Questa pagina è anche in italiano Ce site est aussi en francais
Introduction
Dear graphology friends, I have a tale to recount concerning the origins of my software... would you like to hear it?
It was in 1938, during the school holidays, when I, a fourteen-year old, was in the library, nearby my Father's Law Firm. I was looking for something interesting to read, and the thing that came to hand was a curious little tome, it spoke about a method of investigation which it called practical ... yet scientific. I told myself that I must take a quick look before passing to another book ... it dealt with graphology. The book was by Umberto Koch (1914) a pen name for Girolamo Moretti. I re-read that book carefully several times, graphology stayed with me and over the years it became my main interest outside work. "Practical" and "Scientific" said that book: two magical words for a boy. From that moment I took to speaking about the topic with my school friends, describing to them, sometimes using gestures, the features of our teachers who jotted notes all across our wwwwork.
After
the
war, I took work with the ENI Group of companies, I specialised at
IBM in business
computer science. Back in 1955 the first valve processors were large and clean like
hospital rooms! This paved the way to an administrative career located beside
the Data Processing Center, then embraced duties in general management and for a
time in the personnel field. In Personnel I had chance to bring my
psychoanalytical and graphological skills to fruition.
My
computing background pushed me to find computer solutions for various fields to
serve my company and its employees. I always had my passion for graphology at heart
and little by little, over the years, I went beyond Moretti's methods, past
those of Marchesan and, excited by the ideas of harmony and rhythm, I
passed through Klages and Crépieux-Jamin, and finally arrived at H. de
Gobineau's French grapwwwtry. I studied in order to enrich my psychoanalytical
knowledge but also to work out which of the several methods I
could one day use to computerise graphology. Therefore,
fully appreciating harmony-disharmony in handwriting, something on which all the methods
agreed, I decided to bring about my own "definitive"
or overall method inspired mainly by the musicality of the rhythms
modulated by written gestures.
RPG is a programming language, used in transistor computers which were already working in
nanoseconds. It was 1969 and using RPG, at that time the most advanced language for computers, I began
to write my first program that I called "Graphological Portrait". From the
practical point of view the problems were soon evident: automating
the analysis process with the computing power of the day was extremely slow.
Furthermore I couldn't fully use the massive amount of personality
information with the software. From the earliest programs I felt that the way ahead
was not a simple databank full of phrases "carefully worded...but also
very immobile in a generic sense" that could be called up in accordance
with the signs, instead it had to be the overall architecture of the system that
could produce dynamic phrases "that interpreted the variations in the state of
mind in relation to the setting and situations." In other words the
phrases had to be "adapted to external variables." Nevertheless I convinced
myself that it was worthwhile to go on!
From the outset I decided that I would include all graphological methods available into my thinking. Therefore any software I wrote would take account of all the "signs" defined by the different schools of thought. It pointed to a matricular system with well-formulated algorhythms that cross-referenced signs to each other, finally obtaining, transversally, information on all the possible combinations between themselves. All of this in spite of the fact that the combinations were inspired by different schools.
The
intention was ambitious and from the practical side, it forced me make a
critical review of the various methods
which inspired the individual schools of
graphological thought, with the aim of obtaining a single psychoanalytical method. That meant
finding rules of inter-compatibilty to bring them together, in a unified, ample,
psychoanalytical system, using the various norms of
graphological psycwwwtrics. I realised that by so doing I could dilute the principles on which
the theories of the individual methods were based but my aim went further: by
this I wanted to obtain a
simplified graphological analysis system, not becoming simpler, indeed extremely
well-structured but above all more complete and accessible to all graphologists!
I
had a method in mind that was innovative for traditionals, but that was also
attractive to the others as it was concerned with a full understanding of
similarities and a full look at graphological science. We know that every new
idea is a challenge to tradition and to scientific norms!
With
this methodological attempt, that I called Cybergrafia, I succeeded in creating software that had very specific
users and purposes. This was in great part due to the
continuous compatibility of personal computers and to the evolution of programming
languages in the 1970s. I have
always been somewhat self-critical and fairly restrained so
now these
programs are in the bottom drawer. Perhaps in time I will see again
them more serenely: they were enriching for logical thought, they helped avoided
intuition, and here I have had
the opportunity to cite them in order to
help understand the background of many years of algorithms!
It was in 1985, after fruitful exchanges with graphology-friends trained in the Moretti system I decided to try computerising the Moretti method. In actual fact because of my attempt to put together a global psychoanalytical system I was able to say that the Moretti school was without doubt the most uniform and integrated psychoanalytical system I had come across. It was a system documented by scientific research on the neural system that could claim precise and considered rules very well fine tuned with flexible logic, appropriate for my statistical-mathematic calculations suitable to be applied to psychology.
Having decided
to
rewrite all the previous matricular formulas for Excel for Windows I used
only the signs that go to make up the Moretti system. From
the matricular approach I obtained "category" combinations that
were useful to define, with objectivity, the basic descriptions of
Moretti's character traits, personality traits and other related aspects.
