Thanks a lot Ray,
That schema would work for us. We have (for instance)
<md:typeOfResource collection="yes" xml:lang="en">mixed material</md:typeOfResource>
A manuscript="no" is invalid, as it should be.
Just en passant: For some reason xml:lang isn't allowed on md:typeOfResource (and many other places). Working in a multilingual environment it is an inconvenience not to be allowed on *any* element. Another issue is the ID attribute vs. xml:id... ID should be deprecated and xml:id allowed.
Fra: Metadata Object Description Schema List [[log in to unmask]] på vegne af Denenberg, Ray [[log in to unmask]]
Sendt: 7. oktober 2015 20:31
Til: [log in to unmask]
Emne: Re: [MODS] Critical XML Schema bugs
I have applied changes that I think fix the problem. The fixed version is at
Please test it out. I'd like to leave it out for testing for a couple weeks, and if there are no problems, then make it official.
> -----Original Message-----
> From: Metadata Object Description Schema List
> [mailto:[log in to unmask]] On Behalf Of Sigfrid Lundberg
> Sent: Wednesday, October 07, 2015 3:01 AM
> To: [log in to unmask]
> Subject: [MODS] SV: [MODS] Critical XML Schema bugs
> I believe that there are other attributes that behave the same. Grep of my
> xsd gives these
> They are all fixed :( and misinterpretation of any of them could be serious.
> Fra: Metadata Object Description Schema List [[log in to unmask]]
> på vegne af Thomas Scheffler [[log in to unmask]]
> Sendt: 7. oktober 2015 08:44
> Til: [log in to unmask]
> Emne: Re: [MODS] Critical XML Schema bugs
> Hi Ray,
> thanks for that quick response. I wouldn't have written to the list, if this bug
> does not surface here. It would be great if you are going to fix it soon and
> not waiting for a new release (even fixing older releases). As this bug actually
> invalidates all kind of identifiers (see attachment), I see urgent need for such
> a fix.
> If you agree with me, that the XML Schema is not the standard defining
> document, there are really not many arguments left not to fix it anyway.
> Maybe you can also get rid of xlink:type="simple", which are annoying but at
> least do not alter the information :-)
> kind regards,
> Am 06.10.15 um 18:35 schrieb Denenberg, Ray:
> > Hi Thomas - The MODS EC is aware of this defect, which in fact was
> > introduced several versions ago. Before we published version 3.6 we
> > talked about fixing it, but we decided not to. The defect seemed
> > innocuous and had been around long enough that any problems it may
> > have caused would have already surfaced, and that fixing it now might
> > result in unintended consequences.
> > Is it causing problems? If it is, we'll look further into fixing it.
> > Ray Denenberg Library of Congress
> >> -----Original Message----- From: Metadata Object Description Schema
> >> List [mailto:[log in to unmask]] On Behalf Of Thomas Scheffler
> >> Sent: Tuesday, October 06, 2015 4:05 AM To: [log in to unmask]
> >> Subject: [MODS] Critical XML Schema bugs
> >> Dear MODS users and officials,
> >> recently we updated our code to support MODS 3.6 and the new
> >> nameIdentifier which uses the type definition
> >> "identifierDefinition":
> >> <xs:complexType name="identifierDefinition"> <xs:simpleContent>
> >> <xs:extension base="stringPlusLanguage"> <xs:attribute
> >> name="displayLabel" type="xs:string"/> <xs:attribute name="type"
> >> type="xs:string"/> <xs:attribute name="typeURI" type="xs:anyURI"/>
> >> <xs:attribute name="invalid" fixed="yes"/> <xs:attribute
> >> name="altRepGroup" type="xs:string"/> </xs:extension>
> >> </xs:simpleContent> </xs:complexType>
> >> Problematic is the fact that @invalid is set to fixed="yes". When you
> >> use a schema aware parser to parse a MODS document this will actually
> >> attach a @invalid="yes" on every element of "identifierDefinition"
> >> (like "identifier" or "nameIdentifier") that is missing that
> >> attribute.
> >> XML Schema part 1, section 3.2.1: "[..] fixed indicates that the
> >> attribute value if present must equal the supplied constraint value,
> >> and if absent receives the supplied value as for default."
> >> This bug is not introduced in 3.6 but is also in older versions of
> >> MODS.
> >> Currently all identifier|nameIdentifier get an 'invalid' label.
> >> Obviously this is not way it was intended.
> >> Can you please publish updated versions of affected MODS Schema
> >> documents?
> Thomas Scheffler
> Friedrich-Schiller-Universität Jena
> Thüringer Universitäts- und Landesbibliothek Bibliotheksplatz 2
> 07743 Jena
> Phone: ++49 3641 940027
> FAX: ++49 3641 940022