Top Banner
Seeing Textual Signals Prof. Alvarado MDST 3703 10 September 2013
70

Mdst3703 2013-09-10-textual-signals

Apr 22, 2015

Download

Education

Rafael Alvarado

 
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Mdst3703 2013-09-10-textual-signals

Seeing Textual Signals

Prof. AlvaradoMDST 3703

10 September 2013

Page 2: Mdst3703 2013-09-10-textual-signals

Business

• About the inability to save files– Try disconnecting and reconnecting the drive– Use UVA’s VMWare View Client to access the HIVE

• See http://its.virginia.edu/hive/• Puts you on a Windows system• Can install JEdit there (in your own directory)

• Office hours– My Wednesday and Thursday afternoons are usually

open– Contact me

Page 3: Mdst3703 2013-09-10-textual-signals

Responses

• Studio is hard!• The readings put coding in a different

perspective

Page 4: Mdst3703 2013-09-10-textual-signals

Review

• Plato’s cave is an allegory of media• In digital media, interface and code are related

as shadow to puppet• In studio, we saw how source code could be

rendered differently– JEdit = puppet area (back of the cave)– Web Browser = shadow area (front of the cave)

• What kinds of things varied between the two?

Page 5: Mdst3703 2013-09-10-textual-signals

Review

• The opposition is reproduced as HTML vs CSS– Both are source code, but one defines structure

the other style• Documents exhibit a hierarchy:

1. Structure (elements and attributes)2. Content (“parsed character data”)3. Style (typography, layout)

• Which level would Aristotle consider most important?

Page 6: Mdst3703 2013-09-10-textual-signals

Comments

• All of the scholars we read for today regard text in “geometrical” terms– Although each takes a unique approach

• But whereas Aristotle links his geometry to the point of the play – the former explains the latter – the others don’t link the patterns they discover with a purpose or an effect– Is this true?

Page 7: Mdst3703 2013-09-10-textual-signals

Today, we look at the computer as an aid to reading and interpretation

The computer is the child of logic (codified by Aristotle)

What we find are variant forms of analysis that echo Aristotle

Page 8: Mdst3703 2013-09-10-textual-signals

Is a text a shadow or a puppet?

Page 9: Mdst3703 2013-09-10-textual-signals

Claude Lévi-Strauss

Page 10: Mdst3703 2013-09-10-textual-signals

Structuralism

• Lévi-Strauss was a French structuralist and anthropologist

• He believed that society and culture could be described in mathematical terms, i.e. rules that generate patterns

• Although he did not use computers, he was intrigued by them and modeling his thinking on how he imagined they worked

Page 11: Mdst3703 2013-09-10-textual-signals

Structuralism argues that the visible products of human culture – works of art, language, institutions, etc. – are the results of hidden structures that generate visible behaviors

The best example of this is language

Our speech – the observable part of language – is governed by grammar, or structure, a

hidden set of codes and rules that exist in the brain and shared by a community

Page 12: Mdst3703 2013-09-10-textual-signals

How does Levi-Strauss approach getting at the structure of a text?

Page 13: Mdst3703 2013-09-10-textual-signals

RELATIONS = SENTENCES = INDEX CARDS

BUNDLES OF RELATIONS = SEQUENCES OF CARDS = PASSAGES

Page 14: Mdst3703 2013-09-10-textual-signals

So we chop the text up into units and rearrange them

Page 15: Mdst3703 2013-09-10-textual-signals
Page 16: Mdst3703 2013-09-10-textual-signals
Page 17: Mdst3703 2013-09-10-textual-signals

OVER-RATEDBLOOD RELATIONS

UNDER-RATEDBLOOD RELATIONS

SLAYING OF MONSTERS

DIFFICULTIES WALKING

Page 18: Mdst3703 2013-09-10-textual-signals

To understand the myth, we need to understand the relations between the columns

This will give us the “paradigm” of the text

Page 19: Mdst3703 2013-09-10-textual-signals
Page 20: Mdst3703 2013-09-10-textual-signals

Is this how we think of reading a text?

If not, then what is Levi-Strauss talking about?

Page 21: Mdst3703 2013-09-10-textual-signals

Benjamin Colby

Page 22: Mdst3703 2013-09-10-textual-signals

Colby

• Colby is an American anthropologist• He was one of the first to use a real computer

to do something similar to Levi-Strauss• But his method is dictated by his tools– Words are associated into themes by a thesaurus

of themes– The words in texts are then parsed into this

thesaurus– A pattern of themes emerges

Page 23: Mdst3703 2013-09-10-textual-signals

The IBM 7090, announced in 1958, was the first commercial computer with transistor logic. It was intended mainly for scientific computing, but it was also suitable for business and administrative use.

Page 24: Mdst3703 2013-09-10-textual-signals

Sample thesaurus entries

If a text has a word on the right, then the category on the left is identified as being in the text at that point

Page 25: Mdst3703 2013-09-10-textual-signals
Page 26: Mdst3703 2013-09-10-textual-signals

Colby wants to make observations of this kind

Page 27: Mdst3703 2013-09-10-textual-signals
Page 28: Mdst3703 2013-09-10-textual-signals

Like Levi-Strauss, Colby wants to read these patterns as evidence for deeper structure – paradigms

Can Colby’s method help provide an Aristotelian description of these folktales?

