algorithms in NLP and/or hypnosis?
Kevin Hogan
Algorithms in NLP and hypnosis?
To determine whether there are algorithms in NLP and/or hypnosis, we
first must define both terms. The term algorithm has a generally
accepted definition which I will note below. The term NLP does not have
a generally accepted definition and therefore offer my definition here.
Definition: NLP- The semantic response noted and/or experienced by
someone or somethings stimulus. Neurolinguistic implies the response of
the nervous system by language. Programming implies the act of
installing, changing, or inputting of such language into another,
whether the stimulator is aware of the stimulus or not.
Definition: Algorithm- Systematic mathematical procedure that produces
in a finite number of steps, the answer to a questionor the solution of
a problem. The quesiton or problem in turn must belong to a class that
has an infinite number of numbers or members. There are two classes of
algorithms. “Decidable class” is a class of algorithms that produces a
yes or no response when completed. The second class is called the
“Computable Class.”
An algorithm that leads to the answer (a specific number) is called a
computational procedure. For finite classes of questions, trivial
algorithms always exist (at least in principle). They consist of tables
and values of the answers.
The term algorithm states that a process, procedure, decision or
computation is mathematical and computational in nature where there is
something similar to a “cause and effect” relationship in all cases.
It can be argued that there are algorithms in NLP, possibly only one or
two, and those remain to be clinically tested that indeed they are
algorithms or mere postulates or presuppositions, where one presumes
something to be true to make sense of everything else that follows.
Consider some of the processes of NLP.
Cause: New Behavior Generator Process
Desired Algorithmic Effects: Client utilizes best choices in all cases.
The New Behavior Generator is one of my favorite NLP tools because it
regularly is effective in over half of the cases where I personally
utilize it. However, it is certain, in my work that it is not an
algorithm. We certainly can say that we have given the client more
choices, however, that in itself is not 100% useful. What is 100%
useful is that we give a client more choices and they actually utilize
those choices at/or near 100% of the time. The distinction seems fine,
but is indeed quite important. Therefore, we could argue that the New
Behavior Generator as a process is an algorithm for offering new
choices to a client, but in my experience, it cannot be argued that the
choices are always utilized for superior behavior. The NBG is a very
useful process but it is not an algorithm per se for certain long or
short term change.
Another remarkably useful NLP procedure is that of Reframing. More
success, in my experience with reframing techniques is met than with a
New Behavior Generator. Success is defined in my thinking as long term
change or improvement. Reframing presents excellent results in at least
2/3 of the clients that I use the procedure with. In other words, they
actually experience improvement and/or long term change from the point
where therapy was requested. In fact, if we argue for the algorithm, we
could say that some change occurs in all clients, at least in the short
term. However, this is not proven to be the case in my experience for
the longer term. Results are what I consider excellent with this
process, but not, algorithmic.
Submodality shifts provide a very interesting consideration in this
topic. Certainly, in the short term, finding the driving submodalities
and shifting, them seems to always, 100% of the time, create change in
the short term. In the long term however, the results of therapeutic
intervention don’t match those of reframing in my experience. I believe
that we could say it is always true, and therefore an algorithm that
for short term success (one week or less) submodality shifts always
create change that is verifiable by the clients subjective experience.
Therefore it could be argued that if repeated, say, weekly, you could
experience long term change with almost 100% results. The same could be
said for many techniques in NLP. However, long term with no
reinforcement or second, third, or fiftieth application of the process,
there is certainly no algorithm.
Interestingly there have been some generalizations about submodality
shifts made that are very much inaccurate and potentially harmful. The
generic concept of amplification can cause phobias and other negative
alterations in behavior. It is assumed by many NLP practitioners that
amplification amplifies whatever sensations is being experienced. In
practice this is not true in a significant minority of the cases.
Probably 1/3 of the time. The notion of bringing something closer,
turning the sound up, making it brighter and more colorful is often
useful and often destructive. It is always best to elicit the specific
submodality shift and analogically discover where the shift is
maximized. To do anything else is irresponsible. Amplification is not
an algorithm for positive change. Many, many clients I have store there
best memories as still frames and in black and white where
the more sad memories are motion pictures and in full color. Closeness
is often much more destructive than a picture far away for a positive
or neutral memory. There are no rules, except that experience shows
that one must find the drivers and analogically determine what the peak
experience is felt at. Amplification is not the answer. It is not an
algorithm.
Strategies, which are probably my “favorite” tool of NLP again normally
are useful in therapy, but are not to be confused with algorithms
because they are very context dependent. Dilts discovered the
usefulness of the “NLP spelling strategy” in schools and it has been
proven effective. However, it is not an algorithm in any sense because
it doesn’t always produce positive results. (Dynamic Learning 1996)
Strategies are very useful in discovering how people make poor
decisions and we can often install new strategies, but the notion of
always making positive long term change is nonsense. It doesn’t occur
in experience. Installation of new strategies seems
to greatly help roughly 1/5 of my clients and significantly improves
another 3/5.
Grinder and Bandler noted in 1979 that nothing in NLP is true, it is
simply their hallucination of what is useful. An algorithm is a
verifyably true. What Bandler and Grinder taught was the best of what
they played with in their very brief experiences in unscientific
experiments, which you can read about in books like Frogs into Princes.
In the field of neurolinguistics, started in 1921 by Alfred Korzybski,
we discover that there are “almost” algorithms of mental dis-order.
People who think extensionally are almost always mentally healthier
than those who think intensionally. Those with the ability to abstract
to higher and higher orders are almost always mentally healthier than
those who are “stuck” at one “level”.
