22 Designing of a Depth Classification Schedule: Phases and Steps

Dr M P Satija

Structure:

0. Introduction

0.1 Methodology:

1.Qualities of a Good Classification System

1.1 Steps in Designing

1.2 Level of the Users

1.3 Type of Scheme: Choice of a Model

1.4 Planning the work

1.5 Subject Knowledge

2.Work in the Idea Plane

2.1 Collection of Terms

2.2 Categories and Facets

2.3 Formation of Arrays and Chains

2.3.1 Principles of Helpful Sequence

2.3.1.1 Chains Formation

2.4 Common Isolates (CIs) or Standard Sub-divisions

3.Verbal Plane

4.Notational Plane

4.1 Expressive and ordinal notation

4.2 Citation of facets and the principle of inversion

5.Computer held classification schemes

5.1 Software for designing classification schemes

6.Practical Example: Designing a classification for machine bolts 6.0 Idea Plane

6.1 Characteristics

6.2 Facets

6.3 Arrays and Chains

6.4 Helpful Sequence

6.5 Citation Order

6.6 Verbal Plane: Standardisation of Terms

6.7 Notational Plane

6.8 Rigidly faceted classification

6.9 Freely faceted classification

6.10 Using mixed notation

6.11 Some examples

7. Glossary

8. Bibliography and Further Readings

9. Test questions

1. Introduction

Subject access to information is the central facility of any modern library and information retrieval system. Therefore library personnel must have the requisite expertise in subject cataloguing, classification design and vocabulary control. LIS professional may often be called uponto design subject access or ware house items storage and retrieval systems. We have to learn how to designsuch subject access systems, or knowledge organization tools, as these are called.

0.1 Methodology:

A library classification scheme should be based on some objective methodology. Ranganathan suggested that there should be some summer school for this work. In 1962 DRTCinitiated intensive research in classification design and introduced the topic in the curricula. How to design depth classification schedule was a centraltopic in the curricula of 1960s. Every DRTC student had to prepare one depth schedule for avery specificnarrow subject. Ranganathan combined intuitive, speculativeand rational methods with pragmatic method based on experience.Library Classification Schemes are be of two types:-

1.     General classification

2.     Special subject classification

Construction of classification schedule is a very difficult job; it is not a child’s play, rather it is a very tedious job.Designing a general classification involvesa large number of classificationists and subject specialists. It is not the work of a single person.Therefore, since long no more new general classifications are being designed. We will confine ourselves to construction of depth schedules for special subjects based on facet analysis. It is a skilled,intellectualand time taking job. Despite being a librarian, it is not possible for one to be an “expert” in all subjects. Designing the schedule is not a work of librarians or subject experts alone. It is a team work. It is a collaborative work of subject expert, teachers, librarians and users, too. Of course, apart from knowledge of subject, it needs lot of resources in terms of in subject expertise, money, time and above all patience. Before undertaking this onerous task we must assess our resources of men and material,and havea clearvision of its utility.

1. Qualities of a Good Classification System:

Classification is a tool for knowledge organization. Itsessential and desirable qualities are:

It must be useful to the majority of the library users.

It should be a mapping of the universe of subjects in that area of knowledge.

Its structure and order of subjects should be contemporary.

It should be user oriented and need based.

It shouldbe based on literary warrant.

It must be machine readable.

It should have its system of Book Numbers.

Notation should be user friendly and quickly hospitable to new subjects in future.

It must have provision for phase relations.

If it is a special subject classification then for other subjects it must be sync with some standard general system of classification. For example, a law library may also stock books on Public administration, Political science, Sociology, and many more. It is rightly said an engineering library may also have books on fiction.Therefore for other subjects it must follow some standard general system.

1.1 Steps in Designing

According to Ranganathan entire work is to be divided into three phases, which he calls Planes:Idea, Verbal and Notational planes. However the entire work in these planes can be carried out by the following steps:

Settle the definition and scope of the area of knowledge to be covered by the classification system.

dentify the sources of terms

Extract concepts and terms by applying the canons of characteristics.

Assign terms to pre-defined and named Categories.

Divide terms under categories into facets and isolates.

Arrange facets and isolates into Arrays and Chains.

Make arrays by applying the canons for arrays.

Put isolates in the hierarchical order using principles of Chain formation.

