metadata.opf & AZW3

[ownedbycats]

While trying to investigate whether Amazon formats the DCMI same as epub, I opened up a AZW3 previously acquired from Kindle, imported into Calibre (de'drm'd):

  <metadata xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:calibre="http://calibre.kovidgoyal.net/2009/metadata" xmlns:opf="http://www.idpf.org/2007/opf">
    <dc:title>Third Contact (Envoys Book 1)</dc:title>
    <dc:creator opf:role="aut" opf:file-as="Aldin, Peter J">Peter J Aldin</dc:creator>
    <dc:contributor opf:role="bkp" opf:file-as="calibre">calibre (6.11.0) [https://calibre-ebook.com]</dc:contributor>
    <dc:identifier id="calibre_id" opf:scheme="calibre">50ef213d-0a34-449c-9f3b-02e48a8f203d</dc:identifier>
    <dc:date>2020-11-14T00:00:00-04:00</dc:date>
    <dc:language>eng</dc:language>
    <dc:identifier opf:scheme="MOBI-ASIN">B08KNYV3LX</dc:identifier>

Does this indicate that Calibre inserted them upon either import or opening an un-imported AZW3 in ebook-editor?

Also, what is http://calibre.kovidgoyal.net/2009/metadata? It redirected to the front page. If I change it to https and click past the invalid certificate error it leads to a 404 page.

[jhowell]

Calibre’s editor and the KindleUnpack plugin both attempt to produce an OPF that meets EPUB standards from Kindle formats that are actually structured much differently.

For what Amazon actually uses see https://wiki.mobileread.com/wiki/MOBI#EXTH_Header

[Doitsu]

Originally Posted by ownedbycats

Also, what is http://calibre.kovidgoyal.net/2009/metadata?

Even though it looks exactly like a URL it's actually a so-called XML namespace, which is required when adding custom Calibre metadata elements, e.g. Calibre book ratings, to .opf files.
Unfortunately, Calibre will also add this name space declaration when processing books that don't actually contain custom Calibre metadata entries.

[ownedbycats]

Amusingly, from the wikipedia page:

The Namespaces specification does not say what should happen if such a URL is dereferenced (that is, if software attempts to retrieve a document from this location). One convention adopted by some users is to place an RDDL document at the location.[5] In general, however, users should assume that the namespace URI is simply a name, not the address of a document on the Web.

Thanks