x-logic-www — For discussing development of the X-Logic web-site/portal

[X-logic-www] [X-Logic] Re: X-Logic web site upgrade [14]

[<web...@ge...>]

Subject- Re: X-Logic web site upgrade
Url- http://www.x-logic.org/arch/releases.html

The software running the X-Logic web site has been
given a minor update.
It is now based on PHP-Nuke-4.4. though the web
comment facilities are largely unaffected by this
change.

It is now possible for signed-on users to select
areas of the site for email subscription to 
comments/discussions.
This is done using checkboxes for directories or
files when viewing comments.
Any combination of directories or files can be
checked.

This is an alternative to getting the whole
site through the sourceforge list.
I am still not feeding any traffic towards
X-Logic, but I am planning on retrofitting
this technology to rbjones.com, then resuming
development of both sites and linking traffic
through from rbjones.com to X-Logic.

Roger Jones

[X-logic-www] [X-Logic] Re: OpenMind [12]

[Roger B. J. <rb...@rb...>]

Subject- Re: OpenMind
Url- http://www.rbjones.com/

Two core foundational propositions in X-Logic are, firstly that to give a
semantics to the full range of languages which can be represented in XML
(and that''s pretty much everything) you need a semantic foundation which is
equivalent to (or richer and stronger than) a mathematical foundation
system.

The secondary proposition is just that of the available candidates a
polymorphic higher order set theory is a pragmatic choice.

I don''t have a detailed knowledge of the literature, and I know of no-one
else who advocates mathematical foundation systems as a basis for XML
semantics (though its a pretty obvious idea).

However, the best evidence that this approach is reasonable is the
literature on HOL (which is itself such a foundation system), and I would
guess that the procedings of the TPHOLS conference series would be a good
place to look.

The term semantic embedding is a general term used for techniques for
realising proof support for some language using a semantically correct
mapping into HOL (or some other suitable logical system). Quite a few
languages have been treated in this way, including programming languages,
hardware description languages, and specification languages (such as CSP and
Z).

You have to bear in mind here that I am talking about strict formalisation,
in machine supported languages, not just a mathematical semantics.
Though category theory is often used for mathematical semantics, I am not
aware of it having being used for strictly formal semantics, and if you are
aware of this having been done I would be interested to have a reference.

For more information on HOL and the TPHOL conferences see:
http://www.cl.cam.ac.uk/Research/HVG/HOL/

Another source of information relevant to the connection between HOL and
programming languages is Tobias Nipkow''s isabelle/HOL tutorial, which
virtually presents HOL as a functional programming language.
(http://www.cl.cam.ac.uk/Research/HVG/Isabelle/dist/Isabelle99-1/doc/tutoria
l.pdf)

The foundation I advocate is classical set theory formalised in HOL, which
would be better than plain HOL if you wanted to do category theory in it.

[X-logic-www] [X-Logic] Applying set theory to specification and verification [13]

[<com...@x-...>]

Subject- Applying set theory to specification and verification
Url- http://www.rbjones.com/

I should have referred you to Larry Paulson''s work with Isabelle/ZF, see:
http://www.cl.cam.ac.uk/users/lcp/papers/Sets/

[X-logic-www] [X-Logic] Re: OpenMind [11]

[<com...@x-...>]

Subject- Re: OpenMind
Url- 

Didn''t argue for superiority as foundation.  My point was that such a (categorial) basis provide a better platform for coding mathematical objects and operations.  As you clearly have more knowledge of this area, perhaps you could indicate how you would use set-theoretic constructs as a basis for for such coding.  It may well be there are well understood techniques for doing this of which I am unaware.  If you are aware of such, and have web or other references from which I might learn I would be grateful.

[X-logic-www] [X-Logic] Re: OpenMind [10]

[<com...@x-...>]

Subject- Re: OpenMind
Url- http://www.rbjones.com/

You have to bear in mind that I am talking about foundation system in a technical sense here. i.e. a logic within which mathematics (inter alia) can be developed by conservative extension.

In this sense I know of no well-established categorical foundation system.

If you just mean would it be a good idea to use the concepts of category theory rather than those of set theory for giving semantics to programming languages, the answer is maybe.

I''ve done some work with monads and arrows, which are interesting ways of structuring a semantics.

