11. August 2009 by admin.

During the last week of July 2009 I gave a half-day workshop at Geoweb 2009, I also presented a set of slides on use of XML technologies in BIM Building Information Modelling domain. Starts in paragraph three, continues in future postings. XML, XSLT, OWL, RDF, SVG, MathML, X3D, POVray, Sketchup, FFT, wavelets, cluster clouds and more

At the 2007 Extreme Markup Language Conference, now  called Balisage, I had the pleasure of giving a presentation called ‘OntoClock, The Difference Between Having Ontological Knowledge and Knowing It’.

The presentation was about the OntoClock system, a proof of concept collection of XML technologies working together
in one system or program. OntoClock was an SVG animation of an analog wall clock, the sweep second hand ‘driven’ by
the computer system clock circuitry of the laptop  the program was running on. The point of the OntoClock demo
was not simply that an analog clock SVG animation could  correctly translate dynamic digital time values to the  correct angles for the three clock hands but that the SVG metadata element (of who I am the author) could  gainfully contain multiple kinds of metadata relating both to the visual aspects of SVG output but also to nonvisual (eg human cognitive ontological) aspects simultaneously. The OntoClock SVG file contained the NASA JPL Semantic Web Environmental Technology  Time-Ontology with some added content from myself. Using a reasoner program , such as Jena, the ontology of  time could be related to the motions and (implied) time values depicted by  OntoClock’s three clock hands. By relating sequences of values of particular OntoClock program variables and position and change of position (ie ‘motion’) of the SVG animated clock hands a [BACON-like (Pat Langley, Herbert Simon)] discovery program could through ‘observation’ (such as might be done via a Valgrind type monitor) discover that “change  of time” ’caused’ [concomitant or correlated] ‘change of location’ of animated clock hands! Using a semantic version  of Call Stack/Valgrind the system was able to detect  not only that it _had_ knowledge of time (the OWL ontology in the SVG metadata element) but also that it (the system) detected it discovered its usage and meaning  [by virtue of the reasoner]. In an ontological sense the OntoClock understood not only what time ‘is’ but also that the SVG animation depicts it ‘occurring’!
This might be said to be a kind of metaphorical understanding of time, in the sense of ‘passage of time ‘ represented by ‘movement in space’.

In the period 2008 to August 2009 the OntoClock  capabilities have been expanded to include use of the  metadata system MathML. During the last week of  July 2009 I gave a half-day workshop at Geoweb 2009 where, in part, I explained that by including MathML  content in the metadata element of an SVG based Building Information Modelling (BIM) illustration of a building (eg a house room) and by linking/associating (by means of URI RDF) ontological knowledge of mathematics (via NASA JPL SWEET numerics.owl) the SVG illustration of the room provided not only  picture, that is visible display info, but also ontological knowledge about mathematical concepts and (MathML  based) equations showing what numerical processing to perform.

Instead of merely providing a compiled / ‘binary’ function which DOES the calculation ‘when its start  button is pushed’, by providing the MathML representation of the equations or mathematical  processing involved, it is possible for the computer
to apply / perform reasoning (via Jena, and the like) on the mathematical concepts embodied in / represented by the equations depicted in the MathML. Links from the MathML, such as via its  annotation element, to the mathematical
concepts ontology numerics.owl permit reasoning about mathematical concepts identified in the  MathML equations representation, such as shown in my SVG XML BIM example at Geoweb 2009.

I showed an example of an SVG illustration which had  multiple types of metadata, including MathML with  associations to relevant mathematical concepts in numeric.owl. Applying a reasoning program (Jena)  allows the computer to make inferences about the mathematical concepts as present in the equation / formula depicted in the MathML. The computer can, for example, recognize that a certain mathematical  symbolic subpattern in an equation represents  the  concept of second derivative. It is possible to  depict calculations used in (BIM and other)  modelling by means of their conceptual terms, such as second derivative, or heat design type xyz. That allows conceptual terms, such as  second derivative, Daubechies wavelet and so on, to be used both in human visually operated design programs (CAD, Arch, etc) and also in programmatically based planning programs.  I also showed SVG based BIM building illustration  where the SVG meta data element content included a building ontology (in OWL). A reasoner program  could infer the floors were part of a building, as are walls, doors and windows, but the latter two occur in  walls not ceiling or floors (ordinarily, but with  architectural design based exceptions, such as dormer (windows) and skylights).

