025.431: The Dewey blog

EPC Meeting 132

Giles Martin took the photo of EPC at LC on November 17, 2009. From left to right: Karl Debus-López (Library of Congress, standing in for Beacher Wiggins), Jonathan Furner (UCLA), Welna van Eeden (University of South Africa), Lyn McKinney (Billings [MT] Senior High School), Deborah Rose-Lefmann (Northwestern University), Sandra Singh (University of British Columbia), EPC Vice-Chair Anne Robertson (Australian Committee on Cataloguing), EPC Chair Caroline Kent (British Library), Andrea Kappler (Evansville Vanderburgh [IN] Public Library), and David Farris (Library and Archives Canada).

Meeting 132 of the Decimal Classification Editorial Policy Committee (EPC) was held November 16-17 at the Library of Congress. EPC Chair Caroline Kent (British Library) was re-elected for a second two-year term as chair.

In Table 1 (Standard Subdivisions), EPC approved a proposal to recast T1—0286 Waste technology as T1—0286 Green technology (Environmental technology). In Table 2 (Geographic Areas, Historical Periods, Persons), EPC gave provisional approval to the revision of Asia Minor and Turkey, and requested additional literary warrant information before approving the expansion for Switzerland.

EPC approved selected updates in the following schedules: 004-006 Computer science; 320 Political science; 361-365 Social problems and services; 390 Customs, etiquette, folklore; 400 Language; 569 Fossil mammals and 599 Mammals; 636 Animal husbandry; 646 Sewing, clothing; 690 Buildings; 720 Architecture; 800 Literature and rhetoric; and various historical periods in 900. EPC also reviewed and approved the full schedules for 670 Manufacturing and 680 Manufacture of products for special uses.

Several discussion papers were considered by the committee. One discussion paper addressed current and proposed practices related to representation and use of standard subdivisions with extended or narrowed meanings, and displaced standard subdivisions. Our investigation of such subdivisions was prompted by two streams of inquiry—a set of questions posed by a group of Dewey Section classifiers preparing training materials (Michael Cantlon, Carolyn Turner, and Ruby Woodard), and our ongoing review of practices related to the representation of standard subdivisions in add tables. EPC approved the discussion paper on 281 Eastern churches for external review. We are seeking advice immediately from the Arabic, Greek, and Russian translation teams; we will also be seeking advice from other experts in Orthodox Christianity. Yvonne Jahns (Deutsche Nationalbibliothek), chair of the European DDC Users’ Group (EDUG) 340 Law Working Group, was a guest attendee at the meeting. She participated in the extensive discussion of various 340 Law discussion papers; the full 340 Law schedule will be considered at Meeting 133. Yvonne also presented “20 years SWD: German subject authority data prepared for the future.” EPC considered another set of discussion papers, including one from the EDUG 370 Education Working Group, focused on the 370 Education schedule. EPC reviewed revised proposals for updates to 372.3-372.8 Elementary education in specific subjects, and updates to levels of education in 372, 373, and 378. The full 370 Education schedule will also be considered at Meeting 133.

EPC also heard progress reports on Dewey linked data, mixed translation models, and ongoing efforts to use machine assistance in the development of the abridged edition. Thanks to the efforts of EPC member Welna van Eeden, a representation of the DDC Summaries in Afrikaans was added last week to dewey.info (Afrikaans is the tenth language represented in the service).

EPC scheduled a follow-up electronic Meeting 132A (to be held February 1-26, 2010), during which the committee will review the Editorial Rules for standard subdivisions; further work on the machine-assisted abridged model; updates to T1—0288 Maintenance and repair to address issues raised at Meeting 131; updates to the Table 2 provisions for Asia Minor, Turkey, and Switzerland; a possible expansion at 746.92 Costume to match similar provisions for clothing and accessories elsewhere; the updated history periods for Africa based on comments from outside reviewers; and minor updates proposed by the Swedish translation team. If necessary, we will split the meeting into two parts and schedule Electronic Meeting 132B (possibly as a webinar) at a later date. Meeting 133 will be held June 16-18, 2010, at OCLC headquarters in Dublin, Ohio.

