MTG - Field by Field Analysis and Recommendations
 Share
The version of the browser you are using is no longer supported. Please upgrade to a supported browser.Dismiss

View only
 
ABCDEFGHIJKLMNOPQRSTUVWXYZAAABACADAE
1
Field Label# of Instances# of Coll.# of Val.Collection UsageFree text/CV/DateUsage NotesIssuesProperty RecommendationsValue RecommendationsLinked Data RecommendationsBroader Questions
2
titleTitle815684861239Used in every collection except for road, mma, adaccessFree textMessy, slight variations across items that have the same titles (generally advertising collections)Require field. Set policies for type of collection (archival, doc_phot, etc).
3
titleHeadline78738937420Used in advertising collectionsFree textSame as TitleKeep as Advertising-specific property.
4
alternativeAlternative_Title7468655duc; frankespada; gamble; hasm; italianposters; protfam; quartets; russianpostersFree textUsed to capture English translation when Title is in a non-English language, except in Gamble - there used to capture Chinese title as well as 'Original Image Title'25 instances of value 'None' in Frank Espada collection. Some instances seem more like description than title (eg: "Photographs of the principal individuals involved: James. B. Duke"). Some values that are showing up in alternative are in title field in CONTENTdm, was mapping changed when METS was generated?Set policies for when to use - translations, supplied, etc? Use collection-specific labels to elucidate use.Remove all instances of 'None'. Move descriptive information to Description field.
5
alternativeFirst_Line209711930hasmFree textSame as TitleKeep as legacy field.
6
alternativeRefrain137311306hasmFree textSame as TitleKeep as legacy field.
7
creatorCompany899091014155Used in advertising collectionsMix of free text/authoritativeSometimes authoritative value of proper name, sometimes general heading (eg, 'Museums'), sometimes appears to have been transcribed. Lots of brackets, question marks, etc4626 instances of some form of 'N/A'; 2261 instances of some form of 'unknown'. Lots of 'Various'. Lots of brackets, question marks, etc. Slight variations in terms that result in many different values for the same entity. Keep as Advertising-specific property.Cluster and edit authoritative (or authoritative-like) headings. Remove brackets and question marks. Remove all instances of 'N/A' and 'unknown'. Move 'Various' info to description field.Reconcile against LCNAF.
8
creatorComposer1763836888Used in music collections (hasm; quartets; sheetmusicindex)Mix of free text/authoritativeSometimes authoritative value of proper name, sometimes transcribed. Transcribed values with authoritative values. Question marks.Merge with Creator. Use Composer as collection-specific label. Cluster and edit authoritative (or authoritative-like) headings. Remove question marks. Remove transcribed values, or move to Description.Reconcile against LCNAF.
9
creatorCreator1911926899Used in all collectionsMix of free text/authoritativeSometimes authoritative value of proper name, sometimes transcribed. Transcribed values mixed with authoritative values (heading for GF Handel alongside 'A Lady', lots of initials). Some could be authoritative but are not (Paganini). Lots of quotation marks, allcaps (mostly Mazzoni), brackets, N/A. Set as controlled vocabulary field. Move 26 instances of creator in hasm to contributor field. Cluster and edit authoritative (or authoritative-like) headings. Remove question marks. Remove transcribed values, or move to Description. Excluding Mazzoni will make this a lot easier. Reconcile against LCNAF.
10
creatorInterviewer_Name446129behindtheveilControlled vocabularyInverted personal names. One corporate body (Fox Television). Very cleanLocal controlled vocabulary - most likely not found outside of Duke? Except for Fox.Remap to contributor. Reconcile against LCNAF.
11
creatorLyricist682623009hasm; sheetmusicindexMix of free text/authoritativeSometimes authoritative value of proper name, sometimes transcribed. Transcribed values mixed with authoritative values (heading for Irving Berlin alongside 'Author of O sleep my love', lots of initials). Some could be authoritative but are not (Paganini). Multiple values in field, separated by ';'.Keep as legacy field. Split multi-valued cells. Cluster and edit authoritative (or authoritative-like) headings. Remove question marks. Remove transcribed values, or move to Description.