Decide the citation order of facets, or use PMEST formula.

Dress all the concepts in current and standardized terminology,taking care of synonyms and homonyms.

Assign notation which is expressive, its order of digits is known, and has the capacity and flexibility for future subjects.

Prepare the alphabetical index of terms in schedules and tables using some international standard.

Write an introduction to the system.

Pre-test the classificationunder some controlled conditions.

Circulate it for opinions of prospective users

Take due notice of the genuine feedback and constructive criticism.

Set up policy and machinery for its regular revision.

Prepare a directory or network of its users.

Conduct a workshop for its users to highlight its features and explain its nuances

1.2   Level of the Users

Readers are of primary importance as classification is not an end; it is only a mean. It is meant to serve the readers. Thus scheme must be reader oriented and user friendly. It will be a useless scheme if theuserslevel and their approaches to subject access in libraries are not known. While preparing a scheme, two factors– needs of users and level of users must bedetermined.

1.3 Types of Schemes: Choice of a Model

What kind of classification should be constructed – faceted or enumerative scheme, that is the first question. Now-a-days only option leftis for faceted scheme if we want to constructa Depth classification for the dynamic world of knowledge.A faceted scheme has been described as the basis of all information storage and retrieval. If world is static,and there is nothing to add in future, then we can opt for an enumerative model. But today no world is static. The universe of knowledge is growing very fast; rather its growth is unprecedented.

1.4 Planning the work

We have to properlystudy thesubject in and out from definition, scope to latest trends. For this, first of all its definition is to be settled from standard dictionaries and also subject dictionaries, e.g., Dictionary of Physics, Social sciences, etc. After that we have to decidethe scope or boundary of the subject. As we know boundaries are never agreed to among scholars. They are debatable. They go on changing from time to time, and are different in different systems. There was a time when science was called natural philosophy. Is mechanics a part of mathematics or physics? Such questions must be settledadministratively. It is imperative to draw a clear boundary. We should consult experts before drawing the boundaries of a subject. Once settled,then the Canon of Enumeration must be followed in defining and displaying it.

1.5 Subject Knowledge

It is most essential to have or gain synoptic knowledge of the subject proper and its environment. For background knowledge of a subject, we should read encyclopedias, text books, latest literature of the subject in all media, andbe aware of what is being published in journals and indexed in alerting services. We should keep our eyes on topics which are being discussed in that discipline by reading annual advances in the subject. Help of a subject expert is indispensable at this stage. Scheme should be based on literary warrant and for this literature survey should be done. For literature survey, we have to consult good abstracting journal not one or two issues but issues of the last ten years should be studied. If no abstract is available on that subject then we should study six to ten major journals of last ten years.

2. Work in the Idea Plane

Idea plane involves brain work. It is foundation of the entire structure. Therefore,It is most important of all other planes. S.R. Ranganathan formulated canons of characteristic for breaking of a subject into its constituent concepts.

2.1 Collection of Terms

Keywords from title, abstracts or from the text of the article or key words given by authors should be noted in context in a computer file.Concepts/terms should also be collected from the feature headings of the article. Long list of key terms can easily be extracted from databases.Users queries are another source of candidate terms for inclusion.

2.2  Categories and facets

We have to break the subject into smaller terms on the basis of characteristics. This characteristic is a hammer which breaks the subject area intokey facets. A characteristic is acriteria, or basis of division on which a subject is broken into its constituent concepts. Choosingthe right characteristic is a difficult work.It is intrinsically related to the ultimate purpose of classification. Ranganthan’s canons of characteristics are of great help at this stage. All the key terms should be sorted at two levels. First into broader categories and then into facets within each category. Ranganthan has postulated Five Fundamental-Categories of PMEST,while CRG, London and later BC-2 of Jack Mills (1918-2010) has postulated many more categories which they call facets. Ranganthan’s categories are too broader—in fact there are only three of them,namelyPersonality, Matter and Energy, whereas Space and Time are common categories. Further the concept of Rounds and Levels may be confusing to most of us. Categories formulated by BC-2 are quite in fashion. These in sequence are: Things— Kinds—Parts—Properties–Materials—Processes—Operations—Patient—Product— Byproduct—Agent—Place—Time. Being more explicit these canbe easily followed.

