JULY 2000

U.S. BUREAU OF THE CENSUS
TIGER/Line 1999

Census TIGER(R) and TIGER/Line(R) are registered trademarks of the
U.S. Bureau of the Census.  Windows is a trademark of the Microsoft
Corporation.  ADOBE and Acrobat Reader are trademarks of Adobe Systems
Inc.  ZIP Code and ZIP+4 are trademarks of the U.S. Postal Service.

WHAT ARE TIGER/LINE FILES

The TIGER/Line files were created from the Census Bureau's TIGER
(Topologically Integrated Geographic Encoding and Referencing) data base
of selected geographic and cartographic information. TIGER was developed
at the Census Bureau to support the mapping and related geographic
activities required by the decennial and economic censuses and sample
survey programs.  TIGER/Line files are made available to the public and
are typically used by people to provide the digital map base for their
Geographic Information System or mapping software.

This CD-ROM does not contain software to display or generate maps, nor is
the data on this CD-ROM in the form of a map image.  To create maps with
the data on this CD-ROM, one would typically use a Geographic Information
System package or other mapping software.  The Census Bureau cannot
provide help in how to use this data with specific software packages.
The TIGER/line files are provided in ASCII text format only.  Users are
responsible for converting or translating the files into a format used by
their specific software package.   

For information on how to use the TIGER/Line data with a specific
software package one should contact the company that produced the
software. For information on companies that sell products that use
TIGER/Line data or that provide various TIGER/Line-related services refer
to the "Vendor's List" on the TIGER Page at the Census Bureau's World Wide
Web site described below.

ACCESSING TIGER/LINE FILES

The TIGER/Line files are distributed on CD-ROM (and DVD in the future)
and on the Census Bureau's Internet site (www.census.gov).  This 
README.TXT file references both the disc and Internet distribution
methods.  For those downloading the TIGER/Line files from the Internet, 
unless otherwise noted, you can find the supplementary files mentioned below 
through links from the TIGER/Line download page.

WHAT'S NEW - KEY CHANGES

The TIGER/Line files in this version were originally produced for use by
the Census Bureau in cooperative programs with states and local governments 
in preparation for the 2000 Census.  The files were used by the states and local 
governments participating in the Census 2000 Participant Statistical Areas 
Program (PSAP) and the Traffic Analysis Zone (TAZ) program to verify the correct 
insertion of the statistical area boundaries into the Census TIGER data base that 
they submitted to the U. S. Census Bureau in 1999.   It is important to note 
that, even though each TIGER/Line file represents a whole county, all new 
features in existence may not have been added in this release.  The extent of 
updates in any given county will vary by program, number of participants, type 
of participant, time of participation, and a given participant's degree of 
involvement in a program. 

Some field names have been changed to make them consistent with the field
names that will be used in the Census 2000 statistical data products.  These changes
will be retained in the Census 2000 release of TIGER/Line.

We have removed instances of single addresses in an address range.  This was done
to comply with confidentiality requirements which restrict us from releasing 
individual addresses.  This will result, in some instances, in address ranges with
multiple ZIP+4 codes.

A new Census Feature Classification Code (CFCC) of "P" will appear for some 
street features that would normally have been classified as "A".  These were
features added to the data base as part of the Census 2000 preparation activities
and classified as "provisional."  Not all of the verification work was completed
before this release was created so those features carry the "P" code.  As they 
are verified in future field work the code will be changed to "A" or, if not 
verified, the feature will be removed in subsequent releases.

In previous release of the TIGER/Line files the year portion of the product 
name, as in TIGER/Line 1999, indicated the vintage of the "current" 
governmental unit boundaries.  Because of the delay in producing this release
the "current" boundaries information generally represents the boundaries as
of January 1, 2000.  

DIACRITICAL MARKS

The Census Bureau is no longer using codes to represent diacritical marks found
in some language names.  Beginning with this release of TIGER/Line the Census
Bureau will be using the ISO 8859-1 character set, commonly referred to as Latin-1,
to identify characters with diacritical marks.

TIGER/LINE AND ACCOMPANYING FILES

The 1999 TIGER/Line files contain digital cartographic data and are 
accompanied by documentation and related files. These files are 
distributed on Digital Versital Disc (DVD) and on the Census Bureau's
Internet site (www.census.gov) .  It may also be distributed as disc-on-demand
CD-ROM product.  On the Internet site (select the "TIGER"link from the home 
page) there are links to the directory/folder for each state where the individual 
files are stored.  Other links lead one to the accompanying files.

