LIS 882 Self-assessment

zurichblueInternet και Εφαρμογές Web

21 Οκτ 2013 (πριν από 3 χρόνια και 9 μήνες)

171 εμφανίσεις










LIS 882 Self
-
assessment

By

Janet Leu


Spring 2012












January 27, 2012 (week 3)



Zeng, Marcia Lei and Jian Qin.
Metadata.
Chapter 2, “Current Standards,” p. 12
-
84.



Miller, Steven J.
Metadata for Digital Collections.
Chapter 2,
“Introduction to Resource
Description and Dublin Core,” p. 25
-
58.



Miller, Steven J.
Metadata for Digital Collections.
Chapter 3, “Resource Identification
and Responsibility Elements,” p. 59
-
88.



Miller, Steven J.
Metadata for Digital Collections.
Chapter 4,

“Resource Content and
Relationship Elements,” p. 89.128.



Baca, Martha, ed. Getty Museum. The Getty Standards Program. “Introduction to
Metadata: Pathways to Digital Information.” Online edition, Version 2.1
(
http://www.getty.edu.research/conducting_research/standards/intrometadata/
>



Caplan, Priscilla. “Syntax, Creation, and Storage,”
Metadata Fundamentals for all
Librarians,
p. 12
-
24.



Weibel, Stuart L. “Border Cros
sings: Reflections on a Decade of Metadata Consensus
Building.”
D
-
Lib

Magazine 11,
no 7/8 (2005).
http://www.dlib.org/july05/weibel/07weibel.html


It’s good to know that I didn’t completely bomb Assignment 1, mostly because we went over the

definitions of digital collections, digital libraries, virtual libraries, and institutional repositories

last week.
One of the reasons I took this class was be
cause I noticed that many cataloging

positions now required experienced and knowledge of metadata, but the first class was very

intimidating. After cataloging for almost two years, metadata sounds like the exact opposite of

what I’m used to learning, e
specially when we went through Dublin Core in class today. Besides

realizing I got the DC.Date element wrong because there’s actually a specific way that needs to

be entered (YYYYMMDD),
it’s going to be hard to get used to a metadata schema that has

re
peating elements and no capitalization rules.
Again, that sounds almost strange after a year of

working under the co
nstraints of AACR2, where capitalizations, punctuation, and spacing rules

are very specific and can’t really be bent.


According to
D
-
Lib

Magazine,
I was a senior in high school when the OCLC
-
NCSA

Metadata workshop took place. In 1995, I had no idea how the Internet would change the face of

publishing, libraries, and commerce. Back then, the Internet was a place for me to chat with

str
ange people all over the world and
find material my parents forbade me to look at. One of the

things I noticed in 1995 was the prevalence of scanned imagery in the newsgroups, which may

have been copyright violations. It’s safe to say that none of thes
e images scanned from various

publications and uploaded to internet newsgroups really regarded intellectual property rights,

which is one of the things metadata is used for, especially in an electronic and digital sense.
But

there was the use of metada
ta in the sense of titles, artists, descriptions, and dates, even in 1995,

as even digital objects that may violate copyright and intellectual property rules still needed data

that

described it. I just had no idea what was actually going on those days.


In Anne J. Gilliland’s “Setting the Stage,” I like how she broke down information objects

into

three features: content, context, and structure


all of which can be reflected through

metadata. These three elements describe a resource, and in this digital age, give information

about a resource’s rights and licensing agreements, which seem to go b
eyond the information a

MARC
record can provide. Right now, I don’t have enough experience to determine which

describes a digital object better: legacy cataloging or metadata?
After reading this piece,

along

with Tony Gill’s
Metadata and the Web,
I also

realized that I’ve had experience with metadata

and not known it, when I came across HTML meta tags which make websites easier to find

within a search engine. Or that the descriptive tags on a photo website like Flickr, which
classify

and describe a p
articular photographic image, is its own form of metadata.









February 3, 2012 (week 4)



Zeng, Marcia Lei and jian Qin.
Metadata. Chapter
3, “Schemas


Structure and
Semantics,” p. 87
-
129.