Anyway, the use of a set theoretic foundation does not preclude giving a categorial semantics to a language, if that''s how you want to do it.

As to the unsuitability of set theory, I think you are mistaken. When people give a mathematical semantics to a language they are giving a set theoretic semantics, even if they don''t mention set theory.

Even when Dana Scott was doing domain theory, this was just mathematics, and the natural way to formalise it would be in classical set theory.

Actually there is a bit of work on a foundation system which is ontologically category theoretic on the X-Logic.org site at:
http://www.x-logic.org/pp/gst/pcf-defns.html

This is a first step towards constructing a rather eccentric categorial foundation system. If this were completed it would be logically equivalent to a set theoretic foundation but conceptually rather different.  Pragmatically, I suspect it would be no easier to work with than set theory, but there would be a lot of work to get to the point where one could make a judgement on that.

Roger Jones

[X-logic-www] [X-Logic] Re: OpenMind [9]

[<com...@x-...>]

Subject- Re: OpenMind
Url- 

Wouldn''t a categorial (vice set-theoretic) foundation be a better fit to SGML/XML and current OO development technology?  Won''t argue it as a better foundation - not enough of a mathematician.  But would argue that attempts thus far to create primitive set-theoretic computing constructs have had limited success.  Trends in verification and proof technology, via functional, OO, aspect-oriented, and declarative programming appear to be more suited to a categorial approach.  Only thing I have seen with obvious set-theoretic promise is Johnson-Laird''s *Mental Models* (ca. ''83), and a paper from a AAAI conference some years ago (''88, ''89, or ''91, I think) wherein a logic was described with primitive operations similar to those of Johnson-Laird.  My interest at the time was VV&T of KBS, particularly of the underlying inference mechanisms.

Don''t claim great authority to these comments, but you did invite them.

N. L. Sizemore

[X-logic-www] Apologies and update

[Roger B. J. <rb...@rb...>]

Apologies for the last two rubbish messages, which arose
from my forgetting that the X-Logic site was plumbed into
this mailing list and using it to give a small demo of the 
comment system.

Having sorted out my Unix net access I am now putting the
x-logic cvs in order.
The X-Logic web site will then be upgraded to use the first
prototype from the "mad-portal" project.
This is like the existing site except that logged in users get
fine control over which areas of the web site they receive
email of comments/discussion.
This first mad-portal prototype will also be (probably) the
last to be based on a phpnuke hack.

My focus has moved to building an XML message store
which will be used both for the next web site technology
base and for the X-Logic engine.
This will be (at least initially) a subproject of X-Logic,
but which will have a new directory in the repository
so that it can be made available independently of the
overly complex set of experiments which are part of
the current web site build process.

So getting an X-Logic portal up has slipped in priority.
I am back to leaving the X-Logic project quiet until I
reach some worthwhile software.

Roger Jones
RB...@RB...

[X-logic-www] [X-Logic] Re: Galactic Set Theory and applications [8]

[<com...@x-...>]

Subject- Re: Galactic Set Theory and applications
Url- 

qewrwertwer

[X-logic-www] [X-Logic] Re: OpenMind [7]

[<com...@x-...>]

Subject- Re: OpenMind
Url- 

say some more rubbish

[X-logic-www] [X-Logic] Re: X-Logic - comments and mailing lists [6]

[<com...@x-...>]

Subject- Re: X-Logic - comments and mailing lists
Url- 

Another test comment.
(I need to be sure that recent changes don''t get
digitalspace and sourceforge into a loop).

Should be no s here.

Roger

[X-logic-www] [X-Logic] Re: X-Logic - comments and mailing lists (bug) [4]

[<ope...@op...>]

Well that basically worked. I need to get rid of the <br>s.
<br>

<br>
Not sure what to do about the duplicate identification. ([X-logic-www] [X-Logic])
<br>

<br>
Roger

[X-logic-www] [X-Logic] Re: X-Logic - comments and mailing lists [3]

[<ope...@op...>]

Now that we have an operational comment/discussion
<br>
facility at X-Logic.org I have decided to try out
<br>
gating the comments from X-Logic.org to the sourceforge
<br>
x-logic-www mailing list.
<br>

<br>
This gets over the present defect that though X-Logic
<br>
will mail out its discussions, there is no feature
<br>
to allow users to subscribe for the email.
<br>

