IEEE Guide To Software Requirements Specifications Std 830 1984 SRS
User Manual: Pdf
Open the PDF directly: View PDF .
Page Count: 26
Download | |
Open PDF In Browser | View PDF |
ANSI/IEEE Std 830-1984 IEEE Guide to Software Requirements Specifications SH08714 Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. THIS PAGE WAS BLANK IN THE ORIGINAL Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. ANSI/IEEE Std 830-1984 A n American National Standard IEEE Guide t o Software Requirements Specifications Sponsor Software Engineering Technical Committee of the IEEE Computer Society Approved September 30,1983 IEEE: Standards Board Approved July 20, 1984 American National Standards Institute @ Copyright 1984 by The Institute of Electiiical and Electronics Engineers, Inc 345 East 47th Street, New York, NY 10017, USA N o part of this publictrtion may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior written permission of the publisher. Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. IEEE Standards documents are developed within the Technical Committees of the IEEE Societiies and the Standards Coordinating Committees of the IEEE Standards Board. Members of the committees serve voluntarily and without compensation. They are not necessarily members of the Institute. The standards developed within IEEE represent a consensus of the broad expertise on the subject within the Institute as well as those activities outside of IEEE which have expressed an interest in participating in the development of the standard. Use of an IEEE Standard is wholly voluntary. The existence of an IEEE Standard does not imply that there are no other ways to produce, test, measure, purchase, market, or provide other goods and services related to the scope of the IEEE Standard. Furthermore, the viewpoint expressed at the time a standard is approved and issued is subject to change brought about through developments in the state of the art and comments received from users of the standard. Every IEEE Standard is subjected to review at least once every five years for revision or reaffirmation. When a document is more than five years old, and has not been reaffirmed, it is reasonable t o conclude that its contents, although still of some value, do not wholly reflect the present state of the art. Users are cautioned to check to determine that they have the latest edition of any IEEE Standard. Comments for revision of IEEE Standards are welcome from any interested party, regardless of membership affiliation with IEEE. Suggestions for changes in documents should be in the form of a proposed change of text, together with appropriate supporting comments. Interpretations: Occasionally questions may arise regarding the meaning of portions of standards ,as they relate to specific applications. When the need for interpretations is brought to the attention of IEEE, the Institute will initiate artion t o prepare appropriate responses. Since IEEE Standards represent a consensus of ail concerned interests, it is important to ensure that any interpretation has also received the concurrence of a balance of interests. For this reason IEEE and the members of its technical committees are not able to provide an instant response to interpretation requests except in those cases where the matter has previously received formal consideration. Comments on standards and requests for interpretations should be addressed to: Secretary, IEEE Standards Board 345 East 47th Street New York, NY 10017 USA Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. Foreword (This Foreword is not a part of IEEE Std 830-19(94, IEEE Guide to Software Requirements Specifications.) This guide describes alternate approaches to good practice in the specification of software requirements. The requirements may be explicitly stated by the user or they may be allocated to computer software (that is, programs) by the system requirements analysis process. This guide does not suggest that a hierarchy of software requirements specifications exists, of which each, in turn, defines a smaller subset of requirements. As a guide, this document should help: (1)Software customers to accurately describe what they wish t o obtain. (2) Software suppliers to understand exactly what the customer wants. (3) Individuals to accomplish the following goals: (a) Develop standard software requirements specifications (SRS) outline for their own organizations. (b) Define the form and content of their specific software requirements specifications. (c) Develop additional local supporting items such as an SRS quality checklist, or an SRS writer’s handbook. To the customers, suppliers and other individuals, a good SRS provides several specific benefits. It will accomplish the following goals: (1)Establish the basis for agreement between the customers and the suppliers on what the software product is to do. The complete description of the functions t o be performed by the software specified in the SRS will assist the potential user, to determine if the software specified meets their needs or how the software must be modified to meet their needs. (2) Reduce the development effort. The preparation of the SRS forces the various concerned groups in the customer’s organization to consider rigorously all of the requirements before design begins and reduces later redesign, recoding, and retesting. Careful review of the requirements in the SRS can reveal omissions, misunderstandings, and inconsistencies early in the development cycle when these problems are easier to correct. (3) Provide a basis for estimating costs and schedules. The description of the product to be developed as given in the SRS is a realistic basis for estimating project costs and can be used to obtain approval for bids or price estimates. ‘The SRS also provides a clear description of the required software and makes it easier to estimate and plan the necessary resources. The requirements which, together with a development plan, can be used to measure progress. (4)Provide a baseline for validation and verification. Organizations can develop their validation and verification plans much more productively from a good SRS. As a part of the development contract, the SRS provides a baseline against -which compliance can be measured. (However, that the converse is not true; a standard legal contract cannot be used as an SRS. Such documents rarely contain the detail required and are