The code is indeed very useful to debug issues. When asked for it by YouTube, please provide the actual text rather than a screenshot...I get sent the screenshot a lot and then I have to use OCR software in order to decode it.
Why do you assume it is not base64 encoded? I have always come to believe it is some encrypted data. If I was Google I would probably give every server a UUID and a symmetric key. If an error occurs encrypt (and authenticate) the stacktrace and other debug information with that symmetric key and prepend the UUID. As a developer you can then find out the server that produced the error message, log in via SSH and decrypt the debug information.
But it would be interesting to collect a lot of this error messages to check if they appear to be completely random.
They change completely on every load. Encoding would likely have similar sections between page refreshes, as it doesn't we can assume your theory of encrypted data is correct. It makes sense anyway, it means they can have a user supply a stack trace or similar without exposing any information to them.
More likely a header. It's 3081 byte blob after base64 decoding. If we consider it's encrypted with some block cipher with 128 or 256-bit blocks (IV and/or MAC would be likely to be 128-bit, too), there are 9 bytes for some header and/or padding.
It is surely an encrypted stack trace, encoded in base64, which is probably being spit out by the load balancer proxy. I would not be surprised if Youtube developers used their own Chrome extension for making that binary meat readable.
I have once set out to write a module for Nodejs for the same purpose, but never finished it. Can't really see a downside in this way of reporting errors.
1. Non-standard base64 and
2. Definitely Compressed ... by guess would be snappy compression
3. Possibility of serialization using protocol buffer
4. not sure if such information would be encrypted after such a server failure