Wed Jun 03, 2009

'Natural' artificial intelligence

What if artificial intelligence was a natural outgrowth of the work we do all the time? what would happen would be that UI's would be programmed looser and looser over the years, slowly able to handle more and more variable input. It would be the natural way we organize software. Instead of making software tighter and tighter with validation rules and strong typing, the typing and validation would become more relaxed, there would be fewer and fewer types, or maybe as many types but they (types) would become more and more
complex and powerful.

It would start with parsers to handle numbers and dates in different formats, simple commands in many guises. For example, programs would allow 'bye', 'logout', 'quit', 'log out' and other forms of the program closing command. It would continue with richly embedded characterizations of words so that different programmers would add characterizations and use the ones they needed. Ways would be discovered to merge things rather than having them precisely separate.

Major projects would get going to use this foundation of variability tolerance to give programs an idea of what was going on, then after a while there would be projects focusing on various computer cognition design models. Various different models of cognition would be tried and the few that really worked well would be expanded on. People would expect some level of actual intelligence from programs. It wouldn't be human intelligence, it would be very much stronger in some areas and very much weaker in others. Artificial intelligence would be a day to day thing.

So why has this not happened? Well, for one thing, we as programmers want everything nailed down very tight, with specific objects, validation rules, and strong typing. second, we don't have a standardized way of merging concepts. For another most software is proprietary so the base of intelligence artifacts that would have been built up over the years is prevent. Another problem is no standardized way to combine intelligence artifacts.

The endeme data structure would solve some of these problems. It allows both definition of meanings and matchability between meanings. An endeme is a partcular instance of a combinable orderable list of meanings. For example the list of your three favorite colors in order is an endeme and color and shape are endeme sets. Figuring out how to stack endemes and relate endemes of different endeme sets is something I am still working on, but if thousands of programmers had been working with them for the last 30 years we would have figured it out.

My (half baked) proposal:
  1. 1. start using endemes,
  2. 2. companies allow combinable intelligence artifacts to be published,
  3. 3. Start big bases of word characterizations using endemes.
  4. 4. Figure out how to match, stack, translate and combine them
  5. 5. Companies allow and programmers use highly complex validators in UI's to process user input text.
  6. 6. Come up with more combinable artifacts besides just endemes.
  7. 7. Do some sort of artifical creativity wherever possible, even if it's just variable colors on a UI background
  8. 8. combine the sciences of fuzzy logic, probability, and endemes


Endemes are good for AI analysis and they are good for AC Artificial creativity.


[0] comments (127 views) |  | [0] Trackbacks

Posted by: Jon Grover | Jun 03, 09 | 12:35 pm

Sat May 09, 2009

Peak Years of Various Personal Technologies

Here are the peak years of various personal technologies according to my history theory:

5352 BC - peak of personal *farms?
2661 BC - peak of personal spears?
 703 AD - peak of personal houses?
1377 AD - peak of personal books?
1713 AD - peak of personal *sailboats?
1881 AD - peak of personal guns
1965 AD - peak of personal cars
2007 AD - peak of personal computers
2028 AD - peak of personal *phones?
2038 AD - peak of personal AI's?
2043 AD - peak of personal nanobots?
Each of these peaks is in the exact center of various advancement ages.

The ages are: barbarian age, civilization age, dark age, enlightening ages, federal ages, generator ages, holocaust age, internet age, one world age, microbotic age, sleepless age respectively.


[0] comments (110 views) |  | [0] Trackbacks

Posted by: Jon Grover | May 09, 09 | 3:48 pm

Mon Mar 30, 2009

Name Your Top 9 Dragon Characteristics Poll

Come visit my new poll at Dragon Characteristics at PollDaddy.com. This fits into artificial creativity how? Once I have a list of characteristics I can create random dragons from them. And this could be used for more serious polls. as I think of them. Thanks.

If you liked this, I have added a survey of Elf Characteristics at Zoomerang.

[1] comments (425 views) |  | [0] Trackbacks

Posted by: Jon Grover | Mar 30, 09 | 5:56 pm

Wed Mar 18, 2009

Endemes are a Human to Computer Language

One thing artificial creativity needs is a good human to computer language.

Endemes are a vehicle for computers and humans to communicate. They are a potential mutual language for humans and computers. Of course, all computer languages and applications are mututal languages for humans and computers to communicate, but there is something special about endemes. They allow easier communication for both parties. Computers find applications too rigid to say very much or to have much control over. Humans find computer languages too technical and too primitive to be able to say much in them. A lot can be said with endemes, and computers find them simple enough to understand.

