Sunday, April 30, 2017

White Throated Sparrow

The high whistle of a white throated sparrow takes me all the way back.

Tuesday, April 25, 2017

The Rock Piles blog must be influential

I was just backtracking visitor logs on the Rock Piles site and people are visiting for all kinds of different articles - no doubt the result of searches they are doing. The articles they find are from all different periods of the blog's history. It is an archive not an active news site. As I look at it, the posts cover a reasonably wide variety of subjects centered around ceremonial stonework, but also with arrowheads and other more general archeology or Native American cultural topics - as posted by my guest authors. There are thousands of articles and thousands of readers every month. Right now it stands at around 4.8 thousand readers this month.
I don't know how many of those are repeat visitors but most of them are not. That means I am reaching a large number of people interested enough in such archeological topics to do online searches. I find it sort of weird to put some minor random thought into a post and have someone in Oregon read it the next day. I also find it weird to hear ideas that originated with this blog coming back at me from some other direction - like an echo. For example, I have been hammering away at certain concepts - like rectangular mounds with hollows being burials and *blink* it seems to now be an established fact.

N-segmented text

Continuing the idea from here: if we pass several first order NARs across the segmented text, they may individually or together get a result at any "control" in the segment. The hard part would be higher order NARs getting passed across such a combination of lower order NARs and un consumed VARs in the original segment. I need to develop an infrastructure for managing such overlays of higher and higher order NARs "above" the original segmented text.

The rolling up of the text into the segment (sequence of VARs) and the rolling up of the segment into higher and higher order NARs - leads to the idea that once the text is prepared, all the meaning has already been established. In other words, the rolled up input is the output.

I am grasping but do believe there is a "beautiful theorem" in there somewhere.

Monday, April 24, 2017

Verbs mixed with adjective and the power of the '[ ]' notation for implicitness

I was making up an example of a statement that combined verbs and adjectives and the following, not entirely natural, example came up:

Jon shot a goose that cooked up pretty good

A couple of sort-of interesting things come up as I try to "diagram" it using proto semantics. 

Jon-shot->Goose, [We]-cooked->Goose, [?]_/good

I am using '?' to indicate the ambiguity of whether the cooking or its result was good. Since it is implicit, and since there is a truism that makes them somewhat equivalent, you can see why it is easier to just leave it implicit.

Something like an inserted "We" is needed. Which suggest the general rule of narrative continuity allowing  arbitrary insertion of "I" or "We". This is allowed because they are always in context just as the subject of a story is always in context. [Added: in other words they are global variables]

I think it is good that the proto semantic notation stumbles on exactly the ambiguities that are present in the sentence. The word "that", leaves us uncertain if the entire situation is being described as 'good' or a sub part of it. One does, in fact, sense that ambiguity but also that the ambiguity is not important; which is because of Truism 8 "If an action is described as a success, its outcome is assumed to be good". 

Sunday, April 16, 2017

"Sounds Big: The Effects of Acoustic Pitch on Product Perceptions,"

The article about the article did not mention a discussion of why low pitched sounds are associated with thinking things look larger.

It sounds like the same as the moon on the horizon looking bigger. The connection would be that low pitched sounds carry further so their sources are judged to be further away - hence appearing larger. Same as light being fainter on the horizon. Thanks G.Berkeley.

Tuesday, April 11, 2017

Dear Reader from Italy

Regular readers are so rare, I would be grateful if you would leave a comment. 

Saturday, April 8, 2017

Nesting of narratives and text processsing

The puzzle is how to fit a complex narrative to a text, where the complex narrative nests sub narratives. I had the following thought while riding the Red Line somewhere around Central Square: 

 - First text is processed by being broken up into tokens. The result is called "tokenized text."
 - Next tokens are processed by conversion into a sequence of VARs, including NULL_VAR for tokens that are not recognized. The result is called "segmented text".
 - Now the smallest pieces of sub narrative can be "rolled" over the segmented text to produce a processing of the text into higher and higher order results - you could call them "n-segmented text"....and so on with n increasing with the order of the narrative.

Tuesday, April 4, 2017

More semantic personality profiling - how people summarize and how narratives nest inside narratives

The previous post considered asking different people who had experienced the same thing (a subway ride, buying a dress, etc.) to give a short description. This was to be followed by quantitative study of the notational description of the text - via sentence structure (SnG) and/or via proto semantics. Here is a different question, that can be part of the study:

Ask a participant in the study to give a very brief summary. Then ask for a longer one. Then ask for more details. What forms are present in reversing the summary? You could also ask participants to read a short paragraph and provide a summary, or read a sentence and provide an imaginary flushing out of the story's details.

The purpose of this study is to gather information about how narratives are nested within narratives - and how that varies from person to person.

Saturday, April 1, 2017

How much semantic variation is there person-to-person?

I am railing against the paucity of vocabulary for describing mental states and how it has left us, we who wish to analyze language, with a complete lack of information about how diverse the use is of -say- native English. The problem is that most folks don't know proto semantic notation so they have little hope of capturing specific meanings (absent their representation by words and phrases, i.e. syntax-N-grammar) let alone person-to-person variation. So here is the experiment I would want to do. There is a PhD thesis here:
As much as possible ask 10 or so people for a short description of the same thing that they experienced (a ride on the subway, or a trip to buy groceries, or etc.) The questions to answer are:
  • Do different people use different concepts to describe the same things? How much variety is there?
  • Do different people use the different phrases to describe the same concepts?
  • Do different people use the same phrases to describe different concepts?
The approach is to take a short paragraph from each person. Transcribe the paragraph into proto semantic notation and transcribe it into traditional S-n-G notation. The answer to the questions come from quantifying the comparison of these pieces of notation, person-to-person.

We could call this a profiling of language personality. It is an intrinsically interesting topic - the empirical classification of people by their language use parameters. You could have someone with broad vocabulary saying nothing but a few routine things and someone with a limited vocabulary using a broad variety of concepts and saying a good deal. You could have bell shaped curves. The point is that these interesting things cannot be studied with S-n-G notation alone but they can be studied with that plus proto semantic notation.

Like any other form of profiling, doing it for language personality would be a basis for discrimination. Is it OK to discriminate based on language personality? Clearly not, because your references are always based on some other discrimination. But analyzing differences is not the same as basing discrimination on them.