3:44 p.m.
Ciao everybody, I have a question I would like to submit to this community,
given Drools is my preferred choice for Expert Systems :)
A quick background: I'm not an Expert Systems / Rule engines expert, but
I'm a happy power user "of Rules" since my CS Engineering thesis, a paper
published, and used it with happy results at work and hobby projects.
Currently boarded on a new project with Rules+Processes at work, and still
very happy, so happy, I'm using Drools again in a new hobby project.
[semi-brag-mode off]
Now, in recent times, I cannot ignore what (apparently) is a strong
renaissance of Functional Programming, specifically all this momentum which
Scala and Java 8's Lambda's are getting.
The Question I'm asking myself: is there a specific, or few, areas where
Functional Programming could actually benefit Expert Systems so much to
actually spawn new paradigms of writing Rules? Or benefit in a new
generation of the Rete algorithm?
Or actually, Functional Programming will spawn a completely different
paradigm of system which will go beyond Expert Systems the way are designed
today?
Ignore my rant if you feel it's too generic, but I really wanted to seize a
chance to ask the question to this community :)
Thank you for your feedback,
Ciao,
Matteo
1:06 a.m.
There is no way the irreconcilable difference between a production
rule system (PRS) and functional programming (FP). The former is based
on state and mutable data, in facts and even temporary objects
resulting from evaluation (no matter how the underlying engine works).
There is one (frequently repeated) step in the PRS execution logic
that benefits from FP ideas: RHS evaluation. It's best practice, even
mandatory, to avoid side effects in any methods invoked during this
step. OO programming (as in Java) invites (nominal) violations of this
rule, e.g., during an accumulate that isn't written recursively (and
the current interface prescribes otherwise), but they don't matter
since it is "below the counter".
Removing state and mutability is frequently acceptable - Drools has
sequential mode where facts cannot be changed. However, with all the
apps I've seen and written, the rule design patterns I've identified
and documented state in facts is too valuable.
-W
On 07/03/2013, Matteo Mortari <matteo.mortari(a)gmail.com> wrote:
Ciao everybody, I have a question I would like to submit to this
community,
given Drools is my preferred choice for Expert Systems :)
A quick background: I'm not an Expert Systems / Rule engines expert, but
I'm a happy power user "of Rules" since my CS Engineering thesis, a paper
published, and used it with happy results at work and hobby projects.
Currently boarded on a new project with Rules+Processes at work, and still
very happy, so happy, I'm using Drools again in a new hobby project.
[semi-brag-mode off]
Now, in recent times, I cannot ignore what (apparently) is a strong
renaissance of Functional Programming, specifically all this momentum which
Scala and Java 8's Lambda's are getting.
The Question I'm asking myself: is there a specific, or few, areas where
Functional Programming could actually benefit Expert Systems so much to
actually spawn new paradigms of writing Rules? Or benefit in a new
generation of the Rete algorithm?
Or actually, Functional Programming will spawn a completely different
paradigm of system which will go beyond Expert Systems the way are designed
today?
Ignore my rant if you feel it's too generic, but I really wanted to seize a
chance to ask the question to this community :)
Thank you for your feedback,
Ciao,
Matteo
11:10 p.m.
On Thu, Mar 7, 2013 at 2:44 PM, Matteo Mortari <matteo.mortari(a)gmail.com> wrote:
The Question I'm asking myself: is there a specific, or few,
areas where
Functional Programming could actually benefit Expert Systems so much to
actually spawn new paradigms of writing Rules? Or benefit in a new
generation of the Rete algorithm?
Or actually, Functional Programming will spawn a completely different
paradigm of system which will go beyond Expert Systems the way are designed
today?
Great question, very inspiring just to think of the possibilities. The
biggest opportunity out there right now is for rules-engines to help
bridge the artificial gap that exists between the so called
fundamental models of computation for the different camps. For
example, Java is a Turing style sequential language and Lisp is a
Church style functional language. This shoe-horning of languages into
different camps is arbitrary, boring, mentally restrictive, totally
artificial, and offers little more benefits than easily separating
books on a bookshelf at the bookstore. It is really, really sad to see
the quelling of creativity and reduction of the beautiful parts of
different computational models reduced into tiny, little, lonely camps
that must live in isolation. Digging into rules engines is a pathway,
albeit deeply unpopular, into another important form computational
expression that never really got it's day in the sun like the biggies
today. New and interesting things will benefit all camps when they dig
into the world of production systems and welcome it as a first class
citizen to their programming toolbox. Would it result in new and great
things? Who knows :).
11:19 a.m.
On 11 Mar 2013, at 03:10, Grant Rettke <grettke(a)acm.org> wrote:
On Thu, Mar 7, 2013 at 2:44 PM, Matteo Mortari
<matteo.mortari(a)gmail.com> wrote:
> The Question I'm asking myself: is there a specific, or few, areas where
> Functional Programming could actually benefit Expert Systems so much to
> actually spawn new paradigms of writing Rules? Or benefit in a new
> generation of the Rete algorithm?
> Or actually, Functional Programming will spawn a completely different
> paradigm of system which will go beyond Expert Systems the way are designed
> today?
Great question, very inspiring just to think of the possibilities. The
biggest opportunity out there right now is for rules-engines to help
bridge the artificial gap that exists between the so called
fundamental models of computation for the different camps. For
example, Java is a Turing style sequential language and Lisp is a
Church style functional language. This shoe-horning of languages into
different camps is arbitrary, boring, mentally restrictive, totally
artificial, and offers little more benefits than easily separating
books on a bookshelf at the bookstore. It is really, really sad to see
the quelling of creativity and reduction of the beautiful parts of
different computational models reduced into tiny, little, lonely camps
that must live in isolation. Digging into rules engines is a pathway,
albeit deeply unpopular, into another important form computational
expression that never really got it's day in the sun like the biggies
today.
