Safeguarding Sound Utilizing METS: The Sound Headings Venture.

Uploaded on:
Category: Education / Career
Safeguarding Sound Utilizing METS: The Sound Bearings Venture. Robin Wendler Harvard College Library 7 May 2007. Objectives. " Grow best practices and test rising benchmarks for recorded sound protection and capacity in the computerized area"
Slide 1

Conservation Audio Using METS: The Sound Directions Project Robin Wendler Harvard University Library 7 May 2007

Slide 2

Goals “ Develop best practices and test rising measures for archival sound protection and capacity in the advanced domain” Establish programs for computerized sound safeguarding at every college that will empower us to proceed with this work into the future, and which will deliver interoperable results “ Preserve discriminatingly imperiled, exceedingly important, exceptional field recordings of extraordinary… interest.” Participants Indiana University (Archives of Traditional Music) Harvard University (Archive of World Music)

Slide 3

Parallel Play

Slide 4

Harvard 7-year-old home-developed protection storehouse METS profile made to address interior issues Mixed PC/Mac-based sound studio Pyramix Indiana No conservation vault now; Fedora usage in procedure METS profile made for this venture PC-based sound studio WaveLab Context

Slide 5

Sound Directions: Scope of interoperation today Preservation filing and trade just No end-client conveyance obliged No enlightening metadata obliged Exchange  Ingest  Re-send out

Slide 6

Standards Used in Sound Directions Audio document position AES31-2-2006 (Broadcast Wave) Audio choice rundown AES31-3-1999 under amendment to incorporate markers Archival bundling METS Technical metadata AES Audio Object (in draft) Digital provenance metadata AES Process History (in draft) Indiana utilizing current adaptation Harvard utilizing new draft

Slide 7

Digital Audio Object = What? Tune? Execution? Catch Event? Side/Track? Physical Item? For archival safeguarding, we make one METS for every unique bit of media. This does not forestall presentations in light of different structures.

Slide 8

METS Sections Used M E T S Header Descriptive Metadata Source Metadata Administrative Metadata Technical Metadata File Section Digital Provenance Metadata Structure Map

Slide 9

METADATA AUDIO VERSION Source media techMD: Audio Object techMD: Audio Object (1..n) Audio Decision List (Harvard) Preservation expert (in 1…n records) techMD: Audio Object (1..n) Audio Decision List (Indiana) Preservation expert middle of the road (1..n) Production expert (1..n) techMD: Audio Object (1..n) Audio Decision List Deliverable(s) (1..n) techMD: Audio Object (1..n) SMIL (Harvard) …Plus one digiprovMD for the whole venture

Slide 10

Toolfest Extensive arrangement of little, secluded apparatuses and scripts Calculate checksum Compare checksum Convert markers Convert SMIL De-interleaver Edit ADL header Generate USID Generate UUID Get skillet maps Interleaver JHOVE Marker dump Make MBIT+ditherer MD Make RA maker metadata Make resampler metadata Make RmEditor metadata Mirror undertaking Make RA tech metadata Reverse sound Time code dump Time code/test proselyte Add markers Add container passages Add to process history ADL dump ADL fix ADL information ADL interleaver ADL way substitution ADL source ADL to SMIL ADL to XML BWave link BWave cut BWave alter BWave data

Slide 11

Now we’ve got every one of the parts. How would we make a METS? Populate registry on a document framework Run one instrument (DMART) to develop a sound store bundle mets.xml Including Audio object specialized metadata Process history metadata ADLs Referencing outside records Archival expert sound Production expert sound Deliverable Real Audio SMIL a batch.xml document containing authoritative metadata about the store.

Slide 12

File Groups Harvard M E T S Header <fileSec> <fileGrp ID=“Wave”> … </> <fileGrp ID=“Real”> … </> <fileGrp ID=“AES31”>… </> <fileGrp ID=“SMIL”> … </> </fileSec> Descriptive Metadata Administrative Metadata Indiana <mets:fileSec> <mets:fileGrp ID=“files-audio”> <mets:fileGrp ID=files-sound preservation” USE=“PRESERVATION_MASTER “> … </> <mets:fileGrp ID=files-sound preservationInt” USE=“PRESERVATION_MASTER_INTERMEDIATE”>…</> <mets:fileGrp ID=files-sound production2496” USE=“PRODUCTION_MASTER”> … </> </mets:fileGrp> <mets:fileGrp ID=“files-aes31” USE=“ADL”> … </> <mets:fileGrp ID=“files-doc” USE=“descMD”>… </> <mets:/fileSec> File Section Structure Map

