Skip to the main content

Professional paper

https://doi.org/10.37083/bosn.2023.28.25

Authority data in the COBISS system and why is its importance increasing

Andreja Krajnc Vobovnik orcid id orcid.org/0009-0005-0719-3428 ; Institute of Information Science (IZUM), Maribor, Slovenia


Full text: serbian pdf 1.474 Kb

page 25-37

downloads: 132

cite

Full text: english pdf 1.474 Kb

page 25-37

downloads: 202

cite

Download JATS file


Abstract

With the introduction of the IFLA Library Reference Model (IFLA LRM), authority control is becoming increasingly important, as authority data will play a key role in the network of linked data. Therefore, the preparation of authority data will help to facilitate the implementation of IFLA LRM and will be useful in the transition to RDA (Resource Description and Access).
The paper presents an overview of authority data in the COBISS system, with a focus on the Slovenian COBISS.SI system, and suggests possibilities for the future development of authority records. It presents the status of the authority databases CONOR.SI and SGC, the submission of authority records to the VIAF service, and the acquisition and use of identifiers for personal and corporate body names. It also outlines IZUM’s plans for the transition to a new catalogue that will be in line with the IFLA LRM conceptual model. The basis for the development of the new COBISS catalogue is the selection of as large a set of entities as possible to be realised in authority records and the establishment of relationships between them.

Keywords

authority data; authority control; COBISS.net; VIAF; IFLA LRM

Hrčak ID:

312448

URI

https://hrcak.srce.hr/312448

Publication date:

10.12.2023.

Article data in other languages: serbian

Visits: 844 *




Normativni podaci u COBISS sistemu i zašto će biti sve važniji

Authority data in the COBISS system and why is its importance increasing

Andreja Krajnc Vobovnik

Institut informacijskih znanosti (IZUM), Maribor, Prešernova 17, Slovenija, andreja.krajnc@izum.si

Primljeno / Received 8. 3. 2023.

Prihvaćeno / Accepted 5. 5. 2023.

Dostupan online / Available online: 12. 10. 2023.

Ključne reči / Keywords:

normativni podaci, normativna kontrola, COBISS.net, VIAF, IFLA LRM

authority data, authority control, COBISS.net, VIAF, IFLA LRM

Sažetak / Abstract

Uvođenjem modela IFLA LRM (engl. IFLA Library Reference model) značaj normativne kontrole postaje sve veći, jer normativni podaci igraju ključnu ulogu u mreži povezanih podataka. Stoga će priprema normativnih podataka doprineti lakšoj implementaciji modela IFLA LRM, a koristiće se prilikom prelaska na RDA (engl. Resource Description and Access).

U izlaganju je predstavljen pregled stanja normativnih podataka u COBISS sistemu s naglaskom na slovenačkom sistemu COBISS.SI, a ukazane su i mogućnosti budućeg razvoja normativnih zapisa. Prikazano je stanje u normativnim bazama podataka CONOR.SI i SGC.SI, slanje normativnih zapisa u VIAF servis, kao i pribavljanje i upotreba identifikatora za lična imena i nazive korporacija. Naveli smo i svoje planove za prelazak na novi katalog koji će biti usklađen sa konceptualnim modelom IFLA LRM. Osnov za pripremu novog COBISS kataloga predstavlja izbor što većeg skupa entiteta koji se realizuju u normativnim zapisima i uspostavljanje odnosa između njih.

With the introduction of the IFLA Library Reference Model (IFLA LRM), authority control is becoming increasingly important, as authority data will play a key role in the network of linked data. Therefore, the preparation of authority data will help to facilitate the implementation of IFLA LRM and will be useful in the transition to RDA (Resource Description and Access).

The paper presents an overview of authority data in the COBISS system, with a focus on the Slovenian COBISS.SI system, and suggests possibilities for the future development of authority records. It presents the status of the authority databases CONOR.SI and SGC, the submission of authority records to the VIAF service, and the acquisition and use of identifiers for personal and corporate body names. It also outlines IZUM’s plans for the transition to a new catalogue that will be in line with the IFLA LRM conceptual model. The basis for the development of the new COBISS catalogue is the selection of as large a set of entities as possible to be realised in authority records and the establishment of relationships between them.

Uvod

Normativnu kontrolu moramo razmatrati u kontekstu kataloga biblioteke i njegovih suštinskih funkcija (Wiederhold i Reeve, 2021). U postupcima normativne kontrole katalogizatori u skladu s kataloškim pravilima oblikuju normativni oblik entiteta (tzv. normativnu pristupnu tačku) koji je unikatan za svaki entitet i koji međusobno povezuje sve različite oblike opisivanog entiteta. Time se u katalogu obezbeđuje nedvosmislena identifikacija entiteta (lična imena, nazivi korporacija, naslovi ili sadržinske oznake) i uspostavlja veza s pripadajućim bibliografskim zapisima. Normativni podaci imaju veliki značaj pri katalogizaciji, naročito u okruženju uzajamne katalogizacije, kao što je to slučaj u COBISS sistemu. U okruženju uzajamne katalogizacije i uzajamnih kataloga normativna kontrola je preko potrebna kako se izrazi za iste entitete ne bi duplirali i koristili za povezivanje zapisa iz različitih izvora (Tillet, 1989).

Danas je tehnološki razvoj toliko uznapredovao da se kataloški postupci povlače iz biblioteka i prenose na izdavače, pojedince ili određene specijalizovane grupe. Katalogizatori u bibliotekama, za razliku od ostalih, verifikacijom i validacijom uspostavljaju normativne pristupne tačke za različite vrste entiteta koji se koriste pri opisu izvora (Wiederhold i Reeve, 2021).

Od kraja XX veka, pod okriljem IFLA-e, razvija se novi konceptualni model bibliografskog univerzuma i osnova za razvoj novih bibliografskih informacija koje su prilagođene novim uslovima. Reč je o važnom koraku koji bibliotečku katalogizaciju postavlja na nove temelje (Žumer, 2017). Na osnovu prethodnih modela 2017. godine objavljen je IFLA Library Reference Model (IFLA LRM). Reč je o modelu bibliotečkih podataka koji povezuje korisničke upite sa bibliografskim i normativnim metapodacima definišući identitete i odnose između njih (Wiederhold i Reeve, 2021).

Čemu služi normativna kontrola?

Katalog biblioteke omogućava pregled građe biblioteke i organizovan je s obzirom na svoju funkciju uvažavajući sledeće zahteve korisnika:

  • da li biblioteka ima publikaciju s određenim naslovom i određenog autora?

  • koja dela nekog autora se nalaze u biblioteci?

  • koja izdanja toga dela se nalaze u biblioteci?