+1 :)
I think the work in expert system tools is the only real paradigm shift going on. All
others are just cross pollination of the same ideas.
The downside is to get expert system tools to a level where we can compete easily, as a
more general purpose tool, is very hard. Research
expert system tools stopped in the mid 90s, due to the AI winter:
http://blog.athico.com/2012/05/drools-54-artificial-intelligence.html
We have plenty of ideas, the problem is in balancing trying to build a system suitable for
production use - where people are conservative and don't like change - and a system
which is constantly innovating.
The other issue is in finding people willing to research these ideas with us. It's
very easy for someone to get into scala or jruby or what ever and just start adding
features. It's very hard in our case, because the change is too much, it's very
hard for people to see how to apply what they already know to this new paradigm.
For instance here is an idea I've been mulling over recently (note just random ideas,
not roadmap). We don't have the resources to completely compete against other
languages, and I'm not sure we should do. With Java8 and 9 it starts to become a more
compelling language. Java8 brings lambs and java9 solve the perm gen issue (so we can stop
using MVEL, and just go pure code generation and not worry about blowing the perm gen).
Could we create a system that better layers with java9?
We already want to drop dialects for DRL (it just makes things too hard to explain and
write IDE tooling for). So the idea is to have a single language. This made me think of
the Pizza language, which layered neatly ontop of java, and eventually became scala.
http://pizzacompiler.sourceforge.net
So thinking really long term here. Can we build a "java layer" that provide all
the rule functionality we need - but fit ontop of the java language neatly. We'd
probably need to allow "rule" keyword and have it in Classes, at the method
level. All class members and methods would be available to the rules in that class. Can
probably think of that containing class as a RuleModule. We'd have to rely on CDI and
annotations for declarative wiring of member fields (
http://java.dzone.com/articles/cdi-di-p1). The biggest challenge would be in extending the
java language to support LHS literals too (something I've wanted to do anyway) to
provide LinQ style features, backed underneath by the Drools engine. A list of layered
features might look like:
1) PropertyAccessors (doesn't look like java will ever add this, but DRL needs it to
remain compact)
2) Type Inference (same as above, some argue that lack of contract design is a bad
thing…..)
3) inline casts (needed for units, and keeps DRL compact)
4) "when" literals, like linq - but supports our entire LHS language.
5) rule keyword (similar at method level).
We can use CDI to provide declarative injection of member variables.
public class MyClass {
@Inject
public MyService service; // CDI injected service
public void someMethod() {
}
rule xxx when {
…. // normal LHS here.
} then {
service.call(…..)
someMethod();
}
}
I won't cover LHS literals here, but look at reactive LinkQ, to get some ideas.
There is a lot more to this idea of a layered language, and we can exploit annotations in
all sorts of ways. Similar to my previous email on meta approach to building rule systems.
We could use annotations to wire RuleModules to KnowledgeBases, and vice versa.
One of the challenges of doing this layered option, is that the JDK is GPLd + linking
exception. We'd even need to develop and maintain our own antlr+asm layer that does
100% of java9 + our extensions, or GPLd+linked exception Drools and extend the JDK.
Mark
New and interesting things will benefit all camps when they dig
into the world of production systems and welcome it as a first class
citizen to their programming toolbox. Would it result in new and great
things? Who knows :).
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
2:03 p.m.
On 11 March 2013 16:19, Mark Proctor <mproctor(a)codehaus.org> wrote:
So thinking really long term here. Can we build a "java
layer" that
provide all the rule functionality we need - but fit ontop of the java
language neatly. We'd probably need to allow "rule" keyword and have it in
Classes, at the method level. All class members and methods would be
available to the rules in that class.
There is this production rule system where you can write your rules
embedded in compiling units, and alongside the legacy program units, of a
procedural, modular, strongly type HLL, and where you use expressions in
the language's own syntax in constraints...
We've been using it ever since 1986. Rabbi Akiva was right, wasn't he ;-)
-W
4:47 p.m.
There were a number of research efforts that looked at combining procedural and rule base
programming.
This one is quite interesting:
Combining Rule-Based and Procedural Programming in the XC and XE Programming Lanaugages
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.9.1106&rep=r...
Mark
On 11 Mar 2013, at 18:03, Wolfgang Laun <wolfgang.laun(a)gmail.com> wrote:
On 11 March 2013 16:19, Mark Proctor <mproctor(a)codehaus.org>
wrote:
So thinking really long term here. Can we build a "java layer" that provide all
the rule functionality we need - but fit ontop of the java language neatly. We'd
probably need to allow "rule" keyword and have it in Classes, at the method
level. All class members and methods would be available to the rules in that class.
There is this production rule system where you can write your rules embedded in compiling
units, and alongside the legacy program units, of a procedural, modular, strongly type
HLL, and where you use expressions in the language's own syntax in constraints...
We've been using it ever since 1986. Rabbi Akiva was right, wasn't he ;-)
-W
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
2:04 a.m.
Indeed, this thesis mentions a few features of the system I was
talking about :-)
([13])
-W
On 11/03/2013, Mark Proctor <mproctor(a)codehaus.org> wrote:
There were a number of research efforts that looked at combining
procedural
and rule base programming.
This one is quite interesting:
Combining Rule-Based and Procedural Programming in the XC and XE Programming
Lanaugages
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.9.1106&rep=r...