<br>
I don''t expect much volume.
<br>
By the time there is any volume you will be able to
<br>
selectively subscribe at X-Logic.org to discussion of
<br>
those parts of the site which interest you.
<br>

<br>
Note that the password mailed to this list by X-Logic
<br>
has now been changed.
<br>

<br>
So far as email input to the X-Logic discussions is
<br>
concerned, note that each message comes with a number
<br>
in brackets at the end of the subject.
<br>
This is a unique comment identifier and can be used to
<br>
correctly locate replies to comments in the thread
<br>
structure.
<br>
(this is the task currently assigned to Neil).
<br>

<br>
Since most of the code for doing this is already
<br>
there in .php we need to know how to invoke a php
<br>
script other than by using apache.
<br>
Once this is known the incoming mail processing
<br>
should be trivial.
<br>

<br>
Roger
<br>

<br>

<br>

<br>

<br>

<br>

[X-logic-www] User Password for sourceforge

[<rb...@rb...>]

Welcome to X-Logic.org!

You or someone else has used your email account (x-l...@li...) to register an account at X-Logic.org. The following is the member information:

-Nickname:  sourceforge
-Password:  vetckfivet

Re: [X-logic-www] X-Logic comment system live trial

[Roger B. J. <rb...@rb...>]

Responding to: Neil Nelson Tuesday, November 07, 2000 3:49 PM

> I need to make some time to review your work to date and
> utilize the enhancements you have made to X-logic.org.

No rush.

> Though you seem to have the progress of internal production
> objectives under control, I was wondering how X-logic
> should be marketed/advertised to obtain the larger community
> of which you were previously speaking, the extenral objectives.
> The idea of ``if we build it, then they will come'' perhaps
> needs to be reinforced with ``why is it that people will come
> and then let those people know that they should come''.  If you
> come to X-logic then your life will be better for the following
> reasons.

Yes.
When I started the X-Logic project, which was my first attempt at
starting an OpenSource project, I had intended to make a good
web site first, so that people would understand what the project
was about.

However, I already had lots of words on other web sites, and
I felt that the best way to get an open source project off the
ground is to get some coding done and deliver a prototype.
So without getting very far on the propaganda I started formal
modelling, and then, modelling barely started moved on to
coding in SML.
(I might say that it took me quite a bit of hacking about to
get an "open source" way of building the web site from XML
sources which incorporated the formal modelling stuff, you can
see that the whole thing is already a bit complicated).

There was a specific rationale for getting into code which
relates to the modelling.
X-Logic depends on having XML vocabularies which are
suitable for defining the semantics of other XML vocabularies.
One intended semantic basis was higher order set theory,
and a rich higher order set theory can be defined in XML
by a cascade of language extensions which corresponds
to the HOL logic and then the various theories developed
in that logic.
So I wanted to be able to export theory heirarchies as
language definitions in XML (from ProofPower, Isabelle or
Cambridge HOL).
Then I would get most of the language definition in XML
from the work I had already done on GST.

Anyway, I ran into difficulties, and this ground to a halt.
The difficulties mainly related to my ability to keep going
without having any involvement from anyone else.
So in the summer I had a pause, and then I started looking
at the application domain and getting much more proactive
in trying to establish dialogue.

I only really started looking at X-Logic again when you
started asking questions about it, and I am at present progressing
X-Logic and my portal project with equal vigour.

So far X-Logic has not ben promoted at all, since I never
got to the point where I the story hangs together well enough
(except perhaps in my head).
When it comes to that point, there are three different kinds
of reason why people might take an interest.
(1) If we did go on to making the X-Logic.org site into
a general logic portal, then this would be a reason for
coming to the site independent ot the X-Logic project
itself.

(2) when more of the words are in place I think the
X-Logic site will be full of interesting ideas and that some
people will come just out of interest.
I also think that there are lots of prospective PhD projects
which could be based on or related to X-Logic and so it might be
worth making this clear for any postgrads who are still making
their minds up what to do.

(3) so this leads to the next level of interest, which is that
X-Logic will be an exiting project to work on.
So, again, the web site need to help people identify how
they can contribute and get something out of it.

however, for hackers rather than philosophers the interest
in a project is very much tied up with what it will do in the end.
So this brings us to the applications, and the fact that eventually
people will come to the X-Logic site:

(4) because they want to dowload and use the software!

