As we move closer to the paperless registry, the need for a workable disaster recovery plan becomes greater than ever. While we undergo this slow transition, anytime a computer problem does occur, it's important to use it as a learning experience to make our system better for the future. Last Friday, our Middlesex South Satellite Office shut down along with the entire Middlesex South computer system when the computer server in Cambridge - the key to all recordings and public searches - malfunctioned. It appears that the power supply inside the server overheated and was damaged. Once the power supply was replaced sometime on Saturday, things were restored. Power supplies do not overheat very often, but when they do, there are dire consequences, so we must be ready to react to such a situation. Theoretically, we had prepared for it: if anything happens to the computer server within a registry, the registry is supposed to be able to quickly switch from the damaged local server to the functioning Internet server and to continue recording documents on it. As is the case with most computer operations, when you go from talking about it to actually doing it, things become vastly more complicated. That's one reason that recording was not restored on Friday - the backup plan didn't work as intended. Now we have to find out why and correct it so that the next time a problem occurs, we'll be better prepared to deal with it.