  • da li biblioteka ima publikaciju s određenim sadržajem? (Dimec, 2002)

Normativna kontrola je mehanizam pomoću kojeg se postiže određeni nivo standardizacije u katalogizaciji jer upravo upotreba kontrolisanih elemenata doprinosi tačnijim i doslednijim opisima i na taj način omogućava bibliografsku kontrolu.

Kako bi efikasno odgovorio na ove zahteve, katalog biblioteke mora da uključuje normativnu kontrolu (Tillet, 1989) koja objedinjuje dva postupka (Dimec, 2002):

  1. određivanje elemenata jedinstvenih po obliku po kojima je katalog organizovan i po kojima pretražujemo u katalogu i

  2. njihova veza sa svim alternativnim oblicima tih elemenata.

Normativna kontrola je moguća za lična imena, nazive korporacija, porodična imena, geografska imena, naslove dela i zbirki, kao i za tematske predmetne oznake (Dimec, 2002).

Za sve entitete je značajno to da u publikacijama mogu da se pojave u različitim oblicima, a naša želja je da različite nazive međusobno povežemo (Dimec, 2002). Prednosti normativne kontrole su sledeće (Hartmann i Wiechmann, 2014):

  • standardizacija pristupnih tačaka za opis publikacija,

  • objedinjeni su svi varijantni oblici imena entiteta,

  • svi varijantni oblici mogu da se koriste za pretraživanje i dovode nas do istog rezultata,

  • kooperativno kreiranje i ponovna upotreba normativnih zapisa.

Vodeće kataloške ustanove odgovorne su u određenoj zajednici za izradu normativne datoteke za svoje autore, nazive korporacija, geografska imena, naslove dela i zbirki i tom radu moraju u potpunosti da se posvete (Dimec, 2002).

Normativne baze podataka u mreži COBISS.net

Koncept normativne kontrole primenjuje se već dugo u svetu, a u zemljama u kojima se koristi COBISS u ovoj oblasti dosta zaostajemo. Stoga je IZUM tokom poslednjih 25 godina intenzivno pripremao uslove za uspostavljanje normativne kontrole.

U Sloveniji se od 2003. godine kreiraju zapisi za lična imena u normativnoj bazi podataka CONOR.SI, a od 2018. godine i za korporacije. Normativni zapisi za predmetne oznake Opšteg predmetnog indeksa COBISS.SI (slov. Splošni geslovnik COBISS.SI – SGC) upisani su u bazu podataka SGC i od aprila 2021. godine koriste se za predmetno označavanje u bibliografskim zapisima. Normativni zapisi za dela i ekspresije i dalje su u razvoju (slika 5). Do kraja 2022. godine kreirano je približno 1,46 miliona normativnih zapisa za lična imena, oko 22.000 za nazive korporacija i oko 60.000 za predmetne oznake. Udeo bibliografskih zapisa u bazi podataka COBIB.SI, u kojoj su sva polja za lična imena autora povezana s normativnim zapisima, krajem 2022. godine, nakon skoro 20 godina, dostigao je već 76% (Letno poročilo, 2023). Priprema Opšteg predmetnog indeksa COBISS.SI (SGC) jedan je od većih projekata IZUM-a tokom poslednjih godina. SGC je univerzalni kontrolisani indeks koji sadrži normativne zapise za različite vrste entiteta, od tematskih entiteta, imena, geografskih i vremenskih entiteta do naslova i oblika. Između odrednica uspostavljeni su semantički odnosi jer su u zapise dodate relacije koje usmeravaju na šire, uže i srodne izraze (Zalokar, 2018). Krajem aprila 2023. godine u bazi podataka COBIB.SI bilo je oko 76.000 bibliografskih zapisa povezanih sa SGC. U novim zapisima koji su kreirani u bazi podataka COBIB.SI udeo zapisa koji su povezani sa SGC stalno raste i od maja 2022. godine iznosi oko 18%.

U ostalim državama koje učestvuju u mreži COBISS.net, normativna kontrola ličnih imena uspostavljena je u Albaniji, Bugarskoj, Crnoj Gori, Kosovu i Srbiji. Bosna i Hercegovina i Makedonija su u pripremnoj fazi. U Bugarskoj održavaju i odvojenu bazu predmetnih oznaka, a veza sa bibliografskim zapisima još nije uspostavljena. U svim sistemima koji su uspostavili normativnu kontrolu izuzev Slovenije krajem 2022. godine kreirano je oko 686.000 normativnih zapisa za lična imena i oko 12.000 zapisa za predmetne oznake.

Normativni zapisi za sve vrste entiteta u mreži COBISS.net kreirani su u formatu COMARC/A, a nacionalna pravila za oblikovanje normativnih pristupnih tačaka su različita. Podaci iz normativnih zapisa u bazi podataka CONOR objavljeni su i u COBISS+ (https://plus.cobiss.net/cobiss/si/sl/conor), a primeri za lično ime prikazani su na slikama 1 i 4. U Sloveniji kao osnov za oblikovanje normativne pristupne tačke (lično ime) izabrano je ime pod kojim autor svoja dela najčešće objavljuje. Svi ostali oblici ličnog imena koji se pojavljuju na publikacijama razmatraju se kao varijantni oblici i predstavljaju osnov za oblikovanje varijantnih pristupnih tačka (varijante ličnog imena). U zapisu se nalaze i podaci o jeziku (jezik kojim se koristi osoba/korporacija), nacionalnosti osobe (država), identifikatori (šifra istraživača, ISNI, ORCID), kao i podaci o biografiji i delokrugu osobe koji su objavljeni u javno dostupnim izvorima (napomene).

image1.png

Slika 1. Slovenački normativni podaci za lično ime Josip Osti u COBISS+

Normativna kontrola u sistemu COBISS temelji se na povezivanju bibliografskih zapisa sa normativnim zapisima. U bibliografskom zapisu polja za osobe i korporacije povezane su sa odgovarajućim normativnim zapisom preko identifikacionog broja normativnog zapisa (potpolja 7XX3) u bloku za podatke o odgovornosti. Na osnovu veze između bibliografskog zapisa i normativnog zapisa, pored normativnog oblika imena, programski se dodaju i svi varijantni oblici tog imena (slika 2). Stoga, prilikom pretraživanja po bilo kojem obliku imena dobijamo sva dela nekog autora.

