All
ISBD Consolidated edition may shed some light.
1.3 Other title information
Other title information consists of a word or phrase, or a group of characters,
appearing in conjunction with and subordinate to the title proper, a parallel title or titles of individual works contained in the resource. Other title information can include variant titles appearing on the same source as the title proper.
1.3.2
A statement of other title information can include a statement of responsibility, the name of a publisher, or details relating to other descriptive elements (e.g. an edition statement) when such a statement is linguistically an integral part of the other title information.
Any information appearing as other title information that includes one of the mandatory elements (e.g. a statement of responsibility) is included
either as other title information or elsewhere in the description. Additional other title information is included if it is necessary for identification or otherwise considered important to users of the catalogue.
ISBD examples include:
"situation en 1972"
"pastels, lavis, gouaches, esquisses"
"comédie"
'a novel based on the television series "The Avengers"'
This is reflected in
MARC 21 Bibliographic245 $b Remainder of title ... includes parallel titles, titles subsequent to the first (in items lacking a collective title), and
other title information.
MARC 21 examples:
"La Oficina = Das Büro" (parallel)
"facts or fiction" (other title information)
"[proceedings]" (ISBD 1.3.3.2))
"scale 1:20000" (other title information)
"basic level" (subtitle not included in title proper)
The ISBD namespace declares
isbd:P1006 "has other title information" as a subproperty ("narrower") of
isbd:P1012 "has title".
The results of data consolidation by dumbing up the ISBD and MARC 21 examples are:
M has title "situation en 1972"
M has title "pastels, lavis, gouaches, esquisses"
M has title "comédie"
M has title 'a novel based on the television series "The Avengers"'
M has title "La Oficina = Das Büro"
M has title "facts or fiction"
M has title "[proceedings]"
M has title "scale 1:20000"
M has title "basic level"
This is noise for most applications using ISBD linked data.
I was a member of the group that developed the ISBD namespace. The declaration was made to reflect traditional professional opinion. However, the noisy result of implementing that point of view was not intended by the ISBD Review Group.
I am fortunate to be a member of the group that is developing a potential new version of ISBD that is an implementation of the IFLA LRM. In the first stage of the project, the group is focusing on the LRM Manifestation entity. This will be an opportunity to rectify the error. I think the main options are to remove the declaration from the new ISBD namespace, or to change the definition to explicitly state that 'other title information is a title'.
RDA initially followed the ISBD approach. The 3R Project was an opportunity to review the semantics of the "other title information" element and resolve the problem. The factors that were considered include the treatment of a title as an appellation and an instance of the LRM Nomen entity, and the definition of "other title element" in ISBD, MARC 21, and RDA which avoids saying that it 'is a title'. The 3R Project treated this as a legacy element, retained for continuity and compatibility. I do not think it is really required, because the values that it typically takes are better recorded using the prerecording options:
Record a Manifestation:
manifestation title and responsibility statement to reflect the principle of representation and support the
identify user task. The Toolkit points out that these data can be keyword-indexed but no further consistent data processing can be assured.
Extract and record a value of Manifestation:
title of manifestation (or subtype such as Manifestation:
parallel title proper or Manifestation:
variant title of manifestation. These data can be dumbed-up, etc. and used in alphabetic title browse indexes to support the
find task.
There is no problem here; there was, but it has been resolved. There is no link to instructions for recording a title because the element is not a title. There are no specific instructions for correcting typos because the value of the element is not intended for processing as a title, and the principle of representation can be applied.
The RDA "parallel" elements are also legacy elements that are retained for continuity, etc. They are not really necessary.
The "parallelness" may be recorded in a normalized transcription of a Manifestation: manifestation title and responsibility statement, which may be a clearer reflection of the context ("parallel to what"?).
The "appellation-ness" may be recorded in a "variant" elements.
The selection of one of the values as the 'preferred' or 'proper' value is recorded in a preferred/proper element. RDA allows different agencies to make different choices of preference.
Agency 1 may record:
M has name of publisher "Deutsche Nationalbibliothek"
M has parallel name of publisher "German National Library"
Agency 2 may record:
M has name of publisher "German National Library"
M has parallel name of publisher "Deutsche Nationalbibliothek"
The "parallel" statements can be dumbed-up, according to RDA, to:
M has name of publisher "German National Library"
M has name of publisher "Deutsche Nationalbibliothek"
There is no need for a broader element.