Page 29: Mdst3703 2013-09-10-textual-signals

Colby's theory

TEXTS

CULTURE

Page 30: Mdst3703 2013-09-10-textual-signals

Stephen Ramsay

Page 31: Mdst3703 2013-09-10-textual-signals

“Algorithmic Criticism”

• Ramsay is a UVA graduate student• Teaches English at Nebraska• Developed a method to apply mathematical

graph theory to Shakespeare’s plays

Page 32: Mdst3703 2013-09-10-textual-signals

A

B

C

D

Graph Theory, developed by Euler, allows us to see that one would need to have an even number of bridges to get on and off a given land mass without going over a bridge twice.

Page 33: Mdst3703 2013-09-10-textual-signals

Graph Theory

• Regions and boundaries can be represented by “vertices” and “edges”– AKA nodes and links

• Links can be represented as having a direction or not– Directed vs Undireced

Page 34: Mdst3703 2013-09-10-textual-signals

Many things can be represented as graphs – networks of points and lines

that abstract the relationships between parts

By representing things as graphs, we can transform them in interesting

ways

Page 35: Mdst3703 2013-09-10-textual-signals
Page 36: Mdst3703 2013-09-10-textual-signals
Page 37: Mdst3703 2013-09-10-textual-signals
Page 38: Mdst3703 2013-09-10-textual-signals
Page 39: Mdst3703 2013-09-10-textual-signals

How many colors do you need to create a map in which no adjacent regions have the same color? Graph theory tells us the answer is 4

Page 40: Mdst3703 2013-09-10-textual-signals

What about texts?

Page 41: Mdst3703 2013-09-10-textual-signals

A Comedy of Errors, an early farce

Page 42: Mdst3703 2013-09-10-textual-signals

Richard II, a history

Page 43: Mdst3703 2013-09-10-textual-signals

Cymbeline, a late romance

Page 44: Mdst3703 2013-09-10-textual-signals

Coriolanus, a history, battles as limbs

Page 45: Mdst3703 2013-09-10-textual-signals

Antony and Cleopatra, a history, battles integrated

Page 46: Mdst3703 2013-09-10-textual-signals

Henry IV, Part 1, central place of the Garter Inn

Page 47: Mdst3703 2013-09-10-textual-signals

Henry IV, Part 1, Eastcheap Central

Page 48: Mdst3703 2013-09-10-textual-signals

Measure for Measure, a room in the prison central

Page 49: Mdst3703 2013-09-10-textual-signals

Julius Ceasar, extremely linear

Page 50: Mdst3703 2013-09-10-textual-signals

King Lear, linear then divided

Page 51: Mdst3703 2013-09-10-textual-signals

Henry VI, Part 1

Page 52: Mdst3703 2013-09-10-textual-signals

Henry VI, Part II

Page 53: Mdst3703 2013-09-10-textual-signals

Henry VI, Part III

Page 54: Mdst3703 2013-09-10-textual-signals

Anthony’s path through the play as a subgraph

Page 55: Mdst3703 2013-09-10-textual-signals

Cleopatra’s path

Page 56: Mdst3703 2013-09-10-textual-signals

Antony and Cleopatra

Page 57: Mdst3703 2013-09-10-textual-signals
Page 58: Mdst3703 2013-09-10-textual-signals

Clustering by number of single-incident scenes

Page 59: Mdst3703 2013-09-10-textual-signals

Alignments of tragedy and comedy

Page 60: Mdst3703 2013-09-10-textual-signals

Comedy and tragedy clusters

Page 61: Mdst3703 2013-09-10-textual-signals

Metrics

• the number of unique scene locations• the total number of scenes• the number of single-instance scenes• the number of loops (scene locations that

appear consecutively)• the number of switches (consecutive scene

locations with an intervening location).

Page 62: Mdst3703 2013-09-10-textual-signals

So, Ramsay begins by counting and linking scenes

Then he finds metrics for these graphs (e.g. number of scenes, etc.)

He ends by correlating these metrics to known genres (comedy, romance,

tragedy, history)

Page 63: Mdst3703 2013-09-10-textual-signals

Observations

Page 64: Mdst3703 2013-09-10-textual-signals

In each case, the “meaning” of the text is not what we think of as “meaning”

It is something unconscious

Page 65: Mdst3703 2013-09-10-textual-signals

On the other hand: What do you remember when you read a

book?

Page 66: Mdst3703 2013-09-10-textual-signals

We remember scenes, images, plot lines, values, etc.

We sometimes remember verbatim passages

We don’t normally remember the words

Page 67: Mdst3703 2013-09-10-textual-signals

We get much of our culture through books (and other "cultural models" in Colby's words)

Page 68: Mdst3703 2013-09-10-textual-signals

Like cigarettes, books are a “delivery mechanism”

(not of nicotine, but of culture)

Page 69: Mdst3703 2013-09-10-textual-signals

Text is like this

http://anthonyflo.tumblr.com/post/7590868323/photographer-and-self-described-geek-of-maps

Page 70: Mdst3703 2013-09-10-textual-signals

A text is a signal

Culture is a transmitter