Now I need to figure out how to make endemes work with each other, so that humans can say more than a single word with them.

I am a programmer because I want to be able to talk to my computer. But it is still a dumb machine. It doesn't have the ability to talk to me back. Maybe if I built a computer brain based on endemes it could talk to me back. It might look a little like the endeme index I've been building, except that instead of the links in the index showing the best matching endemes in the list it would link to other endemes based on 'knowledge'.

But then I would have to build an 'index-like' structure that included more than just one endeme set. Ultimately it would have to manage and create its own endeme sets.

This is all possible but it's going to take some work.

[0] comments (150 views) |  | [0] Trackbacks

Posted by: Jon Grover | Mar 18, 09 | 12:23 pm

Sat Feb 28, 2009

Brief History of the last 287 thousand years

Here is a brief timeline of the last 287 thousand years focused on spiritual events.

years     vectors  spiritual hist.  inventions and revolutions               level
_________ ________ ________________ ________________________________________ ___


 287-230K       v4 Adam & Eve?      language revolution 'I'
========= ======== ================ ======================================== N21
 230-201K v1                        penultimate ice age...
 201-172K   v2
 172-144K     v3
 144-115K       v4 Adam & Eve?      team revolution
--------- -------- ---------------- ---------------------------------------- O22
 115-100K v1              
  100-86K   v2
86-72K BC     v3
72-57K BC       v4 Adam & Eve?      technology revolution, Toba
--------- -------- ---------------- ---------------------------------------- M23
57-50K BC v1             
50-43K BC   v2
43-36K BC     v3
36-29K BC       v4 Adam & Eve?      genetic integration revolution
--------- -------- ---------------- ---------------------------------------- L24
29-25K BC v1                        genocide invented
25-22K BC   v2             
22-18K BC     v3          
18-14K BC       v4 Noah             full human & language(am) revs, Med. floods?
========= ======== ================ ======================================== N25
14-13K BC v1       Tower of Babel?
13-11K BC   v2              
 11-9K BC     v3            
8900-7148       v4                  tribal, travel/trade net, society revolution
--------- -------- ---------------- ---------------------------------------- O26
7148-6251 v1       Japheth dominant...
6251-5353   v2              
5353-4456     v3           
4456-3559       v4                  urban and nationhood revolutions
--------- -------- ---------------- ---------------------------------------- M27
3559-3111 v1       Ham dominant...  material production invented
3111-2662   v2                      Egypt United
2662-2213     v3                    Great Pyramid
2213-1765       v4 Abraham          time's arrow and surplus revolutions
--------- -------- ---------------- ---------------------------------------- L28
1765-1541 v1       Shem dominant... invasion and alphabet invented
1541-1316   v2     Exodus, Moses    Joshua, Alphabet, Hammurabi?, Thera erupts
1316-1092     v3   Israel sins      Hammurabi?
 1092-868       v4 David, Solomon   Troy falls?, prophesy, scripture revolutions
========= ======== ================ ======================================== N29
  868-756 v1       Divided kingdom
  756-643   v2     Israel deported  Rome founded, Fall of Egypt & Israel
  643-531     v3   Babylonian captivity, Thales, Buddha, Fall of Judah
  531-419       v4                  Socrates, Confucious, empire&philosophy revs
--------- -------- ---------------- ---------------------------------------- O30
  419-363 v1                        Plato
  363-307   v2                      Aristotle, Alexander the Great
  307-251     v3                    Pyrrhic Wars
  251-195       v4                  Punic Wars, Siege of Syracuse, no revolution
========= ======== ================ ======================================== N29
195-83 BC v1                       
  83-30.5   v2     1st Christmas  
 30.5-143     v3   Jesus raised     the church forms, New convenant
  143-255       v4                  The Christian revolution
--------- -------- ---------------- ---------------------------------------- L28
  255-479 v1       Catholic church  First schism, crisis of 3rd century
  479-703   v2     Islam Invades    Fall of Rome, Krakatoa caldera erupts
  703-928     v3                
 928-1152       v4 Great Schism     The zero revolution
========= ======== ================ ======================================== N29
1152-1264 v1                        Invention of Science, universities founded
1264-1376   v2                      Black Death
1376-1488     v3   Reformation      printing press
1488-1601       v4 Jews expelled    Discovery of America, colonial, balance revs
--------- -------- ---------------- ---------------------------------------- O30
1601-1657 v1                        East India Companies, 30 yrs war
1657-1713   v2                      World War 0 (1519-1714)
1713-1769     v3                    Treaty of Utrecht, age of relative peace
1769-1825       v4 First Horseman (Napoleon?), Industry, federal, American revs
--------- -------- ---------------- ---------------------------------------- M31
1825-1853 v1       Academics reject mass production
1853-1881   v2     Academics reject Japanese industrial revolution
1881-1909     v3   Academics reject telephone, automobile
1909-1937       v4 Einstein         WWI, relativity/physics/cosmology revolution
--------- -------- ---------------- ---------------------------------------- L32
1937-1951 v1       Second Horseman (Hitler,Stalin,WWII?), Computers, cold war
1951-1965   v2     Israel returns   computer languages
1965-1979     v3   Israel survives  personal computer
1979-1993       v4                  Iranian, Usenet and Chinese Industrial revs
========= ======== ================ ======================================== N33
1993-2000 v1                        Graphical web browser, intel agents?, info?
2000-2007   v2                      911
2007-2014     v3   Third Horseman (Financial collapse?)
2014-2021       v4                  Global system, ocean, agents, moon/space rev
--------- -------- ---------------- ---------------------------------------- O34
2021-2025 v1               
2025-2028   v2        
2028-2032     v3        
2032-2035       v4 Fourth Horseman? 
--------- -------- ---------------- ---------------------------------------- M35
2035-2039 v1v2                    
2039-2042     v3v4               
--------- -------- ---------------- ---------------------------------------- L36
2042-2046 v1v2v3v4                   
--------- -------- ---------------- ------------------------------------ N37-L44
2046-2049          Tribulation & Apocalypse
--------- -------- ---------------- --------------------------------------- N45?
2049      v1       Millenial Kingdom
[0] comments (182 views) |  | [0] Trackbacks