The fact that 'there is only one slot for title proper in MARC' is not relevant to the discussion. RDA does not say, and has never said, that the cardinality of 'proper' elements is 1. I assume that MARC 21 is reflecting the authority control paradigm of selecting one value as 'preferred' and the rest as 'variant'. That is not the only way; bi-lingual agencies may want to refine the cardinality to 'within a specific language or script'.
The original RDA definition of 'title proper' is "A chief name of a manifestation ...". The use of the indefinite article indicates an expectation from the get-go that different agencies will make different choices.
The inclusion of "other language or script" in the definitions of parallel elements is an issue. RDA tries to avoid value dependencies: situations where the value recorded in one element is dependent on a value recorded in another element. Value dependecies interfere with semantic dependencies, such as hierarchical arrangement of the elements. This is one of the reasons why it is unlikely that new 'parallel' elements will be added to RDA. Retaining the existing elements means balancing continuity with clarity; the compromise that was applied in the new RDA Toolkit is to standardize parallel value dependencies as 'another language or script'. This can only be interpreted in the context of a local choice by the "agent who creates the metadata" of preferred language etc. Orientation and training will support continuing use of this traditional approach.
So there is a qualifying refinement in "another language or script"; it refines the choice of language or script preferred by an agency, and the full meaning can only be applied within a specific application of RDA.
The concept of 'parallel' as 'translation' is interesting. It may be useful to consider the relationship between 'parallel' elements and 'parallel aggregate', defined as "An aggregate that embodies expressions of a single work". The LRM models the title page, etc. of a manifestation as the expression of a work that is aggregated with the rest of the content in the manifestation. This is usually ignored and not recorded in most applications; the creator of the title, other title information, statement of responsibility, publication statement, etc. (i.e. a publisher) is of no interest. Instead, the content of the title page, etc. is treated as the expression of a metadata work and transcribed and transformed into descriptive metadata according to the principle of representation. A parallel title page, etc. can therefore be regarded as a translation (a different expression of the same metadata work).
------------------------------
Gordon Dunsire
------------------------------
Original Message:
Sent: Jan 28, 2021 09:53 PM
From: Heidrun Wiesenmüller
Subject: Other title information and more in the official Toolkit
Yesterday, I wrote:
-----------
I assume we're agreed that broader and narrower elements (element subtypes) are in a generic relationship, i.e. the relationship between a class and its types or kinds. The German DIN norm for thesauri defines this as a hierarchical relationship between two concepts, of which the subordinate concept has all the attributes of the superordinate concept plus at least one more (specifying) attribute.
I grant that this works for most of the elements in question, e.g. the parallel designation of edition is defined as „A designation of edition in another language or script". So this has all the attributes of designation of edition plus the attribute „in another language or script". (I find this oddly put -- other than what? The definition would only make sense in connection with the 'proper' element. I'll come back to this later.)
-------------
But it has just occurred to me that this is, of course, nonsense.
Rather, the situation is like this: A parallel designation of edition is in another language or script than the corresponding 'proper' designation of edition, i.e. we have a designation of edition in language A or script a and a parallel designation of edition in language B or script b.
So, the parallel element doesn't have all the attributes of the proper element plus the feature "in another language". But rather, these two elements are on the same hierarchical level. They are siblings, because they have all other attributes in common and are only distinguished by language or script.
If a broader/narrower relationship is felt to be necessary in the official Toolkit, then there would need to be a new element covering „designation of edition in any language or script". The children of this would be designation of edition and parallel designation of edition.
Summing up: The presentation of parallel elements as narrower elements of the 'proper' ones isn't logically sound, and this is true for all of them (not only for title proper and parallel title proper). I think this really needs to be looked into and remedied.
Heidrun
------------------------------
Heidrun Wiesenmüller
Stuttgart Media University
Original Message:
Sent: Jan 28, 2021 11:28 AM
From: Heidrun Wiesenmüller
Subject: Other title information and more in the official Toolkit
Gordon,
I did some more thinking about the second part of your mail, where you explained about the relationship between the 'proper' and the parallel elements.
I assume we're agreed that broader and narrower elements (element subtypes) are in a generic relationship, i.e. the relationship between a class and its types or kinds. The German DIN norm for thesauri defines this as a hierarchical relationship between two concepts, of which the subordinate concept has all the attributes of the superordinate concept plus at least one more (specifying) attribute.
I grant that this works for most of the elements in question, e.g. the parallel designation of edition is defined as „A designation of edition in another language or script". So this has all the attributes of designation of edition plus the attribute „in another language or script". (I find this oddly put -- other than what? The definition would only make sense in connection with the 'proper' element. I'll come back to this later.)
