[pandorabots-general] Forum to advance the science of
AI(including the resumption of standard setting activities)
Brian Hoecht
brian_hoecht at msn.com
Thu May 18 07:51:59 PDT 2006
See below.
-----Original Message-----
From:
pandorabots-general-bounces+brian_hoecht=msn.com at list.pandorabots
com
[mailto:pandorabots-general-bounces+brian_hoecht=msn.com at list.pan
dorabots.com] On Behalf Of Noel Bush
Sent: Thursday, May 18, 2006 10:11 AM
To: pandorabots-general at list.pandorabots.com
Subject: Re: [pandorabots-general] Forum to advance the science
of AI(including the resumption of standard setting activities)
Brian,
Will you be helping the foundation restore its status as a
non-profit?
[BH] No. I am an entrepreneur with my own for-profit enterprise
and have no legal or direct financial relationship with the
Foundation, nor do I want one.
It has been listed as "suspended" by the California Secretary of
State for a very long time:
http://kepler.ss.ca.gov/corpdata/ShowAllList?QueryCorpNumber=C234
2616
[BH] I did not know that, but I will encourage Dr. Wallace to
get the foundation off of the suspended list.
Aren't there likely to be problems related to working with an
organization that hasn't maintained compliance? Are
contributions to a suspended 501(c)(3) organization still tax
deductible, for instance?
[BH] I don't really care about that. I support the foundation
to support its work and as a business entity, the expense is
still tax deductible as we derive economic (but not financial)
benefit from its activity.
And if there's any business relationship between a commercial
entity and a principal in the foundation (not saying there is in
your case),
[BH] There is not.
doesn't that pose some potential conflict of interest issues that
can only be exacerbated by the organization being in a
"suspended" status?
[BH] As a CPA from Deloitte & Touche's national head office in
Toronto, I am well versed in the concepts of conflict in both
fact and appearance. I would have to say, that the suspension is
a black mark, but there is nothing about the panel's activities
to advance the science of AI which should cause any conflict that
I can think of.
Personally I felt that the "ALICE AI Foundation" became
problematic several years ago when its founder began actively
directing business to a commercial interest (Franz Inc /
Pandorabots), against the wishes of the board of trustees at that
time.
[BH] I was aware that the two of you had a falling out, but did
not know the details. To my knowledge, Pandorabots doesn't make
any/much $$$ on its AI work. They host bots for free, make a
little on selling T-shirts and a few banner links, but that
hardly makes them unique. Otherwise my understanding is that
Pandorabots is more of a marketing venture for them to showcase
the power of Lisp as (and I am not technically qualified to
provide an opinion on this) it is particularly well suited to the
application and implementation of AI.
I also came to the conclusion that the founder's primary aim for
the organization was to support himself personally -- a laudable
goal, for sure, but not one that would (from my non-professional
understanding) fit the meaning of "charity" that the 501(c)(3)
designation was intended for (though I may be totally wrong about
this). Good grief -- most developers who want to ask for money
just put a Paypal "tip jar" out there.
[BH] We all have the right to earn a living, but as you say it
must be done without running afoul of the legal rules and
conflicts.
[BH] A tip jar? And people wonder why the dot com bubble burst.
I think you just might want to be careful about tying up too
closely with an organization that may not have all its ducks in a
row. It sounds great to start up some kind of process to move
AIML along again (and I think that xAIML will be very
complementary in that regard -- see my post to the pandorabots
list about this). But it might be smarter to start it up under
different auspices -- start from a fresh slate, so to speak. The
"ALICE Foundation" and all of its baggage could end up
dragging down an enthusiastic and energetic effort, with lots of
completely irrelevant matters like this.
[BH] An all volunteer group will have nothing to worry about in
working with the Foundation. The AIML movement owes a debt of
gratitude to the Foundation for being where it is and the
Foundation remains the right forum to advance it under.
AIML can advance, complementary technologies can flourish,
interpreters can proliferate, and botmasters can build more and
better bots, all without the ALICE AI Foundation. The vision, of
course, was different at one time, and likely there still is a
useful function that could be performed by some such
organization. But one with a history as troubled as this one is
probably best to be careful about, at the very least.
[BH] Agreed, except I think the advancement is still best
accomplished via the Foundation, even if it comes from a
volunteer like me. I believe we said this in one of the emails
on the new panel, but here it is again, Dr. Wallace and I are
co-chairing the panel. He is the AIML historian, guru, sage and
expert able to keep the panel true to its founding principles and
rich history. I execute, organize, facilitate, manage and lead.
[BH] No existing problems with the registration authorities are
going to interfere with the important volunteer effort being
facilitated.
[BH] P.S. Thank you for the opportunity to clear the air on
this. It would have been downright weird clarifying the role of
me and my company with respect to this panel in a non-solicited
manner.