12
contributorArranger12332682hasm; sheetmusicindexMix of free text/authoritativeSometimes authoritative value of proper name, sometimes transcribed. Transcribed values mixed with authoritative values.Keep as legacy field. Cluster and edit authoritative (or authoritative-like) headings. Remove transcribed values, or move to Description.Reconcile against LCNAF.
13
contributorArtist7313415adaccess; hasm; sheetmusicindexMix of free text/authoritativeSometimes authoritative value of proper name, sometimes transcribed. What is the difference between this field and Illustrator? Transcribed values mixed with authoritative values. Some inverted, some not. Lots of initials, question marks, illegible. Some could be authoritative but are not (Dashiell Hammett)Keep as legacy field.Cluster and edit authoritative (or authoritative-like) headings. Remove question marks, illegible. Remove transcribed values, or move to Description.Reconcile against LCNAF.
14
contributorChoreographer16213hasm; sheetmusicindexFree textUninverted namesUninverted namesKeep as legacy field. Invert names. Reconcile against LCNAF.
15
contributorContributor18619105behindtheveil;bloomsbury; caribbeansea; duc; dukechapel; hasm; meyermarshall; rubenstein; wlmpcMix of free text/authoritativeTranscribed values mixed with authoritative values. Some names are uninverted. Multiple values in some fields, separated by ';'.Make controlled vocabulary field. (Larger question - same controlled vocabulary as Creator? How do we want to manage named entities?)Split multi-valued cells. Invert names. Cluster and edit authoritative headings. Remove transcribed values, or move to Description. Reconcile against LCNAF.
16
contributorDedicatee258921612hasm; sheetmusicindexFree textUninverted names. Transcribed from item.Transcribed values. Uninverted names. Multiple values in some fields.Keep as legacy field. Split multi-valued cells. Invert names. Cluster and edit authoritative headings. Remove transcribed values, or move to Description. Reconcile against LCNAF.
17
contributorEngraver22952410hasm; sheetmusicindexFree textUninverted names. Transcribed from item.Stuff like "L. Johnson (electrotyper)" and "O.". Lots of initials, parts of names. Uninverted. Multiple values in some fields.Keep as legacy field. Split multi-valued cells. Invert names. Cluster and edit authoritative headings. Remove transcribed values, or move to Description. Reconcile against LCNAF.
18
contributorIllustrator17493472adaccess; hasm; sheetmusicindexMix of free text/authoritativeMostly transcribed. What is the difference between this field and Artist? Lots of initials, parts of names, question marks, illegibles. Messy.Merge with ArtistInvert names. Cluster and edit. Remove transcribed values, or move to Description. Reconcile against LCNAF.
19
contributorLithographer10212257hasm; sheetmusicindexMix of free text/authoritativeSometimes authoritative value of proper name, sometimes transcribed. Lots of initials, variations on the same name, non-name stuff (eg: "Targis (the illustration following the music)")Keep as legacy field. Invert names. Cluster and edit. Remove transcribed values, or move to Description. Reconcile against LCNAF.
20
contributorPerformer251321560hasm; sheetmusicindexMix of free text/authoritativeSometimes authoritative value of proper name, sometimes transcribed. Lots of junk. variations on the same name. Uninverted names. Some could be authoritative but are not (Al Jolson). Keep as legacy field. Invert names. Cluster and edit. Remove transcribed values, or move to Description. Reconcile against LCNAF.
21
contributorPlacement_Company554256822advertising collectionsMix of free text/authoritativeTranscribed - doesn't look normalizedVariations on the same names. Question marks, lots of brackets, illegibles. Keep as Advertising-specific property.Cluster and edit. Remove transcribed values, or move to Description.Reconcile against LCNAF.
22
contributorProducer1481108sheetmusicindexMix of free text/authoritativeMostly uninverted, transcribed.Variations on the same names. Multiple values in some fields, separated with ','. Keep as legacy field. Invert names. Cluster and edit. Remove transcribed values, or move to Description. Reconcile against LCNAF.
