branch in a hierarchy of thoughts, or a step in a sequence of
thoughts. Exploratory Irdakiss does not require but offers an additional convention for
observing a distinction between the first thought and the second thought, instead of
relying on the direction of an arrow, which may be arbitrary. The two thoughts should
be ordered so that the attention of the author is understood to move from the first
thought to the second thought. The first thought is understood to provide a key for the
partial decoding of the second thought, and the meaning unraveled is identified with the
content of the aThought. For example, the first thought, the Key may be thought of as a
file type, or a table structure, or a question, and the second thought, the Code, may be
thought of as a file, or a record in a table, or an answer. Exploratory Irdakiss includes
such an optional convention that can be maintained so that the distinction between the
first and second thought need not rely, for example, on graphical features which may be
arbitrary.

An aThought is meant to be manipulable by the author, but also by an information
system. Therefore an aThought includes prompts and Ids for all three thoughts. The
prompts are included so that the author might recollect, from the aThought, what the
three thoughts are and how they relate. The Ids are included so that the information
system might manipulate the thoughts. Ids are intended to be unique, but this need not
be enforced, for although the aThought must be well formed, the aLove is nothing more
than a set of aThoughts, and there may be inconsistencies among the aThoughts. Also,
changes in thoughts are not necessarily propagated, nor should they always be. The
author dictates the meaning, and the prompts recorded within the aThought, along with
the content, can help the author recollect what they had in mind by relating two
thoughts.

2.2.4 Conceptual constraint: Indicate the structuring type for each relationship

Exploratory Irdakiss requires that the author identify for each aThought a structuring
type. This will often be done by the converter from the chosen software tool. This is
especially important when the aThought is a nontrivial relationship between thoughts.
An author relates thoughts in the context of a global structural constraint of their choice,
either explicitly, themselves consciously, or implicitly, by using a particular software tool,
to impose a particular discipline on the relationships between the thoughts and thereby
affect the development of these thoughts in a certain way. Exploratory Irdakiss assigns
structuring types to each known structural constraint, for example:

· N: Node.
· HU: Hierarchy, unordered. The link is a branch from a tree, and it is not ordered with
respect to its sibling branches.
· HO: Hierarchy, ordered. The link is a branch from a tree, and it is ordered
sequentially with respect to its sibling branches.
· HR: Hierarchy, radial. The link is a branch from a tree, and it is ordered cyclically
with respect to its sibling branches.
· ND: Network, directed. The link is from a network, and is directed.
· NN: Network, nondirected. The link is from a network, and is nondirectional.
· S: Sequence. The link is a step from a sequence.

An author fosters a set of aThoughts as a system, known as an aLove, which aside
from nodes, typically makes use of several additional structuring types, each of which
may be associated with a subsystem of aThoughts. Each subsystem poses its own
conversion problem. The structuring type is the most important feature of the
Exploratory Irdakiss, and is the main purpose for having the standard, because it
respects the discipline that an author has been imposing on themselves while working
with large aggregates.

The structuring type simply records the self-discipline the author has applied, and need
not actually obey the structural constraint, so that structuring types within an aLove may
be invalid, inconsistent, and self-contradictory. They simply record the author’s
intentions. For example, an author may have intended to create steps in a sequence,
but they may have turned out to form a circle.

The following diagram depicts three different aLoves.
[...]

3. Overview of Exploratory Irdakiss

Exploratory Irdakiss defines a minimum set of structural constraints to facilitate
experimentation with the transfer of aggregates of thoughts.

The purpose of the first version of Exploratory Irdakiss is to enable the design and use
of converters for transfer between:
· file formats for software tools for organizing thoughts, as well as various standards
for organizing information
· standard implementations of a conceptual format for an aLove of aThoughts

This version serves its useful purpose without requiring any change in existing software
applications, although vendors may incorporate converters directly into their products.
This version also makes it feasible to design interactive multiconverters for guiding
authors through modifications in structuring type that might allow them to transfer their
aLove to a structurally incompatible software tool.


TM Structure Template; and Request for Association Types

Two requests, which I write more about below:
              ===> I would be very interested to know if anybody has done research
              into the kinds of global structural constraints that are implicit within
              association types (for example, that the associations of the given type
              are expected to form a sequence, or an unordered tree, or a directed
              network, etc.).
              ===> Please let me know of any lists of association types that I might
              study.
 

