Marissa Rhodes
LIS514
Abstracting Journal
Entry #1
Houppert, A.M. (2005). Lost in translation: Common index errors and strategies to avoid them. Key Words 13(4), 125-126.
Given
As the Index Manager at the National Geographic Society, I am responsible for reviewing completed back-of-the-book and magazine indexes. Indexers are a detail-oriented group who want to turn in quality, error-free indexes, but errors occasionally creep in despite our best efforts. Over the past ten years I have seen some of the same errors crop up again and again, many of which could have been avoided with strategic proofing.
Index entries must be well chosen and well crafted, but they must also be correctly spelled, structured, and formatted. If either side of the equation is off-balance, the usefulness of the index will be compromised. The best indexes are very readable as standalone documents, easily followed and deciphered, and accurately point you to where you want to go.
Indicative
Errors diminish the readability and usefulness of an index. Common mistakes may be avoided using the appropriate strategies. MCR
Informative
Errors diminish the readability and usefulness of an index. Redundant page spans may be eliminated by proofing them separately from the headings. Code invertible entries along the way to ensure that the original page spans match those of the inverted entry. See references leading to broader terms may be avoided by consulting a hierarchical database of terms or envisioning related terms in a hierarchical structure. Refer to an appropriate style manual or to the Library of Congress Subject Headings to correct alphabetization discrepancies; adopt a consistent scheme and adhere to it. To avoid having to adjust the length of the index after it is created, obtain a rough estimate from the publisher beforehand and calculate the number of entry lines per book pagetheanticipated length will accommodate.MCR
Critical
The author begins with a couple of peculiar and unnecessary statements. It is questionable whether even the best indexes may be called readable, though the author suggests that solving the following errors will make it so. She goes on to describe the frustration that poor formatting can cause to users even though only two of the five scenarios she discusses have anything to do with formatting. The author highlights, via scenarios, what she considers to be the five most frequent errors made by indexers. She offers strategies which would be more accurately named solutions to each error. The scenarios and solutions are clear and concise. This article is most appropriate for vocational training purposes and would be inadequate for analytical research of a higher order. MCR
Entry #2
Layne, S. S. (1994). Some issues in the indexing of images. Journal of the American Society for Information Science, 45, 8, 583-88.
Given
This article discusses some of the intellectual issues involved in the indexing of visual or pictorial images, postulating that the indexing of images should provide access to images based on the attributes of those images, and provide access to useful groupings of images, not simply access to individual images. The attributes of images can be divided into four categories: “Biographical” attributes, Subject attributes, Exemplified attributes, and Relationship attributes. When creating groupings of images, it is important to consider the following issues or questions: When should the grouping occur? What are the groupings based on? What level of detail is necessary? and What groupings will be useful? More research is needed into the ways images are sought and the reasons that they are useful.
Indicative
Indexing provides access to images, and useful groupings of images, based on their attributes. There are four categories of image attributes, each of which plays a role in description. Groupings of images are useful to searchers whose needs result in indexing complexities but more research is needed to analyze users’ information seeking behaviors. MCR
Informative
Indexing provides access to images, and useful groupings of images, based on their attributes. There are four categories of image attributes. The first is biographical, which concerns the creation and provenance of the image. Second is subject, which reflects the image’s ability to convey meaning via representation. A subject may be of one thing but about another. An image’s subject may also be interpreted generically or specifically, and it can be classified into four facets: time, space, activities/events, and objects. The third attribute category is exemplified which concerns an image’s ability to be an example of a type of thing. Lastly, the relationship category deals with an image’s association with other images, texts or objects. Each attribute category plays a unique role in description. Groupings of images are useful to art historians who tend to juxtapose and compare images, to searchers who are unable to be specific about what they seek, and to searchers whose criteria are so specific and unique that visual identification is the most reasonable option. In forming groups once must consider whether they are to be created at the time of retrieval or prior, during indexing. One may choose to index the image attributes or the attributes of what is represented by the image, deciding which attributes are most important. The indexer must also decide whether to group images whose same attributes share the same value orthose who share the same value for different attributes. The indexer must focus on recall in lieu of precision and consider the possibility of predicting what groupings will be useful, though more research on this topic is required. MCR
Critical
Despite this article being well-organized, the complexity of the content makes this article less readable than it could be. The author is obviously very knowledgeable about the subject but the hierarchical writing style is not ideal. The author creates lists which branch off into other lists, which branch off into even more lists. The information would benefit from an outline-style presentation. This is a very concise but information-packed article that would be helpful a scholastic atmosphere.MCR
Entry #3
Zeng, M. L. (2001). Making indexes for Web sites—a taste of the challenge. Key Words 9 (2),43-45.
Indicative
A professor presents a summary of the problems her students encountered and the solutions they proposed as they executed an assignment to create indexes for dynamic Web documents.MCR
Informative
Dr. Marcia Lei Zeng presents a summary of the problems her students encountered and the solutions they proposed as they executed an assignment to create indexes for dynamic Web documents. There is no persistent URL for frame-based web pages that will invoke both the individual page and the contextual frame simultaneously so linking locators to these pages was not possible without taking the individual page out of context. Linking locators to a specific spot within a larger page is also problematic if the page developer does not know to anchor index entries to this location. Indexing non-text files, the contents of pop-up windows and audio/video files put the onus on the user who was required to invoke these files since the locator links could only take them to the page on which these objects were located. The dynamic nature of the Web obligated students to refrain from indexing content on particularly dynamic pages and just index the page itself. Index layout required special consideration since coherency with the regular site was a priority and the index was to be presented using html, which often resorts to wrapping long entries. MCR
Critical
This article is short, interesting and well-organized. It does not probe the more profound questions within the field of indexing but it does provide a straight-forward list of website indexing issues. The article stops short of suggesting solutions to these problems in some cases but the discussion following the problem description is interesting nonetheless. Most, but not all of the content is focused specifically on website indexing. This article is handy for the practitioner and students looking for practical advice. MCR
Entry #4
Chulick, M. A. G. (2000). The mysteries of fiction indexing. Key Words8 (5),166-170.