often incomplete.) (5) Facilitate transfer. The SRS makes it easier to transfer the software product to new users or new machines. Customers thus find it easier t o transfer the software to other parts of their organization, and suppliers find it easier to transfer it to new customers. (6) Serves as a basis for enhancement. Because the SRS discusses the product but not the project that developed it, the SRS serves as a basis for later enhancement of the finished product. The SRS may need to be altered, but it does provide a solid foundation for continued production evolution. This guide is based on a model in which the result of the software requirements specification process is an unambiguous and complete specification document. In principle, the SRS can be mechanically translated into the specified software program directly. As such, the resulting SRS document itself is the specified software, and the supplier’s only duty (after completing the SRS) would be the mechanical compilation of the SRS into machine code for the target computer. The present state of the art does not support such a compiler with an optimizer of such efficiency to make it practical but this limitation need not, and should not, restrict the intermediate objective of an unambiguous SRS. Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. This guide is consistent with IEEE Std 729-1983,IEEE Standard Glossary of Software Engineering Terminology; ANSI/IEEE Std 730-1981,IEEE Standard for Software Quality Assurance Plans; and IEEE Std 829-1983,IEEE Standard for Software Test Documentation. This guide may be used in conjunction with those standards or separately. This guide was prepared by the Software Requirements Working Group of the Software Engineering Standards Subcommittee of the Technical Committee on Software Engineering of the IEEE Computer Society. At the time the guide was approved, the Software Requirements Working Group had the following membership: A. h5. Davis, Chairperson M. Bariff H. Berlack F. Buckley E. Byrne F. Calm K. Foster S . Frankel T. L. Hannan P. W. Kendra T. M. Kurinara R. A. C. Lane R. Lechner E. Levinson P. Lindemann S.Mroczek A. J. Neumann W. Newson D. Paster B. Pope P. B. Powell G. R. Niedhart J. Russell A. Salwin N. B. Schneidewind D. Schultz R. W. Szczech A. Weigel P. Willis H. Willman A. W. Yonda At the time that it approved this guide, the Software Engineering Standards Subcommittee had the following membership: F. J. Buckley, Chairperson R. J. Abbott A. F. Ackerman L. Beltracchi D. W. Bragg D. W.Burt E. R. Byrne H. Carney J. W. Center A. M. Cicu G. G. Cooke A. J. Cote, Jr P. W.Daggett G. Darling B. Dasarathy A. M. Davis P. A. Denny J. A. Dobbins M. L. Eads J. D. Earls L. G. Egan, J r D. W. Fife J. Flournoy J. J. Forman F. K. Gardner D. Gelperin E. L. Gibbs G. Gladden S.A. GlossSoler J. W. Grigsby R. M. Gross D. A. Gustafson R. T. Gustin T. L. Hannan H. Hecht IL. R. Heselton, I11 S. Horvitz .’1 Howley It. N. Hurwitz S. Hwang .J. H. Ingram cJ. P. Kalasky IX. Kessler ‘l’. M. Kurinara ID. V. LaRosa 13. A. C. Lane (7.R. Lewis 17. C. Lim (3. S . Lindsay IM. Lipow W.M. Lively IM. Lubofsky 13. Lindquist ,4. K. Manhindru 1’. C. Marriott C. F. Martiny IN. McCollough 13. Menkus 13. Meyer 13. F. Miller, Jr (3. S . Morris (3.T. Morun W.G. Murch *J. Nebb (3. R. Niedhart IM. A. Neighbors *J. 0. Neilson ID. J. Ostrom D. J. Pfeiffer R. M. Poston P. B. Powell J. W. Radatz J. C. Rault S. T. Redwine L. K. Reed W. E. Riddle C. W. Rutter, I11 P. E. Schilling N. F. Schniedewind A. D. Schuman L. W. Seagren R. L. Skelton W. Smith H. M. Sneed K. C. Tai B. J. Taute R. H. Thayer G. D. Tice, J r T. L. Tillmans W. S . Turner, I11 E. A. Ulbrich, J r D. Usechak U. Voges R. Wachter J. P. Walter D. Webdale A. H. Weigel N. P. Wilburn W. M. Wong T. Workman A. W. Yonda P. F. Zoll Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. Special representatives to the software engineering standards subcommittee were: J. Milandin: ANSI 21 W. G. Perry: Data Processing Manufa'cturers Association R. Pritchett: EDP Auditors Association T. L. Regulinski: IEEE Reliability Society N. C. Fan: Nuclear Power Engineering Committee, IEEE Power Engineering Society Suggestions for improvement of this guide are welcome. They should be provided to: The Secretary IEEE Standards Board 345 East 47th St New York, New York 10017 At the time the IEEE Standards Board approved this standard on September 20,1983 it had the following members: Edward Chelotti, Vice Chairman James H. Beall, Chairman Sava I. Sherr, Secretary J. J. Archambault John T. Boettger J. V. Bonucchi Rene Castenschiold Edward J. Cohen Len S . Corey Donald C. Fleckenstein Jay Forster Donald H. Heirman Irvin N. Howell Joseph L. Koepfinger* Irving Kolodny George Konomos John E. May Donald T. Michael* John P. Riganati Frank L. Rose Robert W. Seelbach Jay A. Stewart Clifford 0. Swanson Robert E. Weiler W.B. Wilkens Charles J. Wylie *Member emeritus Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. THIS PAGE WAS BLANK IN THE ORIGINAL Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. Contents PAGE SECTION 1. ScopeandOrganization ..................................................... 1.1 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Organization .......................................................... 9 9 9 2 . References ............................................................... 9 3 . Definitions ............................................................... 10 4 . Background Information for Writing a Good SRS .................................. 10 4.1 TheSRS ............................................................. 10 10 4.2 Environment of the SRS ................................................. 4.3 Characteristics of a Good SRS ............................................ 11 4.3.1 Unambiguous . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 4.3.1.1 Natural Language Pitfalls .................................... 11 4.3.1.2 Formal Requiremlents Specifications Languages . . :. . . . . . . . . . . . . . . . 11 4.3.2 Complete ....................................................... 11 4.3.3 Verifiable ....................................................... 12 4.3.4 Consistent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 4.3.5 Modifiable ...................................................... 12 4.3.6 Traceable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 4.3.7 Useable During The Operation and Maintenance Phase .................... 