Re: [PROPOSAL] time zones and reproducible builds (was: How to make groff use local timezone?)

[Dave Kemper]

On 12/21/20, Deri <deri@chuzzlewit.myzen.co.uk> wrote:
> Consider the case where multiple -Z produced groff files,
> produced at different times, are combined into one gropdf run. The only
> sensible time to include
> in the meta-data is the time the pdf was created, why lose potentially
> useful information?

If you need to know when the file was created, why not use the file
system's metadata rather than duplicating that within the file?  Many
tools can look at this metadata without having to know the first thing
about parsing a PDF file.

Are there cases where the timestamp within the file would differ from
the timestamp *of* the file?  This typically happens when a file is
edited after it's created -- which could be common with HTML output,
but would seem to be rare with PDF output.  Any tools that "edit" PDF
files generally write out a new PDF file as a result, rather than
editing the old one in place, right?

(You'd lose the time metadata if, for example, you "cp" the file
rather than "cp -p" it, but that's true of any type of file, not just
PDFs; people who care about time of file creation need to take care to
keep that metadata with it, as a general rule.)