2.3 Formation of Arrays and Chains

After assigning basic concepts to categories and facets sort them into arrays and chains. Making Arrays and Chains is not an easy task. Rather it is the real job. If we doit successfully then we have done the corework of designing a classification. An array is a group of entities of equal rank. Arrays should be exhaustive of its entities and an entity should be assigned to one and only one array(Exclusive).Above all the arrangement of entities in an array should be systematic(logical) and helpful .

2.3.1 Principles of Helpful Sequence:

For this, Ranganathan formulatedCanon of Helpful sequence for arrangement of entities in an order helpful to the majority of users.Principles of Helpful Sequence are as follows:

1.     Principle of Later-in-time/Evolution.

2.     Principle of Spatial contiguity.

3.     Principle of Increasing quantity/Complexity

4.     Principle of Literary warrant/Popularity

5.     Principle of Alphabetical sequence

2.3.2 Chains Formation

Some entities will be arranged hierarchically in chains to show relations of subordination and super-ordination.A chain should be modulated.

2.4 Common Isolates (CIs) or standard sub-divisions:

These are general recurring documentaryaspects which hold good across the spectrum.It denotes only non-subject aspects of the documents occurring again and again in all classes and theirnames usually remain the same. Even in practice, if we are making a depth schedule, the list of CIs canbe borrowed from a standard and well developed classification schemes. Common isolates are mostly the same in UDC, DDC and CC only notation differs. These common isolates can be easily beadapted to a new system. In addition there should be a provision phase and intra-facet relations.

3. Verbal Plane

Language is essential to thought; and words are expressions of ideas.However, meaning of a word is not a fixed point but an area of variable dimensions. As schedule makers, we have to be careful about terminology. In this plane, one has to choose the standardized terms or standard terminology to be given to concepts in the idea Plane.For this, a latest subject dictionary or glossary should be consulted;alternative source are the experts.

• Terms should not be opinionated. Terminology should be neutral; it should not be critical, or evocative. It should be culture sensitive.
• Terms should be current, standardized and used consistently.
• Terms should be written in the context of its upper link which means we should apply the Canon of Context.
• Above all it should be free of homonyms and synonyms

4. Notational Plane

Classification starts with ideas not with notation. Idea is head and notation is its cap. Notation should be assigned in the last to the standardized terms arranged in arrays and chains. While denoting to the standard terms following canons must be applied:

1. Always usemoderately mixed notation. Be not be tempted by the simplicity of pure notation. Pure notation will be ultimately harmful and will not let the schedules grow. On the other hand a highly mixed notation, like the CC, may be detrimental in its use.

2. Notation should be hospitable. It should be expandable internally and externally.

Always keep the gaps. That is keep some digits in reserve at every level.Designate empty, emptying and sectorizing digits. Major divisions should be A/Z which may be sub-divided by numerals. Use numeralsas decimal numbers.

3. Notation should express hierarchy to some extent, though not to the dead end.

4. Allocation of notation should be proportionate according to the subject based on the principle of literary warrant, e.g. CC proportionates half of the notational cake to sciences. That will makethe notation brief.Here is an example of a hierarchical and non expressive notation:-

4.1 Expressive and ordinal notation:-Subject Expressive Notatio

Asia	A	A
South Asia	A5	A5
India	A55	A8
North India	A551	A81
Uttar Pradesh	A5511	A82
Threats toEnvironment	Z	Z
Threat of Depredation	Z1	Z1
Depletion of Resources	Z11	Z2
Deforestation	Z111	Z3
Class Number	Z111A5511	Z3A82

*Gaps in notation for future expansion may be noted.

It is presumed, more expressive a notation less hospitable it is.