On the DVD (or CD-ROM) the TIGER/Line data is stored in the TIGER 
directory in subdirectories by state.  Each state subdirectory is named using the 
two-digit FIPS code and the Post Office abbreviation for the state (ex. 02_AK for 
Alaska).  Within the state subdirectory each county file has the name TGRssccc.ZIP,
where the "ss" is the state FIPS code and "ccc" is the county FIPS code. 

The data for each county (or equivalent) is stored in a single compressed
file that when decompressed results in up to 17 separate files
representing each of the TIGER/Line record types that exist for that
county.  Some counties did not require all of the 17 record types and
therefore will have less than 17 files. If the types of data contained in
record types 4, 6, 7, 8, 9 and Z are not appropriate for a given county then
the files for those record types will NOT be included.  The name of each of
these files uses a modification of the above convention (TGRssccc.RTx)
where "x" is the Record Type.  Appendix A of the TIGER/line documentation
provides a list of counties and their FIPS codes by state.  The data in
these files is in a  standard ASCII format.  The Census bureau does NOT
release the TIGER/Line data in vendor-specific GIS/mapping software 
formats.

Information on the content of each record type can be found in Chapter 6
of the TIGER/Line documentation.  Information on decompressing the data
files is found below under the heading "DATA COMPRESSION".  

The files use the TIGER/Line record formats as described in the
accompanying documentation.  When decompressed they are fixed length
ASCII with record separators (carriage return, line feed). 

ADDITIONAL FILES:

COUNTSss.TXT

Each DVD (or CD-ROM) contains COUNTSss.TXT files in the subdirectory
for each state under the TIGER directory of the disc.  If you are downloading 
from the Census Internet site they can be found in the appropriate state folder.  
The file name will have the appropriate state FIPS code in place of "ss" in the 
file name.  The COUNTSss.TXT files show the counts for the number of 
records for each record type by county for all states that are included on the 
compact disk.  Note that all counties are listed for a given state.  The layout of
COUNTSss.TXT is  as follows:

FIPS state code       16:17
FIPS county code    20:22
Creation date           25:30
Record type            33:33
Record Count         36:43

An important aspect of the COUNTSss.TXT file is the "Creation Date"
information for these files.  This date is the best indication of the vintage
of the TIGER/Line data for that county (or equivalent).  The process of
creating TIGER/Line files involves extracting the appropriate data from
various internal data sets to create what we term a TIGER/Line
benchmark.  Essentially this benchmark is a snapshot in time of our TIGER
data bases which are constantly being updated.  From this benchmark we
create the TIGER/Line files for use in our operations.  Depending on the
nature of the operation for which the files are being created, this process
can proceed over several months as the sets of files for individual states
are created.  The VERSION CODE field in TIGER/Line identifies the
benchmark used and one can determine the vintage of the file from the
information in this field (see page 2-1 of the TIGER/Line documentation). 
However, occasionally errors are found in the benchmark or in the
software that produces the TIGER/Line files.  Corrections are made to the
benchmark file and then a new TIGER/Line is created from the benchmark. 
In these situations the VERSION CODE for the original and corrected files
would be the same but the "Creation Date" information in the
COUNTSss.TXT file would reflect the different vintages.   

README.TXT

The document that you are presently reading.

DOCUMENTATION

The documentation for the TIGER/Line files is contained in the TECHDOC
subdirectory of the DVD (or CD-ROM).   The file name is TIGER99.PDF and is 
provided in the PDF format which requires the ADOBE(TM) Acrobat Reader 
(version 3.0 or 4.0).We are including the installation software for the Windows 95 
(ar40eng.exe)version (4.0) of the reader in the TECHDOC/ACROREAD DVD (or 
CD-ROM) directory. 

Included here are ADOBE's instructions for the installation of the Reader software:
"Copy this software to the hard drive of your computer and run it.  It will install the 
Windows 95 version of the ADOBE Reader software on your computer.  If there is 
a failure at any point during the installation of Acrobat Reader, the installer performs 
a complete uninstall. For this reason, it is  important not to close the installer application 
by clicking its close box in the upper right corner of the background window after 
clicking the "Thank You" dialog box that appears at the end of the installation. If you 
wait for a second or two, the installer will automatically close the background windows 
after the installation is complete."

When the installation is complete you can view the TIGER/Line documentation by 
running the ADOBE Reader and selecting the TIGER99.PDF file from the "Open" 
menu.  To get the ADOBE Reader for other computer platforms visit the ADOBE 
Web site for the proper software (http://www.adobe.com).

