We want to configure our mirrored production/uat/dev ADF Instances to use the same git repo and collaboration branch with 3 separate publish branch's "adf_envhere_publish".

I got pretty far on this and looked like it was going to work beautifully but then ran into a problem with linked services and datasets. Currently, even though global parameters are saved with each factory (separate for prod/uat/dev) you cannot use them in linked services. It would appear you can only use global parameters at the pipeline level and this is a huge pain point that makes no sense. I see no fundamental reason why global parameters should not be useable in linked services, but they're not...

Linked services have the account names or urls hardcoded in them and they're different for each environment, so because we can't parameterize them we can't use one git collab branch for all 3 environments.

We want to work in the same git collab branch so that merges into it (main) can be published in any of the 3 environments and it will work.

And I am aware we can parameterize the linked services but that's not ideal because then we have to parameterize all the datasets etc and specify those parameters every time we use them and that's a royal pain too.