simplifying; getting rid of relation attribute

CSL Development
1

So I’m seriously thinking of getting rid of the relation attribute I
added after flattening the structure. It’s occurred to me that the
only structure where level really matters is titles, and so I can add
a few more title elements. Result would look something like the
below. Any comments/objections?

Bruce

      <template type="chapter">
         <author/>
         <year>
            <prefix> (</prefix>
            <suffix>) </suffix>
         </year>
         <title/>
         <editor/>
         <containerTitle font-style="italic">
            <prefix>, In </prefix>
            <suffix>.</suffix>
         </containerTitle>
         <collectionTitle>
            <suffix>.</suffix>
         </collectionTitle>
         <publisherPlace>
            <prefix>&#160;</prefix>
         </publisherPlace>
         <publisherName>
            <prefix>:</prefix>
         </publisherName>
         <url>
           <prefix>, &#160;</prefix>
         </url>
         <accesDate>
           <prefix> [accessed on </prefix>
           <suffix>]</suffix>
         </accessDate>
         <pages>
            <prefix>, </prefix>
         </pages>
      </template>
      <template type="article">
         <author alternate="containerTitle"/>
         <date>
            <prefix> (</prefix>
            <suffix>) </suffix>
         </date>
         <title/>
         <containerTitle font-style="italic">
            <prefix>&#160;</prefix>
            <suffix>, </suffix>
         </containerTitle>
         <seriesTitle>
            <suffix>, </suffix>
         </seriesTitle>
         <volume/>
         <issue>
            <prefix>(</prefix>
            <suffix>)</suffix>
         </issue>
         <url>
           <prefix>, &#160;</prefix>
         </url>
         <accessDate>
           <prefix> [accessed on </prefix>
           <suffix>]</suffix>
         </accessDate>
         <pages>
            <prefix>, </prefix>
         </pages>
      </template>
      <template type="document-archival">
         <author alternate="title"/>
         <date>
            <prefix> (</prefix>
            <suffix>) </suffix>
         </date>
         <title/>
         <collectionTitle font-style="italic">
            <prefix>&#160;</prefix>
            <suffix>, </suffix>
         </collectionTitle>
         <location/>
         <url>
           <prefix>, &#160;</prefix>
         </url>
         <accessDate>
           <prefix> [accessed on </prefix>
           <suffix>]</suffix>
         </accessDate>
         <pages>
            <prefix>, </prefix>
         </pages>
      </template>
2

Bruce,

It looks fine tome I can not see any problem in it.

David

3

Hi,

apologies for the slightly OT question, but I figured somebody here may
be able to give me some advice:

I seem to recall that there was some discussion somewhere about in-line
text markup in MODS XML, especially regarding markup of math strings.
While full math support would be nice, I’m mostly interested in
preserving italicized or super-/subscripted text parts when exporting
bibliographic records to MODS XML.

Does anybody know of an acceepted solution for this which is valid
within the context of MODS XML?

I appreciate any help or suggestions.

Thanks, Matthias–
refbase.net

4

Hi Matthias,

I seem to recall that there was some discussion somewhere about in-line
text markup in MODS XML, especially regarding markup of math strings.
While full math support would be nice, I’m mostly interested in
preserving italicized or super-/subscripted text parts when exporting
bibliographic records to MODS XML.

Does anybody know of an acceepted solution for this which is valid
within the context of MODS XML?

This was never resolved (nor have I dealt with it in citeproc or CSL
come to think of it!). If you put markup within a title element, it
will not be technically valid.

There are three possibilities:

  1. hit up the LoC again to change this (am cc-ing them)
  2. do it anyway, and don’t worry about if it’s valid
  3. use a custom element within the extension element to contain this (a
    duplicate title)

Each has its problems of course.

BTW, I’d urge, if possible, that you try to preserve semantic
information. E.g. this:

mods:titleSome title with a <xhtml:span
class=“title”>title</xhtml:span></mods:title>

… or even:

mods:titleSome title with a <xhtml:span class=“title”
style=“font-style: italic;”>title</xhtml:span></mods:title>

… would be better than:

mods:titleSome title with a xhtml:ititle</xhtml:i></mods:title>

Not always possible of course, but …

Bruce

5

Hi Bruce, thanks for your quick reply!

I seem to recall that there was some discussion somewhere about
in-line text markup in MODS XML, especially regarding markup of math
strings. While full math support would be nice, I’m mostly
interested in preserving italicized or super-/subscripted text parts
when exporting bibliographic records to MODS XML.

Does anybody know of an acceepted solution for this which is valid
within the context of MODS XML?

This was never resolved
There are three possibilities:

  1. hit up the LoC again to change this (am cc-ing them)

Of course, native support for some kind of in-line markup would be the
best solution. It would be nice if MODS XML could serve as a lossless
format for bibliographic records. In-line text markup would be a
requirement to achieve this goal.

Background: Recently, a rather large institution at my university did
consider to adopt refbase for their institutional workflow. The
capability of refbase to export records as MODS XML was a big plus in
their decision making. But when they realized that crucial record
information such as in-line text markup would not be preserved they went
again with an Endnote-based solution.

  1. do it anyway, and don’t worry about if it’s valid

Yes. But, if possible, I’d like to avoid a non-valid solution.

  1. use a custom element within the extension element to contain this
    (a duplicate title)

Hmm, yes, that sounds doable but I’d loose some interoperability. A
generic solution like 1) would be preferable, of course.

BTW, I’d urge, if possible, that you try to preserve semantic
information. E.g. this:

mods:titleSome title with a <xhtml:span class=“title”
style=“font-style: italic;”>title</xhtml:span></mods:title>

Ok, thanks for the advice, I’ll have to discuss this with my
co-developers.

Best regards, MatthiasOn 22-Jun-2006 at 9:56 -0400 Bruce D’Arcus wrote:

Matthias Steffens ---- @Matthias_Steffens

refbase — a web-based, platform-independent,
multi-user application for managing scientific
literature, references and citations

        http://www.refbase.net
6

I still have this on my list of something to discuss further in the next
major version of MODS. We need to assess the general need for it. I had
made some headway understanding how this might be done when we lost our
programmer. We have several major changes for a version 4.0 and need to
get some discussion going.

Rebecca

7

Hi Rebecca, appreciate the reply.

I still have this on my list of something to discuss further in the
next major version of MODS. We need to assess the general need for it.

I can only speak for the hard sciences (Bruce may contribute with
examples from other fields). But as it is now, the lack of an in-line
text markup solution means that nobody in the hard sciences will be able
to fully use MODS XML as a bibliographic exchange or storage format.

Some examples: In biology, biological species names frequently occur in
the title and abstract of a paper and they must be printed in italics. A
similar situation arises e.g. in earth sciences where isotopic elements
such as “13C” frequently occur in the title and abstract, where the
number “13” in “13C” must be printed as superscript text. Of course,
physics and math sciences have more advanced needs and would require a
full in-line markup language such as MathML.

Best, Matthias