Posted by: Jon Grover | Feb 28, 09 | 2:40 pm

Thu Feb 12, 2009

Kondratieff Waves Indicate God's Inspiration into History

Note, this extension of my history theory does not have Biblical support. It also does not not have Biblical support. It's just to early to tell. I'm here just noodling around with some ideas which may or may not play out. I've been beating up on my history theory for over ten years and have found mounds of evidence to support it. With this extension, I have no evidence yet. It's just an idea. I don't even have it fully fleshed out yet. That will come in time.

My history theory seems to explain things only down to the level of a quarter of an age. below that level the apparent randomness of events seems to defeat its explanatory power. I've been looking for a way to get into finer grained detail, and until now I have been stymied. Mainly because it would be real useful in building my artificially creative time travel game to have a finer grained model than a quarter of an age. Kondratieff waves may have shown me a way, or perhaps not. In any case I still don't know if the pattern is based on draw upwards in the center of the waves, propagation from earlier in history into the waves, or individual branchings at each node inside the wave pattern as I display in the second diagram below.

Kondratieff waves show a connection between the spiritual and the physical world when combined with my history theory. According to some Kondratieff wave theorists, Kondratieff wave cycles are 84 years long. My history theory is that each age is either half as long or twice as long as the one previous depending on whether history advances or declines, and ages exist in objective reality. The ages in my theory are ~ 7*2^(35-level) years long, and potentially reach asymptotic advancement roughly once every 21*2^n years where n is an integer, or if n=3 once every 84 years. Here is a diagram of history using my history theory showing just ages and no waves.

                                                                30    703  1376  2049
                                                                 ^     ^     ^     ^
                                                                 |     |     |     |
                                                                 |     |          ++China
                                                                 |         America||
                                                                                 +++
                                                            Rome          Britain|||
                                                             +=+               +=+-+
                                                  Greece     | |Canonic   Spain| | |
                                                        \+===+ +===+       +===+---+
                                     Mesopotamia         |Grc|enl| |       |Spn|enl|
                                                \+=======+---+---+-+=======+-------+
                                                 |Mesopot|  war  | |Arabia |  war  |
                                 +===============+---------------+---------------+--
                                 |     Egypt     |     urban     |     urban     | 
 +===============================+-------------------------------+------------------
 |         Hatti (trade)         |             trade             |             trade
=+---------------------------------------------------------------+------------------
 .                               .               .       .   .     .       .   .   .
7K BC                          3559 BC         1758 BC  868 419  255AD  1152 1601 2049

                                                              |<- expanded below ->|


The top diagram shows history from about 7000 BC to the present.

