Re: JESS: "publications" / Forgy paper

[James C. Owen]

Exactamundo!  He (Dr. Forgy) is "very" protective about the Rete 2 algorithm.
Seems that he *had* to make the Rete algorithm public since it was funded by
the DOD when he was at CMU, but methinks that he wants to make some money with
the Rete 2 algorithm.  The Rete 2 works on optimizing memory space thereby
increasing the speed.  Actually, he's working on a Rete 3 but I have no idea
what it is; only that it will make the process even faster than Rete 2.
Cool

However, I'll see if we can get his Rete Ph.D. thesis scanned and published -
maybe this weekend.  If so, I'll send it to Dr. Friedman-Hill to publish on the
PST site if he likes.  BTW, Dr. Forgy's email is [EMAIL PROTECTED], not
[EMAIL PROTECTED]  :-)

One other thing; I've been going over the various Java systems (Jess, OPSJ,
JRules, Advisor and JEOPS) the past couple of weeks and trying to run the
manners 128 benchmark on all of them using the same set of data  and,
basically, the same set of rules.  That way only the engine is the difference
factor.  If anyone already has the JEOPS worked out, I would appreciate it if
you would send it to me.  I got CLIPS and converted it straight over to Jess -
which wasn't terribly difficult since they both seem to be be LISP oriented.
(Jim really hates LISP!  All those parentheses and special markers for
variables and just the jumbled way things are handled...)  And I already had
the code for OPSJ, Advisor and JRules.  JEOPS seems to be a different animal
since it is more Java-kinda oriented syntax.  Whatever.  Just a thought.

SDG
jco

-
James C. Owen
Senior KE
Knowledgebased Systems Corporation
6314 Kelly Circle
Garland, TX   75044
972.530.2895


[EMAIL PROTECTED] wrote:

> I think Oliver Hoffmann wrote:
> > James,
> >
> > if you could scan Dr. Forgy's thesis and convert it to pdf, that would be
> > great :)
> >
> > is there also a publication on the Rete II algorithm?
> > http://www.pst.com/
> > claims that it would be much faster...
>
> Nope, it's unpublished and proprietary.
>
> -
> Ernest Friedman-Hill
> Distributed Systems ResearchPhone: (925) 294-2154
> Sandia National LabsFAX:   (925) 294-2234
> Org. 8920, MS 9012  [EMAIL PROTECTED]
> PO Box 969  http://herzberg.ca.sandia.gov
> Livermore, CA 94550
>
> 
> To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
> in the BODY of a message to [EMAIL PROTECTED], NOT to the list
> (use your own address!) List problems? Notify [EMAIL PROTECTED]
> 



To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]

Re: JESS: Modifying facts that are created from a definstancenot working

[Richard Long]

We've created a defadvice for the modify function to automatically update
the shadow facts without resetting the engine.

Here it is, for anyone else who may need this behavior:

(defadvice after modify
(if (?retval isShadow)
  then
   (bind ?_object (fact-slot-value ?retval OBJECT))
   (bind ?_bi (call java.beans.Introspector getBeanInfo (?_object
getClass)))
   ;Now, grab the PropertyDescriptors.  We just need to test the first
one for isBound.
   (bind ?_pd (call ?_bi getPropertyDescriptors))
   (if (< (length$ ?_pd) 2) ;No properties still give 1 descriptor.
 then
  (return ?retval)
   )
   ;If so, we definstance dynamic, if not, then static.
   (if (call (nth$ 2 ?_pd) isBound)
 then
;(printout t "Bound fact modified")
   (undefinstance ?_object)
   (return (definstance (call (?retval getDeftemplate) getBaseName)
?_object dynamic))
 else
   (undefinstance ?_object)
   (return (definstance (call (?retval getDeftemplate) getBaseName)
?_object static))
   )
)
)
- Original Message -
From: <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, February 01, 2002 1:51 PM
Subject: Re: JESS: Modifying facts that are created from a definstance not
working


> I think Villanueva, Ruben wrote:
> [Charset iso-8859-1 unsupported, filtering to ASCII...]
> >   When I try to modify a fact created from a definstance of a Java
> > bean that has property change support the fact does not get modified
> > right away if the defistance function was called with the static
> > parameter.
>
> Yes, by definition. That's how it's supposed to work.
>
> >
> > My understaing is that the modify functionMop is suppose to modify a
> >  fact directly whether the fact was created with property support or
> > not.
> >
>
> Well, no. When you use (modify) on a shadow fact, the properties of
> the Bean are changed, and Jess relies on propertyChangeEvents() to
> then be sent to modify the fact. If you've used static, then there are
> no such events, so the shadow fact isn't updated until you call
> (reset).
>
>
> -
> Ernest Friedman-Hill
> Distributed Systems ResearchPhone: (925) 294-2154
> Sandia National LabsFAX:   (925) 294-2234
> Org. 8920, MS 9012  [EMAIL PROTECTED]
> PO Box 969  http://herzberg.ca.sandia.gov
> Livermore, CA 94550
>
> 
> To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
> in the BODY of a message to [EMAIL PROTECTED], NOT to the list
> (use your own address!) List problems? Notify [EMAIL PROTECTED]
> 
>



To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]

Re: JESS: Modifying facts that are created from a definstancenot working

[ejfried]

I think Villanueva, Ruben wrote:
[Charset iso-8859-1 unsupported, filtering to ASCII...]
>   When I try to modify a fact created from a definstance of a Java
> bean that has property change support the fact does not get modified
> right away if the defistance function was called with the static
> parameter.

Yes, by definition. That's how it's supposed to work.

>
> My understaing is that the modify functionMop is suppose to modify a
>  fact directly whether the fact was created with property support or
> not.
>

Well, no. When you use (modify) on a shadow fact, the properties of
the Bean are changed, and Jess relies on propertyChangeEvents() to
then be sent to modify the fact. If you've used static, then there are
no such events, so the shadow fact isn't updated until you call
(reset).


-
Ernest Friedman-Hill  
Distributed Systems ResearchPhone: (925) 294-2154
Sandia National LabsFAX:   (925) 294-2234
Org. 8920, MS 9012  [EMAIL PROTECTED]
PO Box 969  http://herzberg.ca.sandia.gov
Livermore, CA 94550


To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]

JESS: Modifying facts that are created from a definstance notworking

[Villanueva, Ruben]

  When I try to modify a fact created from a definstance of a Java bean that has 
property change support the fact does not get modified right away if the defistance 
function was called with the static parameter. Using the Simple.java bean that comes 
with Jess, here are the steps that I took and the results.


