User:Jon Awbrey/Symbol Grounding Problem • Peaceful Easy Feeling

From InterSciWiki
Jump to: navigation, search

Peaceful Easy Feeling • Inquiry List

PEF. Note 1


JR = Joe Ransdell

JR: I wonder if someone would attempt a brief statement
    of the symbol-grounding problem?  Something like that
    is at issue here, I think, but I don't recall how that
    problem is usually conceived.

One for the "Standing Under The Soles Of Giants" (SUTSOG) file.
I gave up on this literature a decade ago because its insular
perspective seemed to prevent folks from ever getting around
to reading the past masters, but there are beginning to be
some token references to Peirce in more recent papers,
so hope springs eternal ...

Here is the local (parochial) locus classicus:

| Stevan Harnad (1990)
|"The symbol grounding problem", 'Physica D', 42:335-346.
| http://arxiv.org/html/cs.AI/9906002

Here are some more recent papers with Peirce in their bibs:

| Cangelosi, A., Greco, A., and Harnad, S. (2002)
|"Symbol Grounding and the Symbolic Theft Hypothesis"
| http://eprints.ecs.soton.ac.uk/6470/01/grounding.pdf

| Roy, D. (2004)
|"Grounding Language in the World: Schema Theory Meets Semiotics"
| http://web.media.mit.edu/~dkroy/papers/pdf/aij.pdf

PEF. Note 2


JR = Joe Ransdell

JR: I suppose one might reasonably say that Peirce's view can be regarded as
    an attempt to solve the symbol grounding problem, as Harnad describes it.
    This is speaking vaguely, of course:  I wouldn't want to claim that Peirce
    would describe the problem or its solution exactly as Harnad (or anyone else)
    does, but to say that it involves an attempt at a coordination of the digital
    and the analogical in explicating cognition seems to me a fair description of
    what Peirce was up to.   I take it that you are a devotee of the pure symbolist
    approach, Jon, i.e. you think Peirce was.

The main thing that a reader has to keep in mind when reading this literature,
along with a lot, but not all, of the related literature in AI and Cog Sci,
is that the word "symbol" means something rather different from what the
same string of six letters means in Peirce's work, or in classically
grounded philosophy in general.  A telling symptom of this is that
the adjective "meaningless" does not lie on the tip of the tongue
when speaking of Peirce's symbols, nor does it apply to his usage
of the term "symbol" at all.  Like I repeatedly say, Peirce's
concept of the sign, defined in terms of a 3-adic relation,
anticipates and thereby avoids in advance all of the usual
pseudo-problems of the 2-addicted species of philosophy.

PEF. Note 3


JA = Jon Awbrey
JR = Joe Ransdell

JA: The main thing that a reader has to keep in mind when reading this literature,
    along with a lot, but not all, of the related literature in AI and Cog Sci,
    is that the word "symbol" means something rather different from what the
    same string of six letters means in Peirce's work, or in classically
    grounded philosophy in general.  A telling symptom of this is that
    the adjective "meaningless" does not lie on the tip of the tongue
    when speaking of Peirce's symbols, nor does it apply to his usage
    of the term "symbol" at all.  Like I repeatedly say, Peirce's
    concept of the sign, defined in terms of a 3-adic relation,
    anticipates and thereby avoids in advance all of the usual
    pseudo-problems of the 2-addicted species of philosophy.

JR: Yes, you have been saying that, but you have not made clear to me
    how it is that he manages to pull that off without bringing in the
    auxiliary role of iconism and indexicality, which you think of as
    sullying the purity of the symbol, which supposedly manages to get
    an object without need for association with signs of other types.
    That is the view which Tom Short reduces to absurdity, isn't it?

I've given what serves as an explanation for me
several hundred times by now, for example, here:

| In the pragmatic theory of signs, a "symbol" is a strangely insistent
| yet curiously indirect type of sign, one whose accordance with its
| object depends sheerly on the real possibility that it will be so
| interpreted.  Taking on the nature of a bet, a symbol's prospective
| value trades on nothing more than the chance of acquiring the desired
| interpretant, and thus it can capitalize on the simple fact that what
| it proposes is not impossible.  In this way it is possible to see that
| a formal principle is involved in the meaningful successes of symbols.
| The elementary conceivability of a particular sign relation, the pure
| circumstance that renders it logically or mathematically possible,
| means that the formal constraint it places on its domains is always
| really and potentially there, awaiting its discovery and exploitation
| for the purposes of representation and communication.
|
| Jon Awbrey, "Inquiry Driven Systems:  Inquiry Into Inquiry"
| IDS 068.  http://stderr.org/pipermail/inquiry/2004-May/001507.html
| http://members.door.net/arisbe/menu/library/aboutcsp/awbrey/inquiry.htm

