Computer program should re-assemble notes from the East German Stasi.  By Ned Stafford

A research team in Germany has developed a computer-software system to piece together some 45 million pages of secret police files ripped into 600 million pieces. The files were torn up nearly 18 years ago by panicking agents of communist East Germany's dreaded State Security Service (Stasi).

The pieces of torn documents are scanned on both sides, and the digital images are then analysed by a cluster of 16 computers for 25 features, including colour, shape, texture, handwriting and typeface, Nickolay says. Just like a person doing a jigsaw, the computer then groups the images into clusters with similar features, and finally fits pieces in each cluster together. The software should get better with time, Nickolay notes. "It learns as it processes."

The Fraunhofer Institute's IPK has received 6.3 million euro dollars (US$8.5 million) for a two-year pilot project to reconstruct just 2% of the documents. They opted for a pilot project to see how the IPK's system performs, says Günter Bormann, chief of staff of the agency BStU (Federal Commissioner for the Records of the State Security Service of the former German Democratic Republic), which was formed to analyse Stasi documents. "It is a new system. It is high-tech so there is a risk it will not work," Bormann says.

The government will then decide whether to finance reconstruction of remaining bags. The IPK says piecing together all the 600 million slips of paper by hand would take 30 people 600 to 800 years; their computer program should hopefully be capable of finishing the job in a little more than five years.

Quick rip

The torn documents date from the autumn of 1989, when the communist government of East Germany collapsed and jubilant West Germans and East Germans broke down the hated Berlin Wall. But not all East Germans were dancing in the streets. Stasi agents in ensuing weeks were holed up in offices around East Germany desperately trying to destroy evidence before West German authorities gained access to the files.
"It was a mountain of files," says Bormann. The Stasi lacked enough paper-shredding machines to do the job right, and began tearing documents by hand and stuffing them into bags.

The plan had been to transport bags bulging with documents by trucks to locations where they could be burned, but by January 1990 East German citizens had taken control of Stasi offices and the plan could not be carried out. West German authorities eventually seized still-intact Stasi documents and more than 16,000 bags of ripped documents.

The BStU initially tried to reconstruct torn documents by hand, completing about 350 bags before deciding that the process was too time-consuming.

Together again

Nickolay says the project will begin in June, with an initial set of 10 bags of torn documents to be scanned by IPK and the private firm arvato direct services GmbH, a division of German media giant Bertelsmann AG based in Gütersloh. The scanning will initially be done by hand by the 20 members of the team, but Nickolay says they are "looking for solutions to mechanize" this process. On average, it will take about one day to scan a full bag of torn documents, he says.

Stasi files contain not only information about East Germans, but also about foreigners and spying operations abroad. Under German law, anyone can request that the BStU check to see whether they are mentioned in Stasi documents and, if they are, can gain access to their personal files. Since 1992, 1.5 million people have done so.
Project leader Jan Schneider says the algorithms used for the software could also be used to reconstruct documents shredded into much more uniform pieces by machines. "It wouldn't be too complicated," he says.