13 4.4 Joint Preparation of the SRS ............................................. 13 4.5 SRSEvolution ........................................................ 13 4.6 Tools for Developing an SRS ............................................. 14 14 4.6.1 Formal Specification Methcodologies .................................. 4.6.2 Production Tools ................................................. 14 14 4.6.3 Representation Tools .............................................. 14 5 . SoftwareRequirements ...................................................... 5.1 Methods Used to Express Software Requirements ............................. 14 14 5.1.1 Input/Output Specifications ........................................ 5.1.1.1 Approaches .............................................. 15 5.1.1.2 Difficulties ............................................... 15 5.1.2 Representative Examples ........................................... 15 5.1.3 Models ......................................................... 15 5.1.3.1 Mathematical Models ....................................... 15 15 5.1.3.2 Functional Models ......................................... 5.1.3.3 TimingModels ............................................ 15 5.1.3.4 OtherModels ............................................. 16 5.1.3.5 Cautions ................................................. 16 16 5.2 Annotation of the Software Requirements ................................... 17 5.2.1 Stability ........................................................ 17 5.2.2 Degree of Necessity ............................................... 5.2.3 Annotation Caution ............................................... 17 5.3 Common Pitfalls Encountered in Expressing Requirements ...................... 17 5.3.1 Embedding Design in the SRS ....................................... 5.3.2 Embedding Project Requirements in the SRS ........................... 6 . An SRS Prototype Outline ................................................... 6.1 Introduction (Section 1of the SRS) ........................................ 6.1.1 Purpose (1.1of the SRS) ........................................... 6.1.2 Scope (1.2 of the SRS) ............................................ 6.1.3 Definitions, Acronyms. and Abbreviations (1.3 of the SRS) . . . . . . . . . . . . . . . . 17 17 18 18 18 18 18 Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. SECTION PAGE 6.1.4 References (1.4 of the SIES) ......................................... 18 6.1.5 Overview (1.5 of the SRS) .......................................... 18 6.2 The General Description (Section 2 of the SRS) ............................... 18 6.2.1 Product Perspective (2.1 of the SRS) .................................. 19 6.2.2 Product Functions (2.2 o f the SRS) ................................... 19 6.2.3 User Characteristics (2.3 of the SRS) .................................. 19 6.2.4 General Constraints (2.4 of the SRS) .................................. 19 6.2.5 Assumptions and Dependencies (2.5 of the SRS) ......................... 20 20 6.3 The Specific Requirements (Section 3 of the SRS) ............................. 6.3.1 Information Required as Part of the Specific Requirements . . . . . . . . . . . . . . . . 20 20 6.3.1.1 Functional Requirements .................................... 21 6.3.1.2 Performance Requirements .................................. 6.3.1.3 Design Constraints ......................................... 21 6.3.1.4 Attributes ............................................... 21 6.3.1.5 External Interface Requirements .............................. 22 6.3.1.6 Other Requirements ........................................ 22 6.3.2 Organizing the Specific Requirements .................................. 23 24 6.4 Supporting Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . FIGURES Fig 1 A Functional Model Specifying Any Sequence of Alternating 0’s and 1’s ............. 16 TABLES Table 1 Table 2 Table 3 Table 4 Table 5 Prototype SRS Outline .................................................. Prototype Outline 1for SRS Section 3 ..................................... Prototype Outline 2 for SRS Section 3 ..................................... Prototype Outline 3 for SRS Se’ction3 ..................................... Prototype Outline 4 for SRS Se’ction3 ..................................... 18 23 23 24 24 Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. IEEE Guide to Software Requirements Specifications 2. References 1. Scope and Organization [l]ANSI/IEEE Std 100-1977, IEEE Standard Dictionary of Electrical and Electronics Terms. 1.1 Scope. This is a guide for writing software requirements specifications. I t describer; the necessary content and qualities of a good Software Requirements Specification (SRS) and presents a prototype SRS outline. This guide does not specify industry,-wide SRS standards nor state mandatory SRS requirements. This guide is written undeir the premise that the current state of the art does not warrant or support such a formal standards document. This guide is applicable t o in-house and commercial software products. Special care, however, should be used in its application beca.use: [2] ANSI/IEEE Std 730-1981, IEEE Standard for Software Quality Assurance Plans. [3] ANSI/IEEE Std 729-1983, IEEE Standard Glossary of Software Engineering Terminology. [4] BRUSAW, C. T., ALRED, G. and OLIU, W., Handbook of Technical Writing, New York, St. Martin's Press, 1976. [5] DASARATNY, B., Timing Constraints of Real-Time Systems: Constructs for Expressing Them, IEEE Real-Time Systems Symposium, Dec 1982. (1) This guide is aimed at specifying requirements of software to be developed. Application of this material t o alreadydeveloped software is counter-productive. (2) This guide does not cover the specification of requirements for software being develloped using the techniques of rapid prototyping. [6] DAVIS, A., The Design of a Family of Applications-Oriented Requirements Languages, ZEEE Computer, 1 5 , 5 May 1982, pp 21-28. 