An information-bearing symbol is not a meaningless symbol.
A symbol bears information, in effect, reduces uncertainty
about its objective domain, by virtue of its participation in
a sign relation, which embodies a real constraint among the
domains of signs, objects, and interpretant signs.  That's
all it takes in order for a sign to carry information.
It depends on nothing more than the formal principle
involved in the particular sign relation, together
with the fact that formal constraints or laws have
a full-fledged reality in Peirce's philosophy.

Now, an explanation of sign action and sign meaning
in terms of informational principles will probably not
satisfy those who are looking for an explanation in terms
of causal forces and iconic images, just as an explanation
of molecular bonding in terms of electrical charges and
force fields will probably not satisfy those who are
looking for an explanation in terms of little hooks,
but there is very little that I can do about that.

PEF. Note 4


Incidental Music ---

| 'Cause I gotta peaceful, easy feelin',
|  And I know you won't let me down.
| 'Cause I'm already standin',
|  On the ground.
|
| The Eagles, "Peaceful Easy Feelin"
| http://www.oracleband.net/Lyrics/peaceful_easy_feelin.htm
| http://www.psci.net/~jmarsh/1.html

PEF. Note 5


| Many of our customers are searching
| for urns that naturally fit in with
| their existing decor and surroundings.
|
| http://www.plan4ever.com/urns/cat.phtml?cat=11

Necessity is the mother of natural fitness but
natural fitness is not of necessity the mother.

JA = Jon Awbrey
JR = Joe Ransdell

JA:  Now, an explanation of sign action and sign meaning
     in terms of informational principles will probably not
     satisfy those who are looking for an explanation in terms
     of causal forces and iconic images, just as an explanation
     of molecular bonding in terms of electrical charges and
     force fields will probably not satisfy those who are
     looking for an explanation in terms of little hooks,
     but there is very little that I can do about that.

JR:  What, then, do you make of a statement like this by Peirce,
     which seems on the face of it to be an unequivocal statement
     of the dependence of the symbol on the auxiliary functions of
     icons and indices to do what it does as a symbol?

| A symbol is a sign naturally fit to declare that the set of objects which is
| denoted by whatever set of indices may be in certain ways attached to it is
| represented by an icon associated with it.  To show what this complicated
| definition means, let us take as an example of a symbol the word "loveth".
| Associated with this word is an idea, which is the mental icon of one person
| loving another.  Now we are to understand that "loveth" occurs in a sentence;
| for what it may mean by itself, if it means anything, is not the question.
| Let the sentence, then, be "Ezekiel loveth Huldah".  Ezekiel and Huldah
| must, then, be or contain indices;  for without indices it is impossible
| to designate what one is talking about. Any mere description would leave
| it uncertain whether they were not mere characters in a ballad;  but
| whether they be so or not, indices can designate them.  Now the effect
| of the word "loveth" is that the pair of objects denoted by the pair of
| indices Ezekiel and Huldah is represented by the icon, or the image we
| have in our minds of a lover and his beloved.
|
| C.S. Peirce, 'Collected Papers', CP 2.295, "The Short Logic" (c. 1893)

JR: Do you think of this as a sop thrown to hoi polloi,
    his real view being reserved for those capable of
    working with formal logic (or however you would
    characterize the relevant expertise)?

Really, now.  We can well say that human beings are naturally fit for life
in the temperate zones of the planet Earth, but that hardly means that all
of them live in the middle latitudes, or none can go to the moon and stars.

Make it so ...

PEF. Note 6


BU = Ben Udell
JA = Jon Awbrey

Re: PEF 4.  http://stderr.org/pipermail/inquiry/2005-April/002581.html
In: PEF.    http://stderr.org/pipermail/inquiry/2005-April/thread.html#2578

Every now and then I have to Google a phrase
just to remind myself of how the hoi polloi
talk these days.  Most of the web pages
that I got for "naturally fit" are not
fit to print or even to link to, but
here's yet another interesting one:

http://www.newadvent.org/summa/210903.htm

As far as I know I've never said anything bad about icons or indices,
except that they fall short of world domination.  Nothing about the
existence of atoms, so far as their conception is useful, excludes
the existence of molecules on the one hand or quarks on the other.
The existence of primes does not negate the existence of composites,
indeed, they're all beautiful in their own way.  And I've never made
any claim that "symbols are all you need".  As far as I can tell, all
of the examples of hybrid varieties of symbols involve compound signs,
at least as complex as propositions, or signs that are interpreted by
propositions, and it requires what Peirce called "pure symbols", like
the operators 'not', 'and', 'or', 'if-then', 'of', and a small host
of others, to bind the components of these propositions together.

