The proposed addendum contradicts itself with respect to the direction of elevation post offsets specified. Figures 3 and 4, as well as some of the text, show the elevation post moving to the south and east. However, the formulas and figure 5, show the elevation posts moving north and east. We assume north is correct, and the figures and text should be updated for consistency?
The exact method proposed for how to encode two 8 bit lat/lon offsets and an 8/16/32 bit elevation offset into the TIFF file format is unclear. Please add clarifying details about how to accomplish this while also considering the complexity necessary to read and write such non-typical TIFF formats (two 8 bit channels together with a possibly 16/32 bit channel) using existing standard TIFF manipulation tools and libraries.
Thank you Brian for finding these issues.
Here is a revised version we believe correct the error you mentioned.
Here is our first draft for the Non-uniform Sampled Terrain Elevation, please provide us with your feedback and comments.
We will update it accordingly.