Standards

STANDARDS COMMITTEE
ASIS Votes on Z39.14-199x

by Mark Needleman

Several months ago, as chair of the ASIS Standards Committee, I invited all ASIS members to participate in the standards development process by reviewing NISO Z39.14 Writing Abstracts and helping the committee develop the comments and vote that ASIS would submit as a voting member of the National Information Standards Organization (NISO). Several members accepted my invitation, devoting a lot of time and energy to the process and providing committee members with valuable comments. We used those thoughts to develop the following paper that was submitted to NISO. I will keep the membership informed of the progression of this standard. In addition, I will once again solicit reviewers from the general membership on future votes when feasible.

At the conclusion of this article, I have provided a summary of all votes and actions that have been taken by NISO during the time I have served as chair of the ASIS Standards Committee.

If you are interested in serving on the Standards Committee, please contact me or Richard Hill at ASIS Headquarters with an indication of your interest.

ASIS is voting NO on Z39.14-199x with the following comments.

Major Comments:

  1. The definition of abstract in the definitions section is limited to documents that present studies. Opinion pieces, newspaper articles, status reports and bibliographies are not supposed to have abstracts based on this definition. In an environment where almost every type of thing is accessible through electronic means, the definition needs to be broadened to include all types of documents and the rest of the standard needs to be revised to ensure it conforms to this wider definition.
  2. Is it necessary to list the domains for which abstracts are applicable? The list states "all of the natural, physical and biomedical sciences, as well as within the humanities and the behavioral and social sciences," which raises questions such as "how about engineering or technology which are often not considered 'science'?" The list seems unnecessary and limiting.
Comments on Specific Sections:
  1. Introduction: the comma after "abstract" in the first line is not needed.

  2. Section 3 - Definitions:
    1. The following element is missing:
      a. Abstract authorship: What is the recommended standard for acknowledging the authorship of an abstract when it is used by an access service? The following are examples.
      Author
      Author (revised)
      Author (translated)
      Name of the access service providing the abstract
      Name or initials of the individual abstractor
      This usually appears at the end of the abstract or in a following field of the record. It is recommended that the individual abstractor NOT be indicated (except for in-house purposes) but that the name of the service that provided the abstract or revised or translated it be given. The examples in the appendix should include this attribution.

    2. Access publications and services (rewrite)
      "Print or on-line collections of abstracts accompanied by their bibliographic references that serve as alerting or retrospective access keys, or both, to original documents. Author abstracts are often used verbatim when they are well-written and where permitted by copyright. Otherwise, services either write original abstracts or revise the author's abstract to meet the needs of the service and on-line retrieval."
      [Note: in particular, delete the phrase "when the original one is inferior."]

    3. Electronic abstracts: Is this necessary?

    4. Structured abstracts
      If this is a legitimate form of an abstract, then it should be discussed in the standard. Otherwise, this definition should make it clear that this type of abstract is outside of the scope of this standard. As it is now, the standard contains only this definition and an example in the appendix.
  3. Section 4.1 Purpose
    Replace the phrase "fringe interest" with "marginal interest." An assumption being made here is that users use the abstract as a substitute for the full document. A user may find a document of great interest but only need the abstract at a particular time.

    In addition this section is very limiting. There are other purposes for having abstracts, such as to serve as an introductory overview for users who plan on reading a document, to facilitate free-text searching, to be used in abstracting and indexing databases, etc.

  4. Section 4.2 Location of the abstract
    This section refers to a printed publication. Therefore, the first sentence should be revised to "In a print journal, . . . ."

    How about locating the abstract with the table of contents as an appropriate alternative in some cases? This is appropriate for some journals where the abstract at the beginning of the article would interfere with the style of presentation (e.g., Smithsonian Magazine). Since there is more electronic use of tables of content, the presence of the abstracts there obviously would enhance the value of such a listing.

    For electronic abstracts some consideration should be given and mention made of delimiting them with standard tags or codes to facilitate the electronic handling of abstracts that are part of text files. Such delimiters would also facilitate locating abstracts, particularly in instances when the page format is nonexistent.

  5. Section 5.2 Reports and theses (rewrite)
    "Include an abstract in every separately published report, paper or thesis. If the distribution of the document is restricted (i.e., not publicly available), it is highly desirable that a publicly-available abstract be provided for documentation purposes."

    [Note: The use of "pamphlet" in the definition is puzzling. Also, there are forms of restricted distribution of government documents other than classification (e.g., limited distribution to non-US citizens); the statement should be broad enough to cover all such situations.]

  6. Section 5.5 Standards
    This section goes into detail about the content of an abstract for a standard. Since this is not true for other types of documents described here and since Section 6 addresses the content of abstracts, perhaps this is out-of-place.

  7. Section 6 Types of Abstracts and Their Content The definitions of informative and indicative abstracts are flawed in sections 6.1 and 6.2. It explains that informative abstracts are for experiments and indicative are for all other types. Documents that are not about experiments can be represented in informative abstracts as well.

    This is one of the major flaws in the standard. It relegates all non-empirical documents to the indicative style, when, in fact, many historical, theoretical or philosophical writings would be better served by an informative abstract. This standard appears to disallow the writing of an informative abstract for any document other than one reporting on experimental work. Equal emphasis on the important elements of information to be included in an informative abstract of theoretical work should be provided.

