Gedcom Testbook Scorecard;

Description

The purpose of the scorecard is to present the results obtained in the Gedcom Testbook evaluation process in a numerical format allowing the user determine, to some degree, how well a Gedcom file from one program will import into a different program.

The scorecard is based on the Gedcom 5.5.0 grammar dated May 28th, 1997, and available on the ftp site, gedcom.org. The grammar displays the various Gedcom tags in the correct structural context. This grammar file was created by Jed Allen, formerly of the Office of the Gedcom Coordinator, and intended for use with his GEDCHK program.

The scorecard consists of two parts, import and export. The import section uses the Testbook control file as a standard. It does not use the complete grammar but does use most of the legitimate tags and many of the various structures. The total value obtained when it is scored against the full grammar provides a reference score against which a program's import capabilities may be evaluated. The second part of the scorecard evaluates a program's adherence to the Gedcom grammar by testing the Gedcom file created by the program. The information contained in the Testbook control file is entered into the program. Where and how it is recorded depends on the program's available fields. Some data, e.g. roles, may not entered if the fields are not available. The program's Gedcom file is then used as the standard and scored against the full Gedcom grammar.

Scorecard values are assigned based on relative importance. Three points represents adherence to the standard. Two points shows adherence but the data is imported in such a way as to require editing. One point is given for partial data importation. One or two points are deducted depending on the relative importance of the observed fault. Points are only assigned once per tag/ per structure regardless of how many times the tag is used or a problem appears.

The first column shows the GEDCOM structure, levels and tags. Tags using the same structures are grouped preceding the structure information.

Import:

Next three columns assign values for instances where the tag is not recognized.

The next series of columns deal with instances where the program recognizes the tag bur fails to import the data properly. Values are assigned based on the difficulty in recovering it.

Gedcom 5.5 allows for some grammar structures to be used multiple times for conflicting events, or multiple names. It specifies that the preferred structure appear last in the file. Previous Gedcom specifications required they be first but programs recognizing only a single name or event will process only the last instance. Current specifications indicate preferred events be listed first on page 20 of the lineage linked grammar and last on page 58. Confusion reigns. The report card simply indicates if duplicate tag data can be found in the exception list or the Gedcom file.

In some instances a program will record only part of the data in a structure either ignoring the remainder leaving it to be recovered from the gedcom file or an exception list. This will occur when the receiving field is too small and when the CONT/CONC tags are used with the PAGE tag when recording a source citation. The latter instance is another case of an inconsistency in the Gedcom specification.

The last two columns score those instances when only a reference number is transferred with the data remaining in the Gedcom file and in the case where source data is not transferred from the Gedcom file.

A Note column is provided to indicate problem specifics.

Export:

Developer assigned tags intended to export data between two identical programs are legal in 5.5 but not recommended. These tags are ignored and not scored.

Note column: Used to identify specific problems.