Miller, Steven J.
Metadata for Digital Collections.
Chapter

6, “XML
-
Encoded
Metadata,” p. 149
-
162.



Intner, Sheila S., Susan S. Lazinger, and Jean Weihs.
Metadata and Its Impact on
Libraries.
“Library and Information Related Metadata Schemas,” p. 73
-
108.



Coyle, Karen. “Library Data in a Modern Context.”
Library Tec
hnology Reports
46, 1
(January, 2010): 5
-
13.


I was very glad that Karen D. Miller visited class today and showed us XML. Finally. It’s what

I’ve been wanting to learn since I started the program almost two years ago. Now that I’ve had

some

HTML experience in LIS 753, XML is no longer so intimidating.
It’s a required skill I’ve

seen on many advertisements for cataloging positions, and now that we got to experience it, it no

longer feels like this mysterious big deal.
Too bad we are only
doing XML this one time for

class. I wish we could discuss and do it in more detail.

I also noticed that Ian Richter took the

initiative and checked out a stack of XML books from the library during break. Guess I’ll need

to

find my resources elsewhere.


After reading Karen Coyle’s “Library Data in a Modern Context,” I started wondering

what kinds of changes libraries would face in the future, in the context of technology and

knowledge organization. Coyle states that “The

change that libraries will need to make in

response must include the transformation of the library’s public catalog from a stand
-
alone

database of bibliographic records to a highly hyperlinked data set that can interact with

information resources on th
e World Wide Web. The library data can then be integrated into the

virtual working spaces of the users served by the library” (Coyle, 2010, p. 5). I can understand

how a library’s public catalog is a stand
-
alone database of bibliographic records, in tha
t when

you use the library’s computer to search or browse, or go on the library’s website, you are only

browsing and/or searching within the library’s database of bibliographic records.
This is true

even for consortia, such as SWAN and CARLI. What Coyl
e

proposes sounds like a far
-
reaching

library catalog that goes beyond the confines of a singular
library, and
from what Susan

Covington mentioned in her ER presentation in class, entails abolishing the traditional MARC

bibliographic record in exchange
for linked data.
At the same time, libraries have been sharing

catalogs and cataloging, but users do seem to prefer to search the Internet as an information

platform instead of the library. I agree with Coyle, in that the search options are easier, alo
ng

with the instant gratification, rankings, and social aspects of the web. Since the web is becoming

the main source of information for users, it’s true that “the library needs to be interconnected

with that web of data” (Coyle, 2010, p. 12).
Again,
here we go with that linked data.


I’m glad that we had to read a chapter on XML in
Metadata for Digital Collections
,

along with the brief XML exercise.
I have noticed that XML is one of those practices where it

isn’t

enough to just read about in a book; it’s something that’s actually learned by doing.
The

XML declaration at the beginning of each XML document reminds me of the declarations we

had to put in our XHTML code in 753, and it is good that I took that class

before taking this one.

In figure 6.2.1 on page 155, Dublin Core now makes more sense in XML, when there are

opening and closing tags, and from what we learned in class during week 3, it wouldn’t be that

difficult to apply DC in XML.
On page 158, the example of a MODS record, which is a schema

I know nothing about, looks very similar to DC when expressed in XML.
After seeing these

examples, metadata is slowly beginning to make more sense.







F
ebruary 10, 2012

(week 5)

Readings:



Zeng, Marcia Lei and Jin Qin.
Metadata.
Chater 4, “Schemas


Syntax” p. 131
-
147.



Miller, Steven J.
Metadata for Digital Collections.
Chapter 7, “MODS: The Metadata
Object Description Schema,” p. 163
-
212.



Miller, Steven J.
Metadata for Digital Collection
s.
Chapter 8, “VRA Core: The Visual
Resources Association Core Categories,” p. 213
-
226.



NISO. “Metadata Schemes and Element Sets.” “Creating Metadata,”
Understanding
Metadata, an introduction to metadata,
p. 3
-
10.



Elings, Mary W., and Gunter Waibel. “Metad
ata for All: Descriptive Standards and
Metadata Sharing Across Libraries, Archives and Museums.”
First Monday
12, No.
3(2007).
http://firstmonday.org/htbin/cgiwrap/
bin/ojs/index.php/fm/article/view/1628


What a terrible week fo
r me t
o get sick and miss class.