When creating the fact using definstance with the dynamic parameter the modify 
function seems to work

C:\amber\Jess\Jess60>java -cp .  jess.Main

Jess, the Java Expert System Shell
Copyright (C) 1998 E.J. Friedman Hill and the Sandia Corporation
Jess Version 6.0 12/7/2001

Jess> (defclass simple jess.examples.simple.Simple)
jess.examples.simple.Simple
Jess> (ppdeftemplate simple)
"(deftemplate MAIN::simple extends MAIN::__fact \"$JAVA-OBJECT$ 
jess.examples.simple.Simple\"
 (slot class (default ))
 (slot fraction (default ))
 (slot name (default ))
 (slot serial (default ))
 (slot truth (default ))
 (slot OBJECT (type 2048)))"
Jess>  (bind ?sb (new jess.examples.simple.Simple Dave))

Jess> (definstance simple ?sb dynamic)

Jess> (facts)
f-0   (MAIN::simple (class ) (fraction 0.0) (name 
"Dave") (serial 0) (truth FALSE) (OBJECT 
))
For a total of 1 facts.
Jess> (defrule change_name ?f <- (simple (name "Dave")) => (modify ?f (name "Tom")))
TRUE
Jess> (rules)
MAIN::change_name
For a total of 1 rules.
Jess> (ppdefrule change_name)
"(defrule MAIN::change_name
   ?f <- (MAIN::simple (name \"Dave\"))
   =>
   (modify ?f (name \"Tom\")))"
Jess> (watch all)
TRUE
Jess> (run)
FIRE 1 MAIN::change_name f-0
 <== f-0 (MAIN::simple (class ) (fraction 0.0) (name 
"Dave") (serial 0) (truth FALSE)
(OBJECT ))
 ==> f-0 (MAIN::simple (class ) (fraction 0.0) (name 
"Tom") (serial 0) (truth FALSE) (OBJECT 
))
1
Jess> (facts)
f-0   (MAIN::simple (class ) (fraction 0.0) (name 
"Tom") (serial 0) (truth FALSE) (OBJECT 
))
For a total of 1 facts.
Jess> (jess-version-number)
6.0
Jess>

When creating the fact using definstance with the static parameter the modify function 
does not seem to work

C:\amber\Jess\Jess60>java -cp .  jess.Main

Jess, the Java Expert System Shell
Copyright (C) 1998 E.J. Friedman Hill and the Sandia Corporation
Jess Version 6.0 12/7/2001

Jess> (defclass simple jess.examples.simple.Simple)
jess.examples.simple.Simple
Jess> (ppdeftemplate simple)
"(deftemplate MAIN::simple extends MAIN::__fact \"$JAVA-OBJECT$ 
jess.examples.simple.Simple\"
 (slot class (default ))
 (slot fraction (default ))
 (slot name (default ))
 (slot serial (default ))
 (slot truth (default ))
 (slot OBJECT (type 2048)))"
Jess>  (bind ?sb (new jess.examples.simple.Simple Dave))

Jess> (definstance simple ?sb static)

Jess> (facts)
f-0   (MAIN::simple (class ) (fraction 0.0) (name 
"Dave") (serial 0) (truth FALSE) (OBJECT 
))
For a total of 1 facts.
Jess> (defrule change_name ?f <- (simple (name "Dave")) => (modify ?f (name "Tom")))
TRUE
Jess> (rules)
MAIN::change_name
For a total of 1 rules.
Jess> (ppdefrule change_name)
"(defrule MAIN::change_name
   ?f <- (MAIN::simple (name \"Dave\"))
   =>
   (modify ?f (name \"Tom\")))"
Jess> (watch all)
TRUE
Jess> (run)
FIRE 1 MAIN::change_name f-0
1
Jess> (facts)
f-0   (MAIN::simple (class ) (fraction 0.0) (name 
"Dave") (serial 0) (truth FALSE) (OBJECT 
))
For a total of 1 facts.
C:\amber\Jess\Jess60>java -cp .  jess.Main

Jess, the Java Expert System Shell
Copyright (C) 1998 E.J. Friedman Hill and the Sandia Corporation
Jess Version 6.0 12/7/2001

Jess> (defclass simple jess.examples.simple.Simple)
jess.examples.simple.Simple
Jess> (ppdeftemplate simple)
"(deftemplate MAIN::simple extends MAIN::__fact \"$JAVA-OBJECT$ 
jess.examples.simple.Simple\"
 (slot class (default ))
 (slot fraction (default ))
 (slot name (default ))
 (slot serial (default ))
 (slot truth (default ))
 (slot OBJECT (type 2048)))"
Jess>  (bind ?sb (new jess.examples.simple.Simple Dave))

Jess> (definstance simple ?sb static)

Jess> (facts)
f-0   (MAIN::simple (class ) (fraction 0.0) (name 
"Dave") (serial 0) (truth FALSE) (OBJECT 
))
For a total of 1 facts.
Jess> (defrule change_name ?f <- (simple (name "Dave")) => (modify ?f (name "Tom")))
TRUE
Jess> (rules)
MAIN::change_name
For a total of 1 rules.
Jess> (ppdefrule change_name)
"(defrule MAIN::change_name
   ?f <- (MAIN::simple (name \"Dave\"))
   =>
   (modify ?f (name \"Tom\")))"
Jess> (watch all)
TRUE
Jess> (run)
FIRE 1 MAIN::change_name f-0
1
Jess> (facts)
f-0   (MAIN::simple (class ) (fraction 0.0) (name 
"Dave") (serial 0) (truth FALSE) (OBJECT 
))
For a total of 1 facts.


My understaing is that the modify functionMop is suppose to modify a  fact directly 
whether the fact was created with property support or not.

When I modify a fact that was created without the use of the definstance funtion the 
fact gets modified right away.

C:\amber\Jess\Jess60>java -cp .  jess.Main

Jess, the Java Expert System Shell
Copyright (C) 1998 E.J. Friedman Hill and the Sandia Corporation
Jess Version 6.0 12/7/2001

Jess> (assert (shopping-list milk eggs tp bread butter))

J

Re: JESS: "publications" / Forgy paper

[ejfried]

I think Oliver Hoffmann wrote:
> James,
> 
> if you could scan Dr. Forgy's thesis and convert it to pdf, that would be 
> great :)
> 
> is there also a publication on the Rete II algorithm?
> http://www.pst.com/
> claims that it would be much faster...

Nope, it's unpublished and proprietary.