200 0 aSnokradica fJosip Osti

700 1 3827747 aOsti bJosip f1945-2021 4070 - autor

900 1 3827747 aOsti bJosop f1945-2021

900 0 3827747 aJ. O. f1945-2021

Slika 2. Deo bibliografskog zapisa s vezom na normativne podatke u bazi CONOR.SI

Slanje zapisa u VIAF i identifikatori u normativnim zapisima u bazi podataka CONOR.SI

Virtual International Authority File (VIAF) je servis OCLC-a koji uključuje normativne zapise različitih biblioteka, prevashodno nacionalnih biblioteka, ali i drugih ustanova u globalnoj bazi podataka otvorenog pristupa na internetu (http://viaf.org/) . Članovima VIAF-a novembra 2018. godine pridružila se i Slovenija. U VIAF se dvaput godišnje šalju normativni zapisi za lična imena i nazive korporacija s pripadajućim bibliografskim zapisima koji se u VIAF-u razmatraju kao dodatni elementi za identifikaciju entiteta. Zapisi se u VIAF-u ne kreiraju, već se iz ustanova koje učestvuju objedinjuju u skupove (engl. clusters). Na slici 3 prikazan je primer skupa za lično ime književnika Josipa Ostija u kojem su navedeni normativni oblici imena različnih nacionalnih biblioteka i drugih ustanova : s godinama rođenja i smrti, samo s godinom rođenja, bez godina, kao i ćiriličnim pismom. Učestalost određenog oblika imena vidimo na osnovu broja ikona pored imena. U skupu su navedeni i varijantni oblici imena, autorova dela, koautori, ISNI1 (međunarodni standardni identifikator imena), ako postoji, i drugi podaci (v. link: https://viaf.org/viaf/46778771/ ).

image2.png

Slika 3. Skup za lično ime Josip Osti u VIAF-u

Nakon svakog slanja u VIAF programski upisujemo VIAF ID i ISNI u normativne zapise u bazi podataka CONOR.SI. ISNI je, naime, globalni identifikator javnih identiteta osoba i organizacija i čini deo porodice međunarodnih standardnih identifikatora, kao što su ISBN, ISSN, ISAN, DOI i drugi (ISNI, 2013). Za četiri godine, na osnovu upisa povratnih podataka iz VIAF, ISNI je dobilo 53% poslatih zapisa za lična imena.

U normativnim zapisima za lična imena slovenačkih istraživača, pored ISNI, navodimo još dva identifikatora: šifru istraživača i ORCID. Šifra istraživača je najstariji identifikator koji smo još pre 25 godina počeli da koristimo u sistemu COBISS.SI za identifikaciju istraživača. Šifra je povezana s podacima o istraživačkoj delatnosti pojedinačnog istraživača preko SICRIS sistema (Informacijski, 2022), odnosno E-CRIS u mreži COBISS.net. ORCID (Open researcher and contributor ID) je identifikator srodan identifikatoru ISNI koji se koristi u istraživačkoj sferi za identifikaciju naučnika i istraživačkih organizacija (About, 2022). Istraživači mogu sami da dodaju ORCID u svoj profil u SICRIS sistemu, a u normativne zapise upisuju ga katalogizatori.

Dodavanjem identifikatora u normativne zapise omogućen je pristup drugim izvorima normativnih podataka. Trajni identifikatori smatraju se, naime, ključnim za prelazak sa trenutnih metapodataka na buduće aplikacije, odnosno za povezane podatke (engl. linked data). U COBISS+ na osnovu upisa u normativne zapise već su uspostavljene veze sa servisima SICRIS, ISNI i VIAF, a preko normativne pristupne tačke (odnosno ličnog imena u primeru na slici 4) i sa bibliografskim zapisima u bazi COBIB.SI.

image3.png

Slika 4. Slovenački normativni podaci istraživača Matjaža Perca u COBISS+

IFLA Library Reference Model (IFLA LRM)

Tokom poslednjih 25 godina u bibliotekarstvu je došlo do revolucionarnih promena u razumevanju bibliografskog univerzuma koji katalogizaciju postavlja na nove osnove. Nov konceptualni model pod nazivom Funkcionalni zahtevi za bibliografske zapise (FRBR) objavljen je 1998. godine, a dopunjuju ga docnije razvijeni modeli Funkcionalni zahtevi za normativne podatke (FRAD, 2009) i Funkcionalni zahtevi za predmetne normativne podatke (FRSAD, 2011). Oba modela detaljno opisuju normativni deo bibliografskih podataka (Žumer, 2017). IFLA Library Reference Model (IFLA LRM), koji je u jedan model objedinio i uskladio sva tri prethodna modela, objavljen je 2017. godine.

Bibliografski podaci se nisu promenili dok IFLA LRM model predstavlja novi teorijski okvir za razumevanje prirode bibliografskih podataka. Model se ne usredsređuje na pojedinačni zapis, već na skup bibliografskih podataka u katalozima biblioteka, bazama podataka i na internetu. Usredsređuje se na aspekte koji su najznačajniji pri otkrivanju izvora (engl. resource discovery) koje korisnici izvode. Model ne daje sve detalje, već nudi opštu strukturu. Kao standard IFLA-e predstavlja međunarodni sporazum o strukturi bibliografskih informacija i osnov za:

  • razvoj novih aplikacija,

  • razvoj kataloških standarda,

  • razmenu podataka između različitih zajednica (između biblioteka i van biblioteka),

  • razvoj okruženja povezanih podataka (Oliver, 2021).

IFLA LRM uključuje sve vrste izvora i koristi entitetsko-relacioni okvir. U entitetsko-relacionom modelu koriste se tri elementa: entiteti, atributi i relacije. Entiteti se definišu kao ključni objekti zanimanja korisnika bibliotečko-informacionih sistema. Međusobno su povezani relacijama, a njihova svojstva opisuju atributi. Model uključuje 11 različitih entiteta (tabela 1). Na najvišem nivou je entitet res, a svi ostali su njegove neposredne (drugi nivo) ili posredne (treći nivo) potklase. Entiteti osoba i kolektivni agent predstavljaju potklase entiteta agent (Riva et al., 2017).

Tabela 1. Hijerarhija entiteta IFLA LRM modela (Riva et al., 2017)

Najviši nivo Drugi nivo Treći nivo
LRM-E1 res
LRM-E2 delo
LRM-E3 ekspresija
LRM-E4 manifestacija
LRM-E5 jedinica
LRM-E6 agent
LRM-E7 osoba
LRM-E8 kolektivni agent
LRM-E9 nomen
LRM-E10 mesto
LRM-E11 vremenski raspon

Semantički internet dramatično je izmenio tehnološki kontekst u kojem funkcionišu biblioteke. Od suštinskog je značaja da se bibliotečki podaci mogu koristiti i u internetskim aplikacijama. IFLA LRM je osmišljen tako da podržava pripremu dobro oblikovanih podataka koji imaju konzistentnu strukturu i koje mogu da prepoznaju i ljudi i računari. Podaci nisu samo strukturisani, moraju da sadrže i međusobne relacije. To je suština povezanih podataka: podaci koji su međusobno povezani relacijama i koji su mašinski čitljivi primenom jedinstvenih identifikatora izvora (URI) ili međunarodnih identifikatora izvora (IRI). Povezani podaci mogu da objedinjuju podatke iz različitih domena (Oliver, 2021; Riva i Žumer, 2017).

