For what is defined as "Structured Cabling" (permanent cabling) :
Instead of labeling the patch panel id, I find it best to just keep the patch panels in series, so the second patch panel starts with #49 (assuming 48 port panels). Then on your labeling scheme, you just label IDF # and then your port number. And if you only have 1 IDF, then you only need the port number.
You can get 960 ports into a single rack of patch panels.
Very simple. But you must label the patch panels because you can't rely on the pre-marked port numbers that will start over at 1-48 with each successive patch panel. Very important step!
Labeling the patch cord from the patch panel to the switch is what really should have the self-laminating labels with switch name, slot, port, jack id. Although in some scenarios, it's best to not label the patch cables at all. If it is the kind of environment where multiple people are coming in randomly and moving cables around and doing moves/adds/changes, usually a junior network engineer, you are better off not even labeling them because he'll move someone without changing the label. Then you've got incorrect labeling which is, in my opinion, worse than no labeling. In most idf environments, it's not hard to tug out an unlabeled patch cord to see where it goes in the room, usually the same rack or next rack. When you see a label is switch2 port 22, you are probably going to verify that anyway, right?
I only use patch cable labels in data center scenarios where accidentally unplugging the wrong thing could cause big time problems, and you have tens of thousands of cables running everywhere, and tugging or toning them out isn't practical. These environments are also usually controlled such that you don't have random people coming in and touching things they shouldn't be touching.