But the same is not true for title proper and parallel title proper, because title proper is defined as the preferred title of manifestation (the actual wording is different, I'm using the cross-reference from the Glossary here). This means, at least in our current practice, that there is a built-in cardinality of one: Just like the Highlander, there can only be one preferred title of manifestation in any system. So in my opinion, it is logically unsound to define the parallel title proper as „a title proper in another language or script". It cannot have all the attributes of title proper plus „in another language or script", because if it did, there would be two preferred titles at the same time, and this isn't allowed.
If we imagine a system which doesn't want to distinguish between 'proper' and parallel elements, there wouldn't be much of a problem in the case of designation of edition, when this data were imported into MARC, because this is a repeatable element. E.g., for „Ungekürzte Ausgabe" and „Unabridged version", there would only be a slight mistake in coding, with the English version ending up in the wrong subfield.
But if two titles were marked as „title proper", this wouldn't work as there is only one slot for title proper in MARC. So I think RDA definitely needs an additional hierarchical level here, e.g. a new „main title" element with the subtypes title proper and parallel title proper.
I'd prefer to have a different solution for the other 'proper' and parallel elements, as well. Because due to the way they are presented now, the main thing seems to be lost: that they always appear as a couple, a triple, etc. (it doesn't make sense to talk about a parallel designation of edition if there is no 'proper' one), and that they are related in a special way.
I already pointed out that I find the phrasing „in another language or script" problematic. This kind of definition also does not capture the main thing of the parallelity: the two designations of edition must have the same meaning. If I had „Unabridged version" on the one hand, and „2., überarbeitete Auflage" [2nd, revised edition] on the other, I wouldn't treat the second one as a parallel designation of edition, although it certainly is „a designation of edition in another language or script".
Admittedly, this isn't a very likely example to appear with real-life resources. But still, I think the definitions of the Toolkit fail to explain the character of this phenomenon. I also think the phenomenon would be much easier to understand if 'proper' and parallel elements were presented on the same hierarchical level, as sibling elements.
Heidrun
------------------------------
Heidrun Wiesenmüller
Stuttgart Media University
Original Message:
Sent: Jan 27, 2021 09:14 AM
From: Gordon Dunsire
Subject: Other title information and more in the official Toolkit
All
Why is the element called 'other title information' and not just 'other title'? The label of this long-established element indicates, in my opinion, an acknowledgement that not all instances of other title information are titles.
The Definition and Scope section of Manifestation: other title information gives the definition as "A word, character, or group of words or characters that appears in conjunction with, and is subordinate to, a title proper of a manifestation."
The definition does not imply that other title information is a title.
The Prerecording section states:
"A value of this element may include a phrase that appears on a manifestation as an addition to a Manifestation: title proper that is indicative of the character, contents, and other aspects of a manifestation, including its creation.
A value usually appears in the same metadata Work: recording source as a value of title proper.
A value of this element may include a value of a Manifestation: title of manifestation that is not a title proper."
The Prerecording section points out that other title information may be the source of a title that is not selected as a title proper, but it may also include other information that is not usually considered to be a title or has not utility as a title.
The Prerecording section includes two options for recording the information in another element. The options are not mutually exclusive.
The first option supports the recording of some or all of 'other title information' as a Manifestation: title of manifestation or subtype (so, for example, Manifestation: variant title of manifestation may be used).
The second option supports the recording of a transcription of 'other title information' within the broader element Manifestation: manifestation title and responsibility statement.
Of course, the instructions in the Recording section support the recording of a transcription for the element itself.
Overall, this covers the cases of extracting a title (treated as a nomen) to support access, or transcribing the whole of the 'other title information' as part of a broader element or in a co-extensive element.
The changes between the original Toolkit and the new Toolkit arise from clarifying the distinction between 'recording' and 'transcribing', implementing the new LRM Nomen entity, and retaining continuity.
The element label 'title proper' is retained for continuity and familiarity. It is the same as 'preferred title of manifestation', and there is a cross-reference in the Glossary: preferred title of manifestation See: title proper. The Manifestation: title proper Definition and Scope section gives the definition of title proper as "A nomen that is a title of manifestation that is selected for preference in a specific application or context."
Manifestation: other title information and Manifestation: parallel other title information are not 'disjunct classes'. The RDA entities are examples of disjoint classes; an instance of one of the entities cannot be an instance of another entity. This is stated in the LRM.
