Serge Romanov is unimpressed, Monday, December 13, 2004, by the much vaunted Yad Vashem website database on Holocaust survivor statistics
Who's Who at Yad Vashem and Auschwitz
I read the article on the Holocaust Victims on-line database project by Yad Vashem and your comments on it. I also looked quickly at the site itself, www.yadvashem.org As a computer professional, I would like to mention some problems with the project.
Someone may have paid $22 million for the project; however the technology required is widely available for free. A moderately proficient specialist can create a similarly unimpressive web site with, say, free Apache web server and free MySQL database backend at a vastly smaller fee. Several millions of records are NOT a problem as far as computing power or database technology and/or web serving at (I surmise) a very moderate level is concerned. 2-3 PC computers will make such a project run all right. Approximate and soundex searches are part of most database engines. I may miss the complexities of postprocessing and crosschecking, but equally claims of high technology may be put forward disingenuously.
There seem to be logical problems when such approach to collecting information is used. Submitters are basically taken at their word. Nothing would prevent creation of false entries. How readily it has been done on the Internet I could illustrate with one example. A couple of months ago some Jewish person or organization noticed that a Google (www.google.com) search for "Jew" brings up "www.jewwatch.com" as the first, highest-ranked entry. Immediately a Jewish activist from New York created a web site to coordinate a campaign for overturning the (objective and based on automated algorythms) Google ranking and pushing forward a sympathetic entry from www.wikipedia.org to the first place. Tens of thousands of Jews from all over the world, previously unknown to each other, created links and references to the wikipedia article, keywords were inserted in personal webpages etc. (activists used an old article on how google achieves its ranking to devise the techniques that would raise the wikipedia entry rating) - and about a week later the hated "jewwatch" site dropped to place number two. In the process, some cracking was done, so temporarily the jewwatch site became unavailable, and (as far as I remember) complaints to their ISP were sent, so the jewwatch maintainers then themselves stopped serving their pages for a while.
Now, without discussing merits or otherwise of the jewwatch, and however laughable the aim of the campaign might look, it demonstrates with utmost clarity how an Internet campaign can unite unknown to each other Jews from all over the world around what they would consider a "good cause", even if it seems trivial.
So one really big problem is verification. New entries must be kept separate and somehow cross-checked before merged to with the main database and made available for general access. Automating the cross-checks is not completely possible, i.e. one will have to keep a pool of human workers, trained in historical research and knowing a number of European languages.
In your notes you speculate on whether the Six Million Magic Number problem could arise as a result of this project. It seems, Yad Vashem forestalled possible objections already: in their Frequently Asked Questions text on that site, they plainly state that the number will never reach 6 million:
"When will there be 6,000,000 names in the database? Never. Some Jews left no trace. They were murdered with their entire families, so there was no one left to submit Pages of Testimony for them; or they left no documentary traces; or the traces they left were destroyed. . ."
Then there are problems on the definition of "survivor", which are again unsolvable without human attention to each and every record.
To sum up, a simple web front-end with database search and database update capabilities seems a cheap and nice example of how Internet could be used as a tool in historical research to poll previously unreachable for historians resources and persons.
However without serious and well-trained labour-intensive post-processing the project stands wide open to both errors and intentional abuse (as the Internet-literate generations will immediately spot) and cannot, simply cannot claim to become a Central Database with sufficient degree of veracity.
The above, of course, remains only my personal opinion.
© Focal Point 2004 David Irving