In “Metadata for All: Descriptive Standards and Metadata Sharing Across Libraries,”

there is a quote from Karen Calhoun that really made me think.

It is, “The catalog is in decline,

its processes and structures are unsustainable, and changes need to be swift.” At my volunteer

position, it doesn’t feel like the catalog is in decline at all. No one there is certain as to when

they will adopt RDA,
and I am told that they cannot until Voyager says it is okay for them to do

so.
I can’t understand why there are people out there who think MARC is in decline and linked

data will eventually take over. Susan Ruffalo

certainly doesn’t seem to think so, either.
In the

article, there is mention of uncertainty, especially when it comes to legacy cataloging, but I just

don’t see it. It doesn’t surprise me, however, that whatever metadata is used needs to be utilized

in a cost
-
effective and user
-
friendly way.
At the chart at the end of the article,
seeing
that XML

is the main data format for material culture, bibliographic, and archival systems
reminds me that

I need to practice it more.


In
Metadata for Digital Collections,
I noticed that MODS and Dublin Core are similar,

but MODS has finer granularity and is more detailed. Before learning about MODS, I had no

idea that it was originally an abbreviated XML version of MARC 21, and that they

have a high

level of compatibility. It is quite peculiar to me that MODS doesn’t assume any of the specific

cataloging codes like AACR2 and RDA, since that schema and MODS are so closely tied.

Unlike Dublin Core, MODS has container elements, such as
<originInfo>, and subelements,

such as <place> and <publisher>.
Now that I have a little knowledge of XML, learning MODS

XML structure is not so hard, and when I look at a record within the textbook, I can figure out

what is being described. I also li
ked the Table 7.2 on page 205, which is the side
-
by
-
side

comparison of Qualified Dublin Core and MODS records, as it shows how MODS is more

detailed than Dublin Core with its container elements and multiple subelements. For example,

<creator> in Dubli
n Core is the one repeating element that describes who created a particular

work. In MODS, the container element is <name> and it is described as “personal.” There are

two <namePart> elements, one for the name and the other for the birthdate of the cre
ator. The

<role> this particular person played in the creation of a particul
ar work is also specified.
I am

quite happy to learn about MODS, as I knew absolutely nothing about it during week one.


Since I am not interested in archives and museums, I
probably won’t be coming across

VRA Core as a metadata schema that often, but it still helps to be familiar with it. On page 217,

there is a mapping of VRA 3.0 to Dublin Core, and the first thing I noticed is that there are many

more data elements to V
RA 3.0 than there are to DC. For example, there is the lone DC element

of <dc.identifier>. In VRA 3.0, there are ID Number.Current Repository, ID Number.Former

Repository, ID Number.Current Accession, and ID Number.Former.Accession. For a museum

or a
rchival metadata document, it makes more sense to have
separate
elements that identify

former and current repositories and accessions, than repeating the <dc.identifier> element four

times, as it is more precise.
I also had no idea exactly what granularity was during week one,

and can now safely mention that MODS and VRA 3.0 have a finer granularity than Dublin Core.



F
ebruary 17, 2012 (week 6)

Readings:



Zeng, Marcia Lei and Jian Qin.
Metadata.

Chapter 5, “Metadata Records,” p. 149
-
207.



Ma, Jin. “Metadata in ARL Libraries: A survey of Metadata Practices.”
Journal of
Library Metadata
9.1 (2009).



Oliver, Chris. (2007) “Changing to RDA,”
Felicita Issue #7.
http://www.rda
-
jsc.org/docs/felicitervol53no7p250
-
253.pdf





I am one of those catalogers who are looking forward to changing to RDA. My


Supervisors
have mentioned that they find RDA weird, and won’t be switching from AACR2 to


RDA in the near fu
ture. But there are already records out there on OCLC that have RDA


elements, like “illustrations” instead of “ill.,” and “pages” instead of “p.” I, for o
ne, welcome the


spelling out of words like illustrations and pages.
RDA builds on elements of AACR2, with its


descriptions and access points, but also “encourages the description of relationships between


related resources and between resources and p
ersons or bodies that contributed to the creation of


that resource” (2007, Oliver, p. 251). It would make it even easier for the user to find resources,


and that is a very important aspect. I also like how the “description created can be contained in
a