The fact that we are all very different in our personal history and our
genetic make up makes algorithms for long term improvement almost
impossible. In the short term almost anything can be a facilitator for
short term change. Almost anything that you change in a system will
cause at least short term change in behavior or subjective experience.
In the long term, most processes significantly lose their benefit
unless repeatedly reinforced. We expect this simply because of the
multitude of engrams and experiences that occur after therapy. New
stimulii create new responses that integrate within or over-ride the
previous work of the practitioner.
NLP remains a useful adjunct to hypnotherapy and hypnotherapy a useful
adjunct to NLP. They are both among the most applicable arts and
pseudo-sciences for non-medical improvement that are utilizable by the
masses. Unfortunately, I see no long term algorithms in my work, which
by definition, their should be, because algorithms are essentially,
always “true.” I’m always looking for the ideal tool in every
situation. I have found many utterly useful tools, but await the
algorithms that anyone would lay claim to. In my estimation they simply
don’t exist for long term change. I have found nothing that always
works in every situation or in a seemingly specific situation.
Regression therapy has come very very close, but I can think of several
examples in my own experience where the long term change was not
significant in my opinion.
If anyone knows of an algorithms, share only one. The others will be
worth millions and probably billions!
My best to all in this holiday season,
Kevin Hogan
-------------
Author of The Psychology of Persuasion: How to Persuade Others to Your
Way of Thinking. Pelican Publishing.
John Harrington
Ron,
An algorithm is a set of steps that _always_ work. That is, if you follow the
steps the result is always predictable. It is, ampng other things, a computer
programmer's term.
Kevin was hoping to find algorithms for handling specific client problems, and
was summareizing his findings in the quest. His summary says that he did not
find true algorithms for any treatment modality -- all are less than 100%
effective.
Smiles,
John
Ron <rba...@dcn.davis.ca.us> wrote:
>Kevin Hogan wrote:
>>
>> Algorithms in NLP and hypnosis?
>>
>Could someone simplify and summarize this post for me, especially the
>part about algorithms and their use in NLP?
>Thanks.
>ron
Wolseley B. Shoninger
On Fri, 13 Dec 1996, John Harrington wrote:
> Ron,
>
> An algorithm is a set of steps that _always_ work. That is, if you follow the
> steps the result is always predictable. It is, ampng other things, a computer
> programmer's term.
>
> Kevin was hoping to find algorithms for handling specific client problems, and
> was summarizing his findings in the quest. His summary says that he did not
> find true algorithms for any treatment modality -- all are less than 100%
> effective.
>
As I do like to think in algorithms, I thought I'd jump in here...
The algorithm is a system for handling contingencies, rather than
resolving problems. In this, it is quite easy to write an algorithm for a
given set of problems. the key is that one must, therefore, first, be
able to resolve those problems. The algorithm doesn't solve the problems,
it merely maps the steps to follow in the event of contingencies and
differentials. It offers reasonable responses to those differentials
within a coherent plan of action.
It stands to reason, therefore, that to attain a useful algorithm for a
given problem, find someone who has a high success rate with
that problem, and map out there policy and plan of action,
with regard to these contingencies and differentials.
While algorithms are not "found", but rather written, anyone who deals
effectively with a given problem within the helping professions may be
said to employ an algorithm, though informally. To formally write these
"steps to solution" is a most efficient means of sharing those abilities
with others of the profession, and, often, as a means of empowering the
helpees to help themselves.
However, for the sake of understanding, I think it should be noted that
what Kevin PROBABLY meant was that none of the the solutional
strategies/modalities were 100% effective, rather than that the
algorithms were not effective. Saying the latter would have little
meaning in comparing modalities, as I understand it.
Algorithms are fundamentally just tools of organization [ie for behavior].
Read this letter --> worth keeping Y/N ---> if Y --> save --> worth
responding to Y/N --> if Y --> respond with thank you or flame -->
is the response a flame? --> Y/N --> if Y, press ctrl-alt-del NOW.
Ian Hobson
Hi Kevin,
I don't mean to be rude but why are you using such an unusual and
unhelpful definition of "algorithm" and "NLP"?
>Definition: NLP- The semantic response noted and/or experienced by
>someone or somethings stimulus. Neurolinguistic implies the response of
>the nervous system by language. Programming implies the act of
>installing, changing, or inputting of such language into another,
>whether the stimulator is aware of the stimulus or not.
>
I don't know where to start. This does not make sense.
What is a semantic response? Sematics is the study of how meaning is
encoded by structure. How can a stimulus, note or experience anything?
Neuro - to do with nerves. Linguistic - to do with language.
Nerolinguistic implies using language to alter something about how
people think - not how they respond using language.
Programming is installing, or changing - but not of language but
behaviour and (possibly) beliefs.
The person doing the study is the only one who needs to be aware of what
is going on. Neither the stimulator/therapist or the subject/patient
need know. Satir and Erikson did not understand much of what Grinder and
Bandler noticed.
>Definition: Algorithm- Systematic mathematical procedure that produces
Not systematic, but methodical. Algorithms have nothing necessarily to
do with systematics.
>in a finite number of steps, the answer to a questionor the solution of
>a problem.
Many problems had algorithms that solve some cases, and go on looking in
other cases. These do not necessarily produce an answer, and clearly are
infinite.
You are also in danger of assumung each problem has an answer and an
algorithmic solution. Many problems have no answer, while some have more
than one.
Some problems have no algorithmic solution. You can scramble a password
in ways that make it impossible to unscramble it back.
>The quesiton or problem in turn must belong to a class that
>has an infinite number of numbers or members.
Nonsence.
>There are two classes of
>algorithms. “Decidable class” is a class of algorithms that produces a
>yes or no response when completed.
and they complete in finite time.