Some applications prefer to treat 'parallel' information without distinguishing it from 'proper' information. Other applications want to record only one 'proper' element, and to record 'parallel' elements for the same information presented in a different language. Within that, different applications may want to make a different choice as to which piece is 'proper': the first instance on the source of information? the instance that is in the preferred language of the agent? a throw of the dice?
For that reason, all 'parallel' elements in RDA are subtypes of the 'proper' element. This allows metadata statements that use parallel elements to be 'dumbed-up' to statements that use the 'proper' element:
M has parallel title proper "My title"
implies:
M has title proper "My title"
M has title of manifestation "My title"
M has appellation of manifestation "My title"
All implied statements are valid if the original statement is valid. An application that makes one of these implied statements a 'real' statement can interoperate with another application at the level of the lowest common denominator element.
An element subtype is not a sub-element; the distinction made in the original TK carries over to the new TK (Glossary):
element subtype: A narrower category of an element.
sub-element: An element that is a component of a larger element that aggregates data values from two or more elements.
In short, 'other title information' is not always a 'title' and when it is, it can be recorded as a 'title'.
------------------------------
Gordon Dunsire
Original Message:
Sent: Jan 26, 2021 07:48 PM
From: Mark Ehlert
Subject: Other title information and more in the official Toolkit
Unless I missed it elsewhere in this discussion, "Other title information" and its "Parallel" offspring have no range applied to them. These are attribute elements, not relationship elements pointing to Nomen. And have been since at least the May 2020 version of the text, according to my notes. I see other examples of attribute elements isolated in this manner, like "Sound characteristic."
The question for me is, why aren't the strings recorded for "Other title information" and like elements treated as Nomens? The subtitle "choice and design in the Iliad" has as much 'label-ness' as the title proper "Homer's cosmic fabrication." Granted, subtitles aren't reused as "Title proper" elements often are-think access points for works/expressions.
--
Mark K. Ehlert Alma: NA02
Cataloging and Metadata Librarian Primo VE: NA02
O'Shaughnessy-Frey Library, University of St. Thomas
<http://www.stthomas.edu/libraries/>
"Experience is by industry achieved // And perfected by
the swift course of time"--Shakespeare, "Two Gentlemen of
Verona," Act I, Scene iii
Original Message:
Sent: 1/25/2021 2:19:00 PM
From: Heidrun Wiesenmüller
Subject: Other title information and more in the official Toolkit
So, let's give the new RDA-L a try...!
With the now official Toolkit, I'm often not sure whether things which I find odd are intentional or simply mistakes. At present, I'm exceedingly puzzled by the elements other title information and parallel other title information. In the original Toolkit, both were presented as subordinate to title. See 2.3.1.1 where it says „For the purpose of resource description, titles are categorized as follows: ...". The list which follows includes title proper, parallel title proper, other title information and parallel other title information.
But in the official Toolkit, other title information and parallel title information are not shown as subelements of title of manifestation: Neither does the list at 73.88.03.17 include these elements, nor are they given as narrower elements at the bottom of the page for title of manifestation. I cannot see any reason why other title information should no longer be seen as a kind of title information. So why is it now presented as something completely separate?
And there is another odd thing: If we look at the page for other title information, the only related element is parallel other title information, which is given as a narrower element. Again, I don't understand: To my mind, these are disjunct classes. Either something is classified as other title information or it is classified as parallel other title information.
I just checked: The same situation applies to title proper, where parallel title proper is given as a narrower element. But I'd argue that you cannot have title proper mean two different things at the same time. Either it is a broad element which covers both what we traditionally call title proper and parallel title proper, or it is on a narrower level and makes a clear distinction between title proper and parallel title proper. I think you cannot have it both ways.
Of course we can imagine an application which simply doesn't want to make the distinction between title proper and parallel title proper. That's fine, but in this case I think that one needs to go one level up and use the broader element title of manifestation (or maybe use manifestation title and responsibility statement).
I feel my head starting to spin again. And I'm not really sure whether I was able to make myself clear... could you follow my train of thoughts? And if so, what do you think? Is there a very good reason for this which I was only to dumb to grasp?
Heidrun
------------------------------
--
---------------------
Prof. Heidrun Wiesenmueller M.A.
Stuttgart Media University
Nobelstrasse 10, 70569 Stuttgart, Germany
www.hdm-stuttgart.de/bi
------------------------------