4.2 Citation of facets and the principle of inversion:

The following example has been taken from Aida Slavic (2008)

Time	A1
Space	B1
Agent	C1
Product	D1
…
Thing	M1

Above facets are in the abstract to concrete order but their synthesized class number will be 33M1D1C1B1A1. These facets from M1 to A1 are in the concrete to abstract order. However by the use of inversion principle, their arrangement on the shelf will be in the following order which is from abstract to concrete:

33A1

33B1A1

33D1

33M1

33M1B1A1

33M1C1B1A1

33M1D1

33M1D1C1

33M1D1C1B1A1

5. Computer held classification schemes:

Apart from in a faceted format now all the classification systems are developed in machine readable databases. Dr. Aida Slavic (2008) has listed the following advantages of online classifications:

·        searching and browsing of classification notation is easy

·        facilitates searching notation through associated verbal expression

·        sort and display of entries in many modes is possible

·        there is automatic tracing of hierarchical and associative links and relations

·        tracing of rules to the area of their application is quick

·        navigation between tables, facets and subject areas is easy

·        tracing historical data through a scheme’s lifespan (Historical notes) is possible

·        various outputs, exports and byproducts are made

·        identification of classes independent of notation is possible

·        it is easy to maintain, edit and publish as printed editions of various sizes

Concise UNIMARC Format for Classification Data was published in its draft form in 2000. This format is still awaiting official completion.

5.1       Software for designing classification schemes:

MS Excel and MS Word have some limitations in designing classification. There aresoftware which are specific to the task. L. Varendroff describes some software useful for such purposes. M Gaikaiwari describes the use of acomputer program named SRR to assist librarians in developing, operating and maintaining a freely faceted classification system. ESS 4.0 used by the DDC and MARF of the UDCare highly developed software for designing classification databases.

5.2     Drafting the Schedule

While drafting the schedule, we must have opinions of other librariansand users. This is testing of the schedule to getthe reaction of the users. Only then we will come to know the defects and shortcomings.Worthwhilefeedback from all concerned should be incorporated before finalizing the schedule. Being a professional classificationist, always decide two things:

1. Revision policy and

2.Revision machinery.

Introduction should be written at the end of the task, and must explain the reason for designing it,and mentionwhom it aims at. It must highlight its special features, scope, uses, and how to operate it with amplepractical examples in a simple language.

5.3 Revision

Designing a classification system is one thing;maintainingit is quite another. Lack of currency in terminology and coverage will lead to early disuse and untimely demise. Always keep the system updated and current both in coverage and terminology. Always make use of advances in classification theory in revising the system. There should be a well-defined policy and a permanent machinery for its continuous revision.Authorised revisions should be communicated regularly.Newsletter is a device to keep in touch with the community of users. Changes may be communicated in the newsletter orposted on the website, or blog which may get instant feedback from the reader. Purpose of revision and updating is just like after-sale-service of a machine. Librarians are scared of drastic revisions.

6. Practical Example: Designing a classification for machine bolts

It is reasonably and hopefully believed that the individual librarians will be called upon to design classifications for warehouses to store and retrieve industrial goods for their inventory control. Warehouse management and inventory control is another area of work for librarians. The principles of designing library classification as given by S R Ranganathan are visibly and effectively applicable here.

Let us take an example of designing a faceted classification for machine bolts: A threaded thick pin with a nut used as a fastener.

As usual the entire work can be divided into Idea Plane, Verbal Plane and the Notational Plane.

6.1 Characteristics:

The four Certain and Relevant characteristics associated with a machine bolt are:

Thread Size, Head Shape, Finish, Material

6.2     Facets:

These will work as facets : On going through the literature or the stock the various isolates under different facets are:

6.3 Arrays and Chains:

Under each facet the isolates have to be arranged into arrays and chains:Polished Isolates in Size and Head shape facetsare of equal rank, while those of Finish and Material are to be arranged in arrays and chains.

6.4 Helpful Sequence:

Now we will have to apply the principles of Helpful Sequence to arrange the entities in arrays and chains in some meaningful sequence.

In case of Thread Size the principle of Increasing Quantity is applicable; in Shape facet we apply alphabetical device; and the principle of Favoured Category (literary warrant) is applicable in case of Finish and Material facets. So we get the following rearranged sequence:

6.5 Citation Order:

Now we have to decide the Citation Order of facets, that is, the facet formula in Ranganathan’s terminology. The sequence of facets in a compound subject is of utmost importance.It displays syntactical relations. The facets should be arranged in the sequence of their decreasing importance. The facets arranged in their decreasing importance are:

Size / Material / Shape / Finish 6.6 Standardisation of Terms:

In the Verbal Plane we will review our terminology used to give neutral and current terms to the concepts in the arrays and chains. In the literature the other term used are screws, machine bolts and bolts. We will use the term Bolts for uniformity and consistency. Canon of Context has already been followed along with the canon of helpful sequence.