-
Ernest Friedman-Hill  
Distributed Systems ResearchPhone: (925) 294-2154
Sandia National LabsFAX:   (925) 294-2234
Org. 8920, MS 9012  [EMAIL PROTECTED]
PO Box 969  http://herzberg.ca.sandia.gov
Livermore, CA 94550


To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]

Re: JESS: "publications" / Forgy paper

[Oliver Hoffmann]

James,

if you could scan Dr. Forgy's thesis and convert it to pdf, that would be 
great :)

is there also a publication on the Rete II algorithm?
http://www.pst.com/
claims that it would be much faster...

his email is defunct, it seems (see error quoted below)

:) Oliver

At 03:24 PM 1/02/2002 +, [EMAIL PROTECTED] wrote:
>Hi. This is the qmail-send program at relay.pair.com.
>I'm afraid I wasn't able to deliver your message to the following addresses.
>This is a permanent error; I've given up. Sorry it didn't work out.
>
><[EMAIL PROTECTED]>:
>206.46.164.23 does not like recipient.
>Remote host said: 550 Invalid recipient: <[EMAIL PROTECTED]>
>Giving up on 206.46.164.23.


At 06:47 AM 1/02/2002 -0800, [EMAIL PROTECTED] wrote:

>Another option, of course, is to write to the author of the paper and
>see if you can get a freebie. Dr. Forgy can be reached at
>[EMAIL PROTECTED]

At 11:50 AM 1/02/2002 -0600, James C. Owen wrote:
>Wow!  I've haven't seen this much interest in the paper(s) for a long, long
>time.  It is, however, refreshing, to see that there is considerable interest
>in the underlying Rete algorithm from so many people (er, 
>persons?).  Anyway, I
>contacted Dr. Forgy some time ago (like two weeks?) about both papers (his
>dissertation and the AI publication) and he does not have either one in
>electronic form.  He does have a copy of his thesis in hard copy.  His 1979
>thesis is 178 pages that we "could" scan in, but that is not the best way.
>I've been toying with the idea of using some OCR software to scan in the whole
>thing and then "re-do" the diagrams in a more modern format.  I'm 99.4% sure
>that he would have no objection since the entire thing was done under a DOD
>grant.  And, he's a really nice person.
>
>On the other hand, Ernest is 100% correct that the AI publication is
>copyrighted and we cannot publish, display or show where to get that one.  My
>only suggestion would be to get the publisher to put it out on their web site
>for down loads for a small fee, say $5 or something like that.  Ernest's other
>suggestion is also good; go to your local university library archives and make
>your own copies.  I've done that before and, although it takes some time, it's
>worth the effort.
>
>For most of our group, the discussion that Ernest put together on his web site
>should be sufficient.  It does show the high-level view of the Rete algorithm
>and is quite understandable.  The code for implementing is freely available
>from JEOPS, http://www.di.ufpe.br/~jeops/ so you can see how to do that 
>part as
>well.  One final note:  I've noticed that most who get the thesis glance 
>at the
>first chapter and then put it on the shelf to study when things are less 
>hectic
>and they never get around to actually reading the good stuff.  But, it looks
>nice on your personal library shelf, right?  Just my two cents.  :-)
>
>SDG
>jco
>
>--
>James C. Owen
>Senior KE
>Knowledgebased Systems Corporation
>6314 Kelly Circle
>Garland, TX   75044
>972.530.2895
>
>
>Agustin Gonzalez wrote:
>
> > Maybe it's time for one of us, especially those of us that have the paper
> > and are trained, to create a formal paper describing the public domain RETE
> > algorithm and make the paper publicly available? I can do that if there is
> > enough interest and enough people agrees with me that I won;t have any
> > legal problems (I don't think so because it is public domain).
> >
> > ---
> > Agustin Gonzalez, Principal
> > Town Lake Software
> > www.townlakesoftware.com
> > (512) 248-9839
> >
> > On Friday, February 01, 2002 8:44 AM, James Patterson
> > [SMTP:[EMAIL PROTECTED]] wrote:
> > > I can tell you that it is a waste of time trying to contact the
> > > publisher... An Elsevier representative told me that the issue is out of
> > > print and they do not have reprint service anymore. They offered to
> > > individually reprint the entire journal issue for a couple of hundred
> > > dollars (my boss said no).  I'm not a student and therefore can't get
> > > any cooperation from the local universities.  I even emailed Dr Forgy
> > > (and tried to find a coworker of his during a stint at Ericcson) about
> > > the Elsevier problems (but they didn't respond - and I started feeling
> > > like a stalker so I gave up).
> > >
> > > Good luck,
> > > James
> > >
> > > -Original Message-
> > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
> > > On Behalf Of Oliver Hoffmann
> > > Sent: Friday, February 01, 2002 6:54 AM
> > > To: Jess Mailing List
> > > Subject: JESS: "publications" / Forgy paper
> > >
> > >
> > > Hi :)
> > >
> > > I highly doubt that the contents of published academic papers are
> > > generally
> > > copyright protected. What might be copyright protected is the specific
> > > paper version of a specific paper as distributed by a specific
> > > publisher.
> > > The point of publishing research results is to make them readily
> > > available,
> > >

Re: JESS: Rule Activation

[ejfried]

(defrule r1
 (a ?n)
 (a ?n2&:(< ?n ?n2))
 =>
 (assert (c (+ ?n ?n2

I think Francisco Dario wrote:
> 
> Hi everyone.
> I'm relative new with JESS...
> 
> I have the following problem:
> 
> consider this rule:
> 
> (defrule r1
> (a ?n)
> (a ?n2)
> =>
> (assert (c (+ ?n ?n2
> 
> What I want is a rule that asserts the fact c only when there are two 
> distinct facts ex.: (a 1) and (a 3) generates (c 4)
> 
> The first problem I had, was that when I asserted the first fact and execute 
> the "(run)", the rule "fired". I understood the question, and rewrote the 
> rule:
> 
> (defrule r1
>   ?fact1 <-(a ?n)
>   ?fact2 <-(a ?n2)
>   (test (not (eq ?fact1 ?fact2)))
> =>
> (assert (c (+ ?n ?n2
> 
> This way, the same fact won't activate the rule. But now I have another 
> problem: When I assert the second fact, this new rule is activated twice! I 
> see why. But I want the rule to be activated only once! How can I overcome 
> this problem?
> 
> 
> Thanx for any help.
> Dario
> 
> 
> 
> 
> 
> 
> 
> 
> 
> _
> Chat with friends online, try MSN Messenger: http://messenger.msn.com
> 
> 
> 
> To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
> in the BODY of a message to [EMAIL PROTECTED], NOT to the list
> (use your own address!) List problems? Notify [EMAIL PROTECTED]
> 
> 



-
Ernest Friedman-Hill  
Distributed Systems ResearchPhone: (925) 294-2154
Sandia National LabsFAX:   (925) 294-2234
Org. 8920, MS 9012  [EMAIL PROTECTED]
PO Box 969  http://herzberg.ca.sandia.gov
Livermore, CA 94550


To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]

JESS: Rule Activation

[Francisco Dário]

Hi everyone.
I'm relative new with JESS...