Travelling Pants

On my last visit to Australia, I learned about a project of the Gosford City Library to promote youth services in public libraries in New South Wales, Australia.  It’s the Jean Genies - Travelling Pants! program (PDF file in that last link).  As that document says, “this program has 4 pairs of jeans travelling around NSW public libraries in 2009.  Participating libraries will be decorating the jeans and holding programs or displays based around the theme before sending the jeans on to the next library.”

What Dewey number would we assign to such a program?

• That for jeans (391 Costume and personal appearance being the interdisciplinary number)? – no, you are taken out of there by the reference “For a specific aspect of costume and clothing, see the aspect, e.g., stage costuming 792.026, clothing construction 646.4.”
• That for public libraries (027.4)? – no, you are taken out of that by the reference at 026-027 Specific kinds of institutions, which has hierarchical force at 027.4 Public libraries: “Class a specific operation in a specific kind of institution with the operation.”
• That for public relations (659.2)? – no, you are taken out of there by the reference at 659.29001-659.29999 Subdivisions for organizations producing specific kinds of products and services, or pursuing specific objectives: “however, for public relations in libraries, see 021.7.”

So the Dewey number for the Jean Genies project is 021.7 Promotion of libraries, archives, information centers, where the specific aspects of the use of jeans and of promotion of public libraries are in standing room, so that you cannot add standard subdivisions or the Table 2 number for New South Wales.

The New Editorial Support System: Application and Development Process

When I blogged the data side of our new Editorial Support System (ESS) a couple of months ago, I promised to follow up with blog entries on the ESS from the perspective of the application and the development process.  Funny thing, but since then, we’ve been heavily engaged in work to help get the ESS out the door (some of which is referred to below).   So this blog entry ends up being two entries rolled into one.

Short and sweet on the application side:  The editorial team has collectively decided that the ESS application should be classed in 025.4310285574 (built with 025.431 Dewey Decimal Classification, plus notation 0285 Computer applications from Table 1, plus notation 574 from 005.74, following instructions at T1—0285, to indicate that the ESS operates as a database management system).

Moving on then to the development process of the ESS:  The process that has governed the overall work has been agile software development, which matches a Relative Index term at 005.1 Programming; the class-here note at 005.1 includes software engineering.

The subclasses of 005.1 enumerate the most significant specific tasks involved in the application development process.  For example, 005.1068 Computer programming—Management (built with 005.1 Programming, plus notation 068 Management from Table 1) is the appropriate class number for the work done by the project manager to keep the flow of development tasks moving toward system launch. 

A major component of software development is the analysis of user requirements and the design of software to address user needs.  These aspects of the software development process are classed in 005.12 Software systems analysis and design, where the class-here note specifically indicates that “analysis of a user's problem preparatory to developing a software system to solve it” (i.e., user requirements) should be classed there.

Some standard subdivision concepts have been displaced, both because they have specific meanings in the context of software development and to achieve an appropriate order under 005.1.  For example, while one set of developers built the application and another set worked with the database, the editors tried to keep up with testing the various pieces (e.g., searching/browsing, spellcheck, validation, reports) as they emerged; our testing is classed in 005.14 Verification, testing, measurement, debugging, with testing and measurement having been displaced from 005.10287.

With the number of persons involved in the project, various types of communication were needed to keep everyone informed, including, for example, email (004.692 Electronic mail), videoconferences (384.556 Controlled transmission television), Sharepoint (a collaboration tool that we used primarily for sharing documents and links), and JIRA (an issue tracking system).  (For now, the latter two subjects would be classed in 006.78 Multimedia system programs, but we will need to revisit 006.7 before the whole increasingly-web-based world is classed there.)  In addition, documentation in the form of an online help component within the application will continue to keep users of the system informed; preparation of this component is classed in 005.15 Preparation of program documentation, where development of online help is mentioned in the including note.

