[rfc-i] issue: canonical formats

Leonard Rosenthol lrosenth at adobe.com
Fri Jun 1 11:54:40 PDT 2012

On 6/1/12 10:39 AM, "John R Levine" <johnl at taugh.com> wrote:
>In the discussion so far, one of the few points on which we appear
>to have complete agreement is that the encoding for non-ASCII material is

For the canonical authoring format, I agree 100%.

However, for archival or other formats, that need not be the case (or at
least it hasn't really come up, AFAICT).

>>Additionally, it should be recognized that PDF supports a VERY RICH
>> object-level (as well as document level) metadata model.  One can
>> associate either simplistic "name/value pairs" or an entire chuck of XMP
>> (and XML/RDF-based standard (ISO 16684)) with any set of graphic
>So I hear, but the tool support is a bit thin compared to support for
>things like XML, HTML, and JSON.  Particularly the support in tools that
>don't require buying software from your employer.

Wasn't comparing it or implying it was better. I simply wanted to make
sure that the correct facts were on the table.

And I can think of at least a handful of open source PDF libraries that
support either reading, writing or both this materialŠNot to mention
various commercial solutions from other companies. So no purchases from
Adobe required - though always appreciated :).


More information about the rfc-interest mailing list