1.2 Organization. The remainder of this guide is organized as follows: (1) Section 2 provides the references used throughout the guide. (2) Section 3 provides definitions of specific terms used throughout the guide. (3) Section 4 provides background information for writing a good SRS. (4)Section 5 provides specific guidance for expressing software requirements. ( 5 ) Section 6 discusses each of the essential parts of an SRS and provides alternate prototype outlines. [8] KAIN, R., Automata Theory: Machines and Languages, McGraw Hill, New York, 1972. [ 7 ] FREEDMAN, D. and WEINBERG, G., Handbook of Walkthroughs, Inspections and Technical Reviews, 3rd Ed, Little and Brown Publishers, New York. [9] KOHAVI, Z., Switching and Finite Automata Theory, McGraw Hill, New York, 1970. [ 101 KRAMER, J., Editor, Application Oriented Specifications Glossary of Terms, European Workshop o n Industrial Computer Systems (E WICS), Imperial College, London, England, May 6,1981.' 'Copies of this document are available from EWICS, c / o G. R. Koch, BIOMATIKGmbh, Carl-MezStr 81-83, D-7800 Freiburg, Federal Republic of Germany. 9 Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. IEEE Std 830-1984 IEEE GUIDE TO SOFTWARE [ l l ]MILLS, G., and WALTER, J., Technical Writing, New York, Holt, Rinehart and Winston, 4th Ed, 1978. supplier. The person, or persons, who produce a product for a customer. In the context of this document, the customer and the supplier may be members of the same organization. [ 121 PETERSON, J., Petri Nets, ACM Computing Surveys, 9 , 4 , Dec 1977, pp 223-252. user. The person, or persons, who operate or interact directly with the system. The user(s) and the customer(s) are often not the same person( s) . [13] RAMAMOORHY, C. and SO, H. H., S o f t ware Requirements and Specifications: Status and Perspectives, Tutorial: Software Methodology, RAMAMOORTHY, C. and YEH, R. T., Editors. IEEE Catalog no E H 0 142-0,1978, pp 43-164. 4. Background Information for Writing a Good SRS [14] TAGGART, W. M. Jr, and THARP, M.O., A Survey of Information Requirements Analysis Techniques, ACM Computing Surveys, 9 , 4, Dec 1977, pp 273-290. This section provides background information necessary for writing an SRS. This includes: (1)Examination of the nature of the SRS (2) Environmental considerations surrounding the SRS (3) Characteristics required for a good SRS (4) Recommendations for joint preparation of an SRS ( 5 ) Evolutionary aspects of the SRS (6) The use of automated tools to develop an SRS [ 151 TEICHROEW, D., A Survey o f Languages for Stating Requirements for Computer-Based Information Systems, 1972 Fall Joint Cornputer Conference, 1972, pp 1203-1224. 3. Definitions Except for the definitions listed below, the definitions of all terms used in this guidle conform t o the definitions provided in IEEE Std 729-1983 [ 312, for example, the terms requirement, requirements specification. If EL term used in this guide does not appear in that Standard, then ANSI/IEEE Std 100-1977 [ 11, applies. The terms listed in this section have been adopted from Section 2, [ 101. 4.1 The SRS. The SRS is a specification for a particular software product, program, or set of programs that does certain things. See ANSI/ IEEE Std 730-1981 [ 2 ] , 3.4.2.1. The description places two basic requirements on the SRS: (1)I t must say certain things. For example, software developed from an SRS that fails t o specify that error messages will be provided, will probably fail t o satisfy the customer. (2) It must say those things in certain ways. For example, software developed from an SRS that fails t o specify the format and content of error messages and instead is developed from a vague and nonquantifiable requirement such as All error messages will be helpful, will probably be unsatisfactory. What is helpful for one person can be a severe aggravation t o another person. For recommended contents of an SRS see Section 6. contract. A legally binding document agreed upon by the customer and supplier. This includes the technical, organizational, cost and schedule requirements of a product. customer. The person, or persons, who pay for the product and usually (but not necessarily) decides the requirements. In the context of this document the customer and the supplier may be members of the same organization. language. A means of communication, with syntax and semantics, consisting of a set of representations, conventions and associated rules used t o convey information. 4.2 Environment of the SRS. It is important t o consider the part that the SRS plays in the total software project. The provisions in ANSI/ IEEE Std 730-1981 [ 2 ] , define the minimum required documents for a software project. See [2], 3.4.2. ANSI/IEEE Std 730-1981 [ 21 also identifies the other useful documents. See [ 21 ,3.4.3. partitioning. Decomposition; the separation of the whole into its parts. 2Numbers in brackets correspond t o those of the references in Section 2. 10 Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. IEEE Std 830-1984 REQUIREMENTS SPECIFICATIONS (2) The specification The control total is taken from the last record, might be read as: (a) The control total is taken from the record at the end of the file ( b ) T h e control total is taken from the latest record (c) The control total is taken from the previous record (3) The specification All customers have the same control field, might be read as: (a) All customers have the same value in their control field (b) All customer control fields have the same format (c) One control field is issued for all customers (4)The specification All files are controlled b y a file control block, might be read as: (a) One control block controls the entire set of files (b) Each file has its own block (c) Each file is controlled by a control block, but one control block might control more than one file 4.3.1.2 Formal Requirements Specifications Languages. One way t o avoid the ambiguity inherent in natural language is t o write the SRS in a formal requirements specification language? (1)One major advantage in the use of such languages is the reduction of ambiguity. This occurs, in part, because the formal language processors automatically detect many lexical, syntactic, and semantic errors. (2) One major disadvantage in the use of such languages is the length of time required t o learn them. 4.3.2 Complete. An SRS is complete if it possesses the following qualities: (1)Inclusion of all significant requirements, whether relating to functionality, performance, design constraints, attributes or external interfaces. (2) Definition of the responses of the software t o all realizable classes of input data in all realizable classes of situations. Note that it is important to specify the responses t o valid and invalid input values. (3) Conformity t o any SRS standard that applies t o it. If a particular section of the standard is not applicable, the SRS should Since the SRS has a definite role to play in this documentation scheme, SRS writers shoulld be careful not t o go beyond the bounds of that role. This means the following requirements should be met: (1)The SRS must correctly define all of the software requirements, but no more. (2) The SRS should not describe any design, verification, or project management details, except for required design constraints. Such a properly written SRS limits the range of valid solutions but does not specify any particular design and thus provides the supplier with maximum flexibility. 