MARC record, or a Dublin Core record or many other types of metadata records” (2007, Oliver,


p.251). In short, RDA is proof that AACR2 and MARC are not the be
-
all, end
-
all of cataloging,


and

to me, it’s also an optimistic look into the future of cataloging.
To me, RDA is also evidence


of the possibility that MARC may not be around much longer, and I like how it can be used with


many other metadata schemas.



I
n today’s class, I wondere
d why, according to AACR2, there are three specific types of


materials that are to be cataloged: monographs, continuing resources, and integrated resources.


I’m sure that it was much easier to distinguish amongst these resources before the Internet an
d e
-


books. Maybe it’s because I am old
-
fashioned and prefer monographs to e
-
books, even though I


can be seen reading my Kindle
around campus. But when I hear the word “monograph,” the first


thing that comes to mind is a physical book I can hold in
my hand, not a website’s version of


Bartlett’s
Famous Quotations.
According to the AACR2, a monograph is “a bibliographic


resource that that is complete in one part or intended to be complete with a finite number of


parts
” (AACR2, 2002). Okay, I now see why the website’s version of Bartlett’s
Famous


Quotations

is considered a monograph, as it is complete in one part, even though it can be


accessed

on the web. I think I got this wrong because I am a very literal person. If you tell me


something’s a monograph, I will think you’re talking about a physical book. If you tell me you


will be back in ten minutes, I will look at my watch and hold you

to your word. If you tell me


you’re going to c
all me, I expect you to call me and will even ask you when you intend on doing


so.

That’s why I believe the AACR2 needs to
h
ave an addendum to their definition of a


monograph. In my opinion, they als
o need to include specific
line as to how electronic resources


can be considered monographs, as long as they are complete in one part. Surely I am not the


only one who thinks this.
When I cataloged
Famous Quotations
, I referred to chapter 9 in


AACR
2, and now wonder if I should’ve referred to chapter
2
.



Speaking of integrating resources, I stand firm when I define
The World Factbook

as an


integrating resource. The AACR2’s definition of an integrating resource is
“a bibliographic


resource

that is added to or changed by means of updates that do not remain discrete and are


integrated into the whole (2002, AACR2).
The World Factbook

definitely counts as such, as it is


updated

weekly. Oh, I see. But at least the AACR2 actually lists websites as examples of


integrating resources, and that is another reason why they should be more specific and update


their definition of a monograph, since e
-
books and certain websites are no
w considered


monographs, too.































February 24, 2012 (week 7)


Readings:




Zeng, Marcia Lei and Jian Qin.
Metadata.

Chapter 5, “Metadata Records,” p. 149
-
207.



Tillet
, Barbara. (July 1, 2008). “A Review of the Feasibility of an International
Standard Authority Data Number (ISADN),”
http://archive.ifla.org/VII/d4/franar
-
numbering
-
paper.pdf



Intner, Shelia S., Susan S. Lazinger, and Jean Weihs.
Metadata and Its Impact on
Libraries.
“Creating Library Metadata for Monographic Materials,” p. 109
-
134.

In class, I was very excited to be learning more about XML and RDF.
I’m also very glad

that I

have some experience with HTML and am currently taking LIS 786, which is Advanced

Web Design.
I have been going to Wikipedia to look up their articles in DBpedia. It’s very

interesting to
try and
look at
my favorite Wikipedia entries in DBpedia; many
of the records on

DBpedia have yet to be developed, and I am wondering if this is a good opportunity for those

who are familiar with metadata to contribute to the community. For records on DBpedia that are

already developed,
such as
http://dbpedia.org/page/Warped_Tour
, it is fun to pick out the

elements that I am familiar with now, such as dcterms: subject, and click on the links. I think

DBpedia is a great example of linked data,
and enjoy researching subjects I’m interested in by

clicking on more links to other data. I have also been visiting the LC authorities to look up


records in XML, MODS, and Dublin Core.
At first, the idea of linked data replacing MARC


records intimidated me. Now I think it’s a great way to bring related works and subjects


together, because when you really think about it, many things are connected/related in some way.