Slide 13

One structMap Approach Indiana M E T S Header <mets:structMap TYPE=" legitimate "> <mets:div TYPE=" recording " LABEL=" atm_99003 "> <mets:div TYPE=" face " LABEL=" atm_99003_01 "> <mets:div TYPE=" area " LABEL=" Section 1 " ORDER=" 1 "> <mets:fptr> <mets:area FILEID=" document atm_99003_010101_preservation " BETYPE=" TCF " BEGIN="*0000 " END="*2778 "/> </mets:fptr> <mets:fptr> <mets:area FILEID=" document atm_99003_010101_preservationInt " BETYPE=" TCF " BEGIN="*0000 " END="*2778 "/> </mets:fptr> <mets:fptr> <mets:area FILEID=" record atm_99003_01_production2496 " BETYPE=" TCF " BEGIN="*0000 " END="*2778 "/> </mets:fptr> </mets:div> Descriptive Metadata Administrative Metadata File Section Structure Map

Slide 14

Cross-blur join Indiana M E T S Header <mets:fptr> <mets:par> <mets:area FILEID=" document atm_99003_010101_preservation " ADMID=" fade1 " BETYPE=" TCF " BEGIN="*0264 " END="*2184 "/> <mets:area FILEID=" document atm_99003_010201_preservation " ADMID=" fade2 " BETYPE=" TCF " BEGIN="*0721 " END="*2641 "/> </mets:par> </mets:fptr> Descriptive Metadata Administrative Metadata File Section Structure Map

Slide 15

Alternative structMap Harvard M E T S Header structMap TYPE=" LOGICAL "> <div> <div LABEL=" archival "> <fptr FILEID=" archival_2FAWM_5FDAT_5F172_5Fside_5F1.adl "/> <div LABEL=" archival_2FAWM_5FDAT_5F172_5Fside_5F1Media "> <fptr FILEID=" AWM_5FDAT_5F172_2Fside_5F1l.wav "/>  … <div LABEL=" creation "> <fptr FILEID=" production_2FAWM_5FDAT_5F172_5Fside_5F1.adl "/> <div LABEL=" production_2FAWM_5FDAT_5F172_5Fside_5F1Media ">   <fptr FILEID=" AWM_5FDAT_5F172_2Fside_5F1_.wav "/> … <div LABEL=" deliverable "> <div LABEL=" deliverable_2Fsmil "> <fptr FILEID=" AWM_5FDAT_5F172_2FDeliverable_2F.smil "/> </div> <div LABEL=" deliverable_2Fperformances "> <fptr FILEID=" AWM_5FDAT_5F172_2Fdeliverable_2F1.rm "/> … Descriptive Metadata Administrative Metadata File Section Structure Map

Slide 16

Different desires drive diverse decisions Role of METS for sound Navigation of substance for end clients Navigation of substance for sound architects Interaction of METS and sound gauges Should document references inside AES metadata reflect METS inward structure or unloaded index?

Slide 17

The way it lives up to expectations now Indiana Audio METS SIP Harvard Audio METS SIP Harvard Converts For Ingest Indiana Converts For Ingest Harvard Repository Indiana Repository

Slide 18

The way it ought to work Harvard Audio Object Indiana Audio Object Harvard Repository Indiana Repository Convert to/from Common Profile Convert to/from Common Profile Common Audio METS DIP/SIP

Slide 19

Sound Directions, 2005-2007 Funded by award from National Endowment for the Humanities (U.S.) Thank you!

Slide 21

Interaction of METS and sound benchmarks References inside AES metadata: would it be a good idea for them to be right inside archival bundle or right once unloaded? In what applications/connections will the substance be utilized? End clients Audio engineers

Slide 22

Things Harvard wishes it did another way Don’t keep Mac Creator Codes. We plunk in standard ones, not the ones that quite to records in the bundle. Don’t require any. Don’t keep waveform documents New innovation creates them in less than a moment versus 40 minutes earlier. Keep specialized metadata for disposed of middle of the road substance records as metadata, not as substance. Don’t inquire.

Slide 23

METS Element Harvard <mets xmlns="" xmlns:xlink="" xmlns:xsi="" OBJID="AWM_DAT_172" LABEL="AWM_DAT_172" TYPE="AUDIO_SIP" PROFILE="Harvard" xsi:schemaLocation=""> Indiana <mets:mets xmlns:mets="" xmlns:xlink= xmlns:marc21= http

View more...