4.3 Characteristics of A Good SRS. The previous sections describe the types of informat'ion that should be contained in an SRS. The following concepts deal with particular characteristics. A good SRS is: (1)Unambiguous (2) Complete (3) Verifiable (4)Consistent ( 5 ) Modifiable (6) Traceable (7) Usable during the Operation and Maintenance Phase 4.3.1 Unambiguous. An SRS is unambiguous if - and only if - every requirement stated therein has only one interpretation. (1)As a minimum, this requires that each characteristic of the final product be described using a single unique term. (2) In cases where a term used in a particular context could have multiple meanings., the term must be included in a glossary where its meaning is made more specific. 4.3.1.1 Natural Language Pitfalls. Requirements are often written in a natural language (for example, English). SRS writers who 'use a natural language must be especially careful t o review their requirements for ambiguity. The following examples are taken from Section 2, [71. (1)The specification The data set will contain an end of file character, might be read as: (a) There will be one and only one erid of file character (b) Some character will be designated as an end of file character (c) There will be a t least one end of file character 3F0r detailed discussion on this topic, suggested readingsare [ 6 ] , [ 1 3 ] , [ 1 4 ] , a n d [ 1 5 ] . 11 Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. IEEE Std 830-1984 IEEE GUIDE TO SOFTWARE (4)If a requirement is not expressible in verifiable terms at the time the SRS is prepared, then a point in the development cycle (review, test plan issue, etc) should be identified at which the requirement must be put into a verifiable form. 4.3.4 Consistent. An SRS is consistent if and only if no set of individual requirements described in it conflict. There are three types of likely conflicts in an SRS: (1)Two or more requirements might describe the same real world object but use different terms for that object. For example, a program's request for a user input might be called a prompt in one requirement and a cue in another. ( 2 ) The specified characteristics of real world objects might conflict. For example: (a) The format of an output report might be described in one requirement as tabular but in another as textual. (b) One requirement might state that all lights shall be green while another states that all lights shall be blue. (3) There might be a logical or temporal conflict between two specified actions. For example: (a) One requirement might specify that the program will add two inputs and another specify that the program will multiply them. (b) One requirement might state that A must always follow B, while another requires that A and B occur simultaneously. 4.3.5 Modifiable. An SRS is modifiable if its structure and style are such that any necessary changes to the requirements can be made easily, completely , and consistently. Modifiability generally requires an SRS to: (1)Have a coherent and easy-to-use organization, with a table of contents, an index,, and explicit cross-referencing. ( 2 ) N o t be redundant; that is, the same requirement should not appear in more than one place in the SRS. (a) Redundancy itself is not an error, but it can easily lead t o errors. Redundancy can occasionally help to make an SRS more readable, but a problem can arise when the redundant document is updated. Assume, for instance, that a certain requirement is stated in two places. At some later time, it is determined that the requirement should be altered, but the change is made in only one of the two locations. The SRS then becomes inconsistent. include the section number and an explanation of why it is not applicable. (4)Full labeling and referencing of all figures, tables, and diagrams in the SRS and definition of all terms and units of measure. 4.3.2.1 Use of TBDs. Any SRS tha.t uses the phrase t o be determined (TBD) is not a complete SRS. (1)The TBD is, however, occasionally necessary and should be accompanied by: (a) A description of the conditions causing the TBD (for example, why an answer is not known) so that the situation can be resolved. (b) A description of what must be done to eliminate the TBD. (2) Any project documents that are based on an SRS that contains TBDs, should: (a) Identify the version or state the specific release number of the SRS associated with that particular document. (b) Exclude any commitments dependent upon the sections of the SRS that are still identified as TBDs. 4.3.3 Verifiable. An SRS is verifiable if and only if every requirement stated therein is verifiable. A requirement is verifiable if and only if there exists some finite cost-effective process with which a person or machine can check that the software product meets the requirement. (1)Examples of nonverifiable requirements include statements such as: (a) The product should work well, or The product should have a good human interface. These requirements cannot be verified bt,.cause it is impossible to define the terms good or well. (b) The program shall never enter an infinite loop. This requirement is non-verifiable bc>cause the testing of this quality is theoretica1:ly impossible. (c) The output o f the program shall usually be given within 10 s. This requirement is nonverifiable because the term usually cannot be measured. (2) An example of a verifiable statement is The output o f the program shall be given within 20 s of event X, 60% o f the time; ana' shall be given within 30 s of event X, 100% o f the time. This statement can be verified because it uses concrete terms and measurable quant,ities. (3) If a method cannot be devised to determine whether the software meets a part,icular requirement, then that requirement should be removed or revised. 