Indicative
Author describes her experience indexing serial fiction using Cindex™ software. The complex nature of the novel required her to devise creative solutions regarding term choice and the index’s organizational scheme. MCR
Informative
Author Mary Ann Chulik describes her experience creating a combined index forLes Roberts’s Cleveland-basedMilan Jacovich Mystery Seriesusing Cindex™ software. Indexing serial fiction is useful to literature scholars as well as to recreational readers whose reading experience may be enhanced by the retrieval of information from earlier books as the series progresses.Though Chulik organized the main headings alphabetically, she was obligated to manipulate Cindex™ to organize the event subheadings for characters chronologically since a user would most likely read about plot events in chronological order. This factor also affecting Chulik’s term choice since she was reluctant to spoil plot twists for readers who had not yet read about them. Chulik benefitted from tailoring the term selection process to the series, including Cleveland sports teams since the series is a source of pride to Clevelanders and sports are important to the city. The author concluded that indexing fiction is extraordinarily difficult but rewardin. MCR
Critical
This is a case study about the fascinating and underdeveloped topic of fiction indexing. The author has an excellent and engaging writing style, though it is somewhat informal. The formatting of the article, split under broad, ambiguous headings, is not particularly helpful and she could have elaborated more on her use of the indexing software Cindex™. This article functions as an inspirational piece or as a starting point for serious research on this oft-neglected topic. MCR
Entry #5
Jacobs, C. (1999). If a picture is worth a thousand words, then…The Indexer 21 (3), 119-121.
Given
Images are being recognized as an important method of information transfer. Image indexing thus warrants the same kind of serious attention that is given to text analysis and indexing. Levels of analysis, the interpretation of relationships between the elements in an image, and appropriate vocabulary are the central problems. Commonsense guidelines such as rigorous vocabulary control help, but the essential factor is that image storage and retrieval must be valued.
Indicative
Images are a valid source of information but they are often avoided by indexers. The problematic nature of image indexing explains why information scientists are slow to develop the field. Author insists that indexers take images seriously and makes suggestions for indexers commencing image-indexing projects. MCR
Informative
Images are a valid source of information but indexers continue to neglect the complex subject of image access. Discerning to ofnessof an image is similar to traditional subject indexing but it is more difficult to determine the aboutness of images because they contain layers of representation. Describing the elements of an image is much easier than describing the relationship between those elements. Vocabulary control, or the lack thereof, is also to blame for the neglect of image indexing. Attempts at image indexing are benefited by the indexer understanding his users and the way they will utilize the collection. Vocabulary control and multileveled indexing are also advisable. An analytical framework can help to ensure that the indexer considers all aspects of an image’s aboutness. MCR
Critical
Well-written and interesting, this article is a typical magazine-style piece. The author simplifies the complex issue of image retrieval and provides rational suggestions for image indexers. One of the author’s suggestions is to control vocabulary during word choice, which is a valid piece of advice but it is hardly unique to image indexing. The author implies that information scientists are simply not up to the challenge of analyzing and indexing images for retrieval purposes so they neglect image repositories and downplay their usefulness. She chastises her peers for their perception that images are not a legitimate information source but she does not provide proof that such a perception exists. Nor does she make a case for the usefulness of images as information sources. Though not a hardcore piece of scholarship, this article useful as a starting point for further research. MCR
Entry #6
Graeme, B., & Douglas, A. (1996). Image indexing and retrieval: some problems and proposed solutions. Internet Research: Electronic Networking Applications and Policy 6 (4), 67-76.
Given
Image processing technologies are offering considerable potential for library and information units to extend their databases by the inclusion of images such as photographs, paintings, monograph title-pages and maps. Discusses problems and potential solutions in a structured fashion based on categories of thesauri (text and visual), hybrids, description language and automatic content analysis, with state-of-the-art examples.
Indicative
Increasing technological capabilities have made it possible to index and retrieve images but doing so is more complex than textual indexing and retrieval. Organizations have used textual thesauri to retrieve images for some time though visual and hybrid thesauri are being developed. More recently efforts have been made to develop image description languages and content-based image retrieval systems. Progress is hindered by a lack of research and a failure to launch coordinated and sustained efforts in the discipline. MCR
Informative
Increasing technological capabilities have made it possible to index and retrieve images but doing so is more complex than textual indexing and retrieval because the aboutness of images is not always apparent. The Art and Architecture Thesaurus, ICONOCLASS, and TELCLASS are all textual thesauri that have been used for image retrieval for some time. NASA and UC Berkeley created prototypes for visual and hybrid thesauri while Syracuse University’s system prototype uses visual input and indexing to retrieve images. The Book House design is a prototype based on extensive research on user cognition and image association. The University of London is experimenting with an image description language. More recently, a content-based has been adapted by experimenters who wish to perfect this method of automated content analysis and retrieval. Progress in the field is hindered by a lack of research and a failure to launch coordinated and sustained efforts within the discipline.MCR
Critical
According to the author, this article is intended to discuss the issues arising out of image indexing and the various approaches that difference organizations have taken to address these issues. In reality, the article reads more like a progress report, listing, summarizing and evaluating contemporaneous attempts at image indexing. The article is arranged usefully, separated into parts by category, and it does a good job of contextualizing these advances in the field. Articles of this nature are not conducive to longevity and this article is no exception; it is very dated. The author’s discussion have lost value with time. MCR