Search (14 results, page 1 of 1)

0.030295122 = product of:
  0.12118049 = sum of:
    0.12118049 = sum of:
      0.08411469 = weight(_text_:terminology in 1415) [ClassicSimilarity], result of:
        0.08411469 = score(doc=1415,freq=2.0), product of:
          0.24053115 = queryWeight, product of:
            5.2752647 = idf(docFreq=614, maxDocs=44218)
            0.045596033 = queryNorm
          0.34970397 = fieldWeight in 1415, product of:
            1.4142135 = tf(freq=2.0), with freq of:
              2.0 = termFreq=2.0
            5.2752647 = idf(docFreq=614, maxDocs=44218)
            0.046875 = fieldNorm(doc=1415)
      0.037065797 = weight(_text_:22 in 1415) [ClassicSimilarity], result of:
        0.037065797 = score(doc=1415,freq=2.0), product of:
          0.15966953 = queryWeight, product of:
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.045596033 = queryNorm
          0.23214069 = fieldWeight in 1415, product of:
            1.4142135 = tf(freq=2.0), with freq of:
              2.0 = termFreq=2.0
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.046875 = fieldNorm(doc=1415)
  0.25 = coord(1/4)

Abstract

Domain analysis reveals the contours of knowledge in diverse discourse communities. The Dublin Core Metadata Initiative (DCMI) conferences represent the cutting edge of research in metadata for the digital age. Beak and Smiraglia (2013) discovered a shared epistemology revealed by co-citation perceptions of the domain, a common ontological base, social semantics, and a limited but focused intent. User groups did not emerge from that analysis, raising an interesting question about the content of core thematic extension versus a highly granular intension. We analyzed keywords from the titles by year to identify core and granular topics as they arose over time. The results showed that only 36 core keywords, e.g. "Dublin Core," "Metadata," "Linked Data," "Applications," etc. represents the domain's extension. However, there was much rich terminology among the granularity, e.g., "development," "description," "interoperability," "analysis," "applications," and "classification" and even "domain" pointed to the domain's intension.

Source

Knowledge organization in the 21st century: between historical patterns and future prospects. Proceedings of the Thirteenth International ISKO Conference 19-22 May 2014, Kraków, Poland. Ed.: Wieslaw Babik

0.029709369 = product of:
  0.059418738 = sum of:
    0.041894842 = weight(_text_:headings in 409) [ClassicSimilarity], result of:
      0.041894842 = score(doc=409,freq=4.0), product of:
        0.22113821 = queryWeight, product of:
          4.849944 = idf(docFreq=940, maxDocs=44218)
          0.045596033 = queryNorm
        0.18945095 = fieldWeight in 409, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          4.849944 = idf(docFreq=940, maxDocs=44218)
          0.01953125 = fieldNorm(doc=409)
    0.017523894 = product of:
      0.03504779 = sum of:
        0.03504779 = weight(_text_:terminology in 409) [ClassicSimilarity], result of:
          0.03504779 = score(doc=409,freq=2.0), product of:
            0.24053115 = queryWeight, product of:
              5.2752647 = idf(docFreq=614, maxDocs=44218)
              0.045596033 = queryNorm
            0.14570998 = fieldWeight in 409, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              5.2752647 = idf(docFreq=614, maxDocs=44218)
              0.01953125 = fieldNorm(doc=409)
      0.5 = coord(1/2)
  0.5 = coord(2/4)