IFLA LRM predstavlja veliki korak napred, jer smo njime dobili kompletan model koji će biti, odnosno koji je već osnov za razvoj kataloških pravila i bibliografskih formata (Žumer, 2017). Različite zajednice biblioteka pripremile su svoje pravilnike za katalogizaciju na osnovu IFLA LRM. Hrvatski Pravilnik za opis i pristup građi u knjižnicama, arhivima i muzejima izašao je 2021. godine. U međunarodnom okruženju najprihvaćeniji je Resource description and access (RDA) koji je objavljen 2010. godine, a 2021. godine usklađen je sa IFLA LRM (tzv. 3R projekat). RDA se u svetu naveliko koristi i van anglosaksonskog sveta u različitim zemljama, poput Nemačke, Francuske i skandinavskih zemalja, a u RDA Europe uključene su već 22 zemlje.

Entiteti IFLA LRM i normativni podaci u sistemu COBISS.SI

IFLA LRM je model za buduće bibliografske informacije. Normativni zapisi će se ubuduće koristiti i za entitetno delo, jer entitete tog modela moramo da opišemo kao normativne podatke. Trenutno uređujemo normativne podatke, pri čemu normativne podatke ne izjednačavamo sa entitetima modela IFLA LRM. Međutim, normativni podaci su kvalitetan osnov za budući prelazak na rad s entitetima. Upoređujući model IFLA LRM sa normativnim podacima koje razvijamo u sistemu COBISS.SI utvrdili smo da je mnogo atributa i relacija između njih već zabeleženo u našim normativnim zapisima koji su u normativnim zapisima navedeni kao napomene, a ne kao jedinstveni izrazi atributa. Neki podaci i dalje nedostaju i treba ih dodati.

Normativni zapisi za lična imena i nazive korporacija predstavljaju osnov za entitet agent. Normativni zapisi za dela i ekspresije još su u razvoju. Planiramo da ih za sada još ne oblikujemo u potpunosti prema IFLA LRM modelu, ali samo onoliko koliko to dopušta programska transformacija iz bibliografskih zapisa. U SGC-u predmetne oznake smo podelili na najveći broj entiteta koji se pojavljuju u modelu IFLA LRM. Relacije su uglavnom usklađene s modelom, a s modelom je usklađena i struktura podataka (slika 5).

image4.png

Slika 5. Entiteti IFLA LRM i normativni podaci u sistemu COBISS.SI

Programsko objedinjavanje zapisa u internetskom katalogu COBISS+

U COBISS+ 2015. godine počeli smo s postupkom restrukturiranja COBISS kataloga prema FRBR modelu (docnije uključen u IFLA LRM model). Bibliografske zapise koji prema ovom modelu pripadaju istom delu programski smo objedinili u tzv. skupove. Pritom smo se ograničili na tri odnosa: prevode, revizije i reprodukcije (Krajnc Vobovnik i Mazić, 2017).

Skupovi zapisa prikazuju se pod nazivom Sva izdanja i prevodi i Sva izdanja na tom jeziku. Prilikom analize podataka iz baze COBIB.SI utvrdili smo da smo skupovima obuhvatili više od polovine bibliografskih zapisa i pritom identifikovali 20% dela koja imaju više od jedne manifestacije (najviše 44% kod beletristike) (Krajnc Vobovnik i Mazić, 2017).

Skupovi su prikazani u svim internetskim katalozima u mreži COBISS.net. Na primer, prilikom pretraživanja u COBISS.BH za Selimovićev roman Derviš i smrt dobijamo 104 rezultata za publikacije na različitim medijima (štampane i audio-knjige) i na tri jezika (bosanskom, srpskom i hrvatskom) jer smo pretraživali po originalnom naslovu. Ako izaberemo zapis za štampanu knjigu koja je izašla 1966. godine, a nakon toga link na Sva izdanja i prevodi (slika 6), dobijamo 117 rezultata koji pripadaju istom delu.

image6.png

Slika 6. Sva izdanja i prevodi u COBISS.BH

U skupu dobijamo zapise za publikacije na različitim medijima (štampane i audio-knjige) i na 17 različitih jezika koji se uvrštavaju u skup na osnovu jedinstvenog naslova. Ako potom filterom dalje ograničimo rezultate po jeziku publikacije, npr. nemački, dobijamo 4 zapisa za prevode s naslovom Der Derwisch und der Tod koji su se u skupu našli na osnovu jedinstvenog naslova Derviš i smrt (slika 7).

image7.png

Slika 7. Nemački prevodi dela Derviš i smrt u COBISS.BH

Objedinjavanje zapisa u skupove bio je naš prvi pokušaj da korisnicima jednim pretraživanjem ponudimo i informaciju o svim izdanjima i prevodima traženog dela. Prilikom grupisanja u skupove obuhvatili smo najširi broj bibliografskih zapisa koji u potpunosti ili delimično imaju isti sadržaj. Objedinjavanjem zapisa u skupove napravili smo prvi korak i postavili smernice za dalji razvoj COBISS kataloga (Krajnc Vobovnik i Mazić, 2017).

Normativni zapisi za dela i ekspresije

Dalji razvoj COBISS kataloga vidimo u uvođenju nove kataloške prakse po IFLA LRM modelu. Oblikovanje normativnih zapisa za dela mnogo je preciznije, a naše skupove zapisa koje smo oblikovali u COBISS+ može da rasporedi na više normativnih zapisa za dela i na još više normativnih zapisa za ekspresije koji su međusobno povezani različitim odnosima. Pritom smo već dostigli dva osnovna uslova:

  • format COMARC/B i format COMARC/A dopunili smo poljima za dela i ekspresije i

  • za njih smo u probnoj bazi podataka CONOR.SI oblikovali dve nove maske za unos.

Planiramo da se priprema inicijalne normativne baze podataka za dela i ekspresije odvija u više faza, i to po grupama dela različitog sadržaja i različitih fizičkih karakteristika. Obe grupe imaju svoje specifičnosti koje treba uvažiti pri oblikovanju zapisa. Na osnovu podataka u bibliografskim zapisima po određenim kriterijumima programski ćemo oblikovati normativne zapise za dela i ekspresije.