I have the following problem:

consider this rule:

(defrule r1
(a ?n)
(a ?n2)
=>
(assert (c (+ ?n ?n2

What I want is a rule that asserts the fact c only when there are two 
distinct facts ex.: (a 1) and (a 3) generates (c 4)

The first problem I had, was that when I asserted the first fact and execute 
the "(run)", the rule "fired". I understood the question, and rewrote the 
rule:

(defrule r1
  ?fact1 <-(a ?n)
  ?fact2 <-(a ?n2)
  (test (not (eq ?fact1 ?fact2)))
=>
(assert (c (+ ?n ?n2

This way, the same fact won't activate the rule. But now I have another 
problem: When I assert the second fact, this new rule is activated twice! I 
see why. But I want the rule to be activated only once! How can I overcome 
this problem?


Thanx for any help.
Dario









_
Chat with friends online, try MSN Messenger: http://messenger.msn.com



To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]

Re: JESS: "publications" / Forgy paper

[James C. Owen]

Wow!  I've haven't seen this much interest in the paper(s) for a long, long
time.  It is, however, refreshing, to see that there is considerable interest
in the underlying Rete algorithm from so many people (er, persons?).  Anyway, I
contacted Dr. Forgy some time ago (like two weeks?) about both papers (his
dissertation and the AI publication) and he does not have either one in
electronic form.  He does have a copy of his thesis in hard copy.  His 1979
thesis is 178 pages that we "could" scan in, but that is not the best way.
I've been toying with the idea of using some OCR software to scan in the whole
thing and then "re-do" the diagrams in a more modern format.  I'm 99.4% sure
that he would have no objection since the entire thing was done under a DOD
grant.  And, he's a really nice person.

On the other hand, Ernest is 100% correct that the AI publication is
copyrighted and we cannot publish, display or show where to get that one.  My
only suggestion would be to get the publisher to put it out on their web site
for down loads for a small fee, say $5 or something like that.  Ernest's other
suggestion is also good; go to your local university library archives and make
your own copies.  I've done that before and, although it takes some time, it's
worth the effort.

For most of our group, the discussion that Ernest put together on his web site
should be sufficient.  It does show the high-level view of the Rete algorithm
and is quite understandable.  The code for implementing is freely available
from JEOPS, http://www.di.ufpe.br/~jeops/ so you can see how to do that part as
well.  One final note:  I've noticed that most who get the thesis glance at the
first chapter and then put it on the shelf to study when things are less hectic
and they never get around to actually reading the good stuff.  But, it looks
nice on your personal library shelf, right?  Just my two cents.  :-)

SDG
jco

--
James C. Owen
Senior KE
Knowledgebased Systems Corporation
6314 Kelly Circle
Garland, TX   75044
972.530.2895


Agustin Gonzalez wrote:

> Maybe it's time for one of us, especially those of us that have the paper
> and are trained, to create a formal paper describing the public domain RETE
> algorithm and make the paper publicly available? I can do that if there is
> enough interest and enough people agrees with me that I won;t have any
> legal problems (I don't think so because it is public domain).
>
> ---
> Agustin Gonzalez, Principal
> Town Lake Software
> www.townlakesoftware.com
> (512) 248-9839
>
> On Friday, February 01, 2002 8:44 AM, James Patterson
> [SMTP:[EMAIL PROTECTED]] wrote:
> > I can tell you that it is a waste of time trying to contact the
> > publisher... An Elsevier representative told me that the issue is out of
> > print and they do not have reprint service anymore. They offered to
> > individually reprint the entire journal issue for a couple of hundred
> > dollars (my boss said no).  I'm not a student and therefore can't get
> > any cooperation from the local universities.  I even emailed Dr Forgy
> > (and tried to find a coworker of his during a stint at Ericcson) about
> > the Elsevier problems (but they didn't respond - and I started feeling
> > like a stalker so I gave up).
> >
> > Good luck,
> > James
> >
> > -Original Message-
> > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
> > On Behalf Of Oliver Hoffmann
> > Sent: Friday, February 01, 2002 6:54 AM
> > To: Jess Mailing List
> > Subject: JESS: "publications" / Forgy paper
> >
> >
> > Hi :)
> >
> > I highly doubt that the contents of published academic papers are
> > generally
> > copyright protected. What might be copyright protected is the specific
> > paper version of a specific paper as distributed by a specific
> > publisher.
> > The point of publishing research results is to make them readily
> > available,
> > therefore "copyright" is simply the wrong approach here. More
> > information on http://www.cogsci.soton.ac.uk/~harnad/Tp/resolution.htm
> > I don't want to encourage anyone to re-distribute scanned journal pages
> > without authorization etc., but there are better ways to deal with
> > academic
> > content than waiting for a library employee to get a photocopy 
> > within
> > some weeks. I highly recommend self-archiving content and making these
> > copies available over personal or university web sites and I also highly
> >
> > recommend getting original articles from the people or institutions that
> >
> > created them - everything else is pre-internet and a waste of time and
> > effort.
> >
> > :) Oliver Hoffmann
> >
> > At 04:23 AM 1/02/2002 -0800, [EMAIL PROTECTED] wrote:
> > >Hi Folks,
> > >
> > >This comes up periodically, so please forgive the intrusion while I
> > >discuss it, once again.
> > >
> > >The Charles Forgy paper on the Rete algorithm was published in an
> > >academic journal called "Artificial Intelligence." The contents are
> > >protected by

RE: JESS: "publications" / Forgy paper

[Agustin Gonzalez]

Maybe it's time for one of us, especially those of us that have the paper 
and are trained, to create a formal paper describing the public domain RETE 
algorithm and make the paper publicly available? I can do that if there is 
enough interest and enough people agrees with me that I won;t have any 
legal problems (I don't think so because it is public domain).