This blog entry is dedicated (with heartfelt thanks) to the many persons who have worked diligently with us for well over a year to develop the new ESS.

EPC Meeting 131

Giles Martin took the photo of EPC at OCLC on June 12, 2009.  Standing, from left to right: Andrea Kappler (Evansville Vanderburgh [IN] Public Library), Sandra Singh (University of British Columbia), Lyn McKinney (Billings [MT] Senior High School), and Welna van Eeden (University of South Africa); seated, from left to right: Karl Debus-López (Library of Congress, standing in for Beacher Wiggins), David Farris (Library and Archives Canada), EPC Vice-Chair Anne Robertson (Australian Committee on Cataloguing), and Deborah Rose-Lefmann (Northwestern University).

It seems like forever since I’ve written something here, and I’m suppressing the urge to report chronologically on my activities since the last time you heard from me (so stay tuned for reports on three very interesting meetings [one in Houston and two in Vienna] in which I participated back in April).  Today, I’m jumping forward to last week with a quick summary of Meeting 131 of the Decimal Classification Editorial Policy Committee (EPC), which was held June 10-12 at OCLC.  The meeting was chaired by EPC Vice-Chair Anne Robertson (Australian Committee on Cataloguing).  EPC Chair Caroline Kent (British Library) and new EPC member Jonathan Furner (UCLA [and yes, former DDC assistant editor and creator of 025.431: The Dewey blog]) were unable to attend in person, but e-mailed voluminous comments that were considered by the committee during the meeting.

EPC reviewed the full version of Table 1 (Standard Subdivisions).  One provision was postponed for discussion at Meeting 132 in November 2009: the use of T1—0286 Waste technology to represent green technology.  We'll have more to say about that issue once we sort out the ripple effects throughout the schedules.  In Table 2 (Geographic Areas, Historical Periods, Persons), EPC looked mostly to the past in reviewing provisions for the ancient world in T2—3.  Some of the ancient world proposals have implications for the modern world versions of the Mediterranean and Black Sea regions, Asia Minor, and Turkey.  EPC also approved some adjustments to German administrative units proposed by colleagues at Deutsche Nationalbibliothek, plus the placement of the Barents Sea region, North Calotte, and Sápmi (Lapland).  EPC approved the full version of Table 3 (Subdivisions for the Arts, for Individual Literatures, for Specific Literary Forms), and an expansion for Hakka (Han Chinese who speak Hakka dialect) in Table 5 (Ethnic and National Groups).

EPC approved updates in the following schedules: 004-006 Computer science, 200 Religion, 300 Social sciences, 301-307 Sociology and anthropology, 320 Political science, 330 Economics, 340 Law, 400 Language, 610 Medicine and health, 780 Music, 800 Literature, and 960 History of Africa.  EPC reviewed preliminary versions of 640 Home and family management, 690 Buildings, 710 Civic and landscape art, and 720 Architecture.  On the recommendation of EPC, we plan to post the sewing and clothing sections of 646 shortly for outside comment.  EPC also reviewed a preliminary proposal for updates to sections of 372.3-372.8 Elementary education in specific subjects.  In addition to comments from members and observers on the private EPC mailing list, EPC’s deliberations were informed by reports from the working groups associated with the European DDC Users’ Group (EDUG) in the areas of law, education, archaeology, and technical issues.

EPC reviewed several research initiatives from the OCLC Office of Research and the Dewey editorial team, plus the latest draft specification for Dewey Uniform Resource Identifiers (URIs).  We’ll be back with more information on these initiatives later.

EPC scheduled a follow-up electronic Meeting 131A (to be held August 3-28, 2009), during which the committee will address use of the term “encompassing” in the DDC, plus minor issues in 400, 710, and 780.  Meeting 132 will be held November 16-17, 2009, at the Library of Congress.