Slika 8 prikazuje primer normativnog zapisa za delo Meše Selimovića Derviš i smrt u formatu COMARC/A. Vrsta entiteta je delo kod kojeg su ime autora i naslov dela glavni podaci za autorsko delo. U normativnom zapisu za delo upućujemo i na normativni zapis za autora kod kojeg je upisan i njihov odnos (u našem slučaju kreator). U normativnim zapisima za dela upućujemo i na druga dela Meše Selimovića, odnosno prerade izvornog dela, čime pravimo informacionu mrežu o delu Derviš i smrt:

  • delo je izašlo i u sabranim delima Meše Selimovića

  • video-snimak filma scenarista Zdravka Velimirovića i Borislava Mihajlovića koji je izašao na video-kaseti

001 an - novi zapis bx - normativni zapis ch - ime/naslov

100 bc - privremena cslv - slovenački gba - latinica

101 abos - bosanski

105 aba - latinica

12500 am - odrasli, opšte (lepa književnost)

140 ate - tekstualno delo

154 ba - delo

241_1 31155685 aSelimović, Meša, 1910-1982 tDerviš i smrt

500_1 31155685 aSelimović bMeša f1910-1982 5xxxxa - kreator

541_1 3399268869 aSelimović, Meša, 1910-1982 tSabrana dela u deset knjiga. Knj. 4,

Derviš i smrt: roman czbrana dela

541_1 3399270651 aVelimirović, Zdravko, 1930-2005 aMihajlović, Borislav, 1922-1997

tDerviš i smrt cVideoposnetek 5xc - priredba

Slika 8. Normativni zapis za delo Derviš i smrt u formatu COMARC/A

U normativnim zapisima za ekspresije beležimo intelektualno ili umjetničko izvođenje dela u obliku teksta, notnog ili kartografskog zapisa, zvuka, slike, predmeta, pokreta itd. ili njihove kombinacije. Svaka promjena oblika istog dela (npr. iz štampanog teksta u izgovoreni), sadržaja (npr. dopune i ispravke teksta) i prevoda obično zahteva nastanak novog normativnog zapisa za ekspresiju. (Riva et al., 2017). Slika 9 prikazuje primer normativnog zapisa za audio-knjigu Meše Selimovića Derviš i smrt u formatu COMARC/A. Vrsta entiteta je ekspresija kod kojeg su glavni podaci za autorsko delo ime autora, naslov dela i oblik, što je u našem slučaju audio-knjiga.

001 an - novi zapis bx - normativni zapis ch - ime/naslov

100 bc - privremena cslv - slovenački gba - latinica

101 abos - bosanski

105 aba - latinica

12500 am - odrasli, opšte (lepa književnost)

140 aso - zvučno delo

154 bb - ekspresija

242_1 3399268615 31155685 aSelimović, Meša, 1910-1982 tDerviš i smrt lAudio knjiga

Slika 9. Normativni zapis za ekspresiju za e-knjigu Derviš i smrt u formatu COMARC/A

Zaključak

Uvođenjem IFLA LRM modela značaj normativne kontrole postaje sve veći jer će normativni podaci imati ključnu ulogu u mreži povezanih podataka na koje se model usredsređuje.

Tokom perioda pripreme na novi katalog naš najvažniji zadatak je da i dalje intenzivno pripremamo normativne zapise. Sistemi bez normativne kontrole već zaostaju za savremenim trendovima i potrebama korisnika.

U zemljama sa jakom izdavačkom produkcijom, kao i kod nas, vide se trendovi da se običnom katalogizacijom bave izdavači, pojedinci i određene specijalizovane grupe van bibliotekarstva. Biblioteke moraju na sebe da preuzmu oblikovanje najkvalitetnijih i trajnih informacija, za šta je potrebno mnogo više znanja. To su danas normativni podaci, sutra entiteti, a i jedni i drugi su kamen temeljac bibliografskih informacija.

Biblioteke raspolažu istorijom podataka jer u bibliotekama počiva istorija nekog dela, a izdavači imaju samo trenutne podatke. Istorija koja se nalazi u normativnim podacima (npr. sva izdanja nekog dela) čuva se, naime, u našim fondovima. A bibliotekari su specijalizovani stručnjaci za važniji deo katalogizacije – za oblikovanje normativnih podataka.

Introduction

Authority control must be understood in the context of the library catalogue and its essential functions (Wiederhold and Reeve 2021). In authority control procedures, cataloguers create an authority entity form (called an authority access point) in accordance with the cataloguing rules, which is unique for each entity and interconnects all the different forms of the described entity. This ensures that entities are unambiguously identified in the catalogue (personal names, corporate body names, titles or subject headings) and that they are linked to the corresponding bibliographic records. Authority data is of great importance in cataloguing, especially in a shared cataloguing environment such as COBISS. In a shared cataloguing and union catalogue environment, authority control is essential to ensure that terms for the same entities are not duplicated and are used to link records from different sources (Tillet 1989).

Today, technological development has advanced to such an extent that cataloguing processes are consequently being withdrawn from libraries and transferred to publishers, individuals or certain specialised groups. In libraries, however, cataloguers, unlike others, establish authority access points for the different types of entities used in describing resources through verification and validation (Wiederhold and Reeve 2021).

Since the end of the 20th century, a new conceptual model of the bibliographic universe has been developed under the auspices of IFLA, and the basis for the development of new bibliographic information adapted to the new context. This is an important step that puts library cataloguing on a new footing (Žumer 2017). Based on previous models, the IFLA Library Reference Model (IFLA LRM) was published in 2017. It is a library data model that links user query tasks with bibliographic and authority metadata by defining entities and the relationships between them (Wiederhold and Reeve 2021).

Why is authority control necessary?

The library catalogue provides an overview of library material and is organised according to its function, which is based on the following user requirements:

  • Does the library have a publication with a specific title and a specific author?

  • Which works by a particular author are held in the library?

  • Which editions of this work are in the library?

  • Does the library have a publication with a certain content? (Dimec 2002)

To effectively meet these requirements, the library catalogue must have authority control (Tillet 1989) that combines two processes (Dimec 2002, 93):

  1. Defining the uniformly formed elements, according to which the catalogue is organised and by which you can search, and

  2. Their link to all alternative forms of these elements.

Authority control is a mechanism by which a certain degree of standardisation is achieved in cataloguing, since the use of controlled elements helps to make descriptions more precise and consistent and thus enables bibliographic control.

Authority control is possible by personal, corporate body, family and geographical names, titles of works and collections, and topical subject headings (Dimec 2002).

It is characteristic of all entities that they can appear in publications in different forms, and we want to link the different names together (Dimec 2002). The advantages of authority control are as follows (Hartmann and Wiechmann 2014):

  • Standardisation of access points for describing publications,

  • All variant forms of the entity name are together,

  • All variant forms can be used for searching and lead to the same result,

  • Cooperative creation and reuse of authority records.