>The second class is called the
>“Computable Class.”
I understood that a computable problem not only had a solution, but this
could be reached in finite time (although the method need not be known).
> An algorithm that leads to the answer (a specific number) is called a
>computational procedure.
Why must the answer be a number?
[snip]
>Cause: New Behavior Generator Process
>Desired Algorithmic Effects: Client utilizes best choices in all cases.
>
>
>The New Behavior Generator is one of my favorite NLP tools because it
>regularly is effective in over half of the cases where I personally
>utilize it. However, it is certain, in my work that it is not an
>algorithm. We certainly can say that we have given the client more
>choices, however, that in itself is not 100% useful. What is 100%
>useful is that we give a client more choices and they actually utilize
>those choices at/or near 100% of the time. The distinction seems fine,
>but is indeed quite important. Therefore, we could argue that the New
>Behavior Generator as a process is an algorithm for offering new
>choices to a client, but in my experience, it cannot be argued that the
>choices are always utilized for superior behavior. The NBG is a very
>useful process but it is not an algorithm per se for certain long or
>short term change.
NBG appears to be an algorithm for generating new behaviours, but not an
algorithm for what you want it for. This does not mean it isn't an
algorithm.
[snip]
>Grinder and Bandler noted in 1979 that nothing in NLP is true, it is
>simply their hallucination of what is useful. An algorithm is a
>verifyably true.
You have clearly never tried to verify a significant algoritm as being
true.
Start with something simple. Does adding up a column of figures produce
the same answer as adding from the top down?
>The fact that we are all very different in our personal history and our
>genetic make up makes algorithms for long term improvement almost
>impossible.
You assume they exist, can be coded and taught.
>In the short term almost anything can be a facilitator for
>short term change. Almost anything that you change in a system will
>cause at least short term change in behavior or subjective experience.
>In the long term, most processes significantly lose their benefit
>unless repeatedly reinforced.
So you build in reinforcement.
>
>NLP remains a useful adjunct to hypnotherapy and hypnotherapy a useful
>adjunct to NLP. They are both among the most applicable arts and
>pseudo-sciences for non-medical improvement that are utilizable by the
>masses. Unfortunately, I see no long term algorithms in my work, which
>by definition, their should be, because algorithms are essentially,
>always “true.”
Nope. I find "is it useful?" a better question, than "is it true?"
The question "is it applicable here?" is also very important.
Ian Hobson
Lotus Notes Development skills available in East Midlands and Yorkshire, UK.
LCP qualified. e-mail for details: i...@choice.demon.co.uk
Brian R. Wilcox
Recently, I saw a warning not to open email that says deeyenda b/c it
is a computer virus that is transferred through the net. It can wipe
out a hard drive or scavenge for credit card numbers and other personal
info while it runs under OS/2 or Windows 95.
Has anyone ever thought to design a virus that's sole purpose is to
hunt and kill other viruses kinda like a Norton's Virus protector that
comes to life and transfers from machine to machine.
cyli
Nothing you read in Email can give you a virus. It would have to be an
attachment or something UUencoded and you'd have to .exe it (for the PC,
I don't know what for a Mac.). Just avoid running attachments if you're
really worried. Deeynaa or whatever is another hoax. Check with the
virus Web pages for heaps more information on them. Use a search engine
if you don't already know how to find the pages.
Yes, someone probably has thought of a virus that acts as an anti-virus.
But it's unlikely to do as much good as harm. Different OS's would
react differently. As would different progams. Some would be harmed. And
it's no kinder to infect someone else's machine with an anti-virus than
it is to drop some vodka in their morning orange juice because you think
that's the way your good days start. The analogy is a good way to notice
the ethics involved.
--
"If I die of curiosity, who will entertain you with naive questions?"
I only answer my mail on an average of once every two months. Be
patient.
Sabren (Michal)
On 14 Dec 1996, Brian R. Wilcox wrote:
Brian, :)
think about that for a minute. what does a virus have to do to
hurt your computer? it has to be 1) loaded into memory and 2) executed.
When you recieve mail, and view it on your screen, the text is stored in a
buffer somewhere in your computer's memory. In order to get that text to
execute on your computer, it would have to be injected into an executable
file or else somehow cause the instruction pointer (the little word (pair
of bytes) in your computer that tells it what to do next) to jump all the
way into that text buffer. But the instruction pointer is going to stay
within the windows code, and the code for the reader programs. (It gets a
little complicated with windows, because you probably have several virtual
instruction pointers, but i don't know for sure...)
in any case, there is no way that email can hurt your computer
unless 1) you save it or some part of it to disk AND THEN try to execute
it. 2) your reader program is ALREADY infected with a virus or trojan
horse that causes it to react differently when a particular string (such
as "deeyenda" appears in the message)..
however, that email DOES contain a virus: a virus of the mind.
richard brodie talked about a similar scam in his book by that name (Virus
of the Mind)... actually.. let me rephrase that.. the "deeyenda" message,
if it exists at all, is harmless.. but the WARNING about the deeyenda
virus is ITSELF a virus. It probably encourages you to copy it and
distribute it to your friends, yes? In any case, you've told us about it
here, so you've propagated the mind virus that says "deeyenda is a
computer virus".. interesting, huh?
vaccination is over. :) go back to class.
... michal
> Recently, I saw a warning not to open email that says deeyenda b/c it
> is a computer virus that is transferred through the net. It can wipe
> out a hard drive or scavenge for credit card numbers and other personal
> info while it runs under OS/2 or Windows 95.