And so far as this goes the project is currently weak in terms
of specific credible applications.
My inclination is to focus on the X-Logic query language and
to aim to prototype query engines as early as possible.
This fits in with the "make-a-difference" portal as an application
since the query engine would be the most relevant bit of
X-Logic for that application.

X-Logic query language, to a first approximation is to XML query
languages what logic query languages are to database query languages.
Except that I would expect X-Logic-QL to subsume regular XML
query languages and to deliver a prototype built over a regular
XML query engine which provides the full functionality of the regular
system plus more.
What the more is needs a bit of work to explain.

> This corresponds somewhat to either promoting a system that
> may be widely applied and/or promoting an application that is
> an example of how the system may be applied.

Yes.  We have to do both I think, but the web site at present
does neither very well.

My view is that it will be some time yet.

Roger

[X-logic-www] Strategic discussions

[Roger B. J. <rb...@rb...>]

Since Neil's last message strayed away from web-site
specific issues into more general strategic ones, and I
have not been able to revive the archiving on the other
mailing lists, I have created yet another called
X-Logic-strategy.

I will respond on the things in Neil's last message
to the X-Logic-strategy list if you don't mind
subscribing to that list as well Neil.

I am going to make one more attempt to revive
the archiving on the dev and arch lists, and if this
fails then I will leave them permanently marked
"private" and we can forget about them and invent
equivalents with different names if we need them.

I expect that eventually the X-Logic.org site facilities
will be used instead of the sourceforge forums and
mailing lists, but we aren't there yet.

Roger

[X-logic-www] comments and tasks

[Roger B. J. <rb...@rb...>]

Neil,

I responded to your comment on the X-Logic prospectus,
so go back to that page and select "view 2 comments"
to see how the "thread" looks.

For future comments you should register yourself as
a user with phpnuke (select "your account" from menu
on left after selecting view or add comment, then chose
a nickname and password).
Then if you make sure you are logged in before posting
comments they will be shown as your comments.

On the task think, the two tasks with 0% were the task
I accidentally assigned to you twice.
I have now figured out how to change that and have deleted
the second one.

You probably only saw those two because your own tasks
are listed on your personal page at sourceforge.
If you want to see the rest then you have to go off to the task
manager (and I might have to do something to give you
permission, I'll have a look at that.).

Roger
RB...@RB...

Re: [X-logic-www] X-Logic comment system live trial

[Neil N. <n_n...@pa...>]

Roger,

You are clearly making headway, and I need to get back
to you on your inquiries.

My status with installing the several software packages for
X-logic was that I had downloaded all the required software
and had installed New Jersey ML, Java, and needed to
complete the installation of: Clarke's XT that includes XSLT,
ProofPower, and the remainder of X-logic.

I need to make some time to review your work to date and
utilize the enhancements you have made to X-logic.org.

Though you seem to have the progress of internal production
objectives under control, I was wondering how X-logic
should be marketed/advertised to obtain the larger community
of which you were previously speaking, the extenral objectives.
The idea of ``if we build it, then they will come'' perhaps
needs to be reinforced with ``why is it that people will come
and then let those people know that they should come''.  If you
come to X-logic then your life will be better for the following
reasons.

This corresponds somewhat to either promoting a system that
may be widely applied and/or promoting an application that is
an example of how the system may be applied.

But let me get some things out of the way and then look at
what you have done.

Regards,

Neil

> A prototype comment system for the X-Logic.org web site is
> now in operation.
>
> If you visit the site you should find that at the bottom of the
> right column of almost all pages (the ones in the "standard format")
> is an invitation to add a comment.
> When a comment is added, an invitation to view the comments
> appears above it.
>
> Comments are welcomed, both on X-Logic.org and on the
> comment system.
> I guess, since there is no page for the comment system at X-Logic.org
> comments on that should come here (x-l...@li...)
>
> Roger Jones
> RB...@RB...
>
> _______________________________________________
> X-logic-www mailing list
> X-l...@li...
> http://lists.sourceforge.net/mailman/listinfo/x-logic-www

[X-logic-www] X-Logic comment system live trial

[Roger B. J. <rb...@rb...>]

A prototype comment system for the X-Logic.org web site is
now in operation.