---
Agustin Gonzalez, Principal
Town Lake Software
www.townlakesoftware.com
(512) 248-9839


On Friday, February 01, 2002 8:44 AM, James Patterson 
[SMTP:[EMAIL PROTECTED]] wrote:
> I can tell you that it is a waste of time trying to contact the
> publisher... An Elsevier representative told me that the issue is out of
> print and they do not have reprint service anymore. They offered to
> individually reprint the entire journal issue for a couple of hundred
> dollars (my boss said no).  I'm not a student and therefore can't get
> any cooperation from the local universities.  I even emailed Dr Forgy
> (and tried to find a coworker of his during a stint at Ericcson) about
> the Elsevier problems (but they didn't respond - and I started feeling
> like a stalker so I gave up).
>
> Good luck,
> James
>
> -Original Message-
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
> On Behalf Of Oliver Hoffmann
> Sent: Friday, February 01, 2002 6:54 AM
> To: Jess Mailing List
> Subject: JESS: "publications" / Forgy paper
>
>
> Hi :)
>
> I highly doubt that the contents of published academic papers are
> generally
> copyright protected. What might be copyright protected is the specific
> paper version of a specific paper as distributed by a specific
> publisher.
> The point of publishing research results is to make them readily
> available,
> therefore "copyright" is simply the wrong approach here. More
> information on http://www.cogsci.soton.ac.uk/~harnad/Tp/resolution.htm
> I don't want to encourage anyone to re-distribute scanned journal pages
> without authorization etc., but there are better ways to deal with
> academic
> content than waiting for a library employee to get a photocopy 
> within
> some weeks. I highly recommend self-archiving content and making these
> copies available over personal or university web sites and I also highly
>
> recommend getting original articles from the people or institutions that
>
> created them - everything else is pre-internet and a waste of time and
> effort.
>
> :) Oliver Hoffmann
>
> At 04:23 AM 1/02/2002 -0800, [EMAIL PROTECTED] wrote:
> >Hi Folks,
> >
> >This comes up periodically, so please forgive the intrusion while I
> >discuss it, once again.
> >
> >The Charles Forgy paper on the Rete algorithm was published in an
> >academic journal called "Artificial Intelligence." The contents are
> >protected by copyright law. Although you can make a photocopy for your
> >own use, it's illegal to (for instance) scan it into your computer and
> >post the result on the Internet. Whatever you do, do -not- post a copy
> >to this mailing list, nor even post the URL where it might be
> >downloaded. This goes for any other copyrighted material as well, of
> >course. If you're interested in getting a copy of this paper (or any
> >academic paper):
> >
> >1) Go to your local public library, or the library of your local
> >college or University; or ask a student friend or a friend in another
> >country to do so.
> >
> >2) Check their collection for the appropriate issue of the journal.
> >It's 20 years old, so may be on microfilm or archived in some other
> >way. Ask the librarian for help -- that's why they're there.
> >
> >3) If they have it, great, make yourself a copy.
> >
> >4) If they don't, go to the front desk and ask them to get a copy
> >through the "Interlibrary Loan" program. They should be able to get it
> >within a few weeks.
> >
> >
> >
> >-
> >Ernest Friedman-Hill
> >Distributed Systems ResearchPhone: (925) 294-2154
> >Sandia National LabsFAX:   (925) 294-2234
> >Org. 8920, MS 9012  [EMAIL PROTECTED]
> >PO Box 969  http://herzberg.ca.sandia.gov
> >Livermore, CA 94550
> >
> >
> >To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
>
> >in the BODY of a message to [EMAIL PROTECTED], NOT to the list (use
> >your own address!) List problems? Notify [EMAIL PROTECTED]
> >
>
>
> 
> To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
> in the BODY of a message to [EMAIL PROTECTED], NOT to the list (use
> your own address!) List problems? Notify [EMAIL PROTECTED]
> 
>
>
> 
> To unsubscribe, send the words 'unsubscribe jess-us

Re: JESS: "publications" / Forgy paper

[Oliver Hoffmann]

Ernest,

At 06:47 AM 1/02/2002 -0800, [EMAIL PROTECTED] wrote:

>I think Oliver Hoffmann wrote:
> > Hi :)
> >
> > I highly doubt that the contents of published academic papers are 
> generally
> > copyright protected.
>
>Oliver,
>
>I don't want to start a flame war, or a discussion about intellectual
>freedom or any related topics. I'm not saying that I advocate or decry
>any of the practices outlined here -- but I do want to set the record
>straight. Therefore I should point out that this particular article,
>has the following bit of text on the front page:
>
>   0004-3702/82/-/$02.75 (C) North-Holland
>
>meaning that the contents are copyrighted by the North-Holland

meaning that the stack of paper is (copyright) North-Holland

and not meaning that the intellectual content is (copyright) North-Holland

the intellectual content was published, with the intent of public use

>publishing concern, and they'll gladly sell you a reprint for
>$2.75.

and that is exactely the extend of their (copyright): paper reprints of 
this particular paper version

>Do realize that we're talking about a paper from 1982, and it's
>exceedingly likely that the original was prepared on a
>typewriter. There was no Web on which to post the text so the question
>is moot. Generally, when you submit academic papers to a journal or
>conference nowadays (speaking as someone who has done so) they tell
>you that you may post an electronic copy after publication (sometimes
>after a specific time interval has elapsed) but they virtually always
>require that their copyright notice be included on the paper.

... and this copyright notice refers to paper copies of the original paper 
version

the only thing this (copyright) is good for is for stopping different 
publishers from making money with re-publishing papers

>When you submit a paper you transfer the copyright of the material to
>the journal, and they are therefore legally entitled to dictate what
>can and what cannot be done with the content.

One of the advantages of publishing content is that you don't have to 
repeat the same arguments over and over again. Therefore, I will just refer 
you to
http://www.cogsci.soton.ac.uk/~harnad/Tp/resolution.htm#1.3
for the details of why a journal can not completely determine what you can 
or cannot do with a scientific publication.

>Now, all that said and done, I did a bit of web searching (not clear
>why the OP didn't simply do this first.) North-Holland is part of
>Elsevier Science these days. Here's a web form for ordering reprints:
>
>   http://www.elsevier.nl/homepage/guestbook/?form=reprints
>
>Another option, of course, is to write to the author of the paper and
>see if you can get a freebie. Dr. Forgy can be reached at
>[EMAIL PROTECTED]

I have cc-d this email to Dr. Forgy, maybe he wants to tell us whether he 
thinks the content of his publication can be distributed.