>
> Has anyone ever thought to design a virus that's sole purpose is to
> hunt and kill other viruses kinda like a Norton's Virus protector that
> comes to life and transfers from machine to machine.
... [[SaBReN]]
----------------------------------------------------------------------------
mic...@efn.org | sab...@utdallas.edu | http://www.efn.org/~michal
----------------------------------------------------------------------------
John Harrington
There is one non-program destructive weapon (not a true virus) that can be
downloaded. There are several macros for Word and WordPerfect that can cause a
problem -- sometimes a serious problem, sometimes an annoyance. If you download
a word-processor file and view it in that word processor you can be at risk.
(I am -- well, was -- something of an authority on viruses. It was one of the
subjects I taught.)
Kevin Hogan
Carmine,
I think what you say is critical to the discussion. My concern however
is that flexibility is not necessarily quantifiable, though I
understand precisely what you mean. My only other thought is that I'd
have to ask you, "always?" Whew, that's a lot! I believe when you look
at any textual definition of algorithm, it implies far fewer steps than
flexibility does! On the other, other side of the coin, as far as
success using the tools in the tool box, you are 100% correct.
Flexibility IS the key.
Happy Holidays,
Kevin Hogan
In <01bbea2d$c929ac40$03594ccf@carmineb> "Carmine Baffa"
<cba...@inetnow.net> writes:
>
>There is a set of algorithms in NLP that work
>for me every time, and for anything. It goes
>like this.
>
>If something doesn't work, try something else.
>
>If you know what your outcome is, and you continue
>flexibly calibrating your communication to the response
>you are getting, then you will be effective.
>
>Now, in order to build a model where this is true,
>there may need to be other pieces in place.
>NLP was never designed around any individual
>technique. And, when using any individual
>technique there is always the chance were a
>that individual technique will fail. You cant test
>the validity of NLP based on any technique.
>There are far to many possibilities for Human
>Beings to always respond in accordance with
>any of the techniques which are evangelized
>in so many NLP trainings.
>
>Now, if we instead experience NLP from the foundation
>up; NLP presuppositions, rapport, behavioral flexibility,
>the tote model, etc, we then have an algorithm that
>will hold true. At least in my experience that has
>been the case.
>
>Carmine
>
>John Harrington <mes...@ix.netcom.com> wrote in article
><58rtqi$3...@dfw-ixnews10.ix.netcom.com>...
>> Ron,
>>
>> An algorithm is a set of steps that _always_ work. That is, if you
follow
>the
>> steps the result is always predictable. It is, ampng other things, a
>computer
>> programmer's term.
>>
>> Kevin was hoping to find algorithms for handling specific client
>problems, and
>> was summareizing his findings in the quest. His summary says that he
did
>not
>> find true algorithms for any treatment modality -- all are less
than
>100%
>> effective.
>>
>> Smiles,
>> John
>>
>>
>> Ron <rba...@dcn.davis.ca.us> wrote:
>>
>> >Kevin Hogan wrote:
>> >>
>> >> Algorithms in NLP and hypnosis?
>> >>
>>
Carmine Baffa
Carmine
> >> Algorithms in NLP and hypnosis?
> >>
>
Ian Hobson
In article <58tfjq$s...@dfw-ixnews10.ix.netcom.com>, "Brian R. Wilcox"
<mik...@ix.netcom.com> writes
>Recently, I saw a warning not to open email that says deeyenda b/c it
>is a computer virus that is transferred through the net. It can wipe
>out a hard drive or scavenge for credit card numbers and other personal
>info while it runs under OS/2 or Windows 95.
>
First there was "Good Times". Now this. Good Times was an e-mail,
entitled "Good Times" that warned that if you opened an e-mail entitled
"Good Times" your would catch a virus. Duhhhh!
Many people did - but not in the way they expected. They carefully
forwarded this e-mail to their friends and saved it in "warnings"
databases etc . The bloody thing spread like wildfire. It was a
virus in many definitions of the word, but it needed humans to forward
it in order to spread, and is had no payload. The original author must
have wet himself laughing at the way it went.
YOU CAN NOT CATCH A VIRUS BY OPENING AN INTERNET E_MAIL.
For those who are still worried, read the sig.
>Has anyone ever thought to design a virus that's sole purpose is to
>hunt and kill other viruses kinda like a Norton's Virus protector that
>comes to life and transfers from machine to machine.
Yes. Its called f-prot insn't it? It is spread by FTP.
Ian Hobson
YOU CAN NOT CATCH A VIRUS BY OPENING AN INTERNET E_MAIL.
There is no way standard e-mail packages will execute any code to place the
virus on your machine. An e-mail is text, not instructions.
You can catch a virus by opening an attached file with an application that has
an auto start macro. e.g Word, some spreadsheets etc. Browsing the www with a
Java or x-active aware browser can D/L a Java or A-active applet that could do
damage or give your machine a virus. I've never heard of it being done, and the
site would have to know it was there, but.....
The boot sector on a "non-booting" disc contains code that says "Non-system disk
or disk error, press any key to continue" - this code can carry a boot sector
virus which you can catch if you attempt to boot from the disk.
Any D/Led program (.com or .exe) file can carry a virus.
Non-standard e-mail packages can carry viruses and bombs. Buttons in Lotus Notes
can hold nasty code for example - but the button will not pass through SMTP
gateways into Internet Mail.
YOU CAN NOT CATCH A VIRUS BY OPENING AN INTERNET E_MAIL.
Jim Inman
me...@ix.netcom.com(Kevin Hogan) wrote:
>Carmine,
>I think what you say is critical to the discussion. My concern however
>is that flexibility is not necessarily quantifiable, though I
>Flexibility IS the key.