Previously:

JA: Really, now.  We can well say that human beings are naturally fit for life
    in the temperate zones of the planet Earth, but that hardly means that all
    of them live in the middle latitudes, or none can go to the moon and stars.

JA: Make it so ...

BU quoth:

| Will it be Captain Picard or Syd Barrett?  More incidental lyrics --
| 
| Lime and limpid green, a second scene, the fight between the blue you once knew
| Floating down, the sound resounds around the icy waters underground
| Jupiter and Saturn, Oberon, Miranda, and Titania
| Neptune, Titan, stars can frighten
| 
| Blinding signs flap flicker flicker flicker blam
| Pow pow
| Stairway scare Dan Dare who's there
| 
| Lime and limpid green, the sounds surround the icy waters underground
| Lime and limpid green, the sounds surround the icy waters underground
| 
| -- Syd Barrett, Pink Floyd, "Astronomy Domine"

BU: Seriously, what's confusing me in this discussion is the question of the ROLE
    of Jon's kind of analysis, and whether that ROLE is the agreed subject of the
    argument.  I disbelieve that a mind's semiosis could consist entirely in symbols
    and remain grounded or in touch with reality.  Does that mean that one can't do
    symbolic semiotic and that one can't simply bracket the iconic and indexical
    aspects? I don't see why one can't -- except in regard perhaps to mathematical
    diagrams (which I don't regard as qualitative semblances of objects anyway).
    But can such a study be at the root or foundation of semiotics?  I doubt that
    it can be at the root of the study of semiosis as any kind of complex process
    such as might actually happen.  It could be prior to such a study -- it would
    be more abstract, something spread in the soil.  It's not really according to
    the strict defiinition of the symbol -- the symbol is defined by the effect
    which it will (by grown habit) have on the interpretant, full stop.  It is
    not defined by the effect which it will have on the interpretant such that
    said effect is itself just a symbol -- that's a much narrower criterion,
    the symbol as a separate species rather than a separate "gender" of sign.

BU: Is the argument about whether there's any kind of role at all
    for Jon's kind of pure symbolistic, or is there some particular
    role in the architectonic of semiotic research, regarding which
    the argument is about the pure symbolistic's suitability?

PEF. Note 7


BB = Bill Bailey

Re: PEF 5.  http://stderr.org/pipermail/inquiry/2005-April/002582.html
In: PEF.    http://stderr.org/pipermail/inquiry/2005-April/thread.html#2578

BB: I ought to leave this alone since I don't know what I'm talking about.
    Before I joined this list, I thought I at least understood Peirce's
    definitions of symbols, indexes, and icons.  Oh, for the bliss of
    ignorance!  Now that I'm becoming knowledgeable, I'm battling the
    Aunt Polly spoon-counting syndrome in 'Huckleberry Finn'.  I want
    to throw the whole damn drawer across the room and never count 
    them again.  But Aunt Polly was probably wiser than I am.

BB: If I accept that a symbol is only arbitrarily linked to its
    signification by convention, how can I at the same time accept
    that it is "naturally fit to declare" something?  "Naturally fit"
    doesn't sound like anything arbitrary or purely conventional to me.
    Further, if I accept that its signification functions rest upon
    inherent indexical and iconic properties, how in the world can
    I possible maintain that its definitive characteristics depend
    upon its arbitrary, conventional usages?  I read the passage
    and I think I understand it.  What I don't understand is how
    to reconcile that passage with what I -- and I assume some 
    others -- understood as Peirce's definition of symbol.

In these early accounts, Peirce speaks of symbols being born as indices,
in which the conventional nature is very marked, but still nowhere near
as idiosyntactic as some imagine, since a convention by its etymology
means that many impressions and/or many interpreters are required to
come together and concur on the conferral of the title in question.
So natural fitness is apparently an acquired nature, a maturing
recognizance of the applicable laws of the symbol in question.
Now "arbitrary" means "relating to an arbiter" and "arbiter"
is just another name for an interpreter, judge, or mediator.
The catch is:  It is just a peculiarity of human nature
that judgments always seem more capricious when it's
someone else doing the judging.