If you visit the site you should find that at the bottom of the
right column of almost all pages (the ones in the "standard format")
is an invitation to add a comment.
When a comment is added, an invitation to view the comments
appears above it.

Comments are welcomed, both on X-Logic.org and on the
comment system.
I guess, since there is no page for the comment system at X-Logic.org
comments on that should come here (x-l...@li...)

Roger Jones
RB...@RB...

Re: [X-logic-www] X-Logic website roadmap

[Roger B. J. <rb...@rb...>]

Neil,

I registered you as a developer.
(at least, I hope it was you - n_nelson, if not some stranger is
probably now getting spurious task assignments!)

I made some more tasks.

I accidentally did the one for you twice, but I don't know how
to fix that.
You may be able to, I seem only able to edit the tasks assigned
to me, however knowing sourceforge it will be impossible to
delete tasks!

Don't take any of this too seriously, I'm really trying out the
sourceforge facilities a bit more, there nothing faintly like
commitment involved.

Some of this stuff we are working on here really belongs in
a different portal project I think, but we should regard this
as prototyping and will put the stuff elsewhere when we get
serious about it.

When I get closer to finishing the present exercise you
can have it to play with, so you will be able to look at
the table definitions.

How far did you get in you attempt to do the x-logic build?

Roger

Re: [X-logic-www] X-Logic website roadmap

[Neil N. <n_n...@pa...>]

Roger,

I expect you will need to add me as a developer to the X-logic
project as indicated off the X-logic Summary page under Developers

https://sourceforge.net/project/memberlist.php?group_id=1338

I did some work about three years ago with Perl and MySql.
The task you have in mind is to use Perl on two MySql tables
in phpnuke.  I seem to have Perl and PHP installed from my
RedHat 6.2.  While awaiting your reply, I can review the
available documentation for those two areas and phpnuke.

Regards,

Neil Nelson


> Responding to Neil Nelson  Friday, November 03, 2000 5:42 PM
>
> Neil,
>
> > I need to commend you on your organization.  Clearly I
> > need to find something here to assist you with that seems
> > to be, by your analysis, 3, 4, or 5 that then requires an
> > understanding of phpnuke, which I will look at tomorrow,
> > and see how I can get some effort into your objectives.
>
> 3 & 4 you can't do anything about until I get something up.
> (well you could read the web site).
> 5 you could have a look at.
> You don't need to know much at all about phpnuke to do
> the incoming mail processing job.
> You need to know about one or two tables in the database,
> and you need to know how to access a MySql database from
> a suitable language (perl would do fine).
> Have you done MySql from Perl?
> If not you could check that out.
>
> > By capturing email input with your current ISP apparently
> > means that your are now getting the email, but that you
> > want to get the email in a different manner.  The manner
> > may be that email sent to you or sent to X-logic or other
> > related addresses needs to be additionally sent to a
> > combined address or something along this line?
>
> Well, kind of.
> I want to set up new mail addresses for the portal so that
> mail sent to those addresses gets processed by scripts and
> added into the database.
>
> I have now checked with my ISP and know how to set up
> a new mail address for piping through a script, so all we need
> now is the script!
>
> This also makes the way clear for mailing list syndication.
> Which is harder of course.
>
> There's no rocket science here.
>
> I'm trying out the sourceforge task manager right now,
> just to see whether it is easy to use and actually useful.
> If you are seriously thinking of doing something then
> you should enroll as a developer and have look at what I
> am doing with the task manager.
> Then after a bit more discussion I could assign you a task.
>
> Roger
>
> _______________________________________________
> X-logic-www mailing list
> X-l...@li...
> http://lists.sourceforge.net/mailman/listinfo/x-logic-www

Re: [X-logic-www] X-Logic website roadmap

[Roger B. J. <rb...@rb...>]

Responding to Neil Nelson  Friday, November 03, 2000 5:42 PM

Neil,

> I need to commend you on your organization.  Clearly I
> need to find something here to assist you with that seems
> to be, by your analysis, 3, 4, or 5 that then requires an
> understanding of phpnuke, which I will look at tomorrow,
> and see how I can get some effort into your objectives.

3 & 4 you can't do anything about until I get something up.
(well you could read the web site).
5 you could have a look at.
You don't need to know much at all about phpnuke to do
the incoming mail processing job.
You need to know about one or two tables in the database,
and you need to know how to access a MySql database from
a suitable language (perl would do fine).
Have you done MySql from Perl?
If not you could check that out.