              I am looking forward to participating in the XTM group.  I am not sure
              yet if I will be able to attend in Montreal, but would like to.

              Our laboratory, Minciu Sodas, www.ms.lt, is focusing on the development
              of a standard that will allow individuals to import/export aggregates of
              notes from software tools for organizing them, such as TheBrain,
              www.thebrain.com, MindManager, www.mindmanager.com, Multicentrix,
              www.multicentric.com, Thoughtstream, http://thoughtstream.org, Lucid,
              www.memes.net  Our laboratory's emphasis is on the needs of the users of
              such tools, so that they can work with their accumulated thoughts
              regardless of environment, or at least be able to switch every few
              months or so.

              We have been working through the Infrared Data Association, but the IrDA
              leadership has decided that we fall outside of their focus and that we
              end our working group.  Our work so far is at:
              http://www.ms.lt/importexport.html

              I have to learn more about TopicMaps, but right now I think that they
              are adequate for expressing our standard.  My only reservation so far is
              that I think it would be simpler for us if associations were topics.
              But I presume there are work-arounds for this.  I will be focusing on
              our particular needs, but I hope this may offer insights into how a
              segment of the population, especially individuals, may use topic maps.

              One thing that I have learned is that users of "thought organizers" need
              is not so much an import/export format, but rather, a modeling
              language.  The modeling language is a construct for understanding the
              consequences of transforming (to greater or lesser satisfaction)
              structured aggregates from one tool (user interface) to another tool
              (user interface).  The modeling language should be expressed in more
              than one format, for example, XML, CORBA, Excel...  If it is expressed
              in only one format, then the format will be confused with the modeling
              language, which conceptually can be disastrous, confusing both the user
              and any assisting programmer.

              I feel more comfortable with Topic Maps upon realizing that it, too, is
              really both a modeling language (a way of structuring how we think about
              our information) and a format (HyTime, XTM, ...)  I think it is crucial
              to separate the two aspects, and I get the impression that this is being
              done.  It may be very convenient for us to describe our modeling
              language as a subset of, or template within, the Topic Maps "modeling
              language".

              I would like to organize the creation of a Topic Maps Template that
              would serve our needs, but might also have additional value.  My main
              goal is to identify, and make available, the dozen or so "structural"
              association types that occur in practice.  By structural association
              type (perhaps you can suggest a better term - syntactic?), I mean that
              when I make a structural link, it may presume some global structural
              constraints.  For example, an X "IS A PART OF" Y association may presume
              to be structurally part of a forest of such associations, where the
              branches are unordered.  An X "IS IN" Y association may presume this
              very same structural environment.  In our case, semantically, we are not
              interested in the difference, at least not for our purposes, which is to
              make sure that these structural constraints are made clear to the
              author, and if desired, can be preserved.  In other words, there may be
              thousands of types of semantic associations, but if we are only
              interested in the structural implications, there may be as few as one or
              two dozen structural associations to worry about (different kinds of
              sequences, hierarchies, networks, etc.).  Users of tools for organizing
              thoughts care about the structural associations, rather than the
              semantic associations, because:
              - Users use the structural environment the tool offers, that is, the
              structural constraints it imposes, to shape their writing and thinking.
              - Users do not have the time, energy or interest to mark up or otherwise
              tend to their robustly multiplying thoughts.
              - The ideas most worth writing down are typically the ones that we have
              the least ability to express what we mean.  And it is very disruptive to
              have to work with artificial categories which can destroy that meaning.
              - Users want to have the right to abuse tools, that is, use them for
              strange purposes that their creators may not have semantically intended.

              I will be making a list of such structural link types.
              ===> I would be very interested to know if anybody has done such
              research already.  Although my impression is that this has never been
              done, I suppose it is considered too "simple".  However, I think that
              before delving into the complications of semantic relationships it would
              be a helpful first step to identify the kinds of structural
              relationships that arise, which seem quite limited.  This is especially
              helpful for the kind of import/export of aggregates of thoughts that we
              would like to enable.

              If the research has not yet been done, then I would like to do it very
              simply.
              ===> Please let me know of any lists of association types that I might
              study.  I will go through these lists and try to figure out the
              structural constraints that come up.  I will be glad to share my results
              in the public domain.

              I will write more about the Topic Maps Template I have in mind.

              Yours,

              Andrius Kulikauskas
Wed Jul 5, 2000 4:17pm