23
contributorSponsor97156adaccessMix of free text/authoritativeMostly corporate bodies, names don't look normalizedMostly corporate names but a few have values like 'Industry' and 'Sugar'. Keep as legacy field. Cluster and edit. Clean up.Reconcile against LCNAF.
24
contributorStaging28121sheetmusicindexFree textMostly uninverted personal names.Uninverted names. Not authoritative values (eg., 'Bob Fosse')Keep as legacy field. Invert names. Reconcile against LCNAF.
25
publisherPublisher2015773282hasm; jwtnewsletters; museesdeshorreurs; russianposters; sheetmusicindex; songsheets; wlmpcMix of free text/authoritativeSometimes authoritative value of proper name, sometimes transcribed.Lots of junk. Brackets, question marks, variations in same names, uninverted names.Set as controlled vocabulary field.Split multi-value cells. Invert names. Cluster and edit. Remove transcribed values, or move to Description.Reconcile against LCNAF.
26
dateDate1306364312146most of themFree text/dateNo consistent input format. Brackets, question marks, dashes, nonnumeric stuff, variations of N.D., Set as date data type Split multi-value cells. Remove brackets, reformat consistently. Attempt to normalize using EDTF formats.
27
dateInterview_Date4051136behindtheveilDateConsistent yyyy-mm-dd NoneMerge with Date field. Use Interview_Date as collection-specific labelNone
28
dateInterviewee_Date_of_Birth3761371behindtheveilDateConsistent yyyy-mm-dd NoneKeep as legacy field. This is about the interviewee, not the resource. We need a way to model named entities that aren't digital objects in the repository.
29
dateIssue_Date11481380zinesFree text/dateMix of formats (eg. '1984', '2000 Spring', 'Apr-93') - used as companion to 'Date_Created field in zines collectionNone as it is used with Date_Created field.Out of scope - excluding zines metadata from remediation.If we consider the zines collection out of scope, we don't need to address this field usage immediately.
30
createdDate_Created1448126zinesDateConsistent yyyyNoneOut of scope - excluding zines metadata from remediation.If we consider the zines collection out of scope, we don't need to address this field usage immediately.
31
modifiedDate_Modified41016behindtheveilDateConsistent yyyy-mm-dd What does this refer to? Is this technical metadata?Remove.See property recommendations
32
temporalTemporal_Coverage2135611560adaccess; dsp; duc; eaa; garymonroe; gedney; hasm; italianposters; kwilecki; mma; trumpetFree text/dateMostly yyyy or yyyy-mm or yyyy-mm-dd, but free text tooUsage of 'x', question marks, '/', 's', ('191x0/191x9', '('1932', 'Oct')-10', '1870s/1879'). Also weirdly, 753 instances of the value 'or'. Most weirdnesses seem to be in gedney.Not clear what the purpose of this field is in relation to Date. Generate list of items for which the value in this field is different from the value in the Date field, and try to figure it out.See property recommendations
33
typeDCMI_Type38627327most of themCVDCMI type vocabularyEvery item should have this field. Slight variations in terms (Moving Image vs. MovingImage and Still Image vs. StillImage). Make this the only Type field, require it for all items.Limit values to DCMI type vocabulary. Attempt to apply to all items.Replace with URIs from DCMI namespace.
34
typeGenre877402827CVLargely AAT, some DCMI type erroneously thrown inIncludes DCMI terms. Not confirmed that AAT was consistently used.Merge with Format.Move DCMI terms to Type. Use AAT as controlled vocabulary.
35
typeType12004437216sheetmusicindex accounts for lots of the weirdnessCVAAT, DCMI, maybe local as well?Unclear what CVs are used. AAT, DCMI, maybe LCSH and local headings? Variations in capitalization, misspellings. Some cells contain multiple values separated by ' ; 'Remap to Format. Use Type for DCMI type. Move DCMI terms to Type. Use AAT as controlled vocabulary. Cluster and edit values, map local terms to AAT.