> By capturing email input with your current ISP apparently
> means that your are now getting the email, but that you
> want to get the email in a different manner.  The manner
> may be that email sent to you or sent to X-logic or other
> related addresses needs to be additionally sent to a
> combined address or something along this line?

Well, kind of.
I want to set up new mail addresses for the portal so that
mail sent to those addresses gets processed by scripts and
added into the database.

I have now checked with my ISP and know how to set up
a new mail address for piping through a script, so all we need
now is the script!

This also makes the way clear for mailing list syndication.
Which is harder of course.

There's no rocket science here.

I'm trying out the sourceforge task manager right now,
just to see whether it is easy to use and actually useful.
If you are seriously thinking of doing something then
you should enroll as a developer and have look at what I
am doing with the task manager.
Then after a bit more discussion I could assign you a task.

Roger

Re: [X-logic-www] X-Logic website roadmap

[Neil N. <n_n...@pa...>]

Roger,

I need to commend you on your organization.  Clearly I
need to find something here to assist you with that seems
to be, by your analysis, 3, 4, or 5 that then requires an
understanding of phpnuke, which I will look at tomorrow,
and see how I can get some effort into your objectives.

By capturing email input with your current ISP apparently
means that your are now getting the email, but that you
want to get the email in a different manner.  The manner
may be that email sent to you or sent to X-logic or other
related addresses needs to be additionally sent to a
combined address or something along this line?

Regards,

Neil


> Here's my near term plan for the X-Logic.org website:
>
> 1. Hack phpnuke so that it provides a facility for commenting on web pages.
> 2. Incorporate the facility into the X-Logic.org website
> 3. Use it for discussions about X-Logic architecture while evolving the
> architectural content of the website.
> 4. Tune it up (presentation and usability).
> 5. Add email gateway so that you can continue a discussion thread without
> going into the website.
> 6. Do mailing list syndication (so that X-Logic can gather together
> information from all related mailing lists)
> 7. Use 6 to make a X-Logic "portal".
> 8. Make a separate "make-a-difference portal" project starting with the
> above stuff.
>
> I have done item 1 and am now considering the easiest way to build it in to
> the X-Logic site.
> Ordering on the above is not intended to be linear, some items can be
> addressed concurrently, e.g. 3/4/5.
>
> A big "if" is whether and how I can capture email input with my present ISP.
>
> Feedback and help solicited.
>
> Roger Jones
> RB...@RB...
>
> _______________________________________________
> X-logic-www mailing list
> X-l...@li...
> http://lists.sourceforge.net/mailman/listinfo/x-logic-www

[X-logic-www] X-Logic website roadmap

[Roger B. J. <rb...@rb...>]

Here's my near term plan for the X-Logic.org website:

1. Hack phpnuke so that it provides a facility for commenting on web pages.
2. Incorporate the facility into the X-Logic.org website
3. Use it for discussions about X-Logic architecture while evolving the
architectural content of the website.
4. Tune it up (presentation and usability).
5. Add email gateway so that you can continue a discussion thread without
going into the website.
6. Do mailing list syndication (so that X-Logic can gather together
information from all related mailing lists)
7. Use 6 to make a X-Logic "portal".
8. Make a separate "make-a-difference portal" project starting with the
above stuff.

I have done item 1 and am now considering the easiest way to build it in to
the X-Logic site.
Ordering on the above is not intended to be linear, some items can be
addressed concurrently, e.g. 3/4/5.

A big "if" is whether and how I can capture email input with my present ISP.

Feedback and help solicited.

Roger Jones
RB...@RB...

[X-logic-www] XML

[Roger B. J. <rb...@rb...>]

In response to Neil Nelson DATE: 11/01/2000 09:07:45

I only just saw this, since the one thing I omitted to do when I set up
this list (it appears) is subscribe to it.
(which didn't stop me posting to it, I must fix that)

XML originated as a successor to HTML, but was quickly
seen to have broader application.
The role of XML as an alternative way of coding web pages
is now much less than its role in the exchange of structured data
on the web, but it is also used as a generic method of encoding
arbitrary languages.

For example XML is now used for MathML, i.e for mathematics.
It is also used for UML.

Any language whatsoever which has any connection with the web
is likely to come as or with an XML encoding.

