June 6, 2017
When you are preparing to undergo an ECM migration, there is always a little piece of you just begging for a push button solution to do it for you. However, the reality is that most migrations need some sort of human interaction. But what if there was some sort of technology that could read the minds of everyone in the entire organization to figure out how to best migrate content over to a new system. Maybe some sort of artificial intelligence that could do the migration for you.
How migration is already intelligent
If you stop to think about it, the back-end processes of our Velocity Content Migration solution already works similarly to artificial intelligence. To begin, administrators let our migration solution know the source and target locations. From there, they teach it rules for migrating the content based on metadata values. Then they tell it when to run, how often to run, and when changes are made in the source repository to synchronize those changes to the target repository. Sounds intelligent to me.
However, to take it to the next level and have a migration that is totally automated would require more.
How migration can really be Artificial Intelligence
First of all, it would need to be intelligent enough to understand the organization of content. Perhaps it would be able to determine all content made by an organization through IP addresses or author. Then, it would be able to seek similarities within the metadata to correctly, and accurately move content from one system to another. The difference here is that it would sift through and map metadata on its own. An Artificial Intelligence (AI) migration would also be self-aware and could fix any issues it runs into along the way. It would be able to re-link documents that were linked elsewhere, and it would be able to tell when and if its performance has slowed, why it had done that, and how to fix it. Currently, advanced migration software such as SeeUnity’s Velocity Content Migration can address many of these issues through rules, like the re-linking of documents, however; a human must first know that this will be a problem. The biggest difference is that AI migration could identify the problem without being told by a human.
Is this the future of ECM migration?
We would like to think that AI is the future of everything. But the question here is whether it would be sustainable for all ECM migrations or not. This is where the uncertainty lies. Most ECM migrations are the result of user’s unhappiness with their current system. An AI migration would have to be capable of not just doing a strict copy and paste of your ECM system. It would need to know where you feel a discontent with your system, and find a way to migrate over to the new system to make a process or access to content better.
In the future, we can definitely see migration solutions become more intelligent, and more of the processes will be automated, but we cannot foresee migrations entirely without human interaction.