>Happy Holidays,
>Kevin Hogan
>In <01bbea2d$c929ac40$03594ccf@carmineb> "Carmine Baffa"
><cba...@inetnow.net> writes:
>>
>>> >> Algorithms in NLP and hypnosis?
>>> >>
>>>
>>> >Could someone simplify and summarize this post for me, especially
>the
>>> >part about algorithms and their use in NLP?
>>> >Thanks.
>>>
>>> >ron
>>>
>>>
>>>
Thanks, Ron. I was wondering the same thing. What the heck is this thread about?
Jim - inm...@intplsrv.net
Jim Inman
Saint Jo, Tx
inm...@cooke.net
Kevin Hogan
Hi Ian,
Good to see you today,
In <E$$eIBAKQ...@choice.demon.co.uk> Ian Hobson
<I...@choice.demon.co.uk> writes:
>
>Hi Kevin,
>
>I don't mean to be rude but why are you using such an unusual and
>unhelpful definition of "algorithm" and "NLP"?
>
I use Korzybski's definition of neuro-linguistic, as he coined the
term, I add on the programming which is pretty obvious, and algorithm
is simply defined as it was used in calc. II., that was confirmed with
Encyclopedia Britannica and an unabridged dictionary. I know there are
a lot of people here that went way beyond that, but if someone has a
better more accurate definition of algorithm or nlp, I'd love to have
them.
>I don't know where to start. This does not make sense.
>
>What is a semantic response? Sematics is the study of how meaning is
>encoded by structure. How can a stimulus, note or experience anything?
>Neuro - to do with nerves. Linguistic - to do with language.
>Nerolinguistic implies using language to alter something about how
>people think - not how they respond using language.
I'd have to differ with you on your definition of neuro-linguistic. It
doesn't imply using language to alter, but how they respond.
Programming is the attempt to alter. All communications alters the
human nervous system to some degree, so I can see where you are coming
from!
>
>Programming is installing, or changing - but not of language but
>behaviour and (possibly) beliefs.
>
>The person doing the study is the only one who needs to be aware of
what
>is going on. Neither the stimulator/therapist or the subject/patient
>need know. Satir and Erikson did not understand much of what Grinder
and
>Bandler noticed.
Very true...of course, Korzybski and Hiakawa knew before G&B were born!
(See Science and Sanity, Manhood of Humanity: Science and Art of Human
Engineering, Language in Thought and Action)
>
>>Definition: Algorithm- Systematic mathematical procedure that
produces
>Not systematic, but methodical. Algorithms have nothing necessarily to
>do with systematics.
Could have both definitions I suppose. I thought I had exhausted
possible definitions. Can you cite me a source to check my work?
>>in a finite number of steps, the answer to a questionor the solution
of
>>a problem.
>Many problems had algorithms that solve some cases, and go on looking
in
>other cases. These do not necessarily produce an answer, and clearly
are
>infinite.
I do not believe that meets any definition of an algorithm I have ever
heard of, but would be happy to consider a credible source. Where do I
look?
>
>You are also in danger of assumung each problem has an answer and an
>algorithmic solution. Many problems have no answer, while some have
more
>than one.
I don't believe any problem that is dealt with in therapy has an
algorithmic solution. I have never seen a procedure that always works,
methodical or systematic.
>
>Some problems have no algorithmic solution. You can scramble a
password
>in ways that make it impossible to unscramble it back.
Exactly!
>
>>The quesiton or problem in turn must belong to a class that
>>has an infinite number of numbers or members.
>Nonsence.
My sources are pretty good! Where would I go to find a different set of
definitions?
>
>>There are two classes of
>>algorithms. “Decidable class” is a class of algorithms that produces
a
>>yes or no response when completed.
>and they complete in finite time.
>
>>The second class is called the
>>“Computable Class.”
>I understood that a computable problem not only had a solution, but
this
>could be reached in finite time (although the method need not be
known).
Agree on all except the part in parentheses, I'd have to consider that
and see an example or three!)
>
>> An algorithm that leads to the answer (a specific number) is called
a
>>computational procedure.
>Why must the answer be a number?
What would an example be that isn't?
Please understand, I'm coming from a mathematical background in my
education. There are possibly other definitions, I simply can't think
of one at the moment.
>
>[snip]
>>Cause: New Behavior Generator Process
>>Desired Algorithmic Effects: Client utilizes best choices in all
cases.
>>
>>
>>The New Behavior Generator is one of my favorite NLP tools because it
>>regularly is effective in over half of the cases where I personally
>>utilize it. However, it is certain, in my work that it is not an
>>algorithm. We certainly can say that we have given the client more
>>choices, however, that in itself is not 100% useful. What is 100%
>>useful is that we give a client more choices and they actually
utilize
>>those choices at/or near 100% of the time. The distinction seems
fine,
>>but is indeed quite important. Therefore, we could argue that the New
>>Behavior Generator as a process is an algorithm for offering new
>>choices to a client, but in my experience, it cannot be argued that
the
>>choices are always utilized for superior behavior. The NBG is a very
>>useful process but it is not an algorithm per se for certain long or
>>short term change.
>
>NBG appears to be an algorithm for generating new behaviours, but not
an
>algorithm for what you want it for. This does not mean it isn't an
>algorithm.
Hmmmmm....what good is an algorithm if is not predictable! If 2+2 = 4
80% of the time and 3, 20% of the time, that is ....not really.... an
algorithm.
>[snip]
>>Grinder and Bandler noted in 1979 that nothing in NLP is true, it is
>>simply their hallucination of what is useful. An algorithm is a
>>verifyably true.
>
>You have clearly never tried to verify a significant algoritm as being
>true.