I was also excited to actually see what Metadata looks like

in person when we visited the


LC authorities and got to see it in XML and RDF. That is the best way to learn; the examples of


it in
Metadata
are good, but it’s even better to see it
onscreen.
A good example is

Exhibit 5
-
5
-
4


(Zeng, 2008, p. 18
9)
, which

is a complete Dublin Core m
etadata record expressed in XML.
As


someone who has had some experience with HTML, XML is a lot less intimidating, and just


from looking at the record, it isn’t difficult to understand what’s going on in E
xhibit 5
-
5
-
4. It’s


the metadata for the

Metadata Basics
tutorial webpage


(
http://marciazeng.slis.kent.edu/metadatabasics/index.htm
), and
i
t’s quite easy to find the title,


creator, subject, and everything else someone wants to know about that website from looking at


the metadata record.
Also, one of the electronic resources I cataloged in AACR2 was a website,


and that wa
s quite a struggle. When I examined this particular Dublin Core metadata record


expressed in XML, I felt that the description of a website actually makes more sense in Dublin


Core, than it does AACR2.
They are complementary to each other wh
en expressed in XML, and

RDF, for that matter.


I also like the fact that RDF is not “restricted to encoding information about Web

resources,” (Zeng, 2008, p. 190) and can be used to provide information and relationships

between

people, places, concepts, etc.
It’s also based in logic, with the subject
-
predicate
-
object

expression
. For example, the triple for describing a website consists of the subject, which is the

URL : (
http://
www.loc.gov/index.html
) ; the predicate, which is the word
title,
and the object,

which is the phrase
Library of Congress Home.
These triples can also be encoded in XML,

which is an example that Exhibit 5
-
5
-
5 illustrates (Zeng, 2008, p. 191).
In closing for this week,

I am beginning to find the concepts of Dublin Core, XML, and RDF much easier to use than

AACR2 and MARC, as the former doesn’t look and feel so rigid.








March 2, 2012 (week 8)

Readings:



Zeng, Marcia Lei and Jin Qin.
Meta
data.
Chapter 6, “Metadata Services,” p. 211
-
245.



Gemberling, Ted. “Thema and FRBR’s Third Group,”
Cataloging and Classification
Quarterly,
Volume 48, Issue 5, June 2010, p. 445
-
449.



Kornegay, Rebecca, Heidi Buchanan, and Hildegard Morgan.
Magic Search: Getting the
Best Results from Your Catalog and Beyond.
Chicago: ALA, 2009.



This may sound very obvious, but one of the best things about working with Kate

Harcourt was the opportunity to receive feedback from Director of Original Cataloging at

Columbia University’s library. I chose to work with her because my area of interest is original

cataloging, and if I were to visit New York, I would devote at least

a few days to visit Columbia

University’s library and archival collections.
She did look at two of my records and was slower

at responding than other mentors, but I expected that because of her schedule.


One of the most helpful suggestions she gave
when it came to cataloging electronic

resources was that the chief source of information is the resource itself, using AACR2 rules

9.0B1 and 7.0B1. With electronic resources, it’s easy to get carried away and search other

websites, therefore leaving th
e chief source of information.


Kate also made me more aware of the smaller things one can neglect when cataloging

electronic resources, such as accidentally leaving out the GMD [electronic resources] following

the title proper, from rule 1.1C1. I have

also neglected rule 9.7B3, which is the source of title

and date viewed, which is very important and required. Another thing she pointed out was the

redundancy of “Mode of access: World Wide Web.” According to her, it’s in the rules but no

longer adde
d, since almost everything is web
-
based now and the note doesn’t add any value.

Also, when applying rule 9.7B3, which is much more specific, “mode of access” really does

seem unnecessary.
In
the end, the things I realized I needed to improve upon are m
y accuracy

and attention to details, along with my tendency to go beyond the chief source of information for

electronic resources.

























March 16, 2012 (week 9)

Readings:



Zeng, Marcia Lei and Jian Qin.
Metadata.