One other major Dewey event took place last week—June 12 was the last day for updating Dewey data in the current Editorial Support System (ESS).  The new ESS, along with new data formats based on the MARC classification and authority formats, is scheduled to be launched on July 5 (another reason for our recent periods of silence on this blog).  Rebecca wrote a bit about the new ESS last month—watch for more reports shortly.

Data in Our New Editorial Support System

When I joined the Dewey editorial team over two years ago, planning and design efforts for a next-generation (fourth-generation, in fact) editorial support system (ESS) had already occupied the editors for a lengthy stretch of time, as evidenced by a blog entry on the subject from early 2006.  The day when those efforts will finally come to fruition are fast approaching:  we expect to be using the new system for our editorial work by July.

When the development effort had generated enough documents that we were all losing track of them, someone had the bright idea that we needed to develop a set of categories for organizing them.  What an irony, I thought.  Here we are devising a set of categories for organizing documents when we have the world’s pre-eminent classification scheme at our fingertips.  Why not give the documents DDC class numbers? 

The documents have long since been grouped into a few high-level categories, but giving them class numbers still appeals to me.  As I survey the documents, I note that they fall into three overall areas:  the data, the application, and the development process.  Today we look at the data side of the new ESS, much of which falls within 005.7 Data in computer systems.  Blog entries on the application and development process will be forthcoming in the weeks to come.

As previously blogged (here and here), an important aspect of the new ESS is that, instead of being maintained in a proprietary format, the data will be represented using the MARC classification and authority formats.  (Records for schedule and table numbers, as well as Manual records, will use the classification format, while records for Relative Index headings and mapped headings will use the authority format.)  Works on the MARC classification format are classed in 025.420285572 (built from 025.42 Classification, plus T1—0285 Computer applications, plus 572 from 005.72 Data representation, record formats, as instructed under T1—0285).  Works on the MARC authority format are classed in 025.32220285572 (built as above, except that the base is 025.3222, the comprehensive number for authority files).  Note that, although MARC is in a class-here note at 025.316 Machine-readable record formats, a scatter see reference there instructs that formats for a specific kind of record should be classed in the number for the kind, plus notation T1—0285572, as shown above.

Another data representation nicety in the new system is that diacritics and special characters will no longer be coded differently from all of our “standard” data.  Everything’s Unicode, a topic in standing room at 005.722 Character sets.  All we have to do is drag-and-drop from the Windows Character Map utility. 

Of course, adopting a new data format means needing to convert the data from the old format to the new format.  The including note at 005.72 Data preparation and representation specifically mentions conversion to machine-readable form (implicitly from non-machine-readable form), but is also the number for conversion from one machine-readable form to another.

Coming next:  the features in our new ESS application.

Subject analysis and control

As is probably obvious to many of you, 025.431—a prominent part of the name of this blog—is the number in the Dewey Decimal Classification for the Dewey Decimal Classification.  If “the most significant changes [in the January 2009 New and Changed Entries (in Word and PDF formats)] involve revisions in 025.4,” it should then come as no surprise that we have recently revised our treatment of subject analysis, which includes classification schemes and subject indexing. 

One significant change we made was to provide a number for Abstracting at 025.41; previously, abstracting was treated as an auxiliary technique of subject analysis.  Additionally, comprehensive works on abstracting and subject indexing, such as Explorations in indexing and abstracting pointing, virtue, and power, are classed in the new number.

The distinction formerly made between subject indexing and subject cataloging has been eliminated; the merged subject is now found at 025.47 Subject cataloging and cataloging.  Just as the schedule has long recognized a distinction between general classification systems (in 025.43) and subject-specific classification systems (in 025.46, using notation 001-999 to express the subject), changes now make clearer that general subject indexing and cataloging systems are classed in 025.47, while subject-specific subject indexing and cataloging systems are classed in 025.49 (again using notation 001-099 to express the subject).  For example, A women's thesaurus: An index of language used to describe and locate information by and about women is classed in 025.493054 (built with 025.49 plus 305.4 Women).  General subject cataloging and indexing systems in specific languages are classed in 025.472–025.479, using notation from Table 6 to express the language.  For example, Schlagwortnormdatei (SWD), the German general subject heading system, is classed in 025.4731 (built from 025.47 plus notation T6—31 German).