6.12  Notational Plane

Now we have to assign isolate numbers to all the ranked isolates under different facets. Desirable qualities of a notation are : Simplicity, brevity, mnemonicity. Essential quality is hospitality, if the universe of entities to be classified is a growing one. There are various possible models. The following model, which produces highly simple class numbers, is applicable to a non-expressive rigidly faceted classification where all the facets are always present in a compound subject.

6.8 Rigidly faceted classification:

Similarly, uniformly four digit unique code numbers can be created for any type of machine bolts. Problem arises if any of the facet is missing, e.g. Chromium plated steel bolts of 4 BA size 323 (here shape is missing). This number could also be for Chrome plated Eye head Alloy bolts (size missing). One solution is to use 0 (Zero) for the absent facet. Then the first number will be 3203 and the other will be 0323.

6.9 Freely faceted classification:

Such a classification is only useful for inventory control for a very small store. But in a freely faceted classification it is not necessary that all the facets be present. For such a situation we will have to use facet indicators. Let us fix that the indicator

For material be	;	Semicolon
For Head Shape be	:	Colon
And For Finish be	.	Dot

And presume that Size facet is always present

Let us say our title is:

Round Bolts of 2 BA size with unfinished surface

2 :6 . 7

Steel Hexagonal Bolts of 2 BA Thread Size

2; 2 :3

Unfinished surface Round Head Brass Bolts of 8 BA size

5; 5 :6 . 7

We have to fix the ordinal value of indicator digits. Using the principle of Inversion the ordinal value in their increasing order is . : ;

Use of indicator digits for ordinary users may be difficult – though for a computerized system it poses no difficulty:

6.10  Using mixed notation:

If we use highly mixed notation we can use different species of digits for different facets- so that the kind of specie also acts as a facet indicator.

 

(Here head shape facet has been arranged by the principle of favoured category)

6.11  Some examples:

Hexagonal Bolts of 0 BA Size (Material and finish facets are missing)

A b

Plastic Bolts of unfinished surface (Size and head shape missing)

6 0 5

Hexagonal Unfinished Bolts (Size and material missing)

b 0 5

Unfinished 5 BA Bolts (Material and head shape missing)

D 0 5

Plastic Bolts of 2 BA size (Head shape and finish missing)

B 6

Plastic Bolts of 2 BA size with unfinished surface (Head shape missing)

B 6 0 5

Hexagonal Steel Bolts of 14 BA size with Polishedsurface

J 11 b 0 1

By the principle of inversion the ordinal value of these digits in the increasing order will be

 01-09/a-z/1-9/A-Z

Therefore, Sequence of these class numbers in general to specific order will be

b05, 605, Ab, B6, B605, D05, J11b01

7. Glossary:

Category: A broader class of subjects mostly taken synonymous with facets in the modern classification theory. But in Ranganathan’s CC facets occur within a category.

Citation order: A logical sequence of facets in a compound subject formulated mainly by dependency principle or specific to general order. PMEST is a general and broader citation order. It depicts syntactical relations among facets.

Common auxiliaries: Known by various names in different classification systems, these are non-subject recurring aspects of a document which have the same name and notation and are applicable across the classes. In CC, these are called common isolates, in UDC common auxiliaries and in the DDC standard subdivisions.

Facet analysis: Breaking down a subject into its constituent concepts and assigning them to predetermined categories: or, reversely assigning each of the isolatedconcept to any of the predetermined facets. In essence it is subject analysis.

Helpful Sequence: Any order of subjects which is considered useful to the majority of library users. This order prevails mostly in the main classes, their branches and further subdivisions of equal rank in an array. Ranganathan gave 8 principles for helpful arrangement of entities in an array.

Inversion principle: The order of facets in the facet formula is reverse of their shelving order to achieve general to specific order in filing documents.

Rigidly faceted classification: A faceted classification in which every facet has to be there, or is indicated by a dummy digits to avoid cluttering of facets.

Schedules: The terra firma of library classification system which lists classes and their subdivisions along with their notation and notes.

Semantic relations: Subject relations exhibited by hierarchy or related terms. Semantic relations are a priori.

Syntactical relations: A sort of grammatical, a posteriori relations in a compound subject. These are achieved by an imposed citation order on the facets of a compound subject.