How so?
>
>Start with something simple. Does adding up a column of figures
produce
>the same answer as adding from the top down?
>
>>The fact that we are all very different in our personal history and
our
>>genetic make up makes algorithms for long term improvement almost
>>impossible.
>You assume they exist, can be coded and taught.
Mmmmm...often. But I see none in the strictest sense in changing human
behavior to a specific desired result for long term change that sticks.
>>In the short term almost anything can be a facilitator for
>>short term change. Almost anything that you change in a system will
>>cause at least short term change in behavior or subjective
experience.
>>In the long term, most processes significantly lose their benefit
>>unless repeatedly reinforced.
>So you build in reinforcement.
>
>>
>>NLP remains a useful adjunct to hypnotherapy and hypnotherapy a
useful
>>adjunct to NLP. They are both among the most applicable arts and
>>pseudo-sciences for non-medical improvement that are utilizable by
the
>>masses. Unfortunately, I see no long term algorithms in my work,
which
>>by definition, their should be, because algorithms are essentially,
>>always “true.”
>Nope. I find "is it useful?" a better question, than "is it true?"
>
Now that is a Bandlerism. True is superior to useful simply because if
it is true it is useful to know and utilize. If something is only
useful, it is ....useful....however it is not true because it is not an
algorithm.
>The question "is it applicable here?" is also very important.
Indeed!
Not rude at all! Just a disagreement!
Take care and please fill in anything you can to enhance my
understanding of what you are thinking!
Thanks,
Kevin Hogan
keith
cyli <cy...@visi.com> wrote:
>Nothing you read in Email can give you a virus. It would have to be an
>attachment or something UUencoded and you'd have to .exe it (for the PC,
>I don't know what for a Mac.). Just avoid running attachments if you're
>really worried. Deeynaa or whatever is another hoax. Check with the
>virus Web pages for heaps more information on them. Use a search engine
>if you don't already know how to find the pages.
>Yes, someone probably has thought of a virus that acts as an anti-virus.
> But it's unlikely to do as much good as harm. Different OS's would
>react differently. As would different progams. Some would be harmed. And
>it's no kinder to infect someone else's machine with an anti-virus than
>it is to drop some vodka in their morning orange juice because you think
>that's the way your good days start. The analogy is a good way to notice
>the ethics involved.
just don't drive if you take the above advice to heart.
keith
95 02:09:30 PDT....................*............
ke...@jacky.demon.co.uk...\{} _..oo \{}/*.......
--------------------------[/]/ {}/ [].........
MIME-Version: 1.0 _))))_/() ()__/[]_/_/.........
Content-Type: TEXT/PLAIN; charset=US-ASCII
"Every day try to put a smile on somebodys face, even if you
have to settle for yourself."
Richard Bandler
_
support FREE SPEECH & go to: http://www.McSpotlight.org/
read the truth about the UK longest ever civil trial.
._
Ian Hobson
In article <5902ss$q...@dfw-ixnews9.ix.netcom.com>, Kevin Hogan
<me...@ix.netcom.com> writes
>Hi Ian,
Hi Kevin,
>>I don't mean to be rude but why are you using such an unusual and
>>unhelpful definition of "algorithm" and "NLP"?
>>
>
>I use Korzybski's definition of neuro-linguistic, as he coined the
>term, I add on the programming which is pretty obvious, and algorithm
>is simply defined as it was used in calc. II., that was confirmed with
>Encyclopedia Britannica and an unabridged dictionary. I know there are
>a lot of people here that went way beyond that, but if someone has a
>better more accurate definition of algorithm or nlp, I'd love to have
>them.
>
It is my belief that choice of definitions is a very important choice
that one makes when setting out to do things.
Korzybski's definition is fine is you are studying simple things like
synax and semantics. People are influenced far more by your study than
sentences, and so a broader, "softer" definition is more appropriate
IMO.
[snip]
>>
>>>Definition: Algorithm- Systematic mathematical procedure that
>produces
>>Not systematic, but methodical. Algorithms have nothing necessarily to
>>do with systematics.
>
>Could have both definitions I suppose. I thought I had exhausted
>possible definitions. Can you cite me a source to check my work?
>
Kevin, our languages are only related. You speak American English, and I
speak Britsh English. I was only trying to clarify, because I felt your
statement was ambiguous.
>>>in a finite number of steps, the answer to a questionor the solution
>of
>>>a problem.
>>Many problems had algorithms that solve some cases, and go on looking
>in
>>other cases. These do not necessarily produce an answer, and clearly
>are
>>infinite.
>
>I do not believe that meets any definition of an algorithm I have ever
>heard of, but would be happy to consider a credible source. Where do I
>look?
I'm not prepared to argue semantics, or quote sources. As I have implied
before, if a definition is not suitable for the purpose of analysis,
then it is valid (even helpful) to use another definition that is.
>
[snip]
>>
>>>The quesiton or problem in turn must belong to a class that
>>>has an infinite number of numbers or members.
>>Nonsence.
>
>My sources are pretty good! Where would I go to find a different set of
>definitions?
I would be happy to find sence in the sentence you wrote. I don't. There
is more wrong here than a different set of presupositions.
What do you mean by the "class that has an infinite number of numbers of
members".
[snip]
>>I understood that a computable problem not only had a solution, but
>this
>>could be reached in finite time (although the method need not be
>known).
>
>Agree on all except the part in parentheses, I'd have to consider that
>and see an example or three!)
>
For many years it was known HOW to colour a map in four colours. Only
recently has it been proved that this can always be done. Until it was
proved, no method of colouring the map could be considered to be an
algorithm by your definition, because it was not known to terminate with
the right answer.