Subject indexing and cataloging may be based on either a controlled vocabulary (which remains in 025.47) or natural language (found in a thoroughly revised 025.48).  A new development has been added under 025.48 for Tagging at 025.487. 

EPC Meeting 130

The Decimal Classification Editorial Policy Committee (EPC)* held Meeting 130 at the Library of Congress November 18-19.  The meeting was chaired by Caroline Kent (British Library); Anne Robertson (Australian Committee on Cataloguing) was re-elected vice-chair for another two-year term. 

Prior to Meeting 130, EPC held an electronic pre-meeting, and approved updates to historical periods; modernization of the treatment of radio programs, motion pictures, and television programs regardless of method of distribution; and updates to 510 Mathematics.  All but the last will be distributed to users in early 2009.  EPC also approved, with minor modification, the long-planned updates to groups of people in Table 1 and 305-306—we plan to release the updates in mid-2009.

EPC discussed two sets of updates to Table 2 (Geographic Areas, Historical Periods, Persons)—the conversion of the relocation information and corresponding changes in the reorganization of Scotland, England, and Wales presented in DDC 22; and further updates to the area table for Indonesia.  Both sets of changes require additional consultation with in-country groups.   EPC approved a final round of updates to Table 4 (Subdivisions of Individual Languages and Language Families, for Individual Literatures, for Specific Literary Forms), Table 5 (Ethnic and National Groups), and Table 6 (Languages)—most of the updates will be distributed to users in mid-2009.

In the schedules, EPC approved updates in the following areas: 001 Knowledge; 002 The book; 150 Psychology; 330 Economics; 390 Customs, etiquette, folklore; 400 Language (including the relocation of computational linguistics); 610 Medicine and health; 640 Home and family management; 650 Management and auxiliary services; 660 Chemical engineering & related technologies; 730 Sculpture and related arts; 740 Graphic arts and decorative arts; 750 Painting and paintings; 760 Printmaking and prints; 770.23 Photography as a profession, occupation, hobby; and 790 Recreational and performing arts. EPC also approved updates to the current historical periods in 947 Russia and 968 South Africa. Updates in many of these areas will be released during 2009.  EPC reviewed reports on work under way in 340 Law, and discussion papers on 370 Education and 780 Music. We will be seeking advice from users on a number of topics discussed at Meeting 130, including subjects in primary school curricula, language variations, meals, clothing, musical styles, and literary periods for Galician.

EPC reviewed several research initiatives from the OCLC Office of Research and the Dewey editorial team, plus draft guidelines for classifying photographs for the World Digital Library project—we’ll blog about these activities separately in the coming weeks.

On November 18, EPC held a special dinner in honor of assistant editor Winton Matthews and outgoing member Arlene Taylor.  Winton has indicated that he plans to retire in early 2009—we’ll share the resolution passed by EPC in his honor later on the occasion of his retirement.  Arlene Taylor served on EPC 2000–2008.  The committee's resolution honoring Arlene reads in part:

Whereas Arlene Taylor . . . participated in the development and review of Edition 22 (2003) and Abridged Edition 14 (2004) of the Dewey Decimal Classification, and in the publication of ongoing updates in WebDewey and Abridged WebDewey; . . . focused in particular on the needs of students and teachers of the DDC in her comments; . . . contributed “Teaching the Dewey Decimal Classification System” to a special issue of Cataloging & Classification Quarterly devoted to the DDC; . . . presented “Research in Knowledge Organization” at Meeting 116; . . . participated in the EPC planning retreat, “The Future of the Dewey Decimal Classification,” March 15-18, 2004, at OCLC;  . . . paid special attention to the treatment of groups of people, and in particular prompted improvements to the caption for people in early adulthood and the terminology for children of unmarried parents (the latter formerly “children born out of wedlock”);  . . . Be it resolved that the members of the Decimal Classification Editorial Policy Committee here assembled at the One Hundred Thirtieth Meeting of said committee, the DDC editors, and staff at the Library of Congress Dewey Section and OCLC express to Arlene Taylor their deep gratitude and appreciation for her service and commitment to the Committee and the Classification, their regret that she will no longer be one of their number, and their wishes for her continued success and happiness now that she has retired from the Committee.