The bottom diagram expands the section from Rome to the present. Here are how Kondratieff waves fit into my history theory. A Kondratieff wave lasts about 84 years and each age in my history theory lasts about A = 7*2^n years. So a Kondratieff wave is 3*A where n = 2; I have divided recent history into Kondratieff waves in the top scale and history theory based sections = 7*2^4 in the bottom scale. The First Kondratieff wave centered around the ministry of Jesus Christ. We are presently running toward the end of Kondratieff #24, so things are getting a little dicey economically. Kondratieff wave #24 was based on telecom. This shows that Kondratieff waves are based on the inspiration of God. It might be considered his breaths into the world.


To view this diagram more easily you could copy and paste it into Notepad or some other text editor and use Lucidia Console as the font. Thanks.
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Kondratieff              |  church   |  crisis   |     ?     |     ?     |     ?     |peaceof God| crusades  |     ?     | printing  |  spices   |   cloth   |  telecom  |        
waves (84 yrs)         Jesus   |pax Romana | East Rome |     ?     |     ?     |     ?     |  vikings  |university |     ?     | newspain  |newengland | railroads | biotech      
                         |  v     v     v     v     v     v     v     v     v     v     v     v     v     v     v     v     v     v     v     v     v     v     v     v     v     
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Political                            ?                       ?                       ?                       ?                       ?                       ?                    
waves (168 yrs)          ?               |<      ?      >|       |<      ?      >|       |<      ?      >|       |<      ?      >|       |<      ?      >|       |<      ?      >|
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Religious              Jesus         v    Eastern church,    v    Islam destroys     v    Western Church     v      Reformation      v         Great         v     Armageddon &  
waves (336 yrs)      & church        v  Gnosticism defeated  v       Paganism        v    & Chiristendom     v     & Reconquista     v       Awakenings      v     Jesus Return
----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
                       30.5   115   199   283   367   451   535   619   703   787   872   956  1040  1124  1208  1292  1376  1460  1544  1629  1713  1797  1881  1965  2049
                         |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |
wave number:             1     2     3     4     5     6     7     8     9    10    11    12    13    14    15    16    17    18    19    20    21    22    23    24    25
                         ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^     ^
                         |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |     |
                        /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|\   /|
                        |||   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | 
                       +++++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++ ++
                       | | | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |   | |
                     +-+ | +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +-+   +=+
                     |   |   |   |*      |   |*      |   |       |   |       |   |       |   |       |   |       |   |       |   |       |   |      *|   |*      |   |China
                 +-=-+   |   +-=-+ 157   +---+ 325   +---+       +---+       +---+       +---+       +---+       +---+       +---+       +---+   1755+---+1839   +===+ (enlightenment?)
                 |       |       |       |               |       |               |       |               |       |               |       |           *   |       |Pacifica/Japan (war?)
         +---=---+       |       +---=---+               +-------+               +-------+               +-------+               +-------+         1769  +=======+      2035
         |               |               |               |                               |               |                               |               |America     |2021
=========+                               +---=---=---=---+                               +---------------+                               +===============+(industry)| 2007
  Rome   |                               |               |                               |               |                               | Britain (balance)      | |1993
(balance +===============================+               +---=---=---=---=---=---=---=---+               +===============================+                      | | 1979
 failed)      Canonic (enlightenment)    |                                               v               |      Spain (enlightenment)                           | |1965
                                         +===============================================================+                                                      | 1951
Each character is about 14 years wide                                Arabia (war)        ^                                                                      |1937
                                                                                                                                                                1923
 |       |       |       |       |       |       |       |       |       |       |       |       |       |       |       |       |       |       |       |       |       |
307     195     83     30.5     143     252     367     479     591     703     816     928    1040    1152    1264    1376    1488    1601    1713    1825    1937    2049
       Rome            Jesus           Germans          Rome           Islam           Zero    Nations  Science      Black     America         Peace of       WWI &
       conquers        rises           invade           falls          invades         used    begin    discovered   death     discovered      Utrecht          WWII


Double lines (='s) indicate the main flow of history. Carats (^'s) indicate points of potential infinite advancement aligning with the Kondratieff wave centers had history gone that way. They also indicate the pull upward God has placed on human society since the point Christ set this up in 30.5 AD. There may be other kinds of pull upward at other harmonics of 84 years but I haven't seen them (for example 42, 21, 10.5 etc years). Spiritual ramifications follow the same rule. In 928, the spiritual ramification set off by Jesus finally reached the main stream of history. In 1152 history pulled out of its tailspin.