When the computer has (access to) ontological knowledge  of buildings / architecture and also ontological  knowledge about space (eg space.owl) and time (eg time.owl , and others) it can have this knowledge  associated with mathematical descriptions involved in modelling heat design, stress and loads, pedestrian  and other traffic, even a field-modeled western version of Feng Shui / aesthetics / Hall’s Personal Space  calculations it can analyze building designs not only by  the nuts and bolts CAD/CAM approach, such as finite element analysis, but also conceptually,  for example, knowing (ontologically) about stylistic and pragmatic required features (at the conceptual  (terminological) level the computer can then evaluate how well an architectural description /design addresses or  satisfies those required conceptual level features. A  simple example of ’self-propelled’ architectural design ‘validation’ or ‘verification’ is to have the computer  analyze / review a structural description of a building and check it for ‘completeness’ and ‘correctness’  according to the knowledge in the ontology (of buildings (, time, space, numerics, etc)).

The very beginnings of such computer activity are  already in use in computer gaming where an ontology is  used to be sure all of the required visible things of a room, house or scene are present. This performs far  greater attention to detail / analysis than is done by  a human who does ‘continuity’ in motion picture film  making endeavours. Scenegraphs are often used to do this sort of thing (in graphics productions) but  scenegraphs (traditionally) must be populated by hand
by humans. An ontologically-based validation can populate  the scenegraph programmatically, obviating humans in  this arena.

Posted in OMC Geo-Semantics | No Comments »

8. October 2008 by admin.

I do not subscribe to nor believe in the Oriental notions which underlie Feng Shui, nor am I advocating them in these articles. What I am doing is presenting a Westernized version of at least some of the tenants that comprise Feng Shui based gardens, landscapes and cityscapes. In that respect what we will see here is like a combination of CAD-CAM and Landscaping software, the former providing “physics”, techniques such as finite element analysis and the latter providing a graphical user interface (GUI) which at least provides a picture gallery of elements to be placed and a field or canvas whereupon to place them via drag and drop. The less expensive Landscaping software is 2D and looks a little cartoonish but is usable to produce simple plan view landscapes. Many of these programs provide a grid which may be toggled on or off and the items being dragged can be snapped to the nearest gridpoint. The grid allows easy linear alignment of landscaping elements. In the same way a radial pattern of lines, like the points of a  compass, overlaid by concentric circles provides a no brainer gui to do circular placement rather than rectangular placement. The predefined landscape item drawings and grids means that ones dog could produce some landscape scenario with these packages. (”On the internet no one knows that you’re a dog.” )

For most of these landscape packages there are no predefined calculations which relate the apparent quality of the user’s finished landscaping layout masterpiece. Also on the inexpensive 2D systems it is incumbent on the user’s imagination to picture using real plants etc which correspond to the rather austere gallery items and imagining what your actual house, yard etc would look like with the arrangement of stuff according to your landscaping plan as shown in the program’s gui. Some of these systems provide a cheap third axis by displaying a side view (a cheap elevation) of the still primitively stated graphical items in the program’s landscaping gallery. Rather more expensive landscape programs use 2D photographs of the plants, garden statues etc in the gallery and allow the user to include photographs (in the form of bitmaps) in the output. By providing (varying quality) ability to superimpose gallery items ‘on top of’ pictures of your own house, garage, backyard etc these systems can use a 2D system depicting the x and z axes. (Z axis is height or elevation.) Graphics systems would use the texture mapping process to place a bitmapped-picture (the texture) on top of another picture, which has information about its surface characteristics / modulation. For example a picture of a patch of flowers could be draped over a hillock in the yard and the texture mapping capability of the graphics system would alter the ’shape’ of the flowers patch (picture) as though it were some cloth material draped over the exact shape of the hillock.

One can of course use graphics programs to produce pictures of landscaping scenarios. If one were to use a program such as Blender to produce a landscape picture one could include lighting and other calculations which the graphics package does well. In graphic programs which provide user access to its scene-graph the user can then use rules and calculations such as CAD-CAM software and movie and game making software do. Vector graphics languages such as SVG are quite useful for this sort of thing.

In our discussion we will see that Chi is the most important concept. In the Chinese system Chi is ‘life energy’, often refered to via references such as ‘breath’ (and wind). In these articles on Feng Shui we are not talking about the existence (or non existence) of things like this, we are talking about MODELs of these ideas where western notions of physics and geometry are used to depict the Feng Shui ideas by means of computer graphics. chi (lowercase) is a scalar variable in our article, it represents an amount or magnitude. Magnitude of what is left unstated in so far as the variable chi itself is concerned. What matters is that certain actions, to be explained, can increase or decrease this magnitude, and we will see how this models aspects of Feng Shui.

Our Feng Shui landscaping gui has many of the same things as the programs described just above, grids, predefined gallery of placeable items, a visual layout field to place them on and so on. Additionally our system has a ‘knowledge base’ of rules which model Feng Shui notions. Additionally there is a library of computer functions / programs which perform (modelled) Feng Shui actions. Anyone who has used the simpler landscaping programs will see a large overlap but there are some important differences.

In addition to the chi variable there are also yin and yang scalar variables, Yin and Yang ‘verbs’ or actions which are implemented via items in the predefined library of computer functions / programs. Also there is the shar scalar variable; a radial function, which is used to implement things like a compass and some others which will be discussed in the next article. There we will see a description of how the elements of the model are used to model concepts used in Feng Shui.

Posted in OMC Geo-Semantics | No Comments »

5. October 2008 by admin.

One of the several topics in the inaugural article of this blog ( of August 15 2008, “Hello Geo-Semantic world!”) relates
to the concepts / notions of Edward Hall, who wrote a number of books one of which is “The Hidden Dimension”. He talks about the concept of “personal space”. In the series of articles about Hall’s ideas (presented in this blog) we see that he describes “personal space” and how it relates to buildings, cities and so on. His ideas will be ‘computerized’ for presentation in this set of articles (on Hall). Note that his ideas have nothing to do with Feng Shui.

Look at Worldcomp08 AI paper “Navigation in Dynamic Environments Utilizing Translation of Fuzzy Sets” by Jones et al., Dept Elec. and Comp Eng., University of Denver.
Among the several things of interest in that paper is the use of ‘artificial potential fields’ of attraction (for goals) and repulsion (for obstacles), and the robot controlled by this system follows the path of quickest decent through this potential field to reach the goal.

This notion of ‘layout’ (ie the environment or ‘world’) and affect caused by various items (modelled as emitters, centers of potential fields which ‘extend’ from them) I use to depict some of the Feng Shui concepts too. (Ch’i for example.) And it is also interesting that Edward deBono, in his masterpiece “Atlas of Management Thinking”, also uses some of the same visio-cognitive notions, like inertial ‘continued motion’ of ‘objects’, such as arrowheads in a diagram entering a container. Cast your memory back to the time that you looked at a movie theatre marquee. It looks as though the bar or arrow, which appears by virtue of the lights, is moving. When you look at such a marquee you do see moving things such as bars and arrows. When you  analyze what is actually happening though it is NOT movement but lights going on and off, period. The movement, called “apparent motion”, ‘occurs’ in your mind not your eye or at the marquee lights. These concepts: apparent motion, artificial potential fields, and the like are useful to the western mind in modelling the concepts which we see discussed in our articles on Feng Shui (FS).

Einstein modelled space as a surface which had depressions in it where a mass, such as a planet or a star, was at the bottom of these depressions. Thus any real object moving through space would be affected by the declined surface and ‘move’ (’roll’) toward the mass. In this way the description models / “explains” gravity and mass acceleration.

In this article:
Having once recognized the context that one is Feng Shuing one sets about using the tenants of FS to locate the right stuff in the right places. Recognizing the correct or most relevant FS context associated with a situation helps considerably in selecting the correct FS activity.
This is to say that, to a western mind, FSing activity conveniently groups into categories and these categories can be used as contexts.