(The proof is beyond me!)
Note that it was always possible.
>>
>>> An algorithm that leads to the answer (a specific number) is called
>a
>>>computational procedure.
>>Why must the answer be a number?
>
>What would an example be that isn't?
An algorithm to work out what day of the week Christmas day is on in a
year, results in a day of the week. A hunter's algorithm for catching
food results in enough food to eat. One that searches for a word in a
dictionary finds out weither it is there or not.
>Please understand, I'm coming from a mathematical background in my
>education. There are possibly other definitions, I simply can't think
>of one at the moment.
I also have a Math background. 2i Paure Maths and Stats From Hull in 79.
(Bit rusty now though - grin)
[SNIP]
>>NBG appears to be an algorithm for generating new behaviours, but not
>an
>>algorithm for what you want it for. This does not mean it isn't an
>>algorithm.
>
>Hmmmmm....what good is an algorithm if is not predictable! If 2+2 = 4
>80% of the time and 3, 20% of the time, that is ....not really.... an
>algorithm.
It is not deterministic - but then neither are many things in life.
Supose you were a hunter, and caught half the prey you went after. If I
could show you another way (a better algorithm) that improved the catch
to 80% of the prey you went after, you would, maybe, consider it worth
while?
>
>>[snip]
>>>Grinder and Bandler noted in 1979 that nothing in NLP is true, it is
>>>simply their hallucination of what is useful. An algorithm is a
>>>verifyably true.
>>
>>You have clearly never tried to verify a significant algoritm as being
>>true.
>
>How so?
Researchers in computation have been trying to prove code is an accurate
implementation of the specification for years. All significant projects
are beyond current bran power to understand the specification, and
beyond current comuting power to prove correct.
>>Start with something simple. Does adding up a column of figures
>produce
>>the same answer as adding from the top down?
I was serious. If you can prove that adding a column of figures from top
to bottom *always* produces the same answer as adding it from bottom to
top, then you have proved something worth while knowing. (Paticularly if
you prove the answers can be different!)
[snip]
>>>masses. Unfortunately, I see no long term algorithms in my work,
>which
>>>by definition, their should be, because algorithms are essentially,
>>>always “true.”
>>Nope. I find "is it useful?" a better question, than "is it true?"
>>
>Now that is a Bandlerism. True is superior to useful simply because if
>it is true it is useful to know and utilize. If something is only
>useful, it is ....useful....however it is not true because it is not an
>algorithm.
That is one circular argument - and incorrect.
Firstly, it is impossible to know weither all statements you can make in
any logic system are true or false. Goedel proved this by the neat trick
of defining an algorithm that, given any logical system, produced a
statement in that system whose truth value was unknowable: if it was
true you could prove it was false, and if you assumed it false, you
could prove it was true.
So you can not know the truth or falsity of all statements, even if a
mathematical world. The real world is much messier.
Secondly, Most religions are ways people have of making sence of their
life and social situation. I would argue that these adherants find this
structure of thought useful. Without wishing to enter a flame war,
religions are mutually contradictory, and so can not all be true in any
absolute sence.
[snip]
>
>Not rude at all! Just a disagreement!
>Take care and please fill in anything you can to enhance my
>understanding of what you are thinking!
>
Hope that makes my thinking a bit clearer.
I think we agree that nothing will work 100% and for ever.
The difference between us is what we do about it.
I act, with what I have. You look for a better tool.
Stay well.
Carmine Baffa
Mostly, this post is about wasted bandwidth.
I joined in because I saw a chance to offer
some information about flexibility, and perhaps
to move the discussion in what I perceived to
be a more useful direction.
The entire problem with the subject of this thread
is in it premise. Thinking that if you are testing
any one technique, you are in fact testing NLP.
So if the technique fails, then NLP fails. That
is akin to saying that if a person can't do any
one thing, then that person cant do anything at all.
> >>>
> >>> Ron <rba...@dcn.davis.ca.us> wrote:
> >>>
> >>> >Kevin Hogan wrote:
> >>> >>
> >>> >> Algorithms in NLP and hypnosis?
> >>> >>
> >>>
mbo...@ukans.edu
How, then, would you define an algorithm for an iterative process?
That is, a process whereby you input an "answer" that is assumed to be
close to the correct one, but probably not exact, and the output
response (or answer) from the mathematical process is closer to the
correct solution. However, that, too is put back through the equation
or process and the procedure repeated until such time as the output
does not vary significantly from the input. Here, then, is a case
where you do not necessarily have an "exact" answer, and you may
choose to accept a result which is less precise than if you repeated
the process some number of further times. Nonetheless, the answer is
reasonably correct.
MBovee
UKansas
me...@ix.netcom.com(Kevin Hogan) electronically penned the following
missive:
Kevin Hogan
Hi Matt,
This is one of the exceptions to the rule that I mentioned are fairly
rare. Much like finding the square root of two and the closest result
for pi. In my definition what you suggest fits with what I consider an
algorithm. Many mathematicians wouldn't. The sequence of steps, start
point, end point, etc. are all identical every time the numbers are run
so, I bend to yes.
Take care, Kevin Hogan
Kevin Hogan
Hi Ian,
In <ce2IQHAo...@choice.demon.co.uk> Ian Hobson
<I...@choice.demon.co.uk> writes:
>
>
>In article <5902ss$q...@dfw-ixnews9.ix.netcom.com>, Kevin Hogan
><me...@ix.netcom.com> writes
>>Hi Ian,
>Hi Kevin,
>
>>I use Korzybski's definition of neuro-linguistic, as he coined the
>>term, I add on the programming which is pretty obvious, and algorithm
>>is simply defined as it was used in calc. II., that was confirmed
with
>>Encyclopedia Britannica and an unabridged dictionary. I know there
are
>>a lot of people here that went way beyond that, but if someone has a
>>better more accurate definition of algorithm or nlp, I'd love to have
>>them.
