The article talks about the process of scrambling data and how to keep private data from getting out.
Data-intensive organisations often have to build, fix, and test database applications. Most of the people who hire database developers are private companies, medical and financial companies, and even government agencies. From a business security point of view, commercial and trade secrets must be kept safe. Privacy policies and government laws make it illegal for organisations to give real data to database developers. This is done to protect the people whose information is in a database. If you break these rules and laws, it can be bad for your reputation and you could get in trouble with the law.
Why should you give the developer real data at all? The answer is easy: developers need real-world sample data to build the database, improve its performance, and find and fix any bugs. When developers are given a sample that is too small or too far from the actual data to be used, they make bad decisions. This leads to a database that doesn't work as well as it could or that isn't reliable or easy to use.
The paradox is clear: you can't give the developers real data without getting bad press or getting in trouble with the law, but a database can't really be built without the data. The answer to this problem is to scramble the data.
When data is scrambled, fake records that look like the real ones are put in their place. If a real financial database record says "John Doe, balance $10,000, account #000," a data scrambler will replace it with something random but still plausible, like "Mae Smith, balance $2,345, account #123." This protects the customer's identity by changing the name, and it protects the financial institution's security by changing the balances of the customers' accounts randomly.
Scrambling is the process of moving data from a production database to a test mode database. When used correctly, data scrambling removes the sensitivity from sensitive information. This makes data records that look real and keep the original keys and relationships of the real database. Using scrambled data makes it possible to give developers a fake database that works perfectly well. This lets the developers fully optimise and test the application without putting the system database at risk.
The right way to scramble data is used by DTM Data Generator (www.sqledit.com/dg). It has a "scramble" mode that lets you make a new scrambled table in an existing database or a new one. The information in the scrambled table has been changed, such as names, credit card numbers, medical records, and so on. The replacement records don't look like a random group of characters. Instead, the names are changed to other names, and the credit card numbers are changed to numbers that are the same length and structure.
