Sorry about the long silence everyone, I've been traveling for the past several weeks and simply hadn't seen the notifications.
As Charlie noted, he's made several changes to address the changes we've been discussing and I've incorporated them into the Pull Request. You can see the full set of changes [here](https://github.com/cf-convention/cf-conventions/pull/145/files) and the changes relative to the original PR [here](https://github.com/erget/cf-conventions/pull/2/files). As @dblodgett-usgs notes the rich diff function is really helpful for getting an overview of the changes. I have the impression that we're nearing conversion, is this shared by the rest of the group? I'd like to add my thoughts to what @czender has written concerning the lateral search: Although I do think that it is a bit confusing, especially for people who are used to object-oriented paradigms using inheritance and classes, there are a plethora of data in the wild which implement this and I think that it's good to describe how things should be found. I also agree that there are many data out there which reuse variable names when referencing coordinates, and the amount of such data is set to grow. It is likely that EUMETSAT is producing such data, as we will be disseminating products which contain observations from multiple instruments, all of which use different viewing geometries. Therefore they would have different `lat` and `lon` in the same file. Lateral search won't be needed in order to find the coordinates, but it's nonetheless a good example of multiple uses of the same names. So in my view the current PR, with the new edits from @czender , does a good job of finding the middle ground and ensuring readability across data from the past, present and the future. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/cf-convention/cf-conventions/issues/144#issuecomment-427816650