Defining / recognizing the FS contexts will become clear as we look at the aspects of the computerized version of FS. If you are not at all interested in computerizing / programming or physics or logical development or Cityscapes then you would be happier skipping on to the last article of this series on FS, or perhaps enjoying the articles on the other topics in the inaugural article of this blog, ( of August 15 2008, “Hello Geo-Semantic world!”), and be sure to visit the web site open-meta.com.

Locating the right stuff in the right places — is what FS is all about. So it makes sense to explain what right places are, and what the right stuff is. In this case stuff usually means plants, objects or constructions such as flowing water streams. The selection of what stuff to place in what place is done by use of rules. (Rules: This is one way, using a computer, to depict the tenants / notions which constitute FS.) (An additional way to ‘explain’ things to a computer is to define (one or more) ontology, perhaps in OWL, which shows the relationships among the various concepts comprising the domain which the ontology is about.)

Thus a base set of rules, which provides the knowledge of FS to the computer, can be used to design the FS activity. That is placing and in some cases constructing. In some cases we can use pretty straight forward geospatial constructs to do things and in a few cases we have to resort to so-called ‘fuzzy rules’ (FR) instead of using the usually crisp rules of the system. FR are used to depict or represent notions which are too abstract to represent at the more concrete level of geospatial terms such as geometry, and scientifically accepted causalities. (eg  Gravity, (electro)magnetism, electricity)

We see fuzzy descriptors in constant daily use, take for instance the terms, ‘large bolder’, ‘wide stream’, ‘fast stream’. If we used GML how would we depict ‘large’, ‘wide’, ‘fast’? FRs allow us to capture such abstractions in a consistant meaningful way which the computer can reliably process. We conceptualize with FRs when we (daily) use concepts like ‘near’, ‘big’, etc.
In the next article part on FS we see how to define the basic elements which are used to design / lay out an Feng Shuied space.

Posted in OMC Geo-Semantics | No Comments »

4. October 2008 by admin.

In this article we continue looking at Feng Shui (FS). Using a western viewpoint on understanding the apparent tenants of FS well enough to offer a western expression of it. One of the things many westerners like to do is to use (or seemingly use) western technologies to describe things , to model things, and such. A subset of those folks seem to agree that the best way to do that is to have ’scientifically expressable’ theories and descriptions which should be amenable to (logically) cranking out some form of ‘truth’ (such as philosophy), or at least a ‘considered answer’.

Quite often it is thought that if a computer could be used to perform much or all of the “activity / processing” then the results of that would be ‘fair’ or ‘neutral’. (of course part of the western mentality is the penchant for numbers and philosophers are aware of the bias that that brings to western style deliberation. Picasso tried to visually express what this numeric penchant does to our western perception.)

Given the western penchant for numbers and science what we will see in these articles about FS is a computer graphics and physics version of ‘what FS is’ and ‘how it is done’. Provided with an appropriate ontology and rule base we can use a ‘CAD-CAM system’ perhaps even like one that is used to design super duper passenger aircraft. FS, in western hands, is amenable to being computerized to the extent that its notions can be incorporated into the scene generating software of ‘first person shooter’ 3D computer graphic ‘action programs’ (ie games). Mediums and sensitives would be aghast in
horror in reading that. Taoists would just smile, summoning the effort not to visibly giggle. In the next part of this article on FS we see how the notions of FS are ‘computerized’. (Historically, accounting was a skilled occupation, people scoffed at the idea that it could be automated. What this means is that some future version of Roomba might be able to Feng Shui your living room and other parts of your house.)

Posted in OMC Geo-Semantics | No Comments »

3. October 2008 by admin.

If Feng Shui can be said to be anything it is a collection or ensemble of ideas: collectively referred to as Feng Shui, which loosely into English is “wind and water”.  When Feng Shui (FS) notions are depicted in the Western world, such as in Canada and USA, English words are used.  These words are in books and mostly they are neither literal translations of Chinese notions of Feng Shui nor literal translations of any formal Chinese writings on the same topic.

Instead what we seem to have in the West is a westernized viewpoint which recognizes some similar ideas in the Chinese Feng Shui notions that we westerners hold ourselves. Some things in the Chinese Feng Shui notions seem like or same as our own. Some other notions in the Chinese Feng Shui notions seem definitely unlike or different from our own.

