I'm not being very good about notifying reviewers of the sorts of changes that are going on in the paper.

So... here is a page listing the significant changes information.

Minor changes (Typos, grammar, etc., are not listed.)

**20000408 (April 8th, 2000)**

Changed the discussion of adjustment by obeservations to mention orthogonalization methods as an alternative, and started a section on solving the problem by Orthogonalization Methods. Since that section is unlikely to be followable by anyone that isn't seriously into mathematics, I've left the discussion of solution by normal equations as the prefereed discussion.

**19991123 (November 23rd, 1999)**:

Split the least squares adjustment part into separate discussions of observations by conditions and of observations, to make expanding them out easier.

**19991023 (October 23rd, 1999)**:

Started some rewriting (in, for example, the Proportional Distributions chapter. This is to fix some terminology so that the network adjustment by conditions comes out cleaner.

**19990913 (September 13th, 1999)**:

Split matrix identity stuff off into it's own page.

Added Zhang's Matrix Theory to the bibliography.

**19990904 (September 4th, 1999)**:

Reworked some parts of examples. Made some minor changes to reflect terminology as used in the code library.

**19990826 (August 26th, 1999)**:

I ran several pages by a graduate student recommended by the math department (Parrish Brady <c-bypc@math.utah.edu>) who had lots of comments. Made changes to reflect problems he noticed. (Mostly to the weighted averages page.)

**19990821 (August 21st, 1999)**:

Finished up a first pass with a spelling checker. Automated running it against the directory. Added a number of identities to the matrix page.

**19990816 (August 16th, 1999)**:

Set up to do spelling checks. STARTED to do some corrections.

**19990807 (August 7th, 1999)**:

Changed the Least Squares example derivations to reflect the elements being full 3d weight matrices instead of just real numbers. (Removed a number of assumptions about multiplication commuting.) The derivations of four and five are still being worked on.

Added a section of identities to the matrix review.

Started rework of the first section of the Least Squares Introduction to reflect the changes in the derivations.

**19990810 (August 10th, 1999)**:

Found a non-strained two dimensional example for the weighted averages section. Started rewriting the example to match.

**19990803 (August 3rd, 1999)**:

Minor cleanups. A little work on table of contents stuff.

**19990731 (July 31st, 1999)**:

Reworked parts of the weighted averages section to make it consistent with later examples. Added a section table of contents to it.

**19990730 (July 30th, 1999)**:

Added "Computer Solution of Large Sparse Positive Definite Systems" by George and Liu [George and Liu]to the Bibliography. A very good book, with code.

**19990724 (July 24th, 1999)**:

Mostly just minor editing.

**19990723 (July 23rd, 1999)**:

Rearranged Matrix Review in response to complaints about forward references.

Added a non-trivial loop to the examples in the Least Squares introduction. We now have a non-degenerate loop as an example, and a demonstration that it is exactly the weighted average of the paths to each point.

Fleshed out some more of the text on that page as well.

Added the derivation for that example to the Derivations page. Did some more work on making the derivation of the last example meaningful.

Removed "range maggot" from the glossary in response to complaints.

**19990722 (July 22nd, 1999)**: Based on recommendations
from local reviewers, rearranged examples to make the structure of
the solution matrix more obvious. (More explicitly weighted averages.)

**19990721 (July 21st, 1999)**: split out
code changes from changes in the
paper.

**19990720 (July 20th, 1999)**:Added reportframe.{c,h}
to make test routines more consistent.

**19990717 (July 17th, 1999)**: rearranged code library
in preparation for a symmetric matrix package to match the general
matrix package. Split vector stuff out from matrix stuff a bit.
Corrected/expanded several test programs. Minor directory wide
HTML changes.

Moved the paper into its own subdirectory.

**19990713 (July 13th, 1999)**: Added the test data
appendix.

**19990712 (July 12th, 1999)**: Corrected the code
in daixyz.c that checked reasonableness of a covariance matrix.
(Thank you, Olly.)

**19990710 (July 10th, 1999)**: Made links between the
least squares page and the
Derivation of examples
page for those that want to wade through the actual ugly details.
Added more detail to the least squares page. Started the explanation
of collapsing a traverse to a singe equivalent shot, with the explanation
of the basic net example.

**19990708 (July 8th, 1999)**: Corrected definitions
in the Matrix Summary

**19990706 (July 6th, 1999)**: Lots of minor changes.
Changed the name of the geometry matrix in the
"Least Squares"
and "Using canned packages" sections to reflect
the common practice of using "A" for observation adjustment matrices
and "B" for condition adjustment matrices.

Changed the name of a value matrix to "D" (instead of "V") so it can't be confused with a covariance matrix. Changed data matrices V1, V2, V3, ... to M1, M2, M3 to reflect that they are measurements, and to avoid confusion with the covariance matrices.

Changed my default assumption in the documents to be that coordinates are column vectors instead of row vectors.. Also changed the assumption in the code.

Started a Glossary

Started a Matrix review section

Far from complete.

**Quick Statistics appendix**.

**Regularize terminology**

**Flesh out examples** in least squares discussion.

**Demo program** for adjustment by observations.

Derive data for surveys with mathematically exact answers. (So that programs can actually be compared/tested.)

This page is http://www.cc.utah.edu/~nahaj/cave/survey/intro/changes.html

© Copyright 2000 by John Halleck, All Rights Reserved.

This snapshot was last modified on August 23rd, 2000

And the underlying file was last modified on May 15th, 2000