I’ve been talking a lot about shims lately here, but it’s been rather lopsidedly technical.
Using shims can be extremely helpful in mitigating application compatibility issues and unblocking deployments of Windows Vista. However, there are a few obstacles to navigate before you can really begin using them.
First, you would like to really understand them so you can feel confident regarding the side-effects – explaining the relevant consequences to business and technical decision makers. Hopefully this was a good start, and I typically include this discussion when I talk about shims at conferences as well.
Once you feel comfortable using them, then the next step is learning how to use them. I’ve been blogging about shims, worked with the ACT team to update the documentation in the help file for ACT 5.0.2, and speaking about it publicly at various events (most recently TechEd IT Forum in Barcelona).
But, assuming that you convince people that it’s a good idea to use shims, and you actually know how to use them, you probably want to make sure that you don’t end up in a position where they become something that initially isn’t well managed, which could lead to trouble later. So, I put together a whitepaper discussing best practices from the customers I have worked with who have leveraged this technology, and how they have built policy around when to shim and how to manage custom shim databases. If you have any feedback, please comment!