Content

"Actions have consequences, and this is certainly true of knowledge organization. One reason our colleague Birger Hjoerland (1998) urges epistemological analysis for the problems of information science is that resources might well serve many different purposes for different users, and thus different user groups might have different epistemological relationships with resources. There is a difference between consulting a dictionary for a definition, reading a text for comprehension to increase your knowledge base, reading for pleasure (which, evidently boosts certain endorphins), and synthesizing a scientific report to generate an hypothesis, just to generate a few scenarios. The only commonality in that list is the consultation of a resource. In each case the purpose dictates the activity and is reliant upon a different epistemological aim. No online source of facts is going to suffice if I want something to read that will give me pleasure; no catalog of fine literature is sufficient for the extraction of scientific theory. Hjoerland also suggests that the names we give - to documents, to categories, even to activities - embodies the action of naming, and thereby also the action of facilitating or obfuscating the use of named resources (Hjoerland 2003, 98). Terminology cannot be neutral because the very selection of terms as names either provides a pathway to understanding or a barrier to usage, depending on the epistemological perspective of the user group. I won't go looking for Miss Marple in your dictionary if you call it a dictionary, even though it might contain a perfectly fine list of motives for murder. Likewise, as an information scientist I am not likely to look for research anywhere except in a database that purports to contain peer-reviewed scientific literature. Names have power, and the action of naming is powerful too. We in knowledge organization need to be aware that no matter how elegant our science, the actions based on our research have consequences. A model generated empirically might make an excellent explanation of a specific reality, but if it migrates into the structure of a system for knowledge organization it has the power to help or hinder assignment to categories, not to mention retrieval from those categories.
An important aspect of what we do is facilitating the curatorial aspect of information retrieval or librarianship. What I mean is that our job is not merely to "mark and park," as generations of catalogers famously have said of both resource description and classification, or even to generate parking spaces (to press my metaphor), but rather our job is to place each entity in the best category, each artifact in the best environment, each resource on the best "shelf" to enhance its usability should it actually be sought for retrieval. Hope Olson (2002) has also written about the limits we create when we exercise the power to name. We must be aware of the consequences of our science. In librarianship in the United States at the moment there is a fair amount of hand-wringing about the future, and this anxiety has been fed by the report of Karen Calhoun on the changing nature of the catalog. Calhoun (2006) suggests that the library community should abandon many of its expensive knowledge organization practices - such as the Library of Congress Subject Headings - in favor of integration of search engines into library catalogs. As logical as this seems on the face of it (and as much as we might often have wished LCSH would go away!), purveyors of such notions have either forgotten or rejected the notion of the library as a social instrument, and therefore the order of things in libraries as an extension of that social role. We must also view knowledge organization then as a cultural enterprise, a social act that has consequences. The ontologies we use to devise categorical schemes imply certain realities. If we say there is no music other than Western Art, why, then there must be no point in paying any attention to music of any other sort, right? And if we say that UFOs are a kind of controversial knowledge, we join the community of non-believers who insist that UFOs do not exist. Surely if we thought they were viable phenomena we would create a concrete class for them (see DDC 001.942). Voila, now we know, UFOs do not exist - the DDC says so. And if a gay adolescent searches for literature to help understand and finds that it all falls under "perversion" then we have oppressed yet another youth (see Campbell 2001). Our actions have social consequences.
Librarianship incorporates the tools of knowledge organization as part of its role as cultural disseminator. Subject headings and classification were both intended by their 19`h century promulgators - perhaps most notably Dewey and Cutter - to facilitate learning by grouping materials of high quality together. We might call this enhanced serendipity if we think it happens by accident or act of fate, or we might call it curatorship if we realize the responsibility inherent in our social role. The cataloger's job always has been to place each work sensitively among other works related to it, and to make the relationships explicit to facilitate and even encourage selection (see Miksa 1983). Schallier (2004) reported on the use of classification in an online catalog to enhance just such a curatorial purpose. UDC classification codes were exploded into linguistic strings to allow users to search, not just for a given term, but for the terms that occur around it - that is, terms that are adjacent in the classification. These displays are used alongside LCSH to provide enhanced-serendipity for users. What caught my attention was the intention of the project (p. 271): UDC permits librarians to build virtual library shelves, where a document's subjects can be described in thematic categories rather than in detailed verbal terms. And: It is our experience that most end users are not familiar with large controlled vocabularies. UDC could be an answer to this, since its alphanumeric makeup could be used to build a tree structure of terms, which would guide end users in their searchers. There are other implications from this project, including background linkage from UDC codes that drive the "virtual shelves" to subject terms that drive the initial classification. Knowledge organization has consequences in both theory and application."

0.025655247 = product of:
  0.10262099 = sum of:
    0.10262099 = weight(_text_:headings in 5678) [ClassicSimilarity], result of:
      0.10262099 = score(doc=5678,freq=6.0), product of:
        0.22113821 = queryWeight, product of:
          4.849944 = idf(docFreq=940, maxDocs=44218)
          0.045596033 = queryNorm
        0.46405816 = fieldWeight in 5678, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          4.849944 = idf(docFreq=940, maxDocs=44218)
          0.0390625 = fieldNorm(doc=5678)
  0.25 = coord(1/4)