*Photo at top of EPC at LC on November 18, 2008—from left to right: Beacher Wiggins (Library of Congress), Vice Chair Anne Robertson (Australian Committee on Cataloguing), Welna van Eeden (University of South Africa), Sandra Singh (University of British Columbia), Lyn McKinney (Billings [MT] Senior High School), EPC Chair Caroline Kent (British Library), Andrea Kappler (Evansville Vanderburgh [IN] Public Library), Arlene Taylor (University of Pittsburgh SIS, retired), David Farris (Library and Archives Canada), and Deborah Rose-Lefmann (Northwestern University).

Where does [name your favorite computer-related advance] go?

The November 2008 New and Changed Entries (in Word and PDF formats) present numerous changes in computer science and related areas and have been a long time in the making.  Every EPC meeting of the past two years has considered proposed changes to 004-006.  So, in a nutshell, here’s what’s new:   We have provided classes for handheld computing devices, additional network architectures, specific types of databases, and specific types of multimedia systems and have given instructions on the classing of numerous other current computing topics.  We have distinguished between the Internet (with interdisciplinary works at 004.678) and the World Wide Web (with interdisciplinary works at a newly expanded 025.042, which includes subdivisions for search engines and the semantic web, among other topics).  We have provided guidance for classing multifunctional digital devices, updated terminology and examples throughout 004–006, added a centered entry at 004–006, and eliminated the 004–006 Manual entry. 

That’s a lot to assimilate, so let’s look at some of those changes more closely, beginning with the updated terminology we are using to refer to specific types of computers.  The class 004.12 Mainframe computers is now joined by 004.14 Midrange computers, 004.16 Personal computers, and the newly recognized 004.167 Handheld computing devices.   At the same time, lots of digital devices are now being created with functionality that extends beyond basic computing capability.  We have added several notes that give guidance for treating such multifunctional digital devices.  The basic instruction is to class a multifunctional digital device with its predominant function, but to class works that focus on a specific function other than the predominant function with the specific function; we also include a preference rule for devices with no predominant function.  You can check out versions of the multifunctional device note at 004.11-004.16 Digital computers, 621.383-621.389 Specific communication systems, and  621.3911-621.3916 Digital computers; see if you agree with our sense of the predominant function of PDAs and camera phones, for example.

The 005.7 Data in computer systems area is also witnessing a number of changes.  Data mining has been relocated to 006.312.  Expansions have been added at 005.743 Database design and architecture and 005.745 Data warehousing.  A number of database types are being either newly recognized or relocated, including logic databases, deductive databases, web databases, temporal databases, spatial databases, and constraint databases.

Other computing concepts that now have a specified place within the classification scheme include cluster computing, digital rights management, electronic/digital signatures, embedded computer systems, grid computing, multi-agent systems, peer-to-peer architecture, portable document software, service-oriented architecture, specific types of multimedia systems (e.g., wikis, blogs, online social networks), style sheet languages, web servers, and web services; additional concepts occur as Relative Index terms without being present in the new and changed entries document.  Classifiers working with computer science materials will probably find immediate use for the corresponding class numbers.

Following literary warrant, the DDC had not previously distinguished clearly between the Internet, whose home lies in the networking world, and the World Wide Web, the set of hyperlinked documents that can be accessed via the Internet.  While the networking aspects of the WWW are still classified with 004.678 Internet, interdisciplinary works on the WWW have been relocated to a new subdivision under 025.04 Information storage and retrieval systems, of which the World Wide Web is a remarkable example; class-here concepts at 025.042 include digital libraries and Internet literacy.   This new class has, in turn, further subdivisions:  025.0422 Web sites (where directories of Web sites and works on portals would be classed), 025.0425 (Web) Search and retrieval, 025.04252 Search engines, and 025.0427 Semantic web.