8. Bibliography and Further Readings

• Buchannan, Brian. 1979. Theory of Library Classification. London: Bingley, pp.105-122. Broughton,V. and Slavic,A.2007.Building a faceted classification for the humanities:principles and procedures.Jl. of Documentation 63(5):727-754
• Curwen, Anthony Ca. 1978. Revision of Classification schemes: policies and practices. J. Lib. Ship 10 (1) : 19-38.
• Gaikaiwari, M.2002.An Interactive Application for Faceted Classification Systems. Information Studies 8(1):43-56
• Godert, W.1991.Facet classification inOnlineRetrieval. Int. Classification.18(2):98-105 Gopinath, M. A.1986.Construction of Depth Version of Colon Classification.New Delhi: Wiley Eastern, 163p.
• Hjorland, Birger.2013. Facet analysis: The logical approach to knowledge organization.
• Info Processing and Mgmt.49:545-557.
• Hunter, Eric J.2009.Classifcation Made Simple, 3rd ed. Burlington, VT: Ashgate Publishing,pp.85-93.
• Krishan Kumar. 2004. The Theory of Classification. New Delhi: Vikas, pp. 475–489. Kumbhar, R.2012. LibraryClassification Trends in the 21st Century. Oxford: Chandos,pp.25-37.
• Mahaptra,M.1979.Design of depth classification schedule based on the principles of CC(Ed7). Libri 29(2): 169-18
• Maltby, Arthur. 1975. Sayers Manual of Classification for Librarians, 5th ed. London: Andre Deutsch, pp. 15-108.
• Mills, J. 1962.A Modern Outline of Library Classification. Bombay: Asia, pp. 7-19.
• Mills, J.2004. Faceted classification and logical division in information retrieval.Lib. Trends 52(3):541-57
• Neelameghan, A.1969. Design of Scheme for Classification.Herald of Library Sc. 8,Papers,P,Q,R,ZF,ZG.
• Pocock,H.1997. Classification schemes: development and survival. Cataloguing Australia 23()1/2:10-16
• Ranganathan, S.R. 1952. Inductive vs. Classificatory approach. Annals of Indian Lib.Association.2: 233-245
• Ranganathan, S.R. 1955. International Summer School on Designing of Documentary Classification. Review of Documentation22: 95-98
• Ranganathan, S.R. 1964. Design of Depth Classification. Lib Sc. /Documentation. 1 (1):1-42(Paper A
• Ranganathan, S.R. 1967.Prolegomena to Library Classification, 3rd ed. Bombay, Asia, pp. 543-549.
• Ranganathan, S.R. 1989 Elements of Library Classification, 2nd ed. Ed by B.I. Palmer. New Delhi: UBSPD, pp. 1-10.
• Satija, M.P. 2000. Classification: An essay in Terminology.Knowledge Org 27(4): 221-229.
• Satija, M.P. 2004.A Dictionary of Knowledge Organization.Amritsar : Guru Nanak Dev University, 248 p.
• Slavic, Aida. 2008. Faceted Classification: Management and Use. Axiomathes 18(2) DOI 10.1007/s10516-007-9030-z
• Varendroff, L. 2005. Misconceptions in the record management fraternity. Record Mgmt. Bulletin, 129:6-7.
• Vickery, B.C. 1960.Faceted Classification.A Guide to construction and use of special classification schemes. London:Aslib, 70p. ISBN 978-085-1420103.
• Zins, C.2004. Knowledge organization: an epistemological perspective. Kng.Org. 31(1):49-63

Learn More:

Module LIS/KOP – C/14: Designing of a Depth Classification Schedule

1. Do you know

• No classification is perfect. No single classification is suitable to all types of users.
• In the 19th century the library classifications were designed intuitively without any theory and design methodology, and by a single individual.

2. Points to remember

• Designing a library classification is very costly in terms of time and money.
• Gauge your resources of men and material before venturing two to design a new library classification system.
• Online classification systems have many added features and advantages over the traditional print versions.
• Practically dividing the work into Idea, Verbal and Notational planes is useful, but it cannot be followed rigorously.
• It is better to start with many facets in dividing knowledge than restricting to so called only five fundamental categories.
• Classification should be revised regularly but not too frequently.
• A classification must have a newsletter or its website to communicate with its users.