Chapter 7, “Metadata Quality Measurement
and Improvement,” p. 247
-
266.



Miller, Steven J.
Metadata for Digital Collections.
Chapter 5, “Controlled Vocabularies
for Improved Resource Disovery,” p. 129
-
148.



Intner, Sheila S., Susan S. Lazinger, and Jean Weih
s.
Metadata and Its Impact on
Libraries.

“Integrating Library Metadata into Local Catalogs and Datab
a
ses,” p. 165
-
174.



Schauller, Peter, Joanneum Research. “Content Analysis and MPEG
-
7 Description
Tools,” (37:26 minutes)


Well, my presentation with Krist
in went well. This is after crossing her out on Week 1

because we were figh
ting over who got to do Taxonomies
, Ontologies, Folksonomies, and SKOS

for the ER presentations. We even laughed about that one, and I think we worked well together

because

she has less experience with cataloging and gives me a fresh perspective. I feel like we

learned a lot during our presentation, and figured out what we were doing wrong. She was the

one who suggested we present on that online serials article, which my

mentor described as

“tricky.”
I agree that the digital object in question is now to be described as a monograph, as

even though it is taken from a serial, we are only describing the article itself, because it is

complete in one part.


After presentin
g on Taxonomies, Folksonomies, Ontologies, and SKOS, I am of the mind

that a controlled vocabulary is necessary for classification, because it is more precise and

unambiguous, as opposed to user
-
created folksonomies. In
Metadata For Digital Collections,


controlled vocabularies have synonym and ambiguity control, unlike folksonomies, “as well as

semantic relationships and cross
-
references among terms” (Miller, 2011, p. 129). Even terms

like
search

and
browse

have such strong definitions and expectations on both the user and digital

collection. In
search
, it is up to the user to come up with a search term that may or may not be

correct. It is also up to the digital collection itself to create strong metada
ta that can override the

user’s possible mistake in finding the appropriate search term, while still helping them find what

they’re looking for. In
browse,
it is up to the digital collection to create a controlled vocabulary

so the user may browse the t
erms to find what they’re looking for.
It’s also very true that the

strength of a digital collection’s metadata depends on the “quality of the metadata that structures

and supports the usability and administration of those resources” (Zeng, 2008, p. 247
). If a

digital collection were to not use a controlled vocabulary, then it compromises the quality of the

collection’s metadata and in turn, makes it more difficult for the user. If the quality of a digital

collection’s metadata is weak or inconsiste
nt, then that digital collection may not have repeated

users who see that collection as a respected resource. But unlike legacy cataloging, which is

ruled by AACR2 and RDA, there are many different metadata schemas being used at digital

collections, a
nd the variety of schemas make consistency a challenge. Consistency challenges alo

mean that there may be intracollection and/or intercollection duplicates. An intracollection

duplicate is a local quality problem that needs to be solved by the metadata
creator, but an

intercollection duplicate happens when “multiple collections are integrated into one source”

(Zeng, 2008, p. 259). Search results can be identical or virtually identical, or two
results with

different

titles or descriptions lead to the same source.


As a writer, I am also aware of the impact of words and “the two problems inherent in

‘natural human language’:
ambiguity

and
synonymy” (
M
iller,
2011, p. 131). When it comes to

strong writi
ng and metadata, there is no use for ambiguity, as it is important to be as specific and

precise as possible. With synonymy, however,
means more than one word or phrase can

represent the “same concept, person, place or thing,” (Miller, 2011, p. 132), an
d have equivalent

relationships with each other. To remedy this, there is a “preferred term” that is cross
-
referenced

and linked to other terms. No matter which metadata schema is used, it is important to link the

terms together for the user. Control
led vocabularies also utilize hierarchical relationships to

arrange terms into broader and narrower

ones, which is what a taxonomy is. When a large

number of terms are arranged into categories, it will be easier for users to browse and navigate.

This
reading goes back to my presentation during week 8, which points out the weaknesses of

folksonomies and calls for the need of controlled vocabularies.























March 23, 2012 (week 10)

Readings:



Zeng, Marcia Lei and Jian Qin.
Metadata.