:) Oliver

At 01:54 PM 1/02/2002 +0100, Oliver Hoffmann wrote:
>Hi :)
>
>I highly doubt that the contents of published academic papers are 
>generally copyright protected. What might be copyright protected is the 
>specific paper version of a specific paper as distributed by a specific 
>publisher. The point of publishing research results is to make them 
>readily available, therefore "copyright" is simply the wrong approach 
>here. More information on
>http://www.cogsci.soton.ac.uk/~harnad/Tp/resolution.htm
>I don't want to encourage anyone to re-distribute scanned journal pages 
>without authorization etc., but there are better ways to deal with 
>academic content than waiting for a library employee to get a photocopy 
> within some weeks. I highly recommend self-archiving content and 
>making these copies available over personal or university web sites and I 
>also highly recommend getting original articles from the people or 
>institutions that created them - everything else is pre-internet and a 
>waste of time and effort.
>
>:) Oliver Hoffmann
>
>At 04:23 AM 1/02/2002 -0800, [EMAIL PROTECTED] wrote:
>>Hi Folks,
>>
>>This comes up periodically, so please forgive the intrusion while I
>>discuss it, once again.
>>
>>The Charles Forgy paper on the Rete algorithm was published in an
>>academic journal called "Artificial Intelligence." The contents are
>>protected by copyright law. Although you can make a photocopy for your
>>own use, it's illegal to (for instance) scan it into your computer and
>>post the result on the Internet. Whatever you do, do -not- post a copy
>>to this mailing list, nor even post the URL where it might be
>>downloaded. This goes for any other copyrighted material as well, of
>>course. If you're interested in getting a copy of this paper (or any
>>academic paper):
>>
>>1) Go to your local public library, or the library of your local
>>college or University; or ask a student friend or a friend in another
>>country to do so.
>>
>>2) Check their collection for the appropriate issue of the
>>journal. It's 20 years old, so may be on micr

Re: JESS: "publications" / Forgy paper

[ejfried]

I think Oliver Hoffmann wrote:
> Hi :)
> 
> I highly doubt that the contents of published academic papers are generally 
> copyright protected.

Oliver,

I don't want to start a flame war, or a discussion about intellectual
freedom or any related topics. I'm not saying that I advocate or decry
any of the practices outlined here -- but I do want to set the record
straight.  Therefore I should point out that this particular article,
has the following bit of text on the front page:

  0004-3702/82/-/$02.75 (C) North-Holland

meaning that the contents are copyrighted by the North-Holland
publishing concern, and they'll gladly sell you a reprint for
$2.75. Do realize that we're talking about a paper from 1982, and it's
exceedingly likely that the original was prepared on a
typewriter. There was no Web on which to post the text so the question
is moot. Generally, when you submit academic papers to a journal or
conference nowadays (speaking as someone who has done so) they tell
you that you may post an electronic copy after publication (sometimes
after a specific time interval has elapsed) but they virtually always
require that their copyright notice be included on the paper.

When you submit a paper you transfer the copyright of the material to
the journal, and they are therefore legally entitled to dictate what
can and what cannot be done with the content.

Now, all that said and done, I did a bit of web searching (not clear
why the OP didn't simply do this first.) North-Holland is part of
Elsevier Science these days. Here's a web form for ordering reprints:

  http://www.elsevier.nl/homepage/guestbook/?form=reprints

Another option, of course, is to write to the author of the paper and
see if you can get a freebie. Dr. Forgy can be reached at
[EMAIL PROTECTED] 


-
Ernest Friedman-Hill  
Distributed Systems ResearchPhone: (925) 294-2154
Sandia National LabsFAX:   (925) 294-2234
Org. 8920, MS 9012  [EMAIL PROTECTED]
PO Box 969  http://herzberg.ca.sandia.gov
Livermore, CA 94550


To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]

RE: JESS: "publications" / Forgy paper

[James Patterson]

I can tell you that it is a waste of time trying to contact the
publisher... An Elsevier representative told me that the issue is out of
print and they do not have reprint service anymore. They offered to
individually reprint the entire journal issue for a couple of hundred
dollars (my boss said no).  I'm not a student and therefore can't get
any cooperation from the local universities.  I even emailed Dr Forgy
(and tried to find a coworker of his during a stint at Ericcson) about
the Elsevier problems (but they didn't respond - and I started feeling
like a stalker so I gave up).

Good luck,
James

-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
On Behalf Of Oliver Hoffmann
Sent: Friday, February 01, 2002 6:54 AM
To: Jess Mailing List
Subject: JESS: "publications" / Forgy paper


Hi :)

I highly doubt that the contents of published academic papers are
generally 
copyright protected. What might be copyright protected is the specific 
paper version of a specific paper as distributed by a specific
publisher. 
The point of publishing research results is to make them readily
available, 
therefore "copyright" is simply the wrong approach here. More
information on http://www.cogsci.soton.ac.uk/~harnad/Tp/resolution.htm
I don't want to encourage anyone to re-distribute scanned journal pages 
without authorization etc., but there are better ways to deal with
academic 
content than waiting for a library employee to get a photocopy 
within 
some weeks. I highly recommend self-archiving content and making these 
copies available over personal or university web sites and I also highly

recommend getting original articles from the people or institutions that

created them - everything else is pre-internet and a waste of time and
effort.

:) Oliver Hoffmann

At 04:23 AM 1/02/2002 -0800, [EMAIL PROTECTED] wrote:
>Hi Folks,
>
>This comes up periodically, so please forgive the intrusion while I 
>discuss it, once again.
>
>The Charles Forgy paper on the Rete algorithm was published in an 
>academic journal called "Artificial Intelligence." The contents are 
>protected by copyright law. Although you can make a photocopy for your 
>own use, it's illegal to (for instance) scan it into your computer and 
>post the result on the Internet. Whatever you do, do -not- post a copy 
>to this mailing list, nor even post the URL where it might be 
>downloaded. This goes for any other copyrighted material as well, of 
>course. If you're interested in getting a copy of this paper (or any 
>academic paper):
>
>1) Go to your local public library, or the library of your local 
>college or University; or ask a student friend or a friend in another 
>country to do so.
>
>2) Check their collection for the appropriate issue of the journal. 
>It's 20 years old, so may be on microfilm or archived in some other 
>way. Ask the librarian for help -- that's why they're there.
>
>3) If they have it, great, make yourself a copy.
>
>4) If they don't, go to the front desk and ask them to get a copy 
>through the "Interlibrary Loan" program. They should be able to get it 
>within a few weeks.
>
>
>
>-
>Ernest Friedman-Hill
>Distributed Systems ResearchPhone: (925) 294-2154
>Sandia National LabsFAX:   (925) 294-2234
>Org. 8920, MS 9012  [EMAIL PROTECTED]
>PO Box 969  http://herzberg.ca.sandia.gov
>Livermore, CA 94550
>
>
>To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'

>in the BODY of a message to [EMAIL PROTECTED], NOT to the list (use 
>your own address!) List problems? Notify [EMAIL PROTECTED]
>



To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list (use
your own address!) List problems? Notify [EMAIL PROTECTED]




To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]

Re: JESS: Retracting facts

[ejfried]

I think Dusan Sormaz wrote:
> After playing few more tests I got the picture. :--)

Good. Yes, the idea is that, although  HashMap doesn't have a
"setHashCode()" function, its hashcode changes every time you put an
object into it. Jess therefore has problems with HashMaps, in the same
way that HashMaps have problems with Test objects.