The leading cataloguing institutions in a given community are responsible for creating an authority file for their authors, corporate body names, geographical names, titles of works and collections, and must devote themselves thoroughly to this work (Dimec 2002).

Authority databases in the COBISS.net network

The concept of authority control has long been established in the world, but in the countries where COBISS is used, we have been lagging far behind in this area. For this reason, IZUM has spent the last 25 years intensively preparing the conditions for the establishment of authority control.

In Slovenia, we have been creating records for personal names in the CONOR.SI authority database since 2003 and for corporate body names since 2018. The authority records for the subject headings of the COBISS.SI General List of Subject Headings are registered in the SGC database and have been used for adding subject headings to bibliographic records since April 2021. The authority records for works and forms of expression are still under development (Figure 5). At the end of 2022, around 1.46 million authority records have been created for personal names, around 22,000 for corporate body names and around 60,000 for subject headings. The proportion of bibliographic records in the COBIB.SI database, where all fields for authors’ personal names were linked to authority records, had already reached 76% by the end of 2022, after almost 20 years of use (Annual Report 2023). The development of the COBISS.SI General List of Subject Headings (SGC) is one of IZUM’s major projects in recent years. The SGC is a universal controlled vocabulary and contains authority entries for different types of entities, from subject, name, geographic, temporal subject headings to titles and forms. Semantic relations are established between entries, as relations are added to the entries to point to broader, narrower and related terms (Zalokar, 2018). At the end of 2022, there were around 56,000 bibliographic records linked to SGC in the COBIB.SI database. In the new records created in the COBIB.SI database, the share of SGC-linked records is steadily increasing and as of May 2022 is around 18%.

In the other COBISS.net countries, authority control for personal names is established in Albania, Bulgaria, Montenegro, Kosovo and Serbia (and is under development in Bosnia and Herzegovina and North Macedonia). In Bulgaria, a separate database of subject headings is also maintained, but the link to the bibliographic records is not yet established. In all systems with authority control in place, except Slovenia, a total of around 638,000 authority records for personal names and around 12,000 records for subject headings have been created by the end of 2022.

Authority records for all types of entities are created in the COBISS.net network in the COMARC/A format, and national rules for creating authority access points vary. Data from authority records in the CONOR database is also published in COBISS+ (https://plus.cobiss.net/cobiss/si/sl/conor), examples for personal names are shown in Figures 1 and 4. In Slovenia, the basis for creating an authority access point (Personal name) is the name chosen by the author under which their works are most often published. All other forms of the personal name appearing on publications are considered as variant forms and are the basis for the creation of variant access points (Variants of the personal name). The record also contains information on the language (Language used by the person/corporate body) and nationality of the person (Country), identifiers (Researcher code, ISNI, ORCID) and information on the person+s biography and activities published in publicly available sources (Notes).

image8.png

Figure 1. Slovenian authority data for the personal name Josip Osti in COBISS+

Authority control in COBISS is based on linking bibliographic records with authority records. In a bibliographic record, the Person and Corporate body fields in the Responsibility Data Block are linked to the corresponding authority record via the authority record identification number (subfields 7XX3). By linking the bibliographic record to the authority record, all variant forms of the name are automatically added in addition to the authority form of the name (Figure 2). Therefore, a search for any form of the name will retrieve all works by a given author.

200 0 aSnokradica fJosip Osti

700 1 3827747 aOsti bJosip f1945-2021 4070 - author

900 1 3827747 aOsti bJosop f1945-2021

900 0 3827747 aJ. O. f1945-2021

Figure 2. Part of the bibliographic record with link to authority data in the CONOR.SI database

Sending records to VIAF and identifiers in authority records in the CONOR.SI database

The Virtual International Authority File (VIAF) is OCLC’s service and includes the authority records of various libraries, mainly national, and other institutions in a global database freely accessible online at http://viaf.org/. Slovenia joined the VIAF members in November 2018. Authority records for personal and corporate body names with the corresponding bibliographic records are submitted to VIAF twice a year, which are treated as additional elements for entity identification in VIAF. The records are not created in VIAF, but the records of the participating institutions are grouped in clusters. Figure 3 shows an example of a cluster for the personal name of the writer Josip Osti, where the authority forms of the names of the various national libraries and other institutions are listed: with the years of birth and death, with the year of birth only, without the years, and also in the Cyrillic script. The frequency of a particular form of a name can be seen from the number of icons next to the name. The cluster also lists variant forms of the name, the author’s works, co-authors, ISNI (International Standard Name Identifier2 ), if existing, and other information (see link https://viaf.org/viaf/46778771/ ).

image2.png

Figure 3. Cluster for the personal name Josip Osti in VIAF

After each submission to VIAF, the VIAF ID and ISNI are entered automatically into the authority records in the CONOR.SI database. ISNI is a global identifier of public identities of persons and organisations and is part of the family of international standard identifiers such as ISBN, ISSN, ISAN, DOI and others (ISNI 2013). In four years, ISNI has retrieved 53% of the records sent for personal names, based on the VIAF return data entry.

In the authority records for personal names of Slovenian researchers, we list two identifiers in addition to the ISNI: researcher code and ORCID. The researcher code is the oldest identifier, which we started using 25 years ago in COBISS.SI to identify researchers. The code is linked to the research activity data of the individual researcher via SICRIS (Informacijski, 2022) or E-CRIS on COBISS.net. ORCID (Open Researcher and Contributor ID) is ISNI’s related identifier, which is well established in the research sphere for the identification of scientists and research organisations (About, 2022). Researchers can add ORCID to their SICRIS profile themselves, and it is entered in authority records by cataloguers.

Adding identifiers to authority records allows access to other sources of authority data. Permanent identifiers are seen as key to the transition from current metadata to future applications or linked data. In COBISS+, links to SICRIS, ISNI and VIAF services are already established on the basis of the authority record entry, and also to bibliographic records in the COBIB.SI database via the authority access point (or personal name in the case of Figure 4).

image9.png

Figure 4. Slovenian authority data for the researcher Matjaž Perc in COBISS+

IFLA Library Reference Model (IFLA LRM)

The last 25 years have seen revolutionary changes in the understanding of the bibliographic universe in librarianship, which puts cataloguing on a new footing. In 1998, a new conceptual model, Functional Requirements for Bibliographic Records (FRBR), was published, complemented by the later developed models Functional Requirements for Authority Data (FRAD, 2009) and Functional Requirements for Subject Authority Data (FRSAD, 2011). Both models describe the authority part of bibliographic data in more detail (Žumer, 2017). In 2017, the IFLA Library Reference Model (IFLA LRM) was published, which combined and harmonised all three previous models into one model.