12 Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. IEEE Std 830-1984 REQUIREMENTS SPECIFICATIONS (b) The SRS should contain a record of all special provisions that apply t o individual components such as: (i) Their criticality (for example, where failure could impact safety or cause large financial or social losses). (ii) Their relation t o only temporary needs (for example, t o support a display that may be retired soon). (iii) Their origin (for example, function X is t o be copied from an existing software product in its entirety). (2) Knowledge of this type is taken for granted in the developing organization but is frequently missing in the maintenance organization. If the reason for or origin of a function is not understood, it is frequently impossible t o perform adequate software maintenance on it. (b) Whenever redundancy is necessary, the SRS should include explicit cross-references to make it modifiable. 4.3.6 Traceable. An SRS is traceable if the origin of each of its requirements is clear and if it facilitates the referencing of each :requirement in future development or enhancement documentation. Two types of traceability are recommended: (1)Backward traceability (that is, t o previous stages of development) depends upoln each requirement explicitly referencing its source in previous documents. (2) Forward traceability (that is, to all documents spawned by the SRS) depends upon each requirement in the SRS having a unique name or reference number. When a requirement in the SRS represents an apportionment or a derivative of another requirement, both forward and backward traceability should be provided. Examples include: 4.3.6.1 The allocation of response time t o a data base function from the overall user response time requirement. 4.3.6.2 The identification of a report format with certain functional and user interface requirements. 4.3.6.3 A software product that supports legislative or administrative needs (for example, tax computations, reporting of an overhead ratio). In this case, the exact legislative or administrative document that is being supported should be identified. The forward traceability of the SRS is especially important when the software product enters the operation and maintenance phase. As code and design documents are modlified, it is essential t o be able t o ascertain the complete set of requirements that may be affected by those modifications. 4.3.7 Usable During the Operation and Maintenance Phase. The SRS must address thle needs of the operation and maintenance phLase, including the eventual replacement of the software. (1)Maintenance is frequently carried out by personnel not associated with the original development. Local changes (corrections) can be implemented by means of a well-commented code. For changes of wider scope, however, the design and requirements documentation is essential. This implies two actions (a) The SRS should be modifiable as indicated in 4.3.5. 4.4 Joint Preparation of the SRS. The software development process begins with supplier and customer agreement on what the completed software must do. This agreement, in the form of an SRS, should be jointly prepared. This is important because usually neither the customer nor the supplier is qualified t o write a good SRS by himself. (1)Customers usually do not understand the software design and development process well enough t o write a usable SRS. (2) Suppliers usually do not understand the customer’s problem and field of endeavor well enough t o specify requirements for a satisfactory system. The customer and the supplier need t o work together t o produce a well written and completely understood SRS.4 4.5 SRS Evolution. The SRS may need t o evolve as the development of the software product progresses. (1)It may be impossible t o specify some details at the time the project is initiated. For example, it may be impossible t o define during the Requirements Phase, all of the screen formats for an interactive program in a manner that guarantees that they will not be altered later. (2) Additional changes may ensue as deficien4This guide does not specifically discuss style, language usage, or techniques of good writing. It is quite important, however, that an SRS be well written; for guidance, please refer to general technical writing guides such as [ l ] and [ l l ] . 13 Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. IEEE Std 830-1984 IEEE GUIDE TO SOFTWARE headings and subheadings, the compilation of tables of contents and indexes, etc, all of which help in the production of a more readable SRS. 4.6.3 Representation Tools. Some words in the SRS, especially nouns and verbs, refer specifically to entities and actions in the system. There are several advantages t o identifying them as such. (1)It is possible t o verify that an entity or action always has the same name everywhere in the SRS. Thus calculate trajectory would not co-exist with determine flight path, ( 2 ) It is possible t o identify every place in the specification where a particular entity or action is described. In addition, it may be desirable t o formalize the English structure in some way t o allow automated processing of the content of the SRS. With such constraints it becomes possible to: 4.6.3.1 Display the requirements in some tabular or graphical way. 4.6.3.2 Automatically check the SRS requirements in hierarchical layers of detail, where each layer is complete in itself but may also be expanded upon in a lower hierarchical layer or be a constituent of an upper hierarchical layer. 4.6.3.3 Automatically check that the SRS possesses some or all of the characteristics described in 4.3. cies, shortcomings, and inaccuracies are discovered in the SRS, as the product evolves. Two major considerations in this process are: 4.5.1 The requirements should be specified as completely and thoroughly as possible, even if evolutionary revisions can be forseen as inevitable. For example, the desired screen formats should be specified as well as possible in the SRS as a basis for later design. 