In keeping with our usual practice, all the class numbers given above come from the full edition.  When we put together our monthly posting of new and changed entries, all of the changes affect the full edition; often only a relatively few corresponding changes occur in the abridged edition.  This is because many updates to the scheme occur at a level of granularity that is more specific than the abridged edition.  It is telling that these updates to computer science and related areas follow a much different pattern.  Fully three-eighths of the document involves updates to the abridged edition.  This is because institutions using the abridged edition also have need for classes at the level of specificity, for example, of specific handheld computing devices (004 covers hardware not only from a  technological perspective, but also from the perspective of selection and use) and web search engines.  Many of the terminology changes and notes introduced in the full edition have also occasioned corresponding changes in the abridged edition.

Anyone want to wager how soon before we will provide another mammoth update to computer science and related areas of the schedules?

Computational Linguistics

Ever have difficulty deciding whether material should be classed in 006.35 Natural language processing or in 410.285 Computational linguistics? (It would seem so, since many works have been classed in both numbers.) Since we have also found it difficult to distinguish clearly between the two numbers, we decided to take advantage of a recent major gathering of computational linguists at ACL-08: HLT (ACL = Association of Computational Linguistics; HLT = Human Language Technology) to get their feedback on the treatment of computational linguistics and natural language processing in the DDC.

According to LCSH, the intended distinction between computational linguistics and natural language processing is that Computational linguistics (LCC: P98-98.5; DDC: 410.285; 467 WorldCat records) is for “works on the application of computers in processing and analyzing language,” whereas Natural language processing (Computer science) (LCC: QA76.9.N38; DDC: 006.35; 365 WorldCat records) is for “works on the computer processing of natural language for the purpose of enabling humans to interact with computers in natural language.” Dewey currently adopts this same distinction. The distinction, however, does not reflect current thought.

Computational linguists at ACL-08 tended to agree that “natural language processing” (NLP) and “computational linguistics” (CL) mean pretty much the same thing (or, if different, that the meaning of natural language processing is encompassed within the meaning of computational linguistics). That makes our decision to merge natural language processing and computational linguistics relatively easy.

Deciding where the merged subject should go is much harder. On the one hand, there was agreement that the relative contribution of computer science to computational linguistics is greater than the contribution of linguistics. Similarly, there was agreement that a background in computer science is more essential for computational linguistics than a background in linguistics. Further, computer scientists are much more likely than linguists to embrace computational linguistics as part of their field. From these statements, classing the merged natural language processing / computational linguistics in 006 might seem a no-brainer. On the other hand, however, some of the observations shared suggest that the situation may not be so cut-and-dry: Computational linguistics really belongs in linguistics, but linguists don’t realize it yet. Computer scientists sometimes change the field they apply their skills to (that is, a junior computational linguist might not continue to work in computational linguistics). As a supervisor, you get better results teaching computer science to a linguist than teaching linguistics to a computer scientist.

There are at least two distinctions made in computational linguistics that should inform our decision. The first is a distinction between symbolic and statistical approaches to computational linguistics, the former emphasizing linguistics-based representations of natural language, the latter emphasizing quantitative representations of natural language. Many symbolic approaches could be classed comfortably within linguistics; however, the same could be said of statistical approaches considerably less often.

A second distinction is made in computational linguistics between tasks and applications: Computational linguistics tasks (e.g., part-of-speech tagging, parsing, word sense disambiguation, text segmentation) rely, wholly or in part, on specific properties of language in their processing and analysis and may be combined to form applications of extrinsic value; computational linguistics applications (e.g., question answering, information retrieval, automatic abstracting, machine translation) are comprised of components addressing multiple linguistic properties and are of extrinsic value. Again, one end of our spectrum (in this case, tasks) is much more like linguistics than the other (in this case, applications—unless the application is itself in linguistics, e.g., translation), but all applications carry out some number of tasks.