36
formatFormat77107553dukechapel; eaa; hfc; mma; roadFree text/CVMostly free text but some AAT terms. Physdesc stuff.Should be CV field but is free text. All sorts of weird stuff (eg.: 'half-size; b/w photocopy, cardstock cover'.Limit to AAT terms. Map values to AAT. Remove physdesc stuff to Description if possible. Reconcile against AAT LOD
37
mediumMedium86776953advertising collectionsLocal CV?Mostly a list of local headings for advertising collections, plus some outliers.Improper use of dc:medium. Messy, slight variations, misspellings. Weird stuff like 'Collier's' and 'Charity'. Move most instances to Format. Use this field only to describe physical medium of original.
38
extentExtent22438291275Free textNumber of pages, scans, audiocassetes; slides; dimensions,sheetmusicindex almost entirely values that consist only of numbers, with no unit of measurement indicated. Frank Espada collection uses only the value 'None'.Add unit of measurement to sheetmusicindex values, or remove them.
39
subjectDuke_Opponent15082132rubenstein (only 1 instance); sportsfilmsLocal CVOut of scope - excluding sportsfilms metadata from remediation.Not authorized versions of namesOut of scope - excluding zines metadata from remediation.Map values to authoritative versions of names.Reconcile against LCNAFDo we want Subject fields to be divided into Subject-Topic, Subject-Geographic, Subject-Name?
40
subjectInterviewee_Gender40614behindtheveilUsed to record the gender of the intervieweeReally only two values: male, female. Two values are 'Male; Female' and 'Female; Male'.Do we really need this information in it's own field? UPDATE: Per John Gartrell, this is useful for researchers. As moving this information into a subject field would involve losing meaning (eg, 'men' to describe the collection instead of 'male' to describe the interviewee), I think we should retain the field as a legacy, but revisit developing an metadata application profile for oral history materials at a later date.This is about the interviewee, not the resource. We need a way to model named entities that aren't digital objects in the repository.
41
subjectInterviewee_Occupation7281145behindtheveilLocal CVUsed to record the occupation of the intervieweeLocal terms when authoritative CV could be used insteadEither: move this information to Description field, or merge with Subject-Topic fieldMove to Description field.This is about the interviewee, not the resource. We need a way to model named entities that aren't digital objects in the repository.
42
subjectProduct864371018986advertising collectionsFree text/local CVUsed to record the product being advertisedMix of corporate bodies and general products (Coca Cola, horse racing). Brackets and question marks. Use of 'unknown' and 'n/a'. LOTS of slight variations resulting in multiple values for same entity.Leave as legacy field.LOTS of cluster and edit. Remove bracket, question marks, 'n/a', unknown. Try to use authoritative headings when possible.Reconcile against LCNAFThis is a huge and messy field. I would like to attempt to clean it up but if it turns out to be too complicated we can consider leaving it segregated.
43
subjectRace11012blakeLocal CVUsed to record the race of the person/people depictedNot the most sensitive way to deal with racial characteristicsGet rid of it.Maybe move this information to the Description field?
44
subjectSeries125451480behindtheveil; dukechronicle; dukengineer; frankespada; gamble; garymonroe; hleewaters; jesseandrews; jwtnewsletters; kwilecki; meyermarshall; russianposters; wlmpc; womenstraveldiariesFree textUsed to indicate the archival series from the source collection finding aid, sometimes created for the digital collectionPotentially inconsistently applied. Used differently across collections? (behindtheveil appears to be geographic area, dukechronicle usage indicates edition type, dukengineer is title at time of publication, wlmpc is subject category). In the context of archival collections, how useful is it to have it in the record? Sometimes it isn't exactly as expressed in the finding aid anyways (eg jesseandrews)If we want a field for including series information as pulled from the finding aid, use this field but map it to IsPartOf. Examine other uses and remap to subject, alternative, spatial, as necessary.Keep only uses that indicate location in archival collection (if desired), and use consistently going forward. Ensure that series information is the same as listed in finding aid. Part of larger conversation about how/whether or not to represent archival context in metadata records. We probably need to consult with Rubenstein on this. Potentially Noah can be their proxy.