Abstract

Explicit authority control of works is essentially non-existent. Our catalogs are built on a principle of controlling headings, and primarily headings for names of authors. Our syndetic structure creates a spider's web of networked relationships among forms of headings, but it ends there, despite the potential richness of depth among bibliographic entities. Effective authority control of works could yield richness in the catalog that would enhance retrieval capabilities. Works are considered to constitute the intellectual content of informative artifacts that may be collected and ordered for retrieval. In a 1992 study the author examined a random sample of works drawn from the catalog of the Georgetown University Library. For each progenitor work, an instantiation network (also referred to as a bibliographic family) was constituted. A detailed analysis of the linkages that would be required for authority control of these networks is reviewed here. A new study is also presented, in which Library of Congress authority records for the works in this sample are sought and analyzed. Results demonstrate a near total lack of control, with only 5.6% of works for which authority records were found. From a sample of 410 works, of which nearly half have instantiation networks, only 23 works could be said to have implicit authority control. However, many instantiation networks are made up of successive derivations that can be implicitly linked through collocation. The difficult work of explicitly linking instantiations comes with title changes, translations, and containing relations. The empirical evidence in the present study suggests that explicit control of expressions will provide the best control over instantiation networks because it is instantiations such as translations, abridgments, and adaptations that require explicit linking.

0.020196749 = product of:
  0.080786996 = sum of:
    0.080786996 = sum of:
      0.056076463 = weight(_text_:terminology in 770) [ClassicSimilarity], result of:
        0.056076463 = score(doc=770,freq=2.0), product of:
          0.24053115 = queryWeight, product of:
            5.2752647 = idf(docFreq=614, maxDocs=44218)
            0.045596033 = queryNorm
          0.23313597 = fieldWeight in 770, product of:
            1.4142135 = tf(freq=2.0), with freq of:
              2.0 = termFreq=2.0
            5.2752647 = idf(docFreq=614, maxDocs=44218)
            0.03125 = fieldNorm(doc=770)
      0.02471053 = weight(_text_:22 in 770) [ClassicSimilarity], result of:
        0.02471053 = score(doc=770,freq=2.0), product of:
          0.15966953 = queryWeight, product of:
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.045596033 = queryNorm
          0.15476047 = fieldWeight in 770, product of:
            1.4142135 = tf(freq=2.0), with freq of:
              2.0 = termFreq=2.0
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.03125 = fieldNorm(doc=770)
  0.25 = coord(1/4)

Abstract

Purpose - The purpose of the research reported here is to improve comprehension of the socially-negotiated identity of concepts in the domain of knowledge organization. Because knowledge organization as a domain has as its focus the order of concepts, both from a theoretical perspective and from an applied perspective, it is important to understand how the domain itself understands the meaning of a concept. Design/methodology/approach - The paper provides an empirical demonstration of how the domain itself understands the meaning of a concept. The paper employs content analysis to demonstrate the ways in which concepts are portrayed in KO concept maps as signs, and they are subjected to evaluative semiotic analysis as a way to understand their meaning. The frame was the entire population of formal proceedings in knowledge organization - all proceedings of the International Society for Knowledge Organization's international conferences (1990-2010) and those of the annual classification workshops of the Special Interest Group for Classification Research of the American Society for Information Science and Technology (SIG/CR). Findings - A total of 344 concept maps were analyzed. There was no discernible chronological pattern. Most concept maps were created by authors who were professors from the USA, Germany, France, or Canada. Roughly half were judged to contain semiotic content. Peirceian semiotics predominated, and tended to convey greater granularity and complexity in conceptual terminology. Nodes could be identified as anchors of conceptual clusters in the domain; the arcs were identifiable as verbal relationship indicators. Saussurian concept maps were more applied than theoretical; Peirceian concept maps had more theoretical content. Originality/value - The paper demonstrates important empirical evidence about the coherence of the domain of knowledge organization. Core values are conveyed across time through the concept maps in this population of conference papers.

Content

Vgl. auch den Beitrag: Treude, L.: Das Problem der Konzeptdefinition in der Wissensorganisation: über einen missglückten Versuch der Klärung. In: LIBREAS: Library ideas. no.22, 2013, S.xx-xx.

0.017774476 = product of:
  0.0710979 = sum of:
    0.0710979 = weight(_text_:headings in 490) [ClassicSimilarity], result of:
      0.0710979 = score(doc=490,freq=2.0), product of:
        0.22113821 = queryWeight, product of:
          4.849944 = idf(docFreq=940, maxDocs=44218)
          0.045596033 = queryNorm
        0.3215089 = fieldWeight in 490, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.849944 = idf(docFreq=940, maxDocs=44218)
          0.046875 = fieldNorm(doc=490)
  0.25 = coord(1/4)

Abstract

This paper takes for granted that archival materials will be entered into a catalog in which Library of Congress Subject Headings (LCSH) will be used to provide access. The purposes of subject access are discussed. The matter of selecting the appropriate extent of subject cataloging for archival entities is raised. Archival entities will generally require more detailed subject cataloging than published materials. A scheme for subject analysis of archival materials is presented. LCSH is described briefly, and several archival entities are analyzed and provided with LCSH access points to illustrate the methodology employed. The chief advantages of using LCSH for archival materials are its availability, and its ability to cause archival materials to collocate topically with published materials in integrated online systems.