It appears to us that the best solution would be to drop the distinction between natural language processing and computational linguistics by relocating comprehensive and interdisciplinary works on computational linguistics from 410.285 to 006.35. We would continue to use 410.285 in its broad meaning as computer applications in linguistics; for example, the SIL (initially known as the Summer Institute of Linguistics) software catalog, which supports the work of field linguists, would be classed in 410.28553. This catalog includes, inter alia, fonts, a concordance generator, a tool for drawing syntax trees, interlinear text editors, a Spanish verb conjugator, and a program for learning the International Phonetic Alphabet.

We would love to hear your reactions to this solution. (Or if you have another solution that accounts for the interdisciplinary nature of computational linguistics, we would love to hear that, too.) For best consideration, please either comment on this blog or send email to dewey@loc.gov  by August 15.

Museums

The Washington Post recently published a short evaluative guide to some of the newest museums in Washington, DC, entitled “Worth the Price of Admission?”  The question arises especially because the museums discussed in the article charge admission, while many museums in Washington are free.  The museum themes include crime and punishment, international espionage and spycraft, news and journalism, and Madame Tussaud’s wax models of famous people.

Works that focus narrowly on museum finance are classed in 069.0681 Financial management of museums (built with 069 Museum science plus T1—0681 Organization and financial management), e.g., Data by Discipline: 2003 Museum Financial Information.

Works about museums in a specific area that cover multiple disciplines are classed in 069.09 Historical, geographic, persons treatment plus Table 2 notation for the area, e.g., Britain's Best Museums and Galleries 069.0941 Museums of British Isles, Museums of Chicago: A Guide to Residents & Visitors 069.0977311 Museums of Chicago.  The instructions that permit addition of notation from Table 2 are found in Table 1 at T1—093-T1—099 Treatment by specific continents, countries, localities; extraterrestrial worlds.  These instructions apply because the 09 in 069.09 Historical, geographic, persons treatment is the standard subdivision T1—09 Historical, geographic, persons treatment, as the caption indicates.  The standard subdivision is printed in the schedule to carry special instructions, but the standard instructions still apply.

Among the special instructions at 069.09 Historical, geographic, persons treatment are the following:

Class here specific museums not limited to a specific discipline or subject

Class museums devoted to specific disciplines and subjects with the discipline or subject, plus notation T1—074 from Table 1, e.g., natural history museums 508.074

The class-here note indicates that comprehensive works on a specific museum with general scope, like the British Museum, are classed in 069.09 or its subdivisions, e.g., The British Museum: 250 Years, classed in 069.0942142 Museums of Camden London Borough.  Similarly, comprehensive works on the Smithsonian Institution, a museum complex with 19 museums, are classed in 069.09753 Museums of Washington, DC, e.g., The Smithsonian Institution.

The scatter note under 069.09 indicates that museums that focus on a specific discipline or subject are classed with the discipline or subject, plus notation T1—074 Museums, collections, exhibits from Table 1.  For example, Official Guide to the Smithsonian National Museum of Natural History is classed in 508.074753 Natural history museums in Washington, DC (built with 508 Natural history plus T1—074 Museums, collections, exhibits plus T2—753 Washington, District of Columbia, following instructions at T1—074 Museums, collections, exhibits).  Incidentally, that number is correct because the Smithsonian National Museum of Natural History is not limited in its coverage to the natural history of any specific area; if it were limited, for example, to the natural history of the United States, the number would be 508.73074753 Museums of U.S. natural history in Washington, DC (built with 508 Natural history plus T2—73 United States, following instructions at 508.4-508.9 Treatment by specific continents, countries, localities in the modern world, plus T1—074 Museums, collections, exhibits plus T2—753 Washington, District of Columbia).