Re: Broken iWork 09 FACT

John W Kennedy stated in post 49ab6078$0$5915$607ed4bc@xxxxxx on 3/1/09 9:28
PM:

....
Something's wacky. A real ODT file is most definitely a ZIP file with
(except for some metadata) XML content. So was the SXW format used by
OpenOffice.org 1.0 (but not the SDW format used by pre-OOo versions of
StarOffice). As far as I know, OpenOffice.org invented the trick with
SDW. Then OASIS got together with OOo, and created ODT, based on SDW,
but less OOo-centric. OOo spread***, presentation, and other formats
have similar histories. When Microsoft saw what was happening, and that
ODT, etc., were becoming an ISO standard, they created DOCX to confuse
the issue, and, by stacked committees, voting irregularities, and plain
extortion, made it (or rather a future version that they haven't
actually implemented yet) an ISO standard, too. But they had
overreached, and, facing an international scandal, backed off.

I do not deny that the ODT file is such... but I just tested and changing
the extension to ZIP does not allow it to be unzipped, at least with the
standard tool on OS X. Maybe I will try another tool... there are
variations of ZIP and maybe it is one that Apple does not recognize?
Currently it is a bit of a mystery to me.

The command-line unzip command works just fine.

Just played... even the extension I have to QuickLook into ZIP files
works... Every other utility I tried worked with it. But not just the
"standard" double click. So how is the file different?

Apple quite reasonably chose the same basic idea of ZIPped XML for Pages
files, though they do not use the ODT format, or even import and export
it, which annoys the open-source community no end.

Any idea why Apple uses their own format? I do not know. They have been
pretty good in terms of working with the Open Source community, so I suspect
there is something ODT does not handle as well (for their needs).

I'm sure the .pages format maps better to the Pages code internals,
therefore making it easier for Pages to write and read the files than if
it had to deal with files that weren't specially designed for Pages.

Fair enough.

....
/Some/ XML namespace names do lead to documents, but they are usually
small text documents saying something like, "This document is a dummy
placeholder for XML namespace [so-and-so], which is described at [URL
name]." But that's just for human convenience, and entirely up to the
whim of the designer. The only point of an XML namespace name is to be a
name that is universally unique, and using a dummy URL was a quick and
dirty way to achieve that.

I thought it was supposed to point to a document that described the details
of the format... shows what I know! :)

No, because the reason for the namespace is to mix different XML
languages together in a single document (perhaps you have a single
document that includes typeset math with musical notion alongside your
text). In order to do that, the XML parser has to be able to say "This
bit of the document is MusicXML, this bit is TEI, etc.." The XML parser
does /not/ want to know all the intricate details of how to parse
MusicXML, TEI, and so forth; it just wants to know which subprocessor to
hand the baby to. The namespace name is used for that.

Ok, I knew it did that... but I thought it likely also was there for humans
to see the details of what the "flavor" is. Clearly I was wrong. Thanks
for the correction.


--
[INSERT .SIG HERE]


.