And so it goes ...

PEF. Note 8


JC = John Collier

Re: PEF 1.  http://stderr.org/pipermail/inquiry/2005-April/002578.html 
In: PEF.    http://stderr.org/pipermail/inquiry/2005-April/thread.html#2578

JC: Though I agree with Jon that the debate would be helped by a knowledge
    of Peirce, it should be recognized that the debate is in the context of
    what is now called "neurophilosophy". One of the problems is to discover
    where symbols and meanings come together in the brain. It turns out that
    there are a lot of different neural circuits that deal with representation
    and different types of memory, and nobody really knows yet what is located 
    where, or exactly how it works.  We do know that we use symbols in a general
    way, and that we do it meaningfully and effectively (most of the time).  Part
    of the symbol grounding problem is how a brain like ours can use representations
    in a general way.  This is psychology, and to be fair, the authors listed below
    are psychologists who are well aware of this problem.  Detailed knowledge of
    neurophysiology is required to even understand why the problem is a problem
    that has _not_ been solved by Peirce or anyone else.  To just consider these
    papers in isolation from this other work (which the authors surely do not
    in the context of their journals and meetings) would give a very misleading
    picture of what they are up to.  I also note that "digital" has a special
    meaning in this literature, going back to Dretske's "Knowledge and the Flow
    of Information", though it has some roots in the notion of a binary code as
    well.  I've argued that understanding the nature of binary coding and the
    relation of syntax to semantics from  Peircean view is very useful, but so
    far only a few semioticians have taken much notice.

JC: Perhaps we need more Peircean neurophysiologists around.
    Any volunteers?  I'm trying to educate my flatmate (who was
    hired here to do neurophilosophy in our cog sci program), but
    he keeps saying that neurophysiologists have their own names
    for things, and their own ways of doing things.  The isolation
    goes both ways. Once in a while, though, something resonates.
    When it doesn't, we don't quite get into throwing things at
    each other over dinner, but it gets intense sometimes.  We
    still eat dinner together, though.  My flatmate says I will
    have to relate the work in my paper to current neurophysiology
    before the real scientists will take notice.  Maybe he is right.

I currently view this from a systems theoretic perspective,
from which our favorite species of wetware is just another
brand of intelligent system.  And, yes, there's a lot more
to understand about how intelligent species in general, or
this one in particular, evolve to the point of manifesting
a capacity for inquiry.  And, yes again, a huge subproblem
of this study is figuring out how the dynamic and symbolic
aspects of such critters can be integrated into a coherent
agent or system.  But I still submit that Peirce's special
insights about signs, inference, information, inquiry, and
their interrelations, that shifted the very grounds of the
problem, are 'sine qua non' for any hope of future advance.
Without a more general appreciation of the consequences of
irreducibly triadic relations, the specialized disciplines
are doomed to spin their wheels in all the old dyadic ruts.

PEF. Note 9


JA = Jon Awbrey
JP = Jim Piat

Re: PEF 8.  http://stderr.org/pipermail/inquiry/2005-April/002598.html
In: PEF.    http://stderr.org/pipermail/inquiry/2005-April/thread.html#2578

In part:

JA: Without a more general appreciation of the consequences of
    irreducibly triadic relations, the specialized disciplines
    are doomed to spin their wheels in all the old dyadic ruts.

JP responds, er, cogitates:

JP: While myself not fully understanding the nature of
    a triadic relationship -- I am at least convinced
    that something is missing in the S-R model of
    representation and oh how I wish I could my
    finger on just what it is beyond simply
    repeating to myself a triad a triad.

JP: Maybe something like this -- we as interpreters of symbols
    are the third element that mediates between symbol and object.
    That is to say that symbol and object are joined by virtue of the
    fact that our actions reflect the relationship we attribute to them.
    Two things can only stand in a representative relationship with one
    another with respect to a third thing.  IOWs two things can not stand
    for one another (in the same sense that they can be greater or less
    than one another -- ie opposed to one another) unless a third object
    is introduced.  Representation is necessarily triadic.  Perhaps
    it's as simple as that.  It's simply a matter of defining what
    representation is.  It involves three things.