META DATA

FGDC compliant meta data is included on the DVD/ CD-ROM in the TIGER 
subdirectory.  It is also available from the Census Internet download site. 

DATA COMPRESSION

The TIGER/Line data on this disk has been compressed into an archive file to save 
space and reduce the number of disks required for the data and to facilitate 
downloading from the Internet.  If the files were not compressed it would require 
over 40 CD-ROM disks to hold the data needed to cover the entire United States.   
Use software compatible with PK-ZIP 1.93A or higher to decompress the data.

Info-ZIP's software (Zip, Unzip and related utilities) are examples of packages 
that can be used.  The Info-ZIP software is free and can be obtained as source 
code or executables from various bulletin boards and on line services software 
archive sites.  There are versions for many computer platforms.  One site from 
which this software can be downloaded is: 
ftp://ftp.freesoftware.com/pub/infozip/info-zip.html.

KNOWN PROBLEMS AND ANOMALIES WITH THIS TIGER/LINE DATA SET:

ADDRESS ANOMALIES IN PUERTO RICO

The TIGER/Line files contain some address range coverage for Puerto Rico. 
However, use of this information for geocoding purposes may be
problematic and the data user should proceed with caution.  These address
ranges are preliminary attempts at using Puerto Rico address ranges
in Census Bureau files.  Due to the lack of software or resources to handle
some of the more unique aspects of addressing in Puerto Rico, the address
ranges were entered without the standard edits and quality checks used in
other parts of the United States. Improvements in software and address
standardization for Puerto Rico are expected in the future.  At present, 
there are inconsistencies, overlaps, and duplication of address ranges. 
Address ranges may lack alpha character prefixes or have hyphenated
prefixes. The files also lack the community names used in a four-line
address that the U.S. Postal Service requires to avoid duplicate addresses. 
Errors in the reference files, and other factors may limit the usefulness
of this product for geocoding purposes. 

RECORD TYPE C

There have been a small number of county and county equivalent creations
and deletions made to the Census Bureau's TIGER data base since 1990.
As a result of these changes, 1990 geographic entity data at the national level
are missing in Record Type C.  This refers to geographic areas that are drawn at 
the national level and not within states, such as metropolitan areas.  For example, 
the 1990 Cantwell Alsaka Native village statistical area codes are missing in 
Denali Borough's Record Type C.  This problem, however, is only restricted to 
record type C, and it will not affect the ability of one to map the 1990 spatial 
components of these 1990 geographic entities from other TIGER/Line record 
types.  Specifically, this will affect only 1990 American Indian/ Alaska Native/
 areas, Hawaiian home lands, metropolitan areas, and urbanized areas.  This problem will be
fixed in time  
for the release of the Census 2000 TIGER/Line files. 

ROAD  FEATURE  ANOMALIES

The Census Bureau extracts TIGER/Line files from the TIGER data base
which is continually updated. During the clerical update process for some 
1999 TIGER/Line files errors caused anomalies to be introduced into some
chains represented in Record Types 1 and 2. For these cases road features 
may appear in the 1999 TIGER/Line files unconnected to other road 
features (so-called floating features) or severely skewed in relation to 
surrounding line features of any type.  We correct these errors as we find 
them, however, it is likely that some still exist. 

Another road feature anomaly is the sporadic occurrence of road segments
with a misclassified Census Feature Class Code (CFCC).  The result is that
complete chains for the affected road features will have segments with
different CFCC values assigned erroneously.  This problem could affect
applications that use the CFCC values for network analysis, routing, or for
assigning symbology to a feature when creating a map.  We believe that
these errors were introduced inadvertently during an updating operation to
edit address range information on new as well as existing records.  We are 
continuing to correct these as they are discovered.

QUESTIONS AND ADDITIONAL INFORMATION

Questions about the Census TIGER/Line files and future plans for this and
other TIGER-related products may be directed to the Products and
Services Staff of the Geography Division at (301) 457-1128 or to the 
e-mail address tiger@census.gov.

General information about TIGER-related products is available from the
TIGER page on the Census Bureau's World Wide Web site
(http://www.census.gov/geo/www/tiger/). This Web site also is a source
of information that is of specific interest to TIGER/Line users and of
general interest regarding the geographic programs at the Census Bureau. 
As appropriate, additional information about this product will be posted,
such as ERRATA or USER NOTES.