0.014812064 = product of:
  0.059248257 = sum of:
    0.059248257 = weight(_text_:headings in 5631) [ClassicSimilarity], result of:
      0.059248257 = score(doc=5631,freq=2.0), product of:
        0.22113821 = queryWeight, product of:
          4.849944 = idf(docFreq=940, maxDocs=44218)
          0.045596033 = queryNorm
        0.2679241 = fieldWeight in 5631, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.849944 = idf(docFreq=940, maxDocs=44218)
          0.0390625 = fieldNorm(doc=5631)
  0.25 = coord(1/4)

Abstract

From a historical perspective, one could consider the modern library catalog to be that bibliographical apparatus that stretches at least from Thomas Hyde's catalog for the Bodleian Library at Oxford to the near present. Mai and other recent authors have suggested postmodern approaches to knowledge organization. In these, we realize that there is no single and unique order of knowledge or documents but rather there are many appropriate orders, all of them contextually dependent. Works (oeuvres, opera, Werke, etc.), as are musical works, literary works, works of art, etc., are and always have been key entities for information retrieval. Yet catalogs in the modern era were designed to inventory (first) and retrieve (second) specific documents. From Hyde's catalog for the Bodleian until the late twentieth century, developments are epistemologically pragmatic--reflected in the structure of catalog records, in the rules for main entry headings, and in the rules for filing in card catalogs. After 1980 developments become empirical-reflected in research conducted by Tillett, Yee, Smiraglia, Leazer, Carlyle, and Vellucci. The influence of empiricism on the pragmatic notion of "the work" has led to increased focus on the concept of the work. The challenge for the postmodern online catalog is to fully embrace the concept of "the work," finally to facilitate it as a prime objective for information retrieval.

0.014812064 = product of:
  0.059248257 = sum of:
    0.059248257 = weight(_text_:headings in 5652) [ClassicSimilarity], result of:
      0.059248257 = score(doc=5652,freq=2.0), product of:
        0.22113821 = queryWeight, product of:
          4.849944 = idf(docFreq=940, maxDocs=44218)
          0.045596033 = queryNorm
        0.2679241 = fieldWeight in 5652, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.849944 = idf(docFreq=940, maxDocs=44218)
          0.0390625 = fieldNorm(doc=5652)
  0.25 = coord(1/4)

Abstract

From a historical perspective, one could consider the modern library catalog to be that bibliographical apparatus that stretches at least from Thomas Hyde's catalog for the Bodleian Library at Oxford to the near present. Mai and other recent authors have suggested postmodern approaches to knowledge organization. In these, we realize that there is no single and unique order of knowledge or documents but rather there are many appropriate orders, all of them contextually dependent. Works (oeuvres, opera, Werke, etc.), as are musical works, literary works, works of art, etc., are and always have been key entities for information retrieval. Yet catalogs in the modern era were designed to inventory (first) and retrieve (second) specific documents. From Hyde's catalog for the Bodleian until the late twentieth century, developments are epistemologically pragmatic--reflected in the structure of catalog records, in the rules for main entry headings, and in the rules for filing in card catalogs. After 1980 developments become empirical-reflected in research conducted by Tillett, Yee, Smiraglia, Leazer, Carlyle, and Vellucci. The influence of empiricism on the pragmatic notion of "the work" has led to increased focus on the concept of the work. The challenge for the postmodern online catalog is to fully embrace the concept of "the work," finally to facilitate it as a prime objective for information retrieval.