JP: But how does it come about that one thing can be understood
    or interpreted as standing for another to some interpretant?
    Granted this is an irreducibly triadic affair -- but what is
    the nature of each leg of this relationship.  Without reducing
    the triad to a dyad -- what is going on?   For example the
    interpretant is not standing for the object, nor the object 
    for the interpretant or representamen.  So it seems to me
    legitimate to examine each leg of this asymmetrical triad
    and explicate (in whatever language or symbols one wants)
    just how each leg functions -- what it means or how to
    represent its logical import.

The first thing we have to understand is that some things are just
inherently properties of whole systems, or of things only insofar
as they participate in whole systems.  So far this is an insight
that is common to both Peirce and Saussure, in their own ways.
So non-trivial examples of sign relations necessarily involve
lots of 3-tuples of the form <o, s, i> -- or <s, o, i> if
one prefers.  Thus the properties of sign relations are
properties appropriate to sets, while the combinations
and discombobulations of sign relations are of the
types appropriate to sets.  This renders it very
chancey already to speak of the "legs" of any
sign relation that does not amount to one
single 3-tuple, or what Peirce called
an "elementary 3-adic relation".

The nearest we can get to something 2-adic that is properly derived
from a 3-adic relation is to "project" the whole 3-adic relation on
two of its domains, what we would get by deleting a column from the
corresponding relational data table, and ignoring any duplicates of
the ordered pairs that result.

One of the reasons that I selected an alternate title for this thread,
aside from trying to avoid being tangled in a literature that promises
to remain quite confused about such a symbol thing as what a simple is,
much less the pure and symbol varieties, is to allude to a personal bit
of insight from my undergraduate days, hence the *dustiness of the tune,
to wit:  Sometimes you need to quit trying to figure out how things get
connected, and think about how they got disconnected in the first place.

PEF. Note 10


JA = Jon Awbrey
JP = Jim Piat

Re: PEF 9.  http://stderr.org/pipermail/inquiry/2005-May/002614.html
In: PEF.    http://stderr.org/pipermail/inquiry/2005-May/thread.html#2614

JA: JP responds, er, cogitates:

I was trying to strike a middle ground
between S & R, to wit, sign & relation.

JP: Cogitating, eh  -- you're too kind!  I think that what you say
    below is (on the whole;) a helpful perspective and good advice.
    Really -- kidding aside --  good advice.  And so's that peaceful 
    easy feeling.  Not everything can be resolved --

The application to the problem of semiotic grounding
is to turn the tables on the puzzle and ask instead:
How do objects, signs, interpretants differentially
co-evolve from their primordial indistinctness?
To give it a catchy name, I guess I will call
it the "semiotic take-off problem" (STOP).

I'm still working on it ...

In the mean time here's a primer
that I wrote on the subject of
relational reducibilities:

RAR. Reductions Among Relations (various different formats):
RAR. http://forum.wolframscience.com/showthread.php?threadid=400
RAR. http://forum.wolframscience.com/printthread.php?threadid=400
RAR. http://stderr.org/pipermail/inquiry/2004-April/thread.html#1367

Discussion History

Symbol Grounding Problem • Peirce List (Apr 2005 – May 2005)