Noel
Brian Hoecht wrote:
> I don't think it is any secret to this group that the AI
Foundation is
> putting together a panel to fire back up the standard setting
process
> and general advancement of the science of AI.
>
>
>
> The call to action to join the panel/forum has been received
mostly as
> requests for beta access to the AIML Editor. The Editor is
only the
> means to the end. or a perk if you will for the volunteers.
>
>
>
> Unfortunately, I think the more important message was missed.
The goal
> is, was and remains the advancement of the science of AI -
through the
> Foundation and on a basis committed to its goals and for the
benefit of all.
>
>
>
> Here is the call again if you missed it:
>
>
>
> "If you are highly skilled and experienced in AI, your
foundation needs
> you. Technology and general computer capabilities have kept
expanding
> since 2002. The AIML standards have not kept up. Progress
will not
> happen on its own (translation: Dr. Wallace has a day job), so
if you
> have passion, energy and time that you can commit, we are
creating the
> forum for you to help... and yes, panel/forum members get
access to the
> beta release of the new AIML Editor."
>
>
>
> And yes, we are trying to find a way to give access to the
consumer
> version to everyone who wants a copy, but no it is NOT open
source nor
> will it be (but it will produce standard compliant AIML files
as its
> output and no there is no hidden magic to it beyond it being
more
> efficient than WordPad - multiple *.aiml file manipulation,
GUI, sort,
> relate, test, view, etc.). And yes it is a beta release.
Please be
> patient. We'll solve this as the panel gets going in a fair
and
> transparent manner that is true to the mission statement of the
Foundation.
>
>
>
> So. we are looking for skilled, passionate volunteers. Should
you
> choose to accept this exciting mission here is what you can
expect to
> get to lead:
>
>
>
> 1) a revamping of the ALICE AAA data set to make the creation
of new or
> derivative synthetic personalities more efficient and easier
>
> 2) defining AIML coding best practices then cleaning up the
ALICE AAA
> data set for them
>
> 3) Several how to's for:
>
> a) Embedded javascript
>
> b) Embedded avatar controls
>
> c) Embedded emotive TTS
>
> d) System tags
>
> e) Website + application/db interoperability
>
> 4) Resumption of standard setting activities
>
> 4) Others... (Who knows what the top minds in the field will
come up with)?
>
>
>
> As with the Editor, we are doing these at Ai-Dealer, but don't
want to
> float out to sea on our own. It is why I approached Dr.
Wallace about
> the need to fire back up the activities of the Foundation and
to advance
> the science. It is why I've offered free access to the beta
version of
> the Editor's consumer version to the panel for those
contributing their
> time and effort.
>
>
>
> And to clear up another point, yes there is a corporate version
of the
> AIML Editor available for a modest fee to anyone looking to
implement an
> AI project wanting media exposure via the panel. We are doing
this
> because we want to help bring business community credibility to
the
> field of applied AI. The fee is to help us enhance it in
support of how
> you are going to use it. Members in this group need not
contribute
> their time to the efforts of the volunteer group, saving their
time for
> their own execution. The corporate version includes the items
listed
> under 3a-e above, the consumer version does not.
>
>
>
> As to the xAIML effort, the idea of needing to expand the
capabilities
> of AIML is correct, I just don't think it should be done in
isolation by
> one of the AIML Interpreters. That said, I do like the notion
of <spell
> check> tags and some of the other ideas... but they need the
peer
> review, discussion, agreement and planned release via the
standard
> setting organization.
>
>
>
> As a final note, here is an interesting bit of trivia for the
AI
> community. I grew up in automotive retailing circles and
understand the
> importance of standards. Did you know that in most car dealer
> accounting systems, there are more lines of code dedicated to
> integration with the various OEM's than there is for the
complete
> accounting, sales, service and parts applications? And there
are only
> about 20 OEM's. This happened because each OEM built their own
spec for
> simple things like warranty claims, financial statements,
vehicle
> invoices, parts orders, etc. It took close to a decade to get
standards
> established then implemented, but this organization (see link
below) is
> becoming the dominant force in the industry and is enabling
more,
> better, cheaper and faster integration than the dealers have
ever seen
> before.
>
>
>
> www.starstandards.org <http://www.starstandards.org/>
>
>
>
> So... cynicism aside, having a single inclusive standard is
important.
> Let it go and it will take a decade or more to get it back - if
ever.
> The AI movement is fortunate to have such standards up front.
>
>
>
> BTW - I noted on the Alice AAA page
http://www.alicebot.org/aiml/aaa/
> that the following individuals were the driving force in
transforming
> Alice Silver into Alice AAA. some are on our panel/forum.
others we have
> not heard from. While we are already at nearly 20 members now
it would
> be great to hear from those below whom we have not yet heard.