Mark
On 11 Mar 2013, at 18:03, Wolfgang Laun <wolfgang.laun(a)gmail.com> wrote:
> On 11 March 2013 16:19, Mark Proctor <mproctor(a)codehaus.org> wrote:
> So thinking really long term here. Can we build a "java layer" that
> provide all the rule functionality we need - but fit ontop of the java
> language neatly. We'd probably need to allow "rule" keyword and have it
in
> Classes, at the method level. All class members and methods would be
> available to the rules in that class.
>
> There is this production rule system where you can write your rules
> embedded in compiling units, and alongside the legacy program units, of a
> procedural, modular, strongly type HLL, and where you use expressions in
> the language's own syntax in constraints...
>
> We've been using it ever since 1986. Rabbi Akiva was right, wasn't he ;-)
>
> -W
>
> _______________________________________________
> rules-users mailing list
> rules-users(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
9:13 a.m.
OPS83?
http://repository.cmu.edu/cgi/viewcontent.cgi?article=2478&context=co...
or YES/L1? (seems information on this is out of print and not online either)
http://www.amazon.com/dp/B00070YQSU/ref=r_soa_w_d
"YES/L1: Integrating expert systems technology with traditional programming languages
(Research Report RC. International Business Machines Inc. Research Division)"
I definitely find linq interesting, as it's straight out of the research pages from
these projects - I wonder if the linq/database propel know about these… The first time I
saw it was in this paper "procedural match augments data-driven match"
http://www.aaai.org/Papers/AAAI/1986/AAAI86-037.pdf
They move the "lhs" logic into the consequence block, using the actual
"when" part as a simple goal trigger - allowing the 'lhs' to be used
procedurally, like linq. This allows them to control when a rule is evaluated and that
it's evaluation is atomic, and can have cleanup work done.
Mark
On 12 Mar 2013, at 06:04, Wolfgang Laun <wolfgang.laun(a)gmail.com> wrote:
Indeed, this thesis mentions a few features of the system I was
talking about :-)
([13])
-W
On 11/03/2013, Mark Proctor <mproctor(a)codehaus.org> wrote:
> There were a number of research efforts that looked at combining procedural
> and rule base programming.
> This one is quite interesting:
> Combining Rule-Based and Procedural Programming in the XC and XE Programming
> Lanaugages
>
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.9.1106&rep=r...
>
> Mark
>
>
> On 11 Mar 2013, at 18:03, Wolfgang Laun <wolfgang.laun(a)gmail.com> wrote:
>
>> On 11 March 2013 16:19, Mark Proctor <mproctor(a)codehaus.org> wrote:
>> So thinking really long term here. Can we build a "java layer" that
>> provide all the rule functionality we need - but fit ontop of the java
>> language neatly. We'd probably need to allow "rule" keyword and
have it in
>> Classes, at the method level. All class members and methods would be
>> available to the rules in that class.
>>
>> There is this production rule system where you can write your rules
>> embedded in compiling units, and alongside the legacy program units, of a
>> procedural, modular, strongly type HLL, and where you use expressions in
>> the language's own syntax in constraints...
>>
>> We've been using it ever since 1986. Rabbi Akiva was right, wasn't he
;-)
>>
>> -W
>>
>> _______________________________________________
>> rules-users mailing list
>> rules-users(a)lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/rules-users
>
>
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
2:28 p.m.
Mark,
no, the system I'm talking about is PAMELA, developed here in Vienna.
You may find references to papers citing PAMELA, authored by
F.Barachini and N.Theuretzbacher (one is referenced in the thesis you
quoted, see [13]), but I doubt that you'l find one of the papers on
the web. It was pre-internet days way back then :-)
If you could produce an RBS ranking based on rules fired in
production, I think that PAMELA would be in an excellent position.
There's a three-digit number of installations by now, but they're
running 24/7.
-W
On 12/03/2013, Mark Proctor <mproctor(a)codehaus.org> wrote:
OPS83?
http://repository.cmu.edu/cgi/viewcontent.cgi?article=2478&context=co...
or YES/L1? (seems information on this is out of print and not online
either)
http://www.amazon.com/dp/B00070YQSU/ref=r_soa_w_d
"YES/L1: Integrating expert systems technology with traditional programming
languages (Research Report RC. International Business Machines Inc. Research
Division)"
I definitely find linq interesting, as it's straight out of the research
pages from these projects - I wonder if the linq/database propel know about
these… The first time I saw it was in this paper "procedural match augments
data-driven match"
http://www.aaai.org/Papers/AAAI/1986/AAAI86-037.pdf
They move the "lhs" logic into the consequence block, using the actual
"when" part as a simple goal trigger - allowing the 'lhs' to be used
procedurally, like linq. This allows them to control when a rule is
evaluated and that it's evaluation is atomic, and can have cleanup work
done.
Mark
On 12 Mar 2013, at 06:04, Wolfgang Laun <wolfgang.laun(a)gmail.com> wrote:
> Indeed, this thesis mentions a few features of the system I was
> talking about :-)
> ([13])
> -W
>
> On 11/03/2013, Mark Proctor <mproctor(a)codehaus.org> wrote:
>> There were a number of research efforts that looked at combining
>> procedural
>> and rule base programming.
>> This one is quite interesting:
>> Combining Rule-Based and Procedural Programming in the XC and XE
>> Programming
>> Lanaugages
>>
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.9.1106&rep=r...
>>
>> Mark
>>
>>
>> On 11 Mar 2013, at 18:03, Wolfgang Laun <wolfgang.laun(a)gmail.com> wrote:
>>
>>> On 11 March 2013 16:19, Mark Proctor <mproctor(a)codehaus.org> wrote:
>>> So thinking really long term here. Can we build a "java layer"
that
>>> provide all the rule functionality we need - but fit ontop of the java
>>> language neatly. We'd probably need to allow "rule" keyword and
have it
>>> in
>>> Classes, at the method level. All class members and methods would be
>>> available to the rules in that class.