With
the progression of computers, ever faster and more reliable, I rewrote my software on a
mathematical model that allowed the preservation of the "perfectability" that is the
ability to adapt itself to modification and quickly allow updates and improvements that
I introduced bit by bit. This being on the assumption that graphology, seeking to be
an experimental science, must always be ready to embrace new psycho-graphical and
psycho-analytical directions. The structure of the software, however by now was very
laid out:
the technique of the database was no longer being used, but, drawing on the
conceptual technique of "artificial intelligence", it now had very many
elementary "decision trees" structures that in their schematic
complex imitate the reticular complex of human thought (neural nets). These "decision
trees", were logical structures using Boolean operations to allow rapid
filtering of quite complicated situations, with resultant choices derived from various
situations. Such tree structures (see Note 1) began with intial data (input of the graphological signs) and
by automatic searches for alternative choices (nodes), developed "articulate
phrases" amongst which, in several ways, allowed the creation of
well-formed prose to acheive the desired meaning; the term "suggestions"
is used in the program to define these phrases. Conclusive output of the program: "useful
psychoanalytical notes for the drawing up
of the graphological report" were in fact organised by sub heading from the
many activated "decision trees".
In the 1990s, I decided to release my programs and with this purpose I wrote various articles in which I suggested that businesses should use graphological services. I thus publicised my software in several specialist publications, amongst which were Diritto & Pratica del Lavoro from IPSOA Business School and Direzione del Personale published by AIDP (Associazione Italiana Direttori del Personale)
Foremost
GRAF-2000 seeks to supply the graphologist with "explanations about the
self" inserting, after expressed more meaningful phrases
in the processing: " useful psychoanalytical notes for the drawing up of the
graphological report", the
composition of the signs that subtend to them. It wants moreover to concur with
the graphologist to widen and to deepen the sense of its phrases with psychological
references oriented therefore from allowing it the best customisation and
reformulation to you of the same ones, according to the fine ones which analyses
must be directed. Its I use must answer to the professional
requirements of the sociologist, the psychoanalyst and the graphologist doctor which
find, in the immediate possibility to have a meaningful confirmation of the psychoanalytical
picture that they perceive, an important diagnostic support.
The
result that the program
outputs has been defined, from who recently experienced,
important and collaborated, useful also the graphical expert in order to equip
its skills judicial lawyers and with ulterior elements on the author in question
with writing, of which now will be able, upon request of part, to indicate also
intellectual aptitudes, inclinations and abilities to mean and wish.
Dear friends, if you are still reading, I'd like to express my heartfelt thanks for your time but I still ask you a last favour. Please visit me on the web site www.grafoservice.it and http://www.remet.it/italian/grafologia.htm You have learned all about the development of GRAF-2000, so you can test first-hand how well the program works. And in the light of this, perhaps I'll continue the story another time.
Koch,Umberto (1914) Manuale di
grafologia. Osimo, Officina Tipografica Scarponi,
1. The idea used in writing the program required a rich supply of many elementary structures "decision trees" that in their diagrammatic form are connected to imitate the path followed by human thought which goes across neural networks. These "decision trees" are
essenzialmente strutture logiche che, anche in situazioni complesse, consentono di vagliare, con metodo e rapidità, le scelte espressive meglio rispondenti alle possibili situazioni. Tali strutture, iniziano con i dati di entrata (input dei segni grafologici) ed attraverso momenti di ricerca e di scelta (nodi), sviluppano «frasi articolate» che, tra loro variamente componendosi, pervengono a costruire periodi discorsivi dal senso compiuto.The GRAF-2000 Decision Trees were made possible using the mathematical functions, available in Microsoft EXCEL's HELP , per costruire strutture ad albero che eseguono iterazioni convergenti. Le strutture ad albero ed i loro nodi correlanti sono formule logico-matematiche con funzioni condizionali che si avvalgono della logica booleana per trovare scelte alternative di tipo elementare (yes-no) che pervengano a soluzioni adatte alle più diverse situazioni di vaglio:
"IF
A
is true,the choice is a1, but IF
A
is false then proceed to the alternative B(*) and repeat the
check as made for A"
(*) where A,B,C, etc. are values that can be seen as single signs che loro combinati, nonché other decision trees, whilst a1, b1,c1, etc. are resolving branches that conclude the path of the choice:YES/NO.
Qui sopra lo schema evidenzia due strutture ad albero complesse: A(*) e C(*). Ogni albero di pertinenza nelle sue scelte si ramifica in SI e NO e si correla ai rami degli altri alberi, così da formare lunghi percorsi interattivi, che partendo sempre dai dati elementari segni, pervengono a vari raggruppamenti, sempre più specializzati, detti gruppi categoriali anchessi, a loro volta, costituenti riferimenti che, tra loro correlandosi, compongono ulteriori raggruppamenti di livelli gerarchici superiori. Il programma definisce questi gruppi categoriali specializzati assegnando loro nomi che si richiamano agli aspetti qualitativo-simbolici delle caratteristiche al cui vaglio essi sono finalizzati. Essi vengono utilizzati sia come nodi discriminanti che come indici; ne sono esempi nellelaborato finale, i punteggi su base decimale, indicati a lato di alcuni capitoli, nonché le descrizioni sintetiche dei tratti salienti. Questi gruppi categoriali assumono gradi che sono valori ponderali determinati su base decimale, che risentono nella formulazione di medie, mediane e deviazioni standard, anche della presenza di indici fautori e/o contrastanti.Back to the conference (www.graphodigest.com)
Last update 13 December 2002. Please report any corrections to me at this email address bradlen@graphodigest.com
Hits since 12 August 2002