45
subjectSubject204553448851most of themFree text/local CVSometimes LCSH/LCNAF, sometimes locally defined, sometimes no apparent attempt at regularizing input25 use LCSH/LCNAF (or names are formatted as such). Advertising collections have their own weird list of terms. Gamble, gedney, hasm, mma, sheetmusicindex are all over the place. Make controlled vocabulary field. See broader questionsLimit to LCSH.Reconcile against authoritiesNeed to decide how to treat subjects. Currently separated into vocabulary-specific fields in CONTENTdm, subjects are dumped into a single field when Will creates METS files. We need a way to indicate the authority being used (LCSH, LCNAF, TGM, etc)
46
subjectSubseries2320237jwtnewsletters; kwileckiFree textUsed to indicate the archival subseries from the source collection finding aidThis isn't really subject information. Remap to IsPartOf. Potentially concatenate with Series information. Ensure that series information is the same as listed in finding aid. Part of larger conversation about how/whether or not to represent archival context in metadata records. We probably need to consult with Rubenstein on this. Potentially Noah can be their proxy.
47
subjectTag14415protfamLocal CVShort list of local headingsMerge with Subject.Map values to authoritative versions of names, if possible.
48
descriptionDescription900883951861most of themFree textUsed to add contextual information in a free form manner.A couple of collections have instances of "None" (caribbeansea, jesseandrews). Some contain untranslated foreign languages, others contain translations of foreign language in other fields. Some are very messy (sheetmusic index).I don't know that we need to prescribe use of this field?
49
descriptionIllustrated162612songsheetsYes/NoUsed to indicate presence of illustrationsEither merge with Description field and include information about Illustrations there or don't map field to Description and leave as legacy field (if it's important for users to have this information in its own field).
50
descriptionInstrumentation3018176hasmLocal CVUsed to indicate instrumentationMessy, multiple values per field. Includes initialisms for vocal ranges.Either merge with Subject field and move information about vocal ranges to Description, or don't map field and leave as legacy field (if it's important for users to have this information in its own field).Cluster and edit; map existing terms to the LC medium of performance thesaurus for music Reconcile against LCSH, LCMPT
51
descriptionSetting4128665advertising collectionsLocal CVUsed to indicate setting of billboard (I think)Leave as legacy field.
52
descriptionTime_of_Photo3361336gedneyOne valueUsed to indicate time of day photo was takenOnly one term - 'Night'. Is it necessary for this information to be in a separate field?Move this information to the Description field.
53
descriptionTone3187653advertising collectionsLocal CVUsed to indicate whether photo is color or black and whiteOne erroneous instance. Otherwise uniformly applied.Leave as Advertising specific propertyUse AAT terms 'black-and-white photographs' and 'color photographs'
54
abstractAbstract313vicaFree textOnly used 3 times for one collection. Could use Description field instead?
55
spatialCity_of_Publication99143427hasm; quartets; sheetmusicindexLocal CVUsed to record city of publicationNot normalized values. Some brackets, question marks, instances of 's.l.' and 'n.p.', misspellings, lower/upper case differences. Same place name in different languages (Milan, Milano). Multiple values in some cells.Combine with Country_of_Publication, then merge with Spatial_Coverage.
56
spatialCountry_of_Publication68591148sheetmusicindexLocal CV/free textUsed to record country of publicationMix of country codes and names. Typos, erroneous entries ('1856', 'Oliver Ditson'). Multiple values per cell.Combine with City_of_Publication, then merge with Spatial_Coverage.
57
spatialInterview_Location410159behindtheveilCVUsed to record city where interview took placeKeep as legacy field.LCSH headings are consistently used.
58
spatialInterview_State410111behindtheveilCVUsed to record state where interview took place. LCSHSeems redundant since state information is part of LCSH city headings. But good for broader geographic browsing. There is a better way to do this. Keep as legacy field.LCSH headings are consistently used.