While the bibliographic data has not changed, the IFLA LRM model represents a new theoretical framework for understanding the nature of bibliographic data. The model does not focus on a single record, but on the mass of bibliographic data in library catalogues, databases and on the web. It focuses on the aspects that are most important in resource discovery by users. The model does not give all the details, but a general structure. As an IFLA standard, it represents an international agreement on the structure of bibliographic information and is the basis for the development of:

  • New applications,

  • Cataloguing standards,

  • Data exchange between different communities (between libraries and outside of libraries),

  • Linked data environments (Oliver, 2021).

IFLA LRM includes all types of resources and uses an entity-relational framework. The entity-relational model uses three elements: entities, attributes and relations. Entities are defined as the key objects of interest for users of library information systems. They are linked to each other by relations and their properties are described by attributes. The model includes 11 different entities (Table 1). At the highest level, the entity is Res, and all the others are its direct (second level) or indirect (third level) subclasses. The entities Person and Collective Agent are subclasses of the entity Agent (Riva et al., 2017).

Table 1. IFLA LRM model hierarchy of entities (Riva et al., 2017)

Top level Second level Third level
LRM-E1 Res
LRM-E2 Work
LRM-E3 Expression
LRM-E4 Manifestation
LRM-E5 Item
LRM-E6 Agent
LRM-E7 Person
LRM-E8 Collective Agent
LRM-E9 Nomen
LRM-E10 Place
LRM-E11 Time span

The Semantic Web has dramatically changed the technological context in which libraries operate. The main aspect is that library data can also be used in web applications. IFLA LRM is designed to support the production of well-formed data that has a consistent structure and that can be recognised by both humans and computers. Data is not only structured, but also contains relations between them. This is the essence of Linked Data: data that is linked together by relations and is machine-readable using Uniform Resource Identifiers (URIs) or International Resource Identifiers (IRIs). Linked Data can combine data from different domains (Oliver, 2021, Riva and Žumer, 2017).

IFLA LRM is a major step forward, providing a comprehensive model that will be, or is, the basis for the development of cataloguing rules and bibliographic formats (Žumer, 2017). Different library communities have already developed their own cataloguing policies based on IFLA LRM. In 2021, the Croatian Pravilnik za opis i pristup građi u knjižnicama, arhivima i muzejima (Rules for Description and Access to Cataloguing in Libraries, Archives and Museums) was published. Internationally, the most established is Resource Description and Access (RDA), published in 2010 and aligned with the IFLA LRM in 2021 (known as the 3R project). Globally, RDA is also widely used outside the Anglo-Saxon world in various countries such as Germany, France, and the Scandinavian countries, and 22 countries are already involved in RDA Europe (RDA 2022).

IFLA LRM entities and authority data in the COBISS.SI system

IFLA LRM is a model for future bibliographic information. Authority records will also be used for entity work in the future, as we need to describe the entities of this model as authority data. We are currently editing authority data, and we do not equate authority records with IFLA LRM entities, but they are a quality basis for the future transition to entity work. When comparing the IFLA LRM model with the authority data we develop in the COBISS.SI system, we found that many of the attributes and relationships between them are already recorded in our authority records. However, these are listed in the authority records as notes and not as unique attribute expressions. Some information is still missing and will have to be added.

image10.png

Figure 5. IFLA LRM entities and authority data in the COBISS.SI system

Automatic record merging in the COBISS+ web catalogue

In 2015, we started the process of restructuring COBISS catalogues in COBISS+ according to the FRBR model (which was later incorporated into the IFLA LRM model). Bibliographic records that belong to the same work according to this model were automatically grouped into clusters. In doing so, the clusters were limited to three relationships: translations, revisions and reproductions (Krajnc Vobovnik and Mazić, 2017).

The record clusters are displayed under the title All editions and translations and All editions in this language. When analysing the data in the COBIB.SI database, we discovered that we included just over half of the bibliographic records in the clusters, identifying 20% of works with more than one occurrence form (maximum 44% for fiction) (Krajnc Vobovnik and Mazić, 2017).

The clusters are displayed in all online catalogues in the COBISS.net network. For example, a search in COBISS.BH for Selimović's novel Derviš i smrt yields 104 hits for publications in different media (print and audiobooks) and in three languages (Bosnian, Serbian or Croatian), because we searched by the original title. Selecting the record for a printed book published in 1966 and then the link for All editions and translations (Figure 6) gives 117 hits belonging to the same work.

image11.png

Figure 6. All editions and translations in COBISS.BH

The cluster contains records for publications in different media (printed and audio books) and in 17 different languages, which are classified in the cluster on the basis of the uniform title. If we then further restrict the hits by the language of the publication, e.g. German, via a filter, we get 4 records for translations with the title Der Derwisch und der Tod, which were added to the cluster on the basis of the single title Derviš i smrt (Figure 7).

image12.png

Figure 7. German translations of the work Derviš i smrt in COBISS.BH

Combining records into clusters was our first attempt to provide users with information on all editions and translations of the work they are looking for in a single search. When grouping records into clusters, we have covered the widest possible set of bibliographic records that contain all or part of the same content. By grouping records into clusters, we have taken a first step and set the guidelines for the further development of COBISS catalogues (Krajnc Vobovnik and Mazić 2017).

Authority records for works and expressions

We see the further development of COBISS catalogues in the introduction of a new cataloguing practice according to the IFLA LRM model. The creation of authority records for works is much more precise and we can distribute our record clusters created in COBISS+ across several authority records for works and even more authority records for expressions, which are linked to each other by different relationships. We have already achieved two basic conditions:

  • COMARC/B and COMARC/A formats have been supplemented with fields for works and expressions, and

  • Created two new input masks for them in the CONOR.SI test database.

We plan to develop the initial authority database for works and expressions in several phases, i.e. by groups of works of different content and by different physical characteristics. Both groups have their own specifics that need to be taken into account when creating the records. On the basis of the data in the bibliographic records, authority records for works and expressions will be created automatically following certain criteria.

Figure 8 is an example of an authority record for Meša Selimović's work Derviš i smrt in the COMARC/A format. The entity type is work, where the main data for copyright works are the author's name and the title of the work. In the authority record for the work, we also refer to the authority record for the author, where their relationship (in our case the creator) is also entered. In the authority records for the works, we also refer to other works by Meša Selimović, or adaptations of the original work, thus creating an information network about the work Derviš i smrt:

  • The work was also published in the collected works of Meša Selimović,

  • A video clip of the film by scriptwriters Zdravko Velimirović and Borislav Mihajlović, released on videocassette.