>>>
>>> There is this production rule system where you can write your rules
>>> embedded in compiling units, and alongside the legacy program units, of
>>> a
>>> procedural, modular, strongly type HLL, and where you use expressions
>>> in
>>> the language's own syntax in constraints...
>>>
>>> We've been using it ever since 1986. Rabbi Akiva was right, wasn't
he
>>> ;-)
>>>
>>> -W
>>>
>>> _______________________________________________
>>> rules-users mailing list
>>> rules-users(a)lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/rules-users
>>
>>
> _______________________________________________
> rules-users mailing list
> rules-users(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
9:21 p.m.
Actually Pamela is one of the papers I've been trying to track down, can't get it
on the internet any more. As I first saw it referenced in "Production matching for
large learning systems".
You don't by chance have the paper still?
Barachini, F. (1991) The evolution of PAMELA. Expert Systems, 8(2):87-98
I'm building up a collection of relevant research papers, over at mendelay. That is
one of my missing papers, that I've been unable to track:
http://www.mendeley.com/groups/2918061/rule-systems/papers/
Mark
On 12 Mar 2013, at 18:28, Wolfgang Laun <wolfgang.laun(a)gmail.com> wrote:
Mark,
no, the system I'm talking about is PAMELA, developed here in Vienna.
You may find references to papers citing PAMELA, authored by
F.Barachini and N.Theuretzbacher (one is referenced in the thesis you
quoted, see [13]), but I doubt that you'l find one of the papers on
the web. It was pre-internet days way back then :-)
If you could produce an RBS ranking based on rules fired in
production, I think that PAMELA would be in an excellent position.
There's a three-digit number of installations by now, but they're
running 24/7.
-W
On 12/03/2013, Mark Proctor <mproctor(a)codehaus.org> wrote:
> OPS83?
> http://repository.cmu.edu/cgi/viewcontent.cgi?article=2478&context=co...
>
>
> or YES/L1? (seems information on this is out of print and not online
> either)
> http://www.amazon.com/dp/B00070YQSU/ref=r_soa_w_d
> "YES/L1: Integrating expert systems technology with traditional programming
> languages (Research Report RC. International Business Machines Inc. Research
> Division)"
>
>
> I definitely find linq interesting, as it's straight out of the research
> pages from these projects - I wonder if the linq/database propel know about
> these… The first time I saw it was in this paper "procedural match augments
> data-driven match"
> http://www.aaai.org/Papers/AAAI/1986/AAAI86-037.pdf
>
> They move the "lhs" logic into the consequence block, using the actual
> "when" part as a simple goal trigger - allowing the 'lhs' to be
used
> procedurally, like linq. This allows them to control when a rule is
> evaluated and that it's evaluation is atomic, and can have cleanup work
> done.
>
> Mark
>
> On 12 Mar 2013, at 06:04, Wolfgang Laun <wolfgang.laun(a)gmail.com> wrote:
>
>> Indeed, this thesis mentions a few features of the system I was
>> talking about :-)
>> ([13])
>> -W
>>
>> On 11/03/2013, Mark Proctor <mproctor(a)codehaus.org> wrote:
>>> There were a number of research efforts that looked at combining
>>> procedural
>>> and rule base programming.
>>> This one is quite interesting:
>>> Combining Rule-Based and Procedural Programming in the XC and XE
>>> Programming
>>> Lanaugages
>>>
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.9.1106&rep=r...
>>>
>>> Mark
>>>
>>>
>>> On 11 Mar 2013, at 18:03, Wolfgang Laun <wolfgang.laun(a)gmail.com>
wrote:
>>>
>>>> On 11 March 2013 16:19, Mark Proctor <mproctor(a)codehaus.org>
wrote:
>>>> So thinking really long term here. Can we build a "java layer"
that
>>>> provide all the rule functionality we need - but fit ontop of the java
>>>> language neatly. We'd probably need to allow "rule" keyword
and have it
>>>> in
>>>> Classes, at the method level. All class members and methods would be
>>>> available to the rules in that class.
>>>>
>>>> There is this production rule system where you can write your rules
>>>> embedded in compiling units, and alongside the legacy program units, of
>>>> a
>>>> procedural, modular, strongly type HLL, and where you use expressions
>>>> in
>>>> the language's own syntax in constraints...
>>>>
>>>> We've been using it ever since 1986. Rabbi Akiva was right,
wasn't he
>>>> ;-)
>>>>
>>>> -W
>>>>
>>>> _______________________________________________
>>>> rules-users mailing list
>>>> rules-users(a)lists.jboss.org
>>>> https://lists.jboss.org/mailman/listinfo/rules-users
>>>
>>>
>> _______________________________________________
>> rules-users mailing list
>> rules-users(a)lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/rules-users
>
>
> _______________________________________________
> rules-users mailing list
> rules-users(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
>
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
4:43 p.m.
Who are the primary researchers of expert-systems today?
On Tue, Mar 12, 2013 at 8:21 PM, Mark Proctor <mproctor(a)codehaus.org> wrote:
Actually Pamela is one of the papers I've been trying to track
down, can't get it on the internet any more. As I first saw it referenced in
"Production matching for large learning systems".
You don't by chance have the paper still?
Barachini, F. (1991) The evolution of PAMELA. Expert Systems, 8(2):87-98
I'm building up a collection of relevant research papers, over at mendelay. That is
one of my missing papers, that I've been unable to track:
http://www.mendeley.com/groups/2918061/rule-systems/papers/
Mark
On 12 Mar 2013, at 18:28, Wolfgang Laun <wolfgang.laun(a)gmail.com> wrote:
> Mark,
>
> no, the system I'm talking about is PAMELA, developed here in Vienna.