    Throughout Section 6, the standard presents "dos" and "don'ts" that are a matter of policy and should be determined by the purpose of the abstracts and how they are used. For example, in section 6.1.3 the priority list for what to do about too many results is subjective and should not be recommended universally.

  8. Section 6.1.1 Purpose
    This is the first mention of "self-contained." The explanation for this status, which is included in Section 7.1, should be moved or repeated here.

    The standard is unclear as to who its audience is. Although the standard states that it is intended for both abstractors who work for access services and authors, it may be unlikely that authors are familiar with many of the concepts (e.g., descriptors, identifiers) referred to. If the standard is indeed for use by authors those concepts should be explained as one cannot assume non-professionals are familiar with them.

    In the last sentence, "in order" can be deleted.

  9. Section 6.2 Indicative Abstracts
    Three of the most popular document types for which this type of abstract is written are not included, namely books, conference proceedings and annual reports.

    Last paragraph. Amend 3rd sentence to "These statements may include essential background material and descriptions of the approaches used, aspects of the subject matter discussed or arguments presented in the text."

  10. Section 6.3 Indicative-Informative Abstracts
    This appears to be a catch-all for documents not suited to either an informative or indicative abstract.

  11. Section 7.2 Length
    What is the standard for counting the words in an abstract? All words? This should be stated.

  12. Complete Sentences
    This section also contains information about where collateral information should be located. This information should not be in this section, but should instead be in a separate, easily located enumerated section of the standard.

  13. Section 7.6 Terminology
    We do not agree with the last statement in the first paragraph: "For descriptors, however, such use should be restricted to the meanings specified in the vocabulary control tool from which they are taken." The abstract should contain the language of the document even when it is in conflict with the definitions established in the controlled vocabulary. Abstracting and indexing services should not be changing the wording of author abstracts to meet their own definitions. Just as the titles and the full text of the document contain the language used by the author, so should the abstract. Trying to do otherwise will lead to artificial reconstructions of abstract language that will not match the use of language in the document.

    The second paragraph is very unclear. Here is a suggested rewriting of it:

    "For the purposes of on-line retrieval, abstracts should have terminology that is (1) fully spelled out, (2) refers to the specifics of the document content, (3) places words directly adjacent to other words to represent concepts (e.g., 'middle class and working class' rather than 'middle and working class') and (4) avoids word adjacencies across punctuation that could lead to false retrieval (e.g., 'visibility through the periscope, detection of targets on sonar CRTs')."
    In addition an example should be provided of the standard practice of handling abbreviations the first time they are used - spell the term out the first time it is used followed by the abbreviation, then use abbreviations on subsequent uses.

  14. Appendix
    1. We don't think the use of capital letters for the titles is attractive or easy to read and we doubt if this is common practice in "primary publications." We would recommend instead that the titles be in bold and upper/lower case. We don't have a preference regarding the centering of the titles and don't think the standard should promote this style over any other.
    2. Examples in the appendix are unclear as to the differences between each type of abstract. Features which differentiate one type from another should be highlighted.
    3. Ordering of the appendix: There is something confusing about the way in which the appendix is organized. Also, the appendix should not include additional directions on the recommended use of the abstracts, but should instead refer back to the appropriate section of the standard.

      Proposed reordering of the appendix:

      1. Examples: Informative Abstracts
      2. Examples: Indicative Abstracts
      3. Examples: Indicative-Informative Abstracts
      4. Example: Structured Abstract
      5. Example: Monograph
      6. Examples: Chapters
      7. Examples: Different Styles of Abstracts for the Same Document
      8. Examples: Complete Entries in Access Services (with bibliographic citation)
      9. Citations for Abstracts Used as Examples
General Comments:
  1. As with some other Z39 standards, this one is really only a guideline. It is too bad that there is not a way to indicate this clearly.

  2. The standard uses the word "accurately" in several places where it is unnecessary. For example:
    Foreword: "Basic content must be quickly and accurately identifiable,. . ."
    4.1: ". . . to identify the basic content of a document quickly and accurately; . . .

    The idea of accurately portraying the content is surely implied.

  3. The list of "documents to be abstracted" in Section 5 is too limited. The guidelines should apply to such uses as writing abstracts or summaries for metadata records for data sets, such as the Directory Information Format (DIF) used by the Global Change Directory (NASA) and the Federal Geographic Data Center (FGDC) metadata standard for describing spatial data sets.

  4. It would be useful to know what the motivation was for revising the standard. The forward says that the "revised standard represents no basic departure from the principles and practices of the original edition." If so, why is it is it being revised and what were the purposes of the revision?

  5. It is unclear whether the purpose of extracting text directly from the original document for inclusion in the abstract is acceptable, although this is common practice for access services.

  6. The standard should state what an abstractor should do if the document is poorly written. Also what to do if there is no conclusion - should the abstractor indicate that fact explicitly.

Mark Needleman, chair of the ASIS Standards Committee, works in the University of California Department of Library Automation.


Recent NISO Actions

ASIS Votes on NISO Ballots Pending NISO Votes Other ASIS Actions for NISO