Sure, you can search for alternative sources, hope that the PDF is not just image-scans but indexed and searchable text, and that it has a jump-able table of contents, and that it has not been altered and is up to date. Or you can go through public implementations and try to replicate, infer, and follow implementations or third-party descriptions of the standard. But all of that is error-prone and time and effort investment.
It shouldn’t be like that. A standard should be public so that anyone can implement it, from either side. So anyone can verify and compare against the reference, and call out implementation misalignments.
Sure, you can search for alternative sources, hope that the PDF is not just image-scans but indexed and searchable text, and that it has a jump-able table of contents, and that it has not been altered and is up to date. Or you can go through public implementations and try to replicate, infer, and follow implementations or third-party descriptions of the standard. But all of that is error-prone and time and effort investment.
It shouldn’t be like that. A standard should be public so that anyone can implement it, from either side. So anyone can verify and compare against the reference, and call out implementation misalignments.