> You may find references to papers citing PAMELA, authored by
> F.Barachini and N.Theuretzbacher (one is referenced in the thesis you
> quoted, see [13]), but I doubt that you'l find one of the papers on
> the web. It was pre-internet days way back then :-)
>
> If you could produce an RBS ranking based on rules fired in
> production, I think that PAMELA would be in an excellent position.
> There's a three-digit number of installations by now, but they're
> running 24/7.
>
> -W
>
>
> On 12/03/2013, Mark Proctor <mproctor(a)codehaus.org> wrote:
>> OPS83?
>> http://repository.cmu.edu/cgi/viewcontent.cgi?article=2478&context=co...
>>
>>
>> or YES/L1? (seems information on this is out of print and not online
>> either)
>> http://www.amazon.com/dp/B00070YQSU/ref=r_soa_w_d
>> "YES/L1: Integrating expert systems technology with traditional programming
>> languages (Research Report RC. International Business Machines Inc. Research
>> Division)"
>>
>>
>> I definitely find linq interesting, as it's straight out of the research
>> pages from these projects - I wonder if the linq/database propel know about
>> these… The first time I saw it was in this paper "procedural match augments
>> data-driven match"
>> http://www.aaai.org/Papers/AAAI/1986/AAAI86-037.pdf
>>
>> They move the "lhs" logic into the consequence block, using the actual
>> "when" part as a simple goal trigger - allowing the 'lhs' to be
used
>> procedurally, like linq. This allows them to control when a rule is
>> evaluated and that it's evaluation is atomic, and can have cleanup work
>> done.
>>
>> Mark
>>
>> On 12 Mar 2013, at 06:04, Wolfgang Laun <wolfgang.laun(a)gmail.com> wrote:
>>
>>> Indeed, this thesis mentions a few features of the system I was
>>> talking about :-)
>>> ([13])
>>> -W
>>>
>>> On 11/03/2013, Mark Proctor <mproctor(a)codehaus.org> wrote:
>>>> There were a number of research efforts that looked at combining
>>>> procedural
>>>> and rule base programming.
>>>> This one is quite interesting:
>>>> Combining Rule-Based and Procedural Programming in the XC and XE
>>>> Programming
>>>> Lanaugages
>>>>
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.9.1106&rep=r...
>>>>
>>>> Mark
>>>>
>>>>
>>>> On 11 Mar 2013, at 18:03, Wolfgang Laun <wolfgang.laun(a)gmail.com>
wrote:
>>>>
>>>>> On 11 March 2013 16:19, Mark Proctor <mproctor(a)codehaus.org>
wrote:
>>>>> So thinking really long term here. Can we build a "java
layer" that
>>>>> provide all the rule functionality we need - but fit ontop of the
java
>>>>> language neatly. We'd probably need to allow "rule"
keyword and have it
>>>>> in
>>>>> Classes, at the method level. All class members and methods would be
>>>>> available to the rules in that class.
>>>>>
>>>>> There is this production rule system where you can write your rules
>>>>> embedded in compiling units, and alongside the legacy program units,
of
>>>>> a
>>>>> procedural, modular, strongly type HLL, and where you use
expressions
>>>>> in
>>>>> the language's own syntax in constraints...
>>>>>
>>>>> We've been using it ever since 1986. Rabbi Akiva was right,
wasn't he
>>>>> ;-)
>>>>>
>>>>> -W
>>>>>
>>>>> _______________________________________________
>>>>> rules-users mailing list
>>>>> rules-users(a)lists.jboss.org
>>>>> https://lists.jboss.org/mailman/listinfo/rules-users
>>>>
>>>>
>>> _______________________________________________
>>> rules-users mailing list
>>> rules-users(a)lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/rules-users
>>
>>
>> _______________________________________________
>> rules-users mailing list
>> rules-users(a)lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/rules-users
>>
>
> _______________________________________________
> rules-users mailing list
> rules-users(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
--
Grant Rettke | ACM, AMA, COG, IEEE
grettke(a)acm.org | http://www.wisdomandwonder.com/
Wisdom begins in wonder.
((λ (x) (x x)) (λ (x) (x x)))
4:47 p.m.
I usually try to attend RuleML meet ups and Intellifest (ex October Rules
Fest) there you meet all the Expert System community.
Cheers
On Fri, Mar 15, 2013 at 5:43 PM, Grant Rettke <grettke(a)acm.org> wrote:
Who are the primary researchers of expert-systems today?
On Tue, Mar 12, 2013 at 8:21 PM, Mark Proctor <mproctor(a)codehaus.org>
wrote:
> Actually Pamela is one of the papers I've been trying to track down,
can't get it on the internet any more. As I first saw it referenced in
"Production matching for large learning systems".
>
> You don't by chance have the paper still?
> Barachini, F. (1991) The evolution of PAMELA. Expert Systems, 8(2):87-98
>
> I'm building up a collection of relevant research papers, over at
mendelay. That is one of my missing papers, that I've been unable to track:
> http://www.mendeley.com/groups/2918061/rule-systems/papers/
>
> Mark
> On 12 Mar 2013, at 18:28, Wolfgang Laun <wolfgang.laun(a)gmail.com> wrote:
>
>> Mark,
>>
>> no, the system I'm talking about is PAMELA, developed here in Vienna.
>> You may find references to papers citing PAMELA, authored by
>> F.Barachini and N.Theuretzbacher (one is referenced in the thesis you
>> quoted, see [13]), but I doubt that you'l find one of the papers on
>> the web. It was pre-internet days way back then :-)
>>
>> If you could produce an RBS ranking based on rules fired in
>> production, I think that PAMELA would be in an excellent position.