0.014812064 = product of:
  0.059248257 = sum of:
    0.059248257 = weight(_text_:headings in 2099) [ClassicSimilarity], result of:
      0.059248257 = score(doc=2099,freq=2.0), product of:
        0.22113821 = queryWeight, product of:
          4.849944 = idf(docFreq=940, maxDocs=44218)
          0.045596033 = queryNorm
        0.2679241 = fieldWeight in 2099, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.849944 = idf(docFreq=940, maxDocs=44218)
          0.0390625 = fieldNorm(doc=2099)
  0.25 = coord(1/4)

Abstract

In KO volume 40 number 3 (2013) I included an editorial about keywords-both about the absence prior to that date of designated keywords in articles in Knowledge Organization, and about the misuse of the idea by some other journal publications (Smiraglia 2013). At the time I was chagrined to discover how little correlation there was across the formal indexing of a small set of papers from our journal, and especially to see how little correspondence there was between actual keywords appearing in the published texts, and any of the indexing supplied by either Web of Science or LISTA (Thomson Reuters' Web of ScienceT (WoS) and EBSCOHost's Library and Information Science and Technology Abstracts with Full Text (LISTA). The idea of a keyword arose in the early days of automated indexing, when it was discovered that using terms that actually occurred in full texts (or, in the earliest days, in titles and abstracts) as search "keys," usually in Boolean combinations, provided fairly precise recall in small, contextually confined text corpora. A recent Wikipedia entry (Keywords 2015) embues keywords with properties of structural reasoning, but notes that they are "key" among the most frequently occurring terms in a text corpus. The jury is still out on whether keyword retrieval is better than indexing with subject headings, but in general, keyword searches in large, unstructured text corpora (which is what we have today) are imprecise and result in large recall sets with many irrelevant hits (see the recent analysis by Gross, Taylor and Joudrey (2014). Thus it seems inadvisable to me, as editor, especially of a journal on knowledge organization, to facilitate imprecise indexing of our journal's content.

0.0046332246 = product of:
  0.018532898 = sum of:
    0.018532898 = product of:
      0.037065797 = sum of:
        0.037065797 = weight(_text_:22 in 1420) [ClassicSimilarity], result of:
          0.037065797 = score(doc=1420,freq=2.0), product of:
            0.15966953 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.045596033 = queryNorm
            0.23214069 = fieldWeight in 1420, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.046875 = fieldNorm(doc=1420)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Source

Knowledge organization in the 21st century: between historical patterns and future prospects. Proceedings of the Thirteenth International ISKO Conference 19-22 May 2014, Kraków, Poland. Ed.: Wieslaw Babik

0.0038610206 = product of:
  0.015444082 = sum of:
    0.015444082 = product of:
      0.030888164 = sum of:
        0.030888164 = weight(_text_:22 in 107) [ClassicSimilarity], result of:
          0.030888164 = score(doc=107,freq=2.0), product of:
            0.15966953 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.045596033 = queryNorm
            0.19345059 = fieldWeight in 107, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.0390625 = fieldNorm(doc=107)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Date

10. 9.2000 17:38:22

0.0038610206 = product of:
  0.015444082 = sum of:
    0.015444082 = product of:
      0.030888164 = sum of:
        0.030888164 = weight(_text_:22 in 630) [ClassicSimilarity], result of:
          0.030888164 = score(doc=630,freq=2.0), product of:
            0.15966953 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.045596033 = queryNorm
            0.19345059 = fieldWeight in 630, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.0390625 = fieldNorm(doc=630)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Date

22. 2.2013 11:09:49

0.0038610206 = product of:
  0.015444082 = sum of:
    0.015444082 = product of:
      0.030888164 = sum of:
        0.030888164 = weight(_text_:22 in 636) [ClassicSimilarity], result of:
          0.030888164 = score(doc=636,freq=2.0), product of:
            0.15966953 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.045596033 = queryNorm
            0.19345059 = fieldWeight in 636, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.0390625 = fieldNorm(doc=636)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Date

22. 2.2013 11:43:34

0.0038610206 = product of:
  0.015444082 = sum of:
    0.015444082 = product of:
      0.030888164 = sum of:
        0.030888164 = weight(_text_:22 in 1412) [ClassicSimilarity], result of:
          0.030888164 = score(doc=1412,freq=2.0), product of:
            0.15966953 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.045596033 = queryNorm
            0.19345059 = fieldWeight in 1412, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.0390625 = fieldNorm(doc=1412)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Source

Knowledge organization in the 21st century: between historical patterns and future prospects. Proceedings of the Thirteenth International ISKO Conference 19-22 May 2014, Kraków, Poland. Ed.: Wieslaw Babik

0.0019305103 = product of:
  0.007722041 = sum of:
    0.007722041 = product of:
      0.015444082 = sum of:
        0.015444082 = weight(_text_:22 in 1919) [ClassicSimilarity], result of:
          0.015444082 = score(doc=1919,freq=2.0), product of:
            0.15966953 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.045596033 = queryNorm
            0.09672529 = fieldWeight in 1919, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.01953125 = fieldNorm(doc=1919)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Date

12. 6.2008 20:18:22