In the above there are at least three major categories:

1. textual markup
============
As a successor to HTML and SGML.
XML used for this purpose is likely not to have a regular structure.
This has been the primary actual use of XML in X-Logic to date,
but this is not the primary intended role of XML in X-Logic.

2. structured data
============
This is probably the most popular usage at present on the web, 
and probably the one of greatest commercial significance in the
near to medium term.

In these applications an XML dtd or schema is used to describe
the constraints which ensure the regularity of the structured data.

This kind of usage will be important in X-Logic, but is not the
area in which X-Logic is expected to exell.

3. formal notations
=============
MathML is an example of this.
The concerns the use of XML vocabularies as a representation
for some language more expressive than is needed for transport
of data, but more regularly structured and more semantically
precise than natural languages or textual markup applications.

This is the primary area in which X-Logic has something to say,
but X-Logic provides a framework for combining computation
and reasoning over any XML dialect intended for category (2) or (3).
For languages in category (1) (natural languages) I would advocate
that these are handled by converters between (1) and (3) and that
deduction takes place in (2) and (3).

As far as data access is concerned, there are already XML query
languages defined and implemented (though no standard).
W3C has and XML query language as one of its future deliverables.
A logic query language for XML is one of the applications mentioned
at the X-Logic.org site, and this seems to me an important application
area (possibly commercially significant).

As far as X-Logic is concerned it is probably important to distinguish
between the use of XML in the "architecture" for the languages such
as formal logics which X-Logic is concerned with, and the use of
XML in building web sites.

As far as this list is concerned our primary interest is in the latter,
whereas the x-logic-arch list is the place for the former.
If we can get together a new generation web site then we can
use the comment/forum facilities on the site to hang a discussion
around the content of the web site and develop the story in that way.

Roger Jones
RB...@RB...

[X-logic-www] Re: [X-logic-arch] On Portal Specifications (X-Logic)

[Neil N. <n_n...@pa...>]

Roger,

In not being familiar with XML I took the 5 minute introductory
tutorial at http://java.sun.com/xml/tutorial_intro.html from which
I now have just enough knowledge to be dangerous.  My
impression of XML is that it is a data description language
with the advantages of: being oriented to web pages,
having a good and growing market share, being likely easily
used with Java.

A missing desirable component for XML available to most
database languages is a data access utility, which may be
one of your interests in developing X-logic, in that logic
operations on data via a data description language would
give the  usual data access utility the would be typified by
SQL.

Though it would be inefficient with respect to say, MySql,
to store the more commonly uniform data in XML there
does not seem to be any necessary objection.  And
perhaps there is interest in using the XML data description
language for, say, SQL data and then have an SQL
like language, X-logic, that could be arbitrarily applied to
XML descriptions.

Since XML data is intended to be web-page text it will
not typically be sufficiently uniform to apply the more detailed
SQL commands such as equality and order relations to
that data.  However, equality and order relations could be
applied to the data description elements in XML that would
then order and group classes of data according to that XML
description.  And then as certain data elements are noted as
useful for the detail relations, that increases, in a sense, the
cardinality of the model.  I.e., the cardinality of the model
may be described as the number of distinct elements that it
has, and such that the usual cardinality of uniform data is
high because of the usual easy separation of each element.
Whereas if we could only address the data description, the
cardinality would be low as represented by the smallest
model of that description, which would be the nature of
an XML description with arbitrary text elements in a web
page.  And then there would be intermediate cardinalities/
models where some portion of the data was uniform that
would add properties for the logic/SQL to grab onto.

Just beyond the immediate logic application there is some
potential to apply standard techniques to text data; e.g.,
common search/index methods on key words and
phrases could be used to obtain more properties.  This
area could be extended to cover natural language
processing that would in a model sense maximize the
commonly available properties.  I.e., there would be a
natural model cardinality break point between commonly
and specifically used languages (syntaxes). And then
specific languages such as in mathematics or other, say,
computable languages could also be addressed.  But
in moving beyond the immediate logic and order
relations the number of potential relations increases
faster than any feasible general method that then
requires specific coding to capture those relations
useful at the moment (the application specific functions).

This is just my interm global framework for what we
are trying to do with X-logic.  Does this seem to conform
to the way you are thinking?

Regards,

Neil Nelson