>> There's a three-digit number of installations by now, but they're
>> running 24/7.
>>
>> -W
>>
>>
>> On 12/03/2013, Mark Proctor <mproctor(a)codehaus.org> wrote:
>>> OPS83?
>>>
http://repository.cmu.edu/cgi/viewcontent.cgi?article=2478&context=co...
>>>
>>>
>>> or YES/L1? (seems information on this is out of print and not online
>>> either)
>>> http://www.amazon.com/dp/B00070YQSU/ref=r_soa_w_d
>>> "YES/L1: Integrating expert systems technology with traditional
programming
>>> languages (Research Report RC. International Business Machines Inc.
Research
>>> Division)"
>>>
>>>
>>> I definitely find linq interesting, as it's straight out of the
research
>>> pages from these projects - I wonder if the linq/database propel know
about
>>> these... The first time I saw it was in this paper "procedural match
augments
>>> data-driven match"
>>> http://www.aaai.org/Papers/AAAI/1986/AAAI86-037.pdf
>>>
>>> They move the "lhs" logic into the consequence block, using the
actual
>>> "when" part as a simple goal trigger - allowing the 'lhs'
to be used
>>> procedurally, like linq. This allows them to control when a rule is
>>> evaluated and that it's evaluation is atomic, and can have cleanup work
>>> done.
>>>
>>> Mark
>>>
>>> On 12 Mar 2013, at 06:04, Wolfgang Laun <wolfgang.laun(a)gmail.com>
wrote:
>>>
>>>> Indeed, this thesis mentions a few features of the system I was
>>>> talking about :-)
>>>> ([13])
>>>> -W
>>>>
>>>> On 11/03/2013, Mark Proctor <mproctor(a)codehaus.org> wrote:
>>>>> There were a number of research efforts that looked at combining
>>>>> procedural
>>>>> and rule base programming.
>>>>> This one is quite interesting:
>>>>> Combining Rule-Based and Procedural Programming in the XC and XE
>>>>> Programming
>>>>> Lanaugages
>>>>>
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.9.1106&rep=r...
>>>>>
>>>>> Mark
>>>>>
>>>>>
>>>>> On 11 Mar 2013, at 18:03, Wolfgang Laun
<wolfgang.laun(a)gmail.com>
wrote:
>>>>>
>>>>>> On 11 March 2013 16:19, Mark Proctor
<mproctor(a)codehaus.org> wrote:
>>>>>> So thinking really long term here. Can we build a "java
layer" that
>>>>>> provide all the rule functionality we need - but fit ontop of
the
java
>>>>>> language neatly. We'd probably need to allow
"rule" keyword and
have it
>>>>>> in
>>>>>> Classes, at the method level. All class members and methods
would be
>>>>>> available to the rules in that class.
>>>>>>
>>>>>> There is this production rule system where you can write your
rules
>>>>>> embedded in compiling units, and alongside the legacy program
units, of
>>>>>> a
>>>>>> procedural, modular, strongly type HLL, and where you use
expressions
>>>>>> in
>>>>>> the language's own syntax in constraints...
>>>>>>
>>>>>> We've been using it ever since 1986. Rabbi Akiva was right,
wasn't
he
>>>>>> ;-)
>>>>>>
>>>>>> -W
>>>>>>
>>>>>> _______________________________________________
>>>>>> rules-users mailing list
>>>>>> rules-users(a)lists.jboss.org
>>>>>> https://lists.jboss.org/mailman/listinfo/rules-users
>>>>>
>>>>>
>>>> _______________________________________________
>>>> rules-users mailing list
>>>> rules-users(a)lists.jboss.org
>>>> https://lists.jboss.org/mailman/listinfo/rules-users
>>>
>>>
>>> _______________________________________________
>>> rules-users mailing list
>>> rules-users(a)lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/rules-users
>>>
>>
>> _______________________________________________
>> rules-users mailing list
>> rules-users(a)lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/rules-users
>
>
> _______________________________________________
> rules-users mailing list
> rules-users(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
--
Grant Rettke | ACM, AMA, COG, IEEE
grettke(a)acm.org | http://www.wisdomandwonder.com/
Wisdom begins in wonder.
((λ (x) (x x)) (λ (x) (x x)))
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
--
- MyJourney @ http://salaboy.com <http://salaboy.wordpress.com>
- Co-Founder @ http://www.jugargentina.org
- Co-Founder @ http://www.jbug.com.ar
- Salatino "Salaboy" Mauricio -
10:44 p.m.
There hasn't been a lot of expert system engine research since 1995:
http://blog.athico.com/2012/05/drools-54-artificial-intelligence.html
In general academics have focused more on the logic programming side, referred to as
knowledge representation and reasoning. Developing formalised semantics to represent the
world around us, and mappings between those representations. The mapping/layered aspect
was a key aspect of he RIF standard. Examples of formalised representations would be
Defeasible Logic, http://en.wikipedia.org/wiki/Defeasible_logic, and Description Logic,
http://en.wikipedia.org/wiki/Description_logic. The engine which implements this is just a
side note, and typically they target prolog. Take a look at the last 5 years of RuleML
published papers, and you'll get a good idea.
Academic engines of interest:
Flora-2/XSB, jDrew, silk, opencyc, jess, openrdf, jena.
I inlcude Jess in the above, which is actually still more popular than Drools in Academia
- although it's not open source (only free for non-commercial use).
OWL and semantic ontologies drives a lot of the engine implementation research these
days.
Mark
On 15 Mar 2013, at 20:43, Grant Rettke <grettke(a)acm.org> wrote:
Who are the primary researchers of expert-systems today?
On Tue, Mar 12, 2013 at 8:21 PM, Mark Proctor <mproctor(a)codehaus.org> wrote:
> Actually Pamela is one of the papers I've been trying to track down, can't
get it on the internet any more. As I first saw it referenced in "Production matching
for large learning systems".
>
> You don't by chance have the paper still?
> Barachini, F. (1991) The evolution of PAMELA. Expert Systems, 8(2):87-98
>
> I'm building up a collection of relevant research papers, over at mendelay. That
is one of my missing papers, that I've been unable to track:
> http://www.mendeley.com/groups/2918061/rule-systems/papers/
>
> Mark
> On 12 Mar 2013, at 18:28, Wolfgang Laun <wolfgang.laun(a)gmail.com> wrote:
>
>> Mark,
>>
>> no, the system I'm talking about is PAMELA, developed here in Vienna.
>> You may find references to papers citing PAMELA, authored by
>> F.Barachini and N.Theuretzbacher (one is referenced in the thesis you
>> quoted, see [13]), but I doubt that you'l find one of the papers on
>> the web. It was pre-internet days way back then :-)
>>
>> If you could produce an RBS ranking based on rules fired in
>> production, I think that PAMELA would be in an excellent position.
>> There's a three-digit number of installations by now, but they're
>> running 24/7.
>>
>> -W
>>
>>
>> On 12/03/2013, Mark Proctor <mproctor(a)codehaus.org> wrote:
>>> OPS83?
>>>
http://repository.cmu.edu/cgi/viewcontent.cgi?article=2478&context=co...
>>>
>>>
>>> or YES/L1? (seems information on this is out of print and not online
>>> either)
>>> http://www.amazon.com/dp/B00070YQSU/ref=r_soa_w_d
>>> "YES/L1: Integrating expert systems technology with traditional
programming
>>> languages (Research Report RC. International Business Machines Inc. Research
>>> Division)"
>>>
>>>
>>> I definitely find linq interesting, as it's straight out of the research
>>> pages from these projects - I wonder if the linq/database propel know about
>>> these… The first time I saw it was in this paper "procedural match
augments
>>> data-driven match"
>>> http://www.aaai.org/Papers/AAAI/1986/AAAI86-037.pdf
>>>
>>> They move the "lhs" logic into the consequence block, using the
actual
>>> "when" part as a simple goal trigger - allowing the 'lhs'
to be used
>>> procedurally, like linq. This allows them to control when a rule is
>>> evaluated and that it's evaluation is atomic, and can have cleanup work
>>> done.
>>>
>>> Mark
>>>
>>> On 12 Mar 2013, at 06:04, Wolfgang Laun <wolfgang.laun(a)gmail.com>
wrote:
>>>
>>>> Indeed, this thesis mentions a few features of the system I was
>>>> talking about :-)
>>>> ([13])
>>>> -W
>>>>
>>>> On 11/03/2013, Mark Proctor <mproctor(a)codehaus.org> wrote:
>>>>> There were a number of research efforts that looked at combining
>>>>> procedural
>>>>> and rule base programming.
>>>>> This one is quite interesting:
>>>>> Combining Rule-Based and Procedural Programming in the XC and XE
>>>>> Programming
>>>>> Lanaugages
>>>>>
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.9.1106&rep=r...
>>>>>
>>>>> Mark
>>>>>
>>>>>
>>>>> On 11 Mar 2013, at 18:03, Wolfgang Laun
<wolfgang.laun(a)gmail.com> wrote:
>>>>>
>>>>>> On 11 March 2013 16:19, Mark Proctor
<mproctor(a)codehaus.org> wrote:
>>>>>> So thinking really long term here. Can we build a "java
layer" that
>>>>>> provide all the rule functionality we need - but fit ontop of the
java
>>>>>> language neatly. We'd probably need to allow "rule"
keyword and have it
>>>>>> in
>>>>>> Classes, at the method level. All class members and methods would
be
>>>>>> available to the rules in that class.
>>>>>>
>>>>>> There is this production rule system where you can write your
rules
>>>>>> embedded in compiling units, and alongside the legacy program
units, of
>>>>>> a
>>>>>> procedural, modular, strongly type HLL, and where you use
expressions
>>>>>> in
>>>>>> the language's own syntax in constraints...
>>>>>>
>>>>>> We've been using it ever since 1986. Rabbi Akiva was right,
wasn't he
>>>>>> ;-)
>>>>>>
>>>>>> -W
>>>>>>
>>>>>> _______________________________________________
>>>>>> rules-users mailing list
>>>>>> rules-users(a)lists.jboss.org
>>>>>> https://lists.jboss.org/mailman/listinfo/rules-users
>>>>>
>>>>>
>>>> _______________________________________________
>>>> rules-users mailing list
>>>> rules-users(a)lists.jboss.org
>>>> https://lists.jboss.org/mailman/listinfo/rules-users
>>>
>>>
>>> _______________________________________________
>>> rules-users mailing list
>>> rules-users(a)lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/rules-users
>>>
>>
>> _______________________________________________
>> rules-users mailing list
>> rules-users(a)lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/rules-users
>
>
> _______________________________________________
> rules-users mailing list
> rules-users(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
--
Grant Rettke | ACM, AMA, COG, IEEE
grettke(a)acm.org | http://www.wisdomandwonder.com/
Wisdom begins in wonder.
((λ (x) (x x)) (λ (x) (x x)))
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
10:10 p.m.
On Mon, Mar 11, 2013 at 10:19 AM, Mark Proctor <mproctor(a)codehaus.org> wrote:
For instance here is an idea I've been mulling over recently
(note just random ideas, not roadmap). We don't have
the resources to completely compete against other languages, and I'm not sure we
should do. With Java8 and 9 it
starts to become a more compelling language. Java8 brings lambs and java9 solve the perm
gen issue (so we can
stop using MVEL, and just go pure code generation and not worry about blowing the perm
gen). Could we create a
system that better layers with java9?
Any other JVM languages that would serve you well right now?
8:50 a.m.
On 12 Mar 2013, at 02:10, Grant Rettke <grettke(a)acm.org> wrote:
On Mon, Mar 11, 2013 at 10:19 AM, Mark Proctor
<mproctor(a)codehaus.org> wrote:
> For instance here is an idea I've been mulling over recently (note just random
ideas, not roadmap). We don't have
> the resources to completely compete against other languages, and I'm not sure we
should do. With Java8 and 9 it
> starts to become a more compelling language. Java8 brings lambs and java9 solve the
perm gen issue (so we can
> stop using MVEL, and just go pure code generation and not worry about blowing the
perm gen). Could we create a
> system that better layers with java9?
Any other JVM languages that would serve you well right now?
Anyone that offers
type safety. So Scala, Ceylon, Koitlin would all be contenders. Java is simply the lowest
common denominator that reaches the widest possible mark. But as mentioned, base java is
not enough - we'd need to layer property accessors on it.
Mark
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
10:28 a.m.
There is, but you have to look at things slightly differently - and we have to really work
on a new DRL to better support it, so it's all very experimental and bleeding edge.
It's also probably more lambdas and closures than pure functions.
Where I see lambdas helping is in creating meta control systems for expert systems,
attaching functions to events handlers. I'd quite like to see C# style delegator
operators, for compact ways of assigning event handlers.
I'd then like to start exposing subsystems event models and life cycles in DRL, to
have handlers added. These functions can have state (closures) and control how execution
happens. I sketched a document some time ago on ideas going on in my head - it's very
rough and much is already out of date on how I'd like to do it:
https://community.jboss.org/wiki/RuleModule
So for instance at the moment Drools has "activation-group", which provides hard
coded behaviour. What if instead we exposed the RuleModule (group) and it had events you
could attach to
onRuleAdded
onRuleRemoved
onBeforeRuleFired
onAfterRuleFired
I could add a function to "onAfterRuleFired" which would cancel all the other
waiting rules. So I'd like to actually expose a RuleModule in DRL as declarable
artefact (just a sketch, not working or compilable code)
module MyModule {
onAfterRuleFire() {
conflictSet.cancelAll();
}
}
While the above is declaration based approach, it could use an api approach too
getRuleModule( "MyModule" ).onAfterRuleFire += new RuleModuleEvent() {
conflictSet.cancelAll();
}
I'd more recently sketch out some ideas on controlling rule execution semantics -
based on the DADO research paper for "all" instances to be fired. If you scroll
down to the bottom, you can see a suggested life cycle for rule firings:
https://community.jboss.org/wiki/RuleExecutionSemantics
Other proposed life cycles:
OnBeforeMatchFire
Before a single Match instance is executed
OnAfterMatchFire
After a single Match instance is executed
onBeforeAllFire
Before any Matches in the set have fired
onAfterAllFire
After all Matches in the set have fired
onDeleteMatch
invoked when a Match is deleted (no longer true)
onUpdateMatch
invoked when a Match is updated
onChangeMatch
invoked when a Match is either updated or deleted
I'm also starting to prefer the C# "on" prefix notation for events.
I think as well instead of doing it initially in DRL, a lot can be done in pure Java with
annotations used for wiring and discovery. We can replace or atleast more easily expose
conflict handling and all the "group" stuff we have now. Note a lot of these
events already exist, but they are more for observing. So it's all about re-looking at
our events, and our sub systems, thinking about how they can be better exposed to create a
meta system.
@RuleModule( "MyModule" ) // RuleModule can be optional and inferred from class
name.
public class MyClassName {
@ConflictSet
private ConflictSet set; // Tells it to inject the conflict on instiation
public void onAfterRuleFired(Match match) {
set.canceAll()
}
}
So over all our longer term research is looking into a meta expert system - that looks at
better ways to expose sub systems and their life cycles and using event handlers to allow
the user to control the behaviour. I also think we'll need a sort of macro system, to
make this easily re-usable.
If this is work you are interested in research, then please pop onto irc - we'll be
happy to mentor.
http://www.jboss.org/drools/irc
Mark
On 7 Mar 2013, at 20:44, Matteo Mortari <matteo.mortari(a)gmail.com> wrote:
Ciao everybody, I have a question I would like to submit to this
community, given Drools is my preferred choice for Expert Systems :)
A quick background: I'm not an Expert Systems / Rule engines expert, but I'm a
happy power user "of Rules" since my CS Engineering thesis, a paper published,
and used it with happy results at work and hobby projects. Currently boarded on a new
project with Rules+Processes at work, and still very happy, so happy, I'm using Drools
again in a new hobby project. [semi-brag-mode off]
Now, in recent times, I cannot ignore what (apparently) is a strong renaissance of
Functional Programming, specifically all this momentum which Scala and Java 8's
Lambda's are getting.
The Question I'm asking myself: is there a specific, or few, areas where Functional
Programming could actually benefit Expert Systems so much to actually spawn new paradigms
of writing Rules? Or benefit in a new generation of the Rete algorithm?
Or actually, Functional Programming will spawn a completely different paradigm of system
which will go beyond Expert Systems the way are designed today?
Ignore my rant if you feel it's too generic, but I really wanted to seize a chance to
ask the question to this community :)
Thank you for your feedback,
Ciao,
Matteo
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
4057
days inactive
4066
days old
15 comments
5 participants
participants (5)
-
Grant Rettke -
Mark Proctor -
Matteo Mortari -
Mauricio Salatino -
Wolfgang Laun