>>
>It is my belief that choice of definitions is a very important choice
>that one makes when setting out to do things.
Right, once a word is defined it is far easier to find common ground!
>
>Korzybski's definition is fine is you are studying simple things like
>synax and semantics.
Korzybski simple!? Geez! You're the first person I've heard say that!
People are influenced far more by your study than
>sentences, and so a broader, "softer" definition is more appropriate
>IMO.
OK. We could loosen the definition for this setting.
>[snip]
>>>
>Kevin, our languages are only related. You speak American English, and
I
>speak Britsh English. I was only trying to clarify, because I felt
your
>statement was ambiguous.
:-) I must admit that can happen!
>I'm not prepared to argue semantics, or quote sources. As I have
implied
>before, if a definition is not suitable for the purpose of analysis,
>then it is valid (even helpful) to use another definition that is.
>>
>I would be happy to find sence in the sentence you wrote. I don't.
There
>is more wrong here than a different set of presupositions.
>
>What do you mean by the "class that has an infinite number of numbers
of
>members".
>
Oops. A mis-spelling. OR numbeR of members. (Just like your 'sence' in
the above paragraph!) No secretarial skills here! :-)
>[snip]
>For many years it was known HOW to colour a map in four colours. Only
>recently has it been proved that this can always be done. Until it was
>proved, no method of colouring the map could be considered to be an
>algorithm by your definition, because it was not known to terminate
with
>the right answer.
>
>(The proof is beyond me!)
>
Interesting.
>Note that it was always possible.
>
>An algorithm to work out what day of the week Christmas day is on in a
>year, results in a day of the week.
That is an algorithm.
A hunter's algorithm for catching
>food results in enough food to eat.
This is not.
One that searches for a word in a
>dictionary finds out weither it is there or not.
>
This could be but is not necessarily an algorithm. It depends on the
sequencing.
>I also have a Math background. 2i Paure Maths and Stats From Hull in
79.
>(Bit rusty now though - grin)
Me too. Don't use the stuff enough, you don't remember all the tools.!
>[SNIP]
>
>It is not deterministic - but then neither are many things in life.
>
Right. That which is not very close to being deterministic doesn't fit
most definitions of algorithm. (Is this getting boring yet?!???!!?)
>Supose you were a hunter, and caught half the prey you went after. If
I
>could show you another way (a better algorithm) that improved the
catch
>to 80% of the prey you went after, you would, maybe, consider it worth
>while?
Very worthwhile but not likely to be an algorithm. It would have to
have steps that a computer or any common person could follow and
achieve identical results. It would need a specific sequence that would
recycle until the end is accomplished. I think we misunderstood each
others definition. Specific beginning point. Specific end point. Finite
amount of steps (a very lenient rule!) and a specific set of
actions/sequences that if followed will always bring you very very
close or right to the answer when done, by anyone.
>Researchers in computation have been trying to prove code is an
accurate
>implementation of the specification for years. All significant
projects
>are beyond current bran power to understand the specification, and
>beyond current comuting power to prove correct.
Interesting.
>
>I was serious. If you can prove that adding a column of figures from
top
>to bottom *always* produces the same answer as adding it from bottom
to
>top, then you have proved something worth while knowing. (Paticularly
if
>you prove the answers can be different!)
>
Sure. No question. There is so much in neurolinguistic programming that
is so utterly useful. I would never argue that point. However, the way
I have always understood, "algorithm", and later (this week!) checked,
there is nothing that I see as algorithmic.! There are arguably
algorithms if we loosen the definition though!
>[snip]
>That is one circular argument - and incorrect.
>
>Firstly, it is impossible to know weither all statements you can make
in
>any logic system are true or false. Goedel proved this by the neat
trick
>of defining an algorithm that, given any logical system, produced a
>statement in that system whose truth value was unknowable: if it was
>true you could prove it was false, and if you assumed it false, you
>could prove it was true.
>
Right. This is where everything in the world breaks down. However,
truth is always superior to usefulness. Usefullness doesn't include
truth. Truth includes usefulness.
>So you can not know the truth or falsity of all statements, even if a
>mathematical world. The real world is much messier.
Absolutely. We played the philosophical mind game at some level in
college. Don't recall which. But it was an exercise in futility.
>
>Secondly, Most religions are ways people have of making sence of their
>life and social situation. I would argue that these adherants find
this
>structure of thought useful. Without wishing to enter a flame war,
>religions are mutually contradictory, and so can not all be true in
any
>absolute sence.
This is 100% correct. There can at most be one correct religion and it
probably is "wrong" as well! I think this means we are in agreement and
now simply dealing with semantics for the most part!
>
>[snip]
>>
>Hope that makes my thinking a bit clearer.
>
Yes, it does.!
>I think we agree that nothing will work 100% and for ever.
Yes.
>
>The difference between us is what we do about it.
>
>I act, with what I have. You look for a better tool.
Maybe both. When I do therapy, I don't always look for a better tool.
Same when I'm doing research. BUT, I admit to looking for superior
tools with some regularity. (A touch of pesky perfectionism...a
character flaw?!??!)
>
>Stay well.
You too!
>Ian Hobson
Kevin Hogan
>
>Lotus N
David M. Pierce
To learn about common myths about computer viruses & "Trojan Horses", I
recommend visiting the informative (and entertaining) web page listed
below: