Table 43 (Entries in a file specification dictionary): Observations and possible improvements

[stechio]

Examining the table on topic, I felt the following entries could be improved in readability:

FS entry

FS entry description contains two apparently contradictory statements:

If this entry is present, all other entries in the dictionary shall be interpreted by the designated file system.

and

This entry shall be independent of the F and UF entries.

What does the latter statement mean, considering that the former implies a dependence between FS and the other entries?

Moreover, IMO the FS entry description is a bit redundant and slightly obscure: it ties the interpretation of all other entries in the dictionary to its presence, yet doesn't mention the meaning of its absence. It would be more plain and simple if all the possible states of this entry were unified under the same definition.

PROPOSAL

My proposal is to replace the FS entry description with the following: The name of the file system that shall be used to interpret this file specification. It may be any of the following values: (null) Platform-independent file system (see 7.11.2, "File specification strings") URL URL-based file system (see 7.11.5, "URL specifications") (second-class name) Application-specific file system (see Annex E, "Extending PDF")

F entry

F entry description should be reformulated too, because of its redundancy.

PROPOSAL

My proposal is to replace the F entry description with this terse statement: A backwards-compatible file specification expressed according to the file system specified by FS entry. For cross-platform and cross-language compatibility, the UF entry should be included along with this entry. A PDF reader shall use the UF entry, when present, instead of this one. (PDF 2.0) For unencrypted wrapper [...]

UF entry

UF entry description is apparently in contradiction with F entry: the latter states that any F entry value (file specification string or URL (or application-specific file)) should be accompanied by a corresponding UF entry, whilst the former states that UF entry applies to file specification string only.

PROPOSAL

To my understanding, UF entry should apply to any kind of file specification (ie, file specification string or URL (or application-specific file)); in that case, my proposal would be to replace the UF entry description with this terse statement: A cross-platform and cross-language file specification expressed as a Unicode text string according to the file system specified by FS entry. For backwards compatibility, the F entry should be included along with this entry. A PDF reader shall use this entry, when present, instead of the other one. (PDF 2.0) For unencrypted wrapper [...]