Many
> hands make quick work of the significant task ahead of us. As
you know,
> scrubbing Alice even for a single best practice or task is a
reasonable
> sized effort.
>
> * Joy Harwood
> * Colin Meldrum
> * Dr. Doubly Aimless
> * Chris Hatcher
> * Gary Dubuque
> * Anne Kootstra
> * Lindsay Davies
> * Stefan Zakarias
> * Tony Sharp
> * John Campbell
> * Steve Worswick
> * Gary Poster
> * Soren Gjellerup Christiansen
>
>
>
>
>
>
>
> -----Original Message-----
>
> From:
>
pandorabots-general-bounces+brian_hoecht=msn.com at list.pandorabots
com
>
[mailto:pandorabots-general-bounces+brian_hoecht=msn.com at list.pan
dorabots.com]
> On Behalf Of Dr. Rich Wallace
>
> Sent: Wednesday, May 17, 2006 9:18 PM
>
> To: pandorabots-general at list.pandorabots.com
>
> Subject: RE: [pandorabots-general] Newbie Problem
>
>
>
> Brian, the simple answer to your question is that there does
exist a
> body, the AIML Architecture Committee, to debate, propose and
adopt
> changes and additions to the AIML spec. For various reasons,
that
> committee has been dormant since 2002, and the 1.0.1 spec has
remained
> frozen in place. One reason has been that the guy we relied on
to write
> the spec (who is not a computer scientist, but a "musician")
has decided
> to compete with us rather than cooperate, and has now announced
his own
> confusing attempt to create an "XAIML" standard right here on
this
> mailing list and elsewhere. I personally don't see any
standards
> efforts moving forward until someone starts making more money
from AIML
> applications. So far the various projects have adopted an
"embrace and
> extend" approach, which has been fertile ground for innovation.
>
>
>
> One way to handle a simple macro extension like yours is at the
> authoring level. You can have the botmaster write <sr2/> but
internally
> expand that expression into <srai><star index="2"/></srai> in
the AIML
> file. That way you get the best of both worlds. "AIML
Authoring Tools"
> has been a subject of ongoing discussion for some time.
>
>
>
> Rich
>
>
>
>> Below is a category out of the Alice AAA Reducer.aiml file.
>
>>
>
>> <category>
>
>> <pattern>_ HOWEVER *</pattern>
>
>> <template>
>
>> <srai><star/></srai>
>
>> <srai><star index="2"/></srai>
>
>> </template>
>
>> </category>
>
>>
>
>> My question is this...
>
>>
>
>> If..... <srai><star/></srai> = <sr/>
>
>> Then... <srai><star index="2"/></srai> = <sr2/>???
>
>>
>
>> Then this would be oh so much briefer if stated as:
>
>>
>
>> <category>
>
>> <pattern>_ HOWEVER *</pattern>
>
>> <template>
>
>> <sr/>
>
>> <sr2/>
>
>> </template>
>
>> </category>
>
>>
>
>> Thoughts?
>
>> _______________________________________________
>
>> This is the pandorabots-general mailing list
>
>> To Post, reply to pandorabots-general at list.pandorabots.com
>
>> Unsubscribe and change preferences at
>
>>
http://list.pandorabots.com/mailman/listinfo/pandorabots-general
>
>> Learn netiquette at http://www.dtcc.edu/cs/rfc1855.html
>
>> Learn to read at http://www.literacy.org/
>
>>
>
>
>
> _______________________________________________
>
> This is the pandorabots-general mailing list
>
> To Post, reply to pandorabots-general at list.pandorabots.com
>
> Unsubscribe and change preferences at
>
http://list.pandorabots.com/mailman/listinfo/pandorabots-general
>
> Learn netiquette at http://www.dtcc.edu/cs/rfc1855.html
>
> Learn to read at http://www.literacy.org/
>
>
>
>
>
-----------------------------------------------------------------
-------
>
> _______________________________________________
> This is the pandorabots-general mailing list
> To Post, reply to pandorabots-general at list.pandorabots.com
> Unsubscribe and change preferences at
http://list.pandorabots.com/mailman/listinfo/pandorabots-general
> Learn netiquette at http://www.dtcc.edu/cs/rfc1855.html
> Learn to read at http://www.literacy.org/
_______________________________________________
This is the pandorabots-general mailing list
To Post, reply to pandorabots-general at list.pandorabots.com
Unsubscribe and change preferences at
http://list.pandorabots.com/mailman/listinfo/pandorabots-general
Learn netiquette at http://www.dtcc.edu/cs/rfc1855.html
Learn to read at http://www.literacy.org/
More information about the pandorabots-general
mailing list