Re: [xml-dev] NY Times reference to 'secret coding'

[noah_mendelsohn@us.ibm.com]

Ken North writes:

> which parts of the OOXML spec rely on 'secret coding' ?

I have not read the OOXML specification, and so can't give a straight or 
well informed answer.  That said, I can guess where some of the issues 
might be, and I'd be grateful if someone better informed than I can 
confirm or deny my hunch. 

There are often at least two levels of concern when considering 
compatibility of an office-style file format:  1) given the published 
specifications and an arbitray document instnace, can you extract the 
general semantics -- for example, can you tell that a given page is in two 
column format with a footnote at the bottom vs. 2) can you tell to the 
exact pixel how that document is to be printed on some particular printer, 
and can you predict for that printer exactly how long the footnote can be 
before it wraps to a second page?

While I can't speak authoritatively regarding Microsoft Office or OOXML in 
particular, it wouldn't be surprising to find that a fair amount of the 
information needed for (1), quite possibly all of it, is in the OOXML 
specification.  It is also quite possible that the information needed for 
compatibility at the level of (2) is not provided in the specification. If 
that's true, then to get that level of compatibility or "openness" you 
would need additional information, and presumably the "secret code" for 
the Microsoft Office products embodies at least one set of answers to the 
questions asked in (2).

Note that, as a practical matter, (1) is sometimes sufficient for 
particular purposes but (2) is very often what you need.  Certainly, over 
the years, this is where a lot of the concern has been with "lockin" to 
the widely deployed office applications. If I send you a file for a 
newsletter, for example, and your software correctly determines the number 
of paragraphs, headers, etc., but rearranges them on the page, then quite 
likely you won't consider the file format to be sufficiently interoperable 
for your purposes.

So, seeing that the OOXML specification is asserted to "rely on 'secret 
coding'", I'm curious as to whether I am correctly guessing what the real 
issue is?  Does the OOXML specifiction provide for compatible rendering in 
the sense of (2)?  While I think that's what users typically need for 
interoperability, I'd be somewhat surprised if the OOXML specification 
provided it.

BTW: I assume that because OpenOffice is Open Source, the answers to the 
equivalent questions for ODF can be determined from that source code.

Noah

--------------------------------------
Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142
1-617-693-4036
--------------------------------------