59
spatialInterviewee_Birthplace3811193behindtheveilCVUsed to record interviewee's city or county of birth. LCSHThis is not about the resource, but about the interviewee.Keep as legacy field.LCSH headings are consistently used. This is about the interviewee, not the resource. We need a way to model named entities that aren't digital objects in the repository.
60
spatialInterviewee_Residence410165behindtheveilCVUsed to record interviewee's residence. LCSHThis is not about the resource, but about the interviewee.Keep as legacy field.LCSH headings are consistently used. This is about the interviewee, not the resource. We need a way to model named entities that aren't digital objects in the repository.
61
spatialInterviewee_State_of_Birth98113behindtheveilCVUsed to record the interviewee's state of birth. LCSHInconsistently applied. This is not about the resource, but about the interviewee.Keep as legacy field.LCSH headings are consistently used. This is about the interviewee, not the resource. We need a way to model named entities that aren't digital objects in the repository.
62
spatialRegion_of_Publication2645113zinesLocal CVRegions of U.S.Not authoritative values ('Non-U.S.', 'South U.S.')Out of scope - excluding zines from remediation.
63
spatialSite_Alignment150711507sportsfilmsLocal CV"Home" or "Neutral"Out of scope - excluding sportsfilms from remediation.
64
spatialSpatial_Coverage59069302013CV/free textSometimes free text, sometimes authoritative valuesMessy. Brackets, question marks, typos, misspellings, 'unknown'. Authoritative and nonauthoritative representations of the same places. Sometimes not specific enough (eg. "Patterson").Strongly recommend if applicable. Rename field (Location?)Cluster and edit. Convert to LCSH terms.Reconcile against LC LOD
65
spatialState_of_Publication3018187hasmLocal CV/free textSometimes free text, sometimes authoritative valuesMessy. Inconsistent representation of state names. Sometimes not a state. Multiple values per cell.Merge with Spatial_CoverageSplit multi-valued cells. Convert to LCSH terms
66
spatialVenue954187sportsfilmsLocal CVLocation of game, sometimes name of tournament tooInconsistentOut of scope - excluding sportsfilms from remediation.
67
languageLanguage107021611behindtheveil; broadsides; caribbeansea; dukechapel; dukechronicle; dukengineer; hleewaters; italianposters; jwtnewsletters; kwilecki; meyermarshall; museedeshorreurs; rubenstein; russianposters; wlmpc; womenstraveldiariesCVLanguage of itemNoneMake required/strongly recommend if applicable.Use MARC language codes; display human readable versionReconcile against LC LOD
68
audienceAudience7308111adaccessLocal CVIntended audience of advertisingShort but messy and illogical ('farmers' alongside 'magazine') list of terms. 'Consumers' used 7123 out of 7308 instances. The application of this field needs better definition.Get rid of term as it is used now - current application doesn't make sense.Use Library of Congress Demographic Group Terms?Reconcile against LC LOD
69
identifierBox_Number970610142caribbeansea; dukechapel; frankespada; gedney; jesseandrews; jwtnewsletters; kwilecki; meyermarshall; ronaldreis; rubensteinFree textUsed to indicate box number in archival collectionMostly just contains a number, unit type is indicated by field label. This creates a problem when sharing collection metadata. Also it's mapped to dc:identifier but not really an identifier. Consider getting rid of it. Or, if it's useful to have this as part of the metadata for functionality (retrieval for patrons or for aggregating materials in the digital collection interface), map field to isPartOf and include unit type as part of value.If we keep it, add unit type (Box) to value. Or - can we store an archivespace component ID?Ideally we wouldn't store this information in the item level metadata record but would instead use the structured data from the finding aid to represent this information
70
identifierCall_Number44144quartetsFree textLocal call number of item?If you search these values in the Call Number field from the Rubenstein website they yield no results. Seems like a reasonable field to have but the contents are incomprehensible. Leave as legacy field?
71
identifierEndeca_Identifier41240earlymss; rubensteinFree textKeep as is.