4.5.2 A formal change process should be initiated to identify, control, track, and report projected changes, as soon as they are initially identified. Approved changes in requirements should be incorporated in the SRS in such a way as to: (1)Provide an accurate and complete audit trail of changes. (2) Permit the review of current and superseded portions of the SRS. 4.6 Tools for Developing an SRS. The most obvious way to create an SRS is t o write it in a natural language (for example, English). But because natural languages are rich, although imprecise, a number of more formal methods have been devised t o assist SRS writers. 4.6.1 Formal Specification Methodologies. The degree t o which such formal methodologies may be useful in preparing an SRS depends upon a number of factors: (1)The size and complexity of the program (2) Whether a customer contract requires it (3) Whether the SRS is a vehicle for contracts or merely an internal document (4)Whether the SRS document will become the top level of the design document ( 5 ) What computer facilities are available to support such a methodology No attempt is made here t o describe or endorse any particular tool? 4.6.2 Production Tools. A computer-based word processor is a most useful production aid. Usually, an SRS will have several authors, will undergo several revisions, and will have several reorganizations. A word processor that manages the text as a computer file facilitates this process. Almost all computer systems have a word processor and often a document preparation package is associated with it. This automates paragraphing and referencing, the printing of 5 . Software Requirements Each software requirement in an SRS is a statement of some essential capability of the software t o be developed. The following subsections describe: (1)Methods used t o express software requirements (2) Annotation of the software requirements (3) Common pitfalls encountered in the process 5.1 Methods Used To Express Software Requirements. Requirements can be expressed in a number of ways: (1)Through input/output specifications (2) By use of a set of representative examples (3) By the specification of models 5.1.1 Input/Output Specifications. I t is often effective t o specify the required behavior of a software product as a sequence of inputs and outputs. 5For detailed discussion on this topic, see, for example, [ 6 ] , [ 1 3 ] , [ 1 4 ] , a n d [15]. 14 Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. IEEE Std 830-1984 REQUIREMENTS SPECIFICATIONS 5.1.1.1 Approaches. There are at least three different approaches based on the nature of the software being specified: (1)Some software products (such a:; reporting systems) are best specified by focusing on required outputs. In general, output -focused systems operate primarily on data files. User input usually serves t o provide control information and trigger data file processing. (2) Others are best specified by focusing on input/output behavior. Input/output,-focused systems operate primarily on the current input. They are required t o generate the niatching output (as with data conversion routines or a package of mathematical functions). (3) Some systems (such as process control systems) are required t o remember their behavior so that they can respond t o a m input based on that input and past inputs; that is, behave like a finite state machine. In this case the focus is on both input/output pairs and sequences of such pairs. required inputs and outputs but they do not specify the system’s behavior completely. 5.1.3 Models. Another approach is t o express the requirements in the form of a model? This can be an accurate and efficient way t o express complex requirements. At least three generalized types of models are in common usage: (1)Mathematical (2) Functional (3) Timing Care should be taken t o distinguish between the model for the application; that is, a linear programming model (with a set of linear inequalities and an objective function) and the model for the software which is required to implement the application model. See 5.1.3.5. 5.1.3.1 Mathematical Models. A mathematical model is a model that uses mathematical relations t o describe the software’s behavior. Mathematical models are especially useful for particular application areas, including navigation, linear programming, econometrics, signal processing and weather analysis. A mathematical model might specify the response discussed in 5.1.2 like this: (01)* where * means that the parenthesized character string is repeated one or more times. 5.1.3.2 Functional Models. A functional model is a model that provides a mapping from inputs t o outputs. Functional models, for example, finite state machines or Petri nets can help identify and define various features of the software or can demonstrate the intended operation of the system. A functional model might specify the response, i?reviously described by the rnathematical model, in the form of a finite state machine as shown in Fig 1. In this figure, the incoming arrow points t o the starting state. The double lined box represents the accepting state. The notation X/Y on the lines indicates that when X is accepted as an input, Y is produced as an output . 5.1.3.3 Timing Models. A timing model is a model that has been augmented with timing constraints. Timing models are quite useful for specifying the form and details of the software’s behavior, particularly for real-time systems or for human factors of any system. 5.1.1.2 Difficulties. Most software products can receive an infinite number of sequences as input. Thus, t o completely specify the behavior of the product through input/output sequences would require that the SRS contain an infinitely long set of sequences of inputs and required outputs. With this approach, therefore, it may be impossible to completely specify every conceivable behavior that is required of the software. 5.1.2 Representative Examples. One alternative is t o indicate what behavior is required by using representative examples of that behavior. Suppose, for example, that the system is required t o respond with a “1” every time it receives a “ 0 ” . Clearly, a list of all possible sequences of inputs and outputs would be impossible. However, by using representative sequences one might be able t o fully understand the system’s behavior. This system’s behavior might be described by using this representative set of four dialogues: 0101 010101010101 01 010101 These dialogues provide a good idea of the 6Each of the four sample dialogues given :here (one per line) represents a sequence of one-character user inputs and one-character system outputs. ‘For details on using modeling techniques, see [ 51, P I , [ g l , and [121. 