In the westernized, English version of Chinese Feng Shui notions westerners, like those of us who live in Canada and USA, Australia, New Zealand, and many parts of Europe see what to us is an Asian or “Chinese” set of beliefs which seem to comprise the cornerstones of what Chinese Feng Shui “is” and “how it is done”. Some westerners would call these
particular beliefs superstitions. These same westerners might be surprised what a truly Asian mind might see in the tenants of Christianity and Catholicism. So at this point we are approaching the always contentious grounds of religion / religious belief. With one side calling all other sides superstition, voodoo, hogwash and less gratious things. There are many
flavours of magic and humans have fought and died over them for centuries. This article is not about religion, nor how some westerners (some violently) respond to some of the belief-notions underlying Feng Shui. The article is about discussing concepts / notions of (admittedly western view of) Feng Shui (FS).

Furthermore, the article is about discussing FS in the light of notions associated with Cityscape, and potentially the Cityscape oriented conference at Geoweb 2009. (Let me make it clear right away that the Geoweb folks have no intention of making things like FS formal aspects / parts of the conference, any more than they will make the “correct” way to oil-paint scenes of the city of Rome or ones in Greece.) Next we look at discussion as to how a quasi-formal view of Feng Shui might be made via the western (conceptual) palate.

Posted in OMC Geo-Semantics | No Comments »

15. August 2008 by admin.

I have presented papers at various computer conferences and I will discuss relevant material from those in this blog. One of the things I have been doing in the last few years is doing work in the geo-spatial realm. This realm is of interest and value to me in that being able to do relevant processing of spatial things is at the center of autonomous mobile robotics systems. I do not talk about robotics hardware or building these things because that is not my focus in my area of robotics. It is about reasoning, interpreting, and understanding.

My area of greatest interest is in providing the design of software which can be used to operate an autonomous mobile machine, using its own sensors and sensors from other devices it is in network contact with. Some people call this meta-programming, programs which operate / control other programs and in some cases does modest program generation itself.

I want to launch into a discussion now about some  material which is of interest in the topic domain of the GeoWeb 2009 conference, CityScape. First Cityscape domain is of interest to me because it has a participation in what an autonomous mobile robotics system must be able to functionally cope with, and I believe, that in order to cope best with such a complex environment, which includes many humans, not just things like buildings and so on, it is necessary for the computer to be able to have a modicum of understanding about what these Cityscape things are. A definition of Cityscape is not sufficient for this, the computer rarely can do any useful functioning itself based on just having the text of Cityscape definition placed into it.

This is where the ontologies and various XML technology representations (XML, SVG, KML, OWL, etc) come into play. Unfortunately we cannot educate a computer the way we can humans, just by talking to them (in some natural language such as English). The way we STILL have to ‘educate’ computers is by programming them! Part of the process of educating computers about the Cityscape domain, as opposed to educating people about that domain, is by providing programs and (structured) data.

One view of Cityscapes is it is about super-CAD-CAM programs and data sets but at the building, city and even country sized scale. Many ‘Cityscapers’ would frown on this view because it is much less than the coverage of data and concepts which they intend for this stuff (Cityscapes). Also they would perhaps frown on that view as well because CAD CAM systems tend to be rigid and ’sharp-edged / literal’, and almost none of them have facility for including anything to do with “the human dimension”. It is the human dimension after all that brings buildings, cities, parks and such things into existence. Buildings and cities do not grow like crystals from geological or mineralogical processes. Buildings and cities occur, they are designed and built, only because there are sentient beings, humans, which desire and therefore create these things.

Caves and forests were nice for a while and what nature provided us was used and appreciated. However, we could ’see’, with our mind, that something better could be had if we constructed it ourselves instead of sticking to only what nature provided us. It is interesting that we still use almost exclusively nature provided materials to build these items with: dirt, sand, minerals, metals taken from the ground and processed into a different form than nature seemed to provide. Even our electric power largely comes from minerals and metals (coal, uranium) from the ground.

It is the understanding of various things which the human mind has / can learn that has allowed us to greatly surpass any civilization that remained solely with naturally provided shelter and energy. Even our food is still the stuff of the ground, we have just organized the husbandry, occurrence, location and processing of what nature provides to suit larger group sizes.