72
identifierFolder49422189broadsides; caribbeanseaFree textUsed to indicate folder number or label in archival collectionNot really an identifier. Broadsides contains folder number, caribbeansea folder label.Caribbeansea folder label information could go in a description field. Broadsides folder number info could go into an isPartOf field. If we keep it, add unit type (Folder) to value. Or - can we store an archivespace component ID?
73
identifierIdentifier360673199eaa; garymonroe; hasm; italianposters; museedeshorreurs; wlmpc; womenstraveldiariesFree textGenerally used to indicate location in archival collection (Box 1 etc)Not really an identifier. Inconsistently applied. eaa contains only numbers, no unit type. museedeshorreurs only contains 'none'. It's unclear what the values used in hasm represent.Remove all instances of none. Move information about location in archival collection to isPartOf. Prepend with unit type if missing. We need to have a comprehensive strategy for managing identifiers, as well as representing context for archival collection.
74
identifierInterview_Number4101410behindtheveilFree textInterview numberNot sure what the interview number refers toDetermine whether or not this is a meaningful identifier - in the online interface the 'Item Identifier' is the DPC ID (I think). Merge with Identifier; use Interview_Number as collection-specific label
75
identifierIssue_Number22011142zinesFree textIssue numberOut of scope - excluding zines from remediation.
76
identifierNegative_Number407213746gedneyFree textPhotographer's negative number (I think)Keep as is.Keep as is.
77
identifierOCLC_Number44144quartetsOCLC/Worldcat ID numberWhy is this recorded for this collection but not others? Do other Tripod2 materials have Worldcat records?
78
identifierPrint_Number180751691frankespada, hmp, jesseandrews, kwilecki; ronaldreisFree textPhotographer's print numberKeep as is.
79
identifierRoll_Number50331708gambleFree textPhotographer's roll numberNot really an identifier? Consists of just a number or number/letter combo.Either remap to IsPartOf or merge with Identifier and use Roll_Number as collection-specific label.
80
identifierVolume2468290dukengineer; dukechronicleFree textVolume number Not really an identifierRemap to IsPartOf
81
hasversionDigitized22586120726roadFree textIncludes link to digitized version in OAAA
82
ispartofIs_Part_Of4730443diap; esr; hasm; rubensteinFree textIs Part Of' displays in the interface. In diap, sometimes value is a collection guide title, but sometimes just a personal name, with no indication as to what this means. In hasm, values are date ranges with no indication as to what this means. In rubenstein, values are collection slugs - what is rubenstein?Figure out a more discernable label. Use consistently as this is a required field for DPLA (their label is 'Collection').When value is a collection guide, include link to said guide. Figure out what the personal names in diap refer to, what date ranges in hasm refer to.This relates to how Source/Series/Subseries/Box/Folder information is represented in metadata.
83
provenanceProvenance56801748Free textProvenance of physical item/collection158 instances of 'None'. Not used in all collections for which we presumably have provenance information?Remove instances of 'None'. Some normalization/cleanup.
84
sourceSource99434blake; hleewaters; hmpFree textName of archival collection from which digital collection was derivedThis appears to be the same usage as IsPartOf fields.Could go either way but I think we should just use IsPartOf for the archival collection, as this is what DPLA MAP uses for Collection. Move all instances to IsPartOfThis relates to how IsPartOf/Series/Subseries/Box/Folder information is represented in metadata.
85
sourcePublication92344635adaccess; eaa; mma; protfamFree textName of publication which included adMessy, multiple iterations of same entity. Mix of authoritative and transcribed values.Merge with Source; use Publication as collection-specific label.Cluster and edit. Reconcile against authorities.
86
rightsRights121911927Less than half, should be all of them.Free textFree text rights statement.Inconsistently applied. 831 have value of 'None' (caribbeansea). Leave as isLeave as is, other than a little bit of tidying. Out of scope for this phase of the project.Going forward, this is going to be a broader conversation probably involving Kevin Smith and other people in the library.
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Loading...