[0] comments (196 views) |  | [0] Trackbacks

Posted by: Jon Grover | Feb 12, 09 | 6:45 pm

Tue Feb 03, 2009

Resource Attributes of Endeme Sets

Endeme Sets frequently (perhaps always) consist of a list of items geared to use up or fill up a particular resource. The EndemeSet class I have built contains a resource member to contain this information. The resources can be thought of as budgets, or just as resources. Here are some examples:

Example endeme sets



Example budgets



Example resources



Defn: An endeme set is a combinable orderable list of possible characteristics in a resource.

To some extent an endeme set defines what "all possible" means.

[0] comments (175 views) |  | [0] Trackbacks

Posted by: Jon Grover | Feb 03, 09 | 1:09 pm

Fri Jan 23, 2009

Column Objects May Breed Information Oriented Software

I've discovered in doing artificial creativity designs that I am working with information rather than data. The distinction is not well defined. I see information as data that has meaning to people, and data as information that does not have meaning to people, only to a computer. The endeme data structure that I am now using for many of my artificial creativity designs is meant to hold information rather than data. Information is meaningful to a human. Data is meaningful to a computer.

I have become interested in the following way to shift from data programming to information programming. Create classes for columns in a table rather than for rows. The standard approach is to make classes for rows in a table so that each column ends up being a field in the class. This is strong typing, it also resists a generic approach to analyzing the data in the class or the table. Working from the other axis is harder and more prone to the problems associated with weak typing and mis-modeling the natural objects in a table (the rows). However, one begins with a string for the column name, an approach to the display of that data, an analysis approach, and an approach to the display of that analysis. Combining these things as fields in a 'column' class forces a more generic approach. And it is still 'object oriented', thus somewhat controllable.

Of course taking this course I expect to be crashing and burning sometime quite soon. Maybe it is through failure that this direction will succeed in a business setting. If the system becomes a failure, then the people asking for things from the system will ask for less. Just tell them you are going to make this part of the system broken, then try to meet their needs. You will be able to meet most of them with a lot of pain and in the process as long as you avoid making row objects you will be able to do refactoring on an entirely different plane.

You see, dividing things up by column rather than row requires that the functions applied to each column will need to be excessively generic. This means lots of extra development and once this is refactored you will get generic classes, objects and functions that will be of use when looking at any XML or table, or at least building blocks. You're going to be suffering through all sorts of difficulties, using every trick in the book both old techniques and new to get it to work. People will see you struggling and will ask for fewer little things and you can tell them that you can do big things for them but they won't be perfect. Imperfection can be the beginning of information programming. To make this manageable just do it with part of your project small enough so that if you completely crash and burn you can rewrite that part from scratch, and large enough so that you have a bunch of fields with diversity to play with. You're going to be forced into writing small parsers, using the strategy pattern, lots of delegates, and even using endemes.

I've been thinking about the difficulties of strong typing combined with the necessity of strong typing. I have come to the conclusion that artificial creativity programming works best off of a base of information programming. That we have so few standard information data structures, is part of the problem that artificial creativity faces. If the base isn't there, you can't build on it. I continue my search to find information data structures.
Thanks.

[0] comments (97 views) |  | [0] Trackbacks

Posted by: Jon Grover | Jan 23, 09 | 12:57 pm

Tue Jan 13, 2009

Don't Knock Hackneyed Creativity

The benefits of hackneyed creativity.

When a human writes a book or creates something that is generally derivative and formulaic, we call that hackneyed. The person is called a hack. We think of it as uncreative. Don't knock it. There is no point in waiting for for computers to be able to do 'true' creativity before we declare success. If a computer does hackneyed creativity easily and consistently, that is a substantial advance in artificial creativity. It will also be helpful the the 'hacks' tell us their formulas so we can embed those formulas in software.

Once we have a solid foundation of hackneyed artificial creativity, then we can move on to the next step of more creative creativity, what people often call 'true' creativity. This of course presumes that the hackneyed creative system's software is available. We can only reach 'true' creativity with computers by building on hackneyed creativity if the hackneyed creativity software is accessible.
Thanks.