> 
> 
> Dusan Sormaz
> 
> At 08:09 AM 2/1/02 -0500, you wrote:
> >Ernest,
> >
> >a small observation. This test example does not work because hashCode has 
> >been reimplemented inTest class, not because hash code for Test object 
> >(used in HashMap) has been changed. If method hashCode is deleted (or 
> >commented out, then everything is OK).
> >So HashMap uses hashCode() method to retrieve the key. There is no method 
> >setHashCode () in Object class.
> >
> >Dusan Sormaz
> >
> >At 01:19 PM 1/31/02 -0800, you wrote:
> 
> ***
> * Dusan Sormaz, PhD, Assistant Professor
> * Ohio University
> * Industrial and Manufacturing Systems Engineering Department
> * 280 Stocker Center, Athens, OH 45701-2979
> * phone: (740) 593-1548
> * fax:   (740) 593-0778
> * e-mail: [EMAIL PROTECTED]
> * url: http://www.ent.ohiou.edu/~sormaz
> ***



-
Ernest Friedman-Hill  
Distributed Systems ResearchPhone: (925) 294-2154
Sandia National LabsFAX:   (925) 294-2234
Org. 8920, MS 9012  [EMAIL PROTECTED]
PO Box 969  http://herzberg.ca.sandia.gov
Livermore, CA 94550


To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]

Re: JESS: Retracting facts

[Dusan Sormaz]

After playing few more tests I got the picture. :--) 

Dusan Sormaz
At 08:09 AM 2/1/02 -0500, you wrote:
Ernest,
a small observation. This test example does not work because hashCode has
been reimplemented inTest class, not because hash code for Test object
(used in HashMap) has been changed. If method hashCode is deleted (or
commented out, then everything is OK).
So HashMap uses hashCode() method to retrieve the key. There is no method
setHashCode () in Object class.
Dusan Sormaz
At 01:19 PM 1/31/02 -0800, you wrote:


***
* Dusan Sormaz, PhD, Assistant
Professor 

* Ohio University
* Industrial and Manufacturing Systems Engineering Department
* 280 Stocker Center, Athens, OH 45701-2979
* phone: (740) 593-1548 
* fax:   (740) 593-0778  
* e-mail: [EMAIL PROTECTED] 
* url:
http://www.ent.ohiou.edu/~sormaz

***

Re: JESS: Retracting facts

[Dusan Sormaz]

Ernest,
a small observation. This test example does not work because hashCode has
been reimplemented inTest class, not because hash code for Test object
(used in HashMap) has been changed. If method hashCode is deleted (or
commented out, then everything is OK).
So HashMap uses hashCode() method to retrieve the key. There is no method
setHashCode () in Object class.
Dusan Sormaz
At 01:19 PM 1/31/02 -0800, you wrote:
Jess assumes (and in this case, the
assumption is unwarranted) that
any objects in the slots of a fact will have constant hashCode
values
-- i.e., the hashCode of the contents of a slot will not change
while
the object is in a slot. This is true for "value" objects like
Strings
and Integers, and, actually, most objects. Some classes work
differently, however. HashMap, like many containers, defines
equals()
and hashCode() to both change their results depending on the
contents
of the container. This drives Jess nuts, as Jess goes to great
lengths
to index things according to their identity and hashCode, and if
these
things change unexpectedly, chaos ensues.
You have to read between the lines of manual sections 3.4 and the
second paragraph of 4.7 to see some hint of this; I should probably
spell this out more clearly.
Before you start to think there's something "wrong" with Jess
for
behaving this way, note that HashMap itself gets just as upset if
you
meddle with its contents. The following program puts some objects
into
a HashMap, then changes the hashCode of the keys. When you try to
retrieve those objects using the modified keys, HashMap can't find
them! Once you use an object as a key in a HashMap, you must treat
it
as immutable. 
You could perhaps make Jess work properly by using a trivial
"wrapper"
for the HashMap before putting it into a slot, as long as that
wrapper
had a constant hashCode.

--Example program --
import java.util.HashMap;
public class Test {
    private int m_i;
    public Test(int i) {
    m_i = i;
    }
    public void setHashCode(int i) {
    m_i = i;
    }
    public int hashCode() {
    return m_i;
    }
    public boolean equals(Object o) {
    if (o instanceof Test) {
    return
m_i == ((Test) o).m_i;
    } else
    return
false;
    }
    public static void main(String[] argv) {
    HashMap map = new
HashMap();
    Test t1 = new Test(1);
    Test t2 = new Test(2);
    
    map.put(t1,
"tee-one");
    map.put(t2,
"tee-two");
   
System.out.println(map.get(t1));
   
System.out.println(map.get(t2));
    
    t1.setHashCode(9);
    t2.setHashCode(23);
    
   
System.out.println(map.get(t1));
   
System.out.println(map.get(t2));
    }
}

I think Marc Shlomo L. Hronec wrote:
> Hi,
> 
> I've experienced problems with retracting facts under the
following
> circumstances: a) when it contained a reference to a java object and
b) when
> I accessed the member methods of that object. It can't be retracted
but it
> can be modified: that produces two modified copies (!). The same
problem
> appears to exist with definstanced facts.
> 
> See the sample code and output below:
> 
> (clear)
> (reset)
> (defclass map java.util.HashMap)
> (watch all)
> (definstance map (new java.util.HashMap) static)
> (retract 1)
> (facts)
> 
> (definstance map (new java.util.HashMap) static)
> (call (fact-slot-value (fact-id 2) OBJECT) put "key1"
"val1")
> (call (fact-slot-value (fact-id 2) OBJECT) get
"key1")
> (retract 2)
> (facts)
> 
> (clear)
> (reset)
> (deftemplate map (slot object))
> (watch all)
> (assert (map (object (new java.util.HashMap
> (retract 1)
> (facts)
> 
> (assert (map (object (new java.util.HashMap
> (call (fact-slot-value (fact-id 2) object) put "key1"
"val1")
> (call (fact-slot-value (fact-id 2) object) get
"key1")
> (retract 2)
> (facts)
> 
> (modify 2 (object nil))
> (facts)
> 
> 
> ; OUTPUT 
> 
> Jess> TRUE
> Jess> java.util.HashMap
> Jess> TRUE
> Jess>  ==> f-1 (MAIN::map (class
) (empty
> TRUE) (OBJECT ))
> 
> Jess>  <== f-1 (MAIN::map (class
) (empty
> TRUE) (OBJECT ))
> TRUE
> Jess> f-0   (MAIN::initial-fact)
> For a total of 1 facts.
> Jess>  ==> f-2 (MAIN::map (class
) (empty
> TRUE) (OBJECT ))
> 
> Jess> Jess> "val1"
> Jess> TRUE
> Jess> f-0   (MAIN::initial-fact)
> f-2   (MAIN::map (class
) (empty TRUE)
> (OBJECT ))
> For a total of 2 facts.
> Jess> TRUE
> Jess> TRUE
> Jess> TRUE
> Jess> TRUE
> Jess>  ==> f-1 (MAIN::map (object
))
> 
> Jess>  <== f-1 (MAIN::map (object
))
> TRUE
> Jess> f-0   (MAIN::initial-fact)
> For a total of 1 facts.
> Jess>  ==> f-2 (MAIN::map (object
))
> 
> Jess> Jess> "val1"
> Jess> TRUE
> Jess> f-0   (MAIN::initial-fact)
> f-2   (MAIN::map (object
))
> For a total of 2 facts.
> Jess> 
> Jess> f-0   (MAIN::initial-fact)
> f-2   (MAIN::map (object nil))
> f-2   (MAIN::map (object nil))
> For a total of 3 facts.
> Jess>

