The differences between JOT and the data format used in the UNIPEN project (UNIPEN file format) are due to the very different goal of these standards.
The UNIPEN format is (mainly) for technical and scientific use. It is not a terse format but contains a lot of data (coordinates and annotations) to suit the needs of people testing handwriting recognition algorithms on huge amounts of data. Therefore it requires a lot of memory and a powerful processor. There are no provisions for ink width or ink color. Such provisions could be easily added, since the format is extensible, consisting of fields tagged with ASCII keywords. However, color and ink width are more relevant for an application-oriented standard, like JOT.
JOT grew out of the insight that ink, i.e., sequences of 2D vectors, is a data type not very well covered in existing data standards for applications. Here, the goal is to have a terse and sufficient standard to use in applications running on small platforms like PDAs and pen-based notebook computers.
Unlike the situation in speech and audio, where strong standards have emerged, the situation in pen computing and graphical tablets is very diverse. There are dozens of binary and ASCII file types used by manufacturers of digitizer tablets and by researchers. What is worse, even if you have an X and Y coordinate pair, it is essential to know how it was sampled: sampling rate, resolution, accuracy, time of sample point, type of sampling, origin (LL, UL) etc. Large international standardisation rounds - like for the Office Document Interchange Format (ODIF) - systematically forget about 'ink' as a very basic and usefile type of data. The ink data type combines the graphical aspects of line drawing functions with the temporal aspects of media like sound or video.
Back to the Pen Computing - Data Standards
since 15/Mar/1996