15 Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. IEEE Std 830-1984 IEEE GUIDE TO SOFTWARE I -” Fig 1 A Functional Model Specifying Any Sequence of Alternating Os and I s Whenever an SRS uses a model: (a) It means that the model provided an especially efficient and accurate way t o specify the requirements (b) I t does not mean that the implementation of the software product must be based on that model. A model that works effectively for explaining requirements in a written document may not be optimal for the actual software implementation. A timing model might add these constraints t o the model shown in Fig 1. (1)The stimulus 0 will occur within 30 13 of the arrival in state S1 (2) The response 1 will occur within 2 si of the arrival in state S2 5.1.3.4 Other Models. In addition t o the aforementioned, specific applications hiave particularly helpful models. For example, a compiler specification might employ attribute grammars, or a payroll system might use tables. I t is t o be noted that the use of a formal requirements language for an SRS usually implies a need for the use of a particular model. 5.1.3.5 Cautions. Whatever type of model is used: (1)It must be rigorously defined, either in the SRS or in a document referenced in the SRS. This definition should specify (a) The required ranges of the model’s param eters (b) The values of constraints it uses (c) The required accuracy of results (d) The load capacity (e) The required execution time (f) Default or failure response (2) Care must be taken t o keep a model clefinition within the domain of requirements. 5.2 Annotation of the Software Requirements. Typically, all of the requirements that relate t o a software product are not equally important. Some requirements may be essential, especially for lifecritical applications, while others may be just nice to have. (1)Each requirement in the SRS should be annotated t o make these differences in relative importance clear and explicit. (2) Annotating the requirements in this manner, helps: (a) Customers t o give more careful consideration to each requirement, which often clarifies any hidden assumptions they may have. (b) Developers to make correct design decisions and devote appropriate levels of 16 Authorized licensed use limited to: Indian Institute of Technology Indore. Downloaded on January 23,2018 at 03:24:05 UTC from IEEE Xplore. Restrictions apply. IEEE Std 830-1984 REQUIREMENTS SPECIFICATIONS 5.3.1 Embedding Design in the SRS. Embedding design specifications in the SRS unduly constrains the software designs and artificially places potentially dangerous requirements in the SRS. (1)The SRS must specify what functions are t o be performed on what data t o produce what results at what location for whom. The SRS should focus on the services to be performed. The SRS should not normally specify design items such as (a) Partitioning the software into modules (b) Allocating functions t o the modules (c) Describing the flow of information or control between modules (d) Choosing data structures (2) It is not always practical to consider the design as being completely isolated from the SRS. Security or safety considerations may impose requirements that reflect directly into design constraints; for example, the need t o (a) Keep certain functions in separate modules (b) Permit only limited communication between some areas of the program (c) Compute check sums for critical quantities In general, it must be considered that the selection of an appropriate high-level design for the software may require vast amounts of resources (perhaps as much as 10%t o 20% of the total product development cost). There are two alternatives : (1)Ignore the warning in this guide and specify the design in the SRS. This will mean that either a potentially inadequate design is stated as a requirement (because insufficient time was spent in arriving at it), or an exorbitant amount of time is spent during the requirements phase (because an entire design analysis is performed before SRS completion). (2) Use the advice in 5.1.3 of this guide. State the requirements using a model design used solely to assist in the description of the requirements and not intended t o serve as the actual design. effort t o the different parts of the software product. 5.2.1 Stability. One method of annotating requirements uses the dimension of stability. A requirement may be considered stable when it is thought that the needs which it addresses will not change during the expected life of the software, or it may be considered volatile and subject t o change. 5.2.2 Degree of Necessity. Another way t o annotate is t o distinguish classes of requirements as mandatory, desirable, and optional. (1) Mandatory implies that the software will not be acceptable unless these requirements are provided in an agreed manner. (2) Desirable implies that these are requirements that would enhance the software product, but would not make it unacceptable if ithey are absent. ( 3 ) Optional implies a class of functions that may or may not be worthwhile, which gives the supplier the opportunity t o propose soinething which exceeds the SRS. 5.2.3 Annotation Caution. Prior to annotating the requirements, a thorough understanding of the contractual implications of such annotations, should be obtained. 5.3 Common Pitfalls Encountered in Expressing Requirements. An essential point about the SRS is that it should specify the results that must be achieved by the software, not the means of obtaining those results. (1) The basic issues that the requirements writer must address are these: (a) Functionality - what the software is supposed to do (b) Performance - the speed, availability, response time, recovery time of various software functions, etc (c) Design Constraints Imposed o n an Implementation -any required standards in effect, implementation language, policies for d
Source Exif Data:File Type : PDF File Type Extension : pdf MIME Type : application/pdf PDF Version : 1.3 Linearized : No Page Count : 26 Subject : Modify Date : 2018:01:22 22:24:05-05:00 Author : IEEE Title : IEEE guide to software requirements specifications - IEEE Std 830-1984 Producer : Adobe PDF Library 4.0; modified using iText® 5.5.10 ©2000-2015 iText Group NV (AGPL-version) Creator : Acrobat Capture 3.0 Create Date : 2004:05:11 12:14:44+05:30 Style : Searchable Image (Exact)EXIF Metadata provided by EXIF.tools