Dramatis Personae
BB = Bill Bailey
BM = Bernard Morand
BU = Benjamin Udell
CM = Christophe Menant
IS = Inna Semetsky
JA = Jon Awbrey
JC = John Collier
JH = John Hanna
JP = Jim Piat
JR = Joseph Ransdell
KM = Kirsti Määttänen
VA = Victoria N. Alexander
  1. http://lyris.ttu.edu/read/archive?id=152232 JR
  1. http://lyris.ttu.edu/read/archive?id=152604 JA
  2. http://lyris.ttu.edu/read/archive?id=152633 JR
  3. http://lyris.ttu.edu/read/archive?id=152640 JA
  4. http://lyris.ttu.edu/read/archive?id=152655 JR
  5. http://lyris.ttu.edu/read/archive?id=152682 JA
  6. http://lyris.ttu.edu/read/archive?id=152737 JR
  7. http://lyris.ttu.edu/read/archive?id=152759 BB
  8. http://lyris.ttu.edu/read/archive?id=152766 BU
  9. http://lyris.ttu.edu/read/archive?id=152768 BU
  10. http://lyris.ttu.edu/read/archive?id=152769 BB
  11. http://lyris.ttu.edu/read/archive?id=152772 BU
  12. http://lyris.ttu.edu/read/archive?id=152781 JR
  13. http://lyris.ttu.edu/read/archive?id=152804 KM
  14. http://lyris.ttu.edu/read/archive?id=152819 BM
  15. http://lyris.ttu.edu/read/archive?id=152847 JR
  16. http://lyris.ttu.edu/read/archive?id=152852 JR
  17. http://lyris.ttu.edu/read/archive?id=152908 BB
  1. http://lyris.ttu.edu/read/archive?id=152777 JA
  2. http://lyris.ttu.edu/read/archive?id=152786 BU
  1. http://lyris.ttu.edu/read/archive?id=152779 JA
  2. http://lyris.ttu.edu/read/archive?id=152798 JA
  3. http://lyris.ttu.edu/read/archive?id=152835 BU
  4. http://lyris.ttu.edu/read/archive?id=152988 JA
  5. http://lyris.ttu.edu/read/archive?id=153260 JA
  6. http://lyris.ttu.edu/read/archive?id=153267 JP
  7. http://lyris.ttu.edu/read/archive?id=153767 JA
  8. http://lyris.ttu.edu/read/archive?id=153791 JP
  9. http://lyris.ttu.edu/read/archive?id=153823 JR
  10. http://lyris.ttu.edu/read/archive?id=153844 JA
  1. http://lyris.ttu.edu/read/archive?id=152972 JR
  2. http://lyris.ttu.edu/read/archive?id=152991 BB
  3. http://lyris.ttu.edu/read/archive?id=153007 JR
  4. http://lyris.ttu.edu/read/archive?id=153012 BB
  5. http://lyris.ttu.edu/read/archive?id=153256 JC
  6. http://lyris.ttu.edu/read/archive?id=153270 VA
  7. http://lyris.ttu.edu/read/archive?id=153278 BB
  8. http://lyris.ttu.edu/read/archive?id=153349 IS
  9. http://lyris.ttu.edu/read/archive?id=153374 BM
  10. http://lyris.ttu.edu/read/archive?id=153536 JC
  11. http://lyris.ttu.edu/read/archive?id=153542 BB
  12. http://lyris.ttu.edu/read/archive?id=153624 JC
  13. http://lyris.ttu.edu/read/archive?id=153653 JC
  14. http://lyris.ttu.edu/read/archive?id=153668 JR
  15. http://lyris.ttu.edu/read/archive?id=153680 BB
  16. http://lyris.ttu.edu/read/archive?id=153689 JA
  17. http://lyris.ttu.edu/read/archive?id=153699 BB
  18. http://lyris.ttu.edu/read/archive?id=153865 JC
  19. http://lyris.ttu.edu/read/archive?id=153866 JC
  20. http://lyris.ttu.edu/read/archive?id=153965 CM
  21. http://lyris.ttu.edu/read/archive?id=154024 BB
  22. http://lyris.ttu.edu/read/archive?id=154109 JH
  23. http://lyris.ttu.edu/read/archive?id=155402 JC
  24. http://lyris.ttu.edu/read/archive?id=155481 CM
  1. http://lyris.ttu.edu/read/archive?id=153674 JR

Peaceful Easy Feeling • Inquiry List (Apr 2005 – May 2005)

  1. http://web.archive.org/web/20081120225656/http://stderr.org/pipermail/inquiry/2005-April/002578.html
  2. http://web.archive.org/web/20121113153333/http://stderr.org/pipermail/inquiry/2005-April/002579.html
  3. http://web.archive.org/web/20121113153343/http://stderr.org/pipermail/inquiry/2005-April/002580.html
  4. http://web.archive.org/web/20081120185541/http://stderr.org/pipermail/inquiry/2005-April/002581.html
  5. http://web.archive.org/web/20081120190827/http://stderr.org/pipermail/inquiry/2005-April/002582.html
  6. http://web.archive.org/web/20081120194913/http://stderr.org/pipermail/inquiry/2005-April/002584.html
  7. http://web.archive.org/web/20060217225239/http://stderr.org/pipermail/inquiry/2005-April/002589.html
  8. http://web.archive.org/web/20121113153355/http://stderr.org/pipermail/inquiry/2005-April/002598.html
  9. http://web.archive.org/web/20120726142113/http://stderr.org/pipermail/inquiry/2005-May/002614.html
  10. http://web.archive.org/web/20120726142115/http://stderr.org/pipermail/inquiry/2005-May/002615.html