Contents

• Contents
• Overview of the Budapest Sociolinguistic Interview Data
  • Brief historical background
    • This was followed
  • Selection of the linguistic variables
  • Classification of BSI tasks
  • Production tasks
    • Oral sentence completion (written clue)
    • Oral sentence completion (no clue)
    • Word elicitation
    • Reading passages
    • Minimal pairs
    • Word list
    • The reporter's test
  • Judgement tasks
    • ``Same or different?''
    • ``Which is correct?''
    • ``How do YOU say it?''
    • ``Demográfia''
    • The staple-removal test
  • Guided conversation
    • List of conversation modules
      • Personal background (BIO)
      • Games (J T)
      • Childhood (GYE)
      • Fights, scuffles (VER)
      • Dating (SZE)
      • Marriage (H Z)
      • Danger of death (HAL) XX
      • Fear of death (F‚L)
      • Dreams ( LM)
      • Family (CSA)
      • Religion (VAL) XX
      • Friendship (BAR)
      • Street crime (B´´UN)
      • School (ISK)
      • Jobs, employment (MUN) XX
      • Abortion (ABO) XX
      • Nuclear plants (ATO)
      • Leisure time (SZI) XX
      • Jokes (HUM)
      • Ethnic minorities (ETN) XX
      • Language (NYE)
      • Informant's choice
  • References
• Transcribing the tape recorded material
  • What to transcribe
  • In what form
  • The structure of the input data
    • The conversation
    • The card-based data
      • Itemise the data
      • Set up variants
  • References
• Relational database systems
  • Structure of output data
  • Data entry
• A revised system
  • Redesigned data model
  • The graphical user interface
• Data retrieval
  • Current limits
  • A common multimedia interface
    • Digitizing the tape recordings
      • Limitations of the tape recordings
      • Making a digital recording
      • Linking the data with the sound files
    • Converting the data files
      • Processing the database files
      • Handling the reading passages
      • Adjusting the transcript
    • Weaving everything together
• Future work
  • Elaboration of the relational database system
  • SGML coding
  • Implementing the database in a Client/Server setting
• Transcription rules
  • Codes
    • Brackets
    • The uncertainty of the transcriber
    • Missing elements
    • Pauses
    • Hesitation
    • Non-conforming suffix
    • Hypercorrect ik verb form
    • -suk/-sk, -szuk/-szk
    • -n k
    • -e interrogative particle
    • -ba/-be, -ban/-ben
    • l-, t-, d- deletion
    • Consonant clusters
    • Overlapping speech
    • Slips, self-corrections, false starts
    • Response giving suffix only
    • Quotation
    • Extralinguistic remarks
    • Foreign words
    • Lenghtened variant consonants
  • Instructions
    • Codes within words
    • Long pauses
    • Pronounciation variants: vowels
    • Pronounciation variants: consonants
    • Dialect speech
    • 25#25”25#25”25#25”
    • Syllable deletion
    • Pauses
  • Items to be standardized
  • Items not to be transcribed or standardized
  • Dictionary (not to be standardised or explained)
  • Form conventions of transcribed text
    • Division of the transcription
    • The format of text lines

Preface

The present volume aims to give an overview of the way the data collected in the Budapest Sociolinguistic Interview (BSI) is processed. It will be readily appreciated that a large scale project like the BSI which generated an enormous amount of highly delicate linguistic datacollected through an array of various tasks, poses a challenge to data processing. It was felt that such a complex undertaking was worth devoting a separate volume to it. The core of the present paper was presented at the ``Workshop on Spoken Language Analysis'' held on 27-28 May 1994, in Venice.

After a brief overview of the BSI data, the paper will give a survey of our evolving attempts to accomodate the BSI data. During its history of more than ten years now, the BSI database system has undergone a major revision. This was precipitated by rapid technological advancement and necessitated by certain flaws in the design of the data model. Despite its recognized shortcomings, the earlier database version is also described in some detail because we feel it may still serve a useful purpose in showing how a first attempt may be made. If the tone of presentation is felt overly self-critical, it was a deliberate policy to help anyone embarking on a similar enterprise to avoid the pitfalls.

The emphasis will be on the data model and data entry - in line with the practical concerns of the BSI projects so far. However, Chapter 5, discusses data retrieval including a description of the multimedia edition of a complete interview on CD-ROM. The paper ends with a discussion of future word. It contains suggestions as to how the technology developed for the CD-ROM needs further elaboration in order to make all the BSI data accessible in a common graphic interface through the World Wide Web. While the bulk of the paper deals with the part of the BSI data that was numerically coded and processed in the database, Appendix A contains a detailed guide to the codes and conventions used in the transcription of the quided conversation. Familiarity with this part will prove indispensible for any serious study of the transcripts.

Budapest, January 1998

Tam s V radi

References

==1 Kontra, Mikl¢s & V radi Tam s. 1997. The Budapest Sociolinguistic Interview: Version 3 (Working Papers in Hungarian Sociolinguistics No 2, December 1997). Budapest: Linguistics Institute of the Hungarian Academy of Sciences. 54 pp.

==1 V radi, Tam s. 1998. Manual of The Budapest Sociolinguistic Interview Data (Working Papers in Hungarian Sociolinguistics No 4, January 1998). Budapest: Linguistics Institute of the Hungarian Academy of Sciences. 101 pp.