It is our knowledge and logical powers that have allowed us to achieve such things as Cityscapes. Some of this knowledge can be captured and represented in computers in ways that allow the computer to do useful processing with it , not merely display it for human eyes.
Taxonomies and ontologies are means of organizing data such that the computer can perform reasoning with it. Relational data base technology allows us to store large amounts of raw data but the processing capabilities of data base systems is rather limited in the areas where the computer is meant to operate with the data, rather than humans. Logical processing by machine, using ‘reasoners’ (reasoning programs), employing ontological data can refine it into ‘knowledge’ because these kinds of systems are able to apply logic (such as first order predicate logic, description logic) to these ontologies and make inferences!

This blog will have discussion about ‘the human dimension’ of Cityscape, from the viewpoint of computer systems technology. Amongst the things that we will see is discussion of

a) the use and representation of space and time (vis-a-vis Cityscape oriented domain),
b) Edward T Hall, and his concepts, now called “Personal Space”, which he called proxemics when he first wrote about them,
c) the connection of Hall’s ideas to the concepts at play in “Feng Shui” (pronounced fung-schway); how A-STAR (A*) and ‘field potential’ modelling does and does not represent notions at play in “Feng Shui”,
d) how architects incorporate human sensibilities into their designs,
e) the cross-ties of Hall’s concepts and Feng Shui notions with OASIS HUML, and offshoots of this technology such as animation of 3D faces, with discussion of those graphics / representation systems,
f) ontologies of space and time, different _points-of-view_ (POV) about what they ‘are’ represented in the various ontologies of those domains. This sub-discussion shows how it is possible to represent more than one ‘ground truth’ in logical systems. (Humans do this on a daily basis.) This is a ‘heavy technical’ co-version of item a), talking about / using: OWL, XML, XSLT, KML, SVG data sets.
g) how SVG can be used to do 2 1/2D and 3D visuals. The use and importance of the SVG Metadata element. SVG Filter technology and texture-mapping. transformation into KML.
h) 3D (”first person”) computer games and how they are made more like real life, so-called AI techniques used so that, for example, human-figures move in ways that are like real humans, not just like algorithm-operated cartoons; they can’t fly nor walk through walls, everything that is supposed to be in physical existence in the game is affected by gravity, and has mass (inertia). Bridge building engineers have and use such knowledge in the bridges, as do highway engineers. Without incorporating the knowledge, which is discussed, about the ‘world’ (like gravity) such games would have ludicrous / impossible  ‘actions’ happening. Such knowledge must be included in a Cityscape system, in addition to knowledge / representation of ‘the human dimension’.

There needs to be developed a computer subsystem which can do for rational-activities what scene-graphs do for the visual constituents of computer visual displays. (OASIS HUML, was designed in part to look at ways of doing this.) Schank and Abelson did some development in this arena, that will be discussed and the limitations of that approach explained. ‘Eidochronic Behaviour Grammar’ a paper in ‘Science’ (a while ago) will be discussed and why it never went anywhere will be discussed, as well as ‘Tale-Spin’ and why it hasn’t seemed to have gone anywhere. Knowing the things that dont work (and why) is at least as important as knowing about the things that do (work). And, yes, this stuff is relevant to the geo-semantic web.

Story-Telling seems to have made a come back in the public’s interest. Functional discussion of this activity will be discussed in this blog as it does relate to what the computer ‘might spawn’ at the location of some geo-reference in addition to or instead of ‘a (KML type) marker’ [with (its) content].

For instance, some museums have instituted a facility whereby as vistors pass near to an exhibit a description and or narrative about that object ‘plays’ (audibly) in an short range RF receiver which the visitor carries for that purpose (loaned to visitor while in museum). Why not have a podcast/soundtrack available at the point of a geo-reference, or a video ’short’, why not a travelog or campfire/history story. Isnt it easy enough to have URI links (to podcasts, etc etc) from a geo-reference!? Already street-scenes are available for viewing as a result of ‘opening-up’ some geo-reference point in a viewer. Back in 1999 I made an SVG animation that had a synchronized soundtrack that ‘played’ during the animation. Why not link-up SVG and KML and do things like this for geo-program users?

Check out open-meta.com for information.

Posted in OMC Geo-Semantics | No Comments »