JESS: "publications" / Forgy paper

[Oliver Hoffmann]

Hi :)

I highly doubt that the contents of published academic papers are generally 
copyright protected. What might be copyright protected is the specific 
paper version of a specific paper as distributed by a specific publisher. 
The point of publishing research results is to make them readily available, 
therefore "copyright" is simply the wrong approach here. More information on
http://www.cogsci.soton.ac.uk/~harnad/Tp/resolution.htm
I don't want to encourage anyone to re-distribute scanned journal pages 
without authorization etc., but there are better ways to deal with academic 
content than waiting for a library employee to get a photocopy  within 
some weeks. I highly recommend self-archiving content and making these 
copies available over personal or university web sites and I also highly 
recommend getting original articles from the people or institutions that 
created them - everything else is pre-internet and a waste of time and effort.

:) Oliver Hoffmann

At 04:23 AM 1/02/2002 -0800, [EMAIL PROTECTED] wrote:
>Hi Folks,
>
>This comes up periodically, so please forgive the intrusion while I
>discuss it, once again.
>
>The Charles Forgy paper on the Rete algorithm was published in an
>academic journal called "Artificial Intelligence." The contents are
>protected by copyright law. Although you can make a photocopy for your
>own use, it's illegal to (for instance) scan it into your computer and
>post the result on the Internet. Whatever you do, do -not- post a copy
>to this mailing list, nor even post the URL where it might be
>downloaded. This goes for any other copyrighted material as well, of
>course. If you're interested in getting a copy of this paper (or any
>academic paper):
>
>1) Go to your local public library, or the library of your local
>college or University; or ask a student friend or a friend in another
>country to do so.
>
>2) Check their collection for the appropriate issue of the
>journal. It's 20 years old, so may be on microfilm or archived in some
>other way. Ask the librarian for help -- that's why they're there.
>
>3) If they have it, great, make yourself a copy.
>
>4) If they don't, go to the front desk and ask them to get a copy
>through the "Interlibrary Loan" program. They should be able to get it
>within a few weeks.
>
>
>
>-
>Ernest Friedman-Hill
>Distributed Systems ResearchPhone: (925) 294-2154
>Sandia National LabsFAX:   (925) 294-2234
>Org. 8920, MS 9012  [EMAIL PROTECTED]
>PO Box 969  http://herzberg.ca.sandia.gov
>Livermore, CA 94550
>
>
>To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
>in the BODY of a message to [EMAIL PROTECTED], NOT to the list
>(use your own address!) List problems? Notify [EMAIL PROTECTED]
>



To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]

JESS: Forgy paper

[ejfried]

Hi Folks, 

This comes up periodically, so please forgive the intrusion while I
discuss it, once again.

The Charles Forgy paper on the Rete algorithm was published in an
academic journal called "Artificial Intelligence." The contents are
protected by copyright law. Although you can make a photocopy for your
own use, it's illegal to (for instance) scan it into your computer and
post the result on the Internet. Whatever you do, do -not- post a copy
to this mailing list, nor even post the URL where it might be
downloaded. This goes for any other copyrighted material as well, of
course. If you're interested in getting a copy of this paper (or any
academic paper):

1) Go to your local public library, or the library of your local
college or University; or ask a student friend or a friend in another
country to do so.

2) Check their collection for the appropriate issue of the
journal. It's 20 years old, so may be on microfilm or archived in some
other way. Ask the librarian for help -- that's why they're there.

3) If they have it, great, make yourself a copy. 

4) If they don't, go to the front desk and ask them to get a copy
through the "Interlibrary Loan" program. They should be able to get it
within a few weeks.



-
Ernest Friedman-Hill  
Distributed Systems ResearchPhone: (925) 294-2154
Sandia National LabsFAX:   (925) 294-2234
Org. 8920, MS 9012  [EMAIL PROTECTED]
PO Box 969  http://herzberg.ca.sandia.gov
Livermore, CA 94550


To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the list
(use your own address!) List problems? Notify [EMAIL PROTECTED]

Re: JESS: I want a paper about Rete Algorithm

[Mahesh Gandhe]

 hi friends,
I  got the paper from the AI journal (1987), I will post its information soon.
Regards,
Mahesh Gandhe.
  Harvinder Singh <[EMAIL PROTECTED]> wrote: 

Dear Friends, 
It is better if someone shares the most of popular production system algorithms! 
Harvinder 
  ÀÌ°æÇö[Lee Kyoung Hyoun,ì°ÌÝúÛ] <[EMAIL PROTECTED]> wrote: 




Hello,
I study in Expert System..
and I study in JESS
I want a paper about Rete Algorithm 
 
This is Title about Rete algorithm...
Forgy, C.L.: Rete: A Fast Algorithm for the Many Pattern/Many Object    Pattern Match Problem    Artificial Intelligence, 19(1982) 17-37
Please..
 
Thank you for your interest...
 
Have a happy days..
 I will give you every place where you set your foot(Joshua 1:3)  Lee Kyoung Hyoun([EMAIL PROTECTED])   Opentel Co., Ltd.   Telecom Tech. R&D Lab.   Tel : 82-2-579-7857~8   http://www.opentel.co.kr- 


Do You Yahoo!?Yahoo! Auctions Great stuff seeking new owners! Bid now!Do You Yahoo!?
Yahoo! Auctions Great stuff seeking new owners! Bid now!