Sallyann Amdur Sack
(January 2009)

During 2007/08, the Destroyed Communities project worked on two tracks.

Track one was directed to identifying a small number of communities which could serve as pilot studies for the project as a whole. For that purpose, the active participation of nonprofessional family historians was enlisted and four communities in Lithuania and Poland were selected (Pushalotas, Lithuania; and Ostrow Mazewiecka, Staszow, Zdunska Wola, Poland). These communities all have active genealogical research groups who are working with the project director in order to implement their part in the project.

In the case of Ostrow Mazewiecka (OM), a formal agreement was signed with representatives of the OM Research Family and with JRI-Poland on the one hand, and the IIJG on the other hand, detailing the terms of the joint effort. In each case extensive use is being made of genealogy trees and other available 20th century data, as well as the considerable experience and knowledge of the individual genealogists researching the Jewish communities of the towns selected. At present, attentions are being directed at the identification and solution of problems in readying the pilot projects for the overall effort.

Pilot studies of additional communities may be added as work proceeds. In all cases, a two-fold end-result is aimed at: development of family trees for individual Shoah victims and creation of merged family trees representing the web of kinship that existed among the members of the individual communities on the eve of the Shoah.

Track two, involving the major research effort thus far, has focused on developing the advanced computer programmes required to integrate and analyze diverse datasets of genealogical information – that is, developing the algorithms and software needed to generate family trees, primarily by merging data in large, existing databases.

To that end, a team of experts was assembled, consisting of Dr. Alexander Beider (Paris), a mathematician and the world expert on Jewish names, Logan Kleinwaks, a mathematician and computer expert, Gary Mokotoff, presently the publisher of AVOTAYNU who has a strong computer background, Dr. Stephen Morse (San Francisco) who developed of the first Pentium chip, and Jean-Pierre Stroweis, a software engineer and former president of the Israel Genealogical Society.

The team has met twice in intensive, two-day meetings in Newark, New Jersey (July 2007) and Bethesda, MD (April 2008). It concluded that development of the necessary software is technically feasible and that initial progress could best be made by forming two subgroups of two members each–Alexander Beider and Steven Morse to develop the algorithms needed for phonetic matching of names, Jean-Pierre Stroweis and Logan Kleinwaks to work on algorithms to enable the identification of relationships between different individuals recorded in separate databases (or indeed within the same one).

Beider and Morse successfully concluded their work with the development of the Beider-Morse Phonetic Matching (BMPM) algorithm, a tool that searches name lists for names that are phonetically equivalent to a desired name. It is the first significant improvement in indexing genealogical databases since the Daitch-Mokotoff (D-M) Soundex was created 23 years ago, and represents a major break-through, both for this project and for name identification in general.

In parallel, a conceptual approach for developing algorithms needed to enable identification of relationships between individuals has been devised and was demonstrated by Kleinwaks and Stroweis at the April 2008 meeting.

Click here for a fuller description of BMPM.