001 an - new record bx - authority record ch - name/title

100 bc - provisional cslv - Slovenian gba - Latin

101 abos - Bosnian

105 aba - Latin

12500 am - adult, general

140 ate - textual work

154 ba - work

241_1 31155685 aSelimović, Meša, 1910-1982 tDerviš i smrt

500_1 31155685 aSelimović bMeša f1910-1982 5xxxxa - creator

541_1 3399268869 aSelimović, Meša, 1910-1982 tSabrana dela u deset knjiga. Knj. 4,

Derviš i smrt : roman czbrana dela

541_1 3399270651 aVelimirović, Zdravko, 1930-2005 aMihajlović, Borislav, 1922-1997

tDerviš i smrt cVideoposnetek 5xc - adaptation

Figure 8. Authority record for the work Derviš i smrt in the COMARC/A format

Authority records for expressions record the intellectual or artistic performance of a work in the form of text, musical notation or a cartographic record, sound, pictorial representation, object, movement, etc., or a combination of these. Any change in the form of the same work (e.g. from printed text to spoken word), in its content (e.g. additions and corrections to the text) and in translations usually requires the creation of a new authority record for the expressions (Riva et al., 2017). Figure 9 is an example of an authority record for Meša Selimović's audio book Derviš i smrt in the COMARC/A format. The entity type is an expression where the main data for copyright works are the author's name, the title of the work and the format, which in our case is an audiobook.

001 an - new record bx - authority record ch - name/title

100 bc - provisional cslv - Slovenian gba - Latin

101 abos - Bosnian

105 aba - Latin

12500 am - adult, general

140 aso - sounds work

154 bb - expression

242_1 3399268615 31155685 aSelimović, Meša, 1910-1982 tDerviš i smrt lAudiobook

Figure 9. Authority record for the expression for the audio book Derviš i smrt in the COMARC/A format

Conclusion

With the introduction of the IFLA LRM model, the importance of authority control is increasing, as authority data will play a key role in the network of linked data on which the model focuses.

In the run-up to the new catalogue, our most important task is to continue to work intensely on the creation of authority records. Systems without authority control are already lagging behind modern trends and user needs.

In countries with a strong publishing production, and also in our country, there is a trend for simple cataloguing to be handled by publishers, individuals or certain specialised groups outside librarianship. Libraries, on the other hand, have to take on the task of creating the highest quality and most sustainable information, which requires much more knowledge. Today, that is authority records, and tomorrow, it will be entities, both of which are stable building blocks of bibliographic information.

Libraries have access to the history of data, because libraries actually hold the history of a work, while publishers only have current data. The history contained in the authority data (e.g. all editions of a work) is held in library holdings. And librarians are specialists in the most important part of cataloguing – the creation of authority data.

Bibliografija / Bibliography

About ORCID. (2022). ORCID. https://info.orcid.org/what-is-orcid/.

ISNI for Libraries. (2013). ISNI. https://www.loc.gov/aba/pcc/documents/ISNIs%20for%20libraries%202013-08-26.docx.

model for bibliographic information. IFLA. https://www.ifla.org/wp-content/uploads/2019/05/assets/cataloguing/frbr-lrm/ifla-lrm-august-2017.pdf

Riva, P. & Žumer, M. (2017). The IFLA Library Reference Model, a step toward the Semantic Web. World Library and Information Congress: 83th IFLA General Conference and Assembly, August 19th–25th 2017, Wroclaw, Poland: Libraries, Solidarity, Society, 1-8. http://library.ifla.org/1763/1/078-riva-en.pdf.

Notes

[1] International Standard name Identifier – https://isni.org

[2] 1 International Standard name Identifier – https://isni.org

References

 

Dimec, Z. 2002Normativna kontrola iz preteklosti v prihodnost: ali se sedanjost odloča za mednarodno sodelovanje. Knjižnica. 46(3):91–104. https://knjiznica.zbds-zveza.si/knjiznica/article/view/5622

 

Hartmann, S., & Wiechmann, B. 2014RDA and authority files: impact on GND.[Presentation]. IFLA Satellite Meeting on RDA, 2014. Frankfurt am Main, Germany.: https://wiki.dnb.de/download/attachments/132747829/iflaVortragHartmannWiechmannRdaImpact.pdf?version=1&modificationDate=1516195759000&api=v2

 

Informacijski sistem o raziskovalni dejavnosti v Sloveniji. 2022Institut informacijskih znanosti, IZUM. https://cris.cobiss.net/ecris/si/sl/introISNI for Libraries. 2013ISNI. https://www.loc.gov/aba/pcc/documents/ISNIs%20for%20libraries%202013-08-26.docx

 

Krajnc Vobovnik, A., & Mazić, G. 2017Združevanje bibliografskih zapisov v COBISS+: začetek novih katalogov v sistemih COBISS = Clustering bibliographic records in COBISS+. Knjižnica. 6112:87–100. https://knjiznica.zbds-zveza.si/knjiznica/article/view/6163

 

Letno poročilo o delu IZUM za leto 2022. 2023Institut informacijskih znanosti, IZUM. https://izum.si/wp-content/uploads/2023/03/Letno_porocilo_2022_vse_sprejeto_na_UO_2023-02-28.pdf

 

Oliver, C. 2021Introducing RDA: a guide to the basics after 3R. (2nd ed.). ALA Editions.

 

Riva, P., Le Bœuf, P. & Žumer, M. 2017IFLA library reference model: a conceptual model for bibliographic information. IFLA. https://www.ifla.org/wp-content/uploads/2019/05/assets/cataloguing/frbr-lrm/ifla-lrm-august-2017.pdf Riva, P. & Žumer, M. 2017The IFLA Library Reference Model, a step toward the Semantic Web. World Library and Information Congress: 83th IFLA General Conference and Assembly. 2017(Wroclaw, Poland). Libraries, Solidarity, Society,. 1–8. http://library.ifla.org/1763/1/078-riva-en.pdf

 

Tillett, B. B. (Ed.). , editor. 1989Authority control in the online environment: considerations and practices. 9. p. 3Haworth Press.;

 

Wiederhold, R. A., & Reeve, G. F. 2021Authority Control Today: Principles, Practices, and Trends. Cataloging & Classification Quarterly. 5923:129–158. https://www.tandfonline.com/doi/full/10.1080/01639374.2021.1881009

 

Zalokar, M. 2018Nov mejnik v razvoju normativne baze podatkov Splošni geslovnik COBISS.SI. (SGC). Blog COBISS.http://blog.cobiss.si/2018/08/23/nov-mejnik-sgc/

 

Žumer, M. 2017IFLA Library Reference Model (IFLA LRM), pomemben korak k sodobnim bibliografskim informacijskim sistemom = IFLA Library Reference Model (IFLA LRM), an important step towards modern bibliographic information systems. Knjižnica. 6112:9–22. https://knjiznica.zbds-zveza.si/knjiznica/article/view/6160


This display is generated from NISO JATS XML with jats-html.xsl. The XSLT engine is libxslt.