[0] comments (361 views) |  | [0] Trackbacks

Posted by: Jon Grover | Jan 13, 09 | 12:37 pm

Tue Dec 23, 2008

Breaking the Qual/Quant Barrier

Usually information is either qualitative like profession and gender or quantitative like age and weight. These characteristics can generally not be compared or have operations on them that combine them. In addition qualitative data can not have quantitative operations on it, and quantitative data is resistant to qualitative operations. I call this the qual/quant barrier.

The Endeme system begins to break down the qual/quant barrier.

Endemes:


An endeme is a list of characteristics which are placed in a particular order indicating the value of what is endemized. The list of characteristics is drawn from an 'endeme set' and all of the characteristics in the endeme set are used up in each endeme. So an endeme contains the complete list just in a different order for each one.

So, a person's height could be stored with other physical measurements like their girth, shoe size, hat size, etc with fudge factors for each measurement in the endeme set. Colors could be stored together as an endeme showing the strength for each. Weights of items in a pack could be stored as an endeme.

other examples:
- height,length,width
- weight,mass
- color,shade,saturation
- face dimensions
- distance
- ages,dates
- power,energy
- speed

Quantity is still fudged because the data are stored relative to each other so exact quantities will not be obtainable, just approximations. If you keep 'units' plus 'averages' of each characteristic in the endeme set this will allow these data to be mixed in the same endeme and reconstituted as needed from and endeme of that set. This means that the characteristics will have to be selected so that their values tend to be orthogonal and even opposing to each other. Percentages of a resource are perfect example.

An example that will not work would be physical characteristics related to size. It doesn't do to have all characteristics that have a positive relationship to a person's weight for example weight, height, girth, stride etc. because their values would all be high or low together and the endeme only shows relationships, not actual quantities directly, so some of the items if they are related would be actually forced to be too low in the list, if say they all need to be high in the list.

A good example that would work would be resource distributions in an army pack. There is a finite resource (weight), and the various kinds of items would each use up a certain percentage, so when one went higher another would tend to go lower. And the endeme would properly account for this. Colors work, but only if darkness and desaturation and things like that are included, so the colors have counterweights.

OK, now back to bringing quality and quantity together.


Endemes combine qualitative and quantitative data. So there might be a breakpoint in each endeme or endeme set that says where the qualitative (combination) stuff ends. We already should have one saying where zero is. The fields qual/quant ized have to be comparable to each other and you have to accept that their quantitative values will be heavily fudged.

On the qualitative side, simply set a breakpoint below which characteristics do not apply in a particular instance. For example if you say only the first six colors in the 22 characteristic list are 'used', then qualitatively the color indicated is a combination of those six.

On the quantitative side, store units for each characteristic and make sure the characteristics are not 'encumbered'. Weight and gender might not be combinable because gender has no measurable nature, but profession does Lets say profession was divided into rough measures of how technical, how people oriented, and how task oriented. These could be included along with weight and hat size in the same endeme (given units for a characteristics), and since they tend not to have relationships with each other, most (not all) endemes would contain fairly accurate quantitative data for each and could be matched with other endemes for other people using the same characteristics. True, someone who was high on most characteristics - worked intensely with people, technical things, tasks, and had a large weight and hat size would be stored inaccurately but this is the price you pay for the benefit of having quantitative and qualitative information mergeable, matchable, comparable, and relatable.

Number of Characteristics in an Endeme Set


This is why you have a large number of characteristics in one endeme. The more characteristics, the lower chance of inaccuracy. 22 (17-26) seems to be a good number for characteristics. Beyond that I suspect that characteristics are hard to chose that don't relate to each other excessively and below that I suspect that the inaccuracy effect I just mentioned will surface too often for the information to be usable. In other words the lower the number of characteristics the more likely it is that the data in particular instances is encumbered; the higher the number of characteristics the more likely it is that the characteristics themselves are encumbered. Because some characteristics will generally be encumbered (you are going to want to have height and weight in your list for other reasons), you generally need to get out near 22 characteristics to have enough to resolve this. The sweet spot seems to be 22 or 23 items.

[0] comments (558 views) |  | [0] Trackbacks

Posted by: Jon Grover | Dec 23, 08 | 6:13 pm

  NEXT page