Comment 16 for bug 1538011

That's a really good point about the end user impact of this change for user facing metadata. I'm inclined to agree with melwitt on this, for user facing data, if a user has specified a key to be a certain case, it should stay that case.

For aggregate metadata and this bug specifically, it makes sense to research why the unique constraint even exists at all on that key field. Once we know that, we can decide the best course to move forward.

In the meantime, I have a fix for the metadata key issue that illustrates the lowercase path, but I'm not comfortable with it until we've come to a decision about how we want to deal with the metadata case sensitivity issue for all of our metadata if anything for the sake of consistency.