Chapter 8, “Achieving Interoperability,” p.
267
-
285.



Miller, Steven J.
Metadata for Digital Collections.
Chapter 9, “Metadata Interoperability,
Shareability, and Quality,” p. 227
-
250.



NISO.
“Interop
erability and Exchange of Metadata,” “Future Directions,”

Understanding
Metadata, an introduction to metadata,
p. 11
-
12.



Caplan, Priscilla. “Approaches to Interoperability,”
Metadata Fundamentals for All
Librarians.
P. 33
-
44.



I am glad we spent some class time discussing

metadata mapping and crosswalks

i
n preparation

for Assignment 5
, as no matter what the schema, there are similar elements in Dublin Core and

ONIX, such as Title, Creator, Date, etc. The largest difference is that ONIX has many more

elements, and is more intricate and specific. Intero
perability also relates to metadata conversion

and processing, which involves correcting or removing punctuation “that will interfere with

indexing in their aggregated repository,” (Miller, 2011, p. 234). That way, content is

normalized. Table 9.3 on
page 235 of Miller’s monograph is a good example of the crosswalk

from MODS to Dublin Core, and it is interesting to see the similarities between the two

(<subject> vs Subject, <identifier> vs Identifier) and the differences
(Creator & Contributor vs

<
name> & <namePart>).
The crosswalk is also an example of interoperability actions that

“take place before operational level metadata records are created” (Zeng, 2008, p. 269). These

interoperability

actions concentrate on the elements, which is what the crosswalk does. At the

record level, interoperability entails converting, reusing, and/or combining metadata records

with other types of metadata records to create new records. Once at the record
level, it is

already too late to consider interoperability at the schema level, and this happens when “a

distinctive metadata schema is developed or adopted for a project when metadata records have

already been created before of interoperability was car
efully considered” (Zeng, 2008, p. 275).


I also agree with NISO, in that interoperability is a challenge when there are various

metadata standards, initiatives, extensions, and profiles. Personally, I am not so sure if we need

all these different meta
data standards, but many of them were created to serve to specific

purposes of a particular digital collection, and each digital collection has its own audiences and

needs. That is why I’m glad RDF exists as a data model that can integrate multiple met
adata

schemes and
wish
we spent a little
more time was spent on it.




















March 30, 2012 : week 11



Zeng, Marcia Lei and Jian Qin
.
Metadata.
Chapter 9, “Metadata Research Landscape,”
p. 289
-
299.



Miller, Steven J.
Metadata for Digital Collections.
Chapter 10, “Designing and
Documenting a Metadata Scheme,” p. 251
-
302.



Kennedy, Marie R. “Nine Questions to Guide You in Choosing a Metadata Schema.”
Journal of Digital Information 9, no. 26 (2008).
http://journals.tdl.org/jodi/article/view/226



The OAI Execute and the OAI Technical Committee. “The Open Archives Initative
Protocol for Metadata Harvesting.”
http://www.openarchives.org/OAI/openarchivesprotocol.html


I guess being able to air our grievances during last week’s class
had its advantages, because I

was very glad that we went o
ver MODS in class today.
I chose MODS as part of my final

project, and after having more experience with it, I think I made the right decision.
It is more

granular than Dublin Core, but not as in
-
depth as MARC records or other metadata schemas, like

O
NIX. In my experience, I thought it was best for describe a single digital object within a digital

collection.
My group described a negative of an image produced by William P. Gottlieb, and

filled in the blanks a
ccording to the MODS template, then post
our results on Blackboard.


My favorite reading this week was the Marie R. Kennedy article. Unfortunately, I didn’t

ask myself these questions before choosing MODS for the final project, nor did I when I chose

ONIX for assignment 5. But when choosing

a schema, it is always very important to know who

will be using the collection. Will it be for the general public, or will it be for a more specialized

audience, such as law students? It’s also very important to know who will be cataloging the

collection, and how much time and money is available.
Just from doing records in MODS and

ONIX, I have realized it can be time consuming on the cataloger, so there may need to be more

than one person involved with the project. I think it’s fair to assu
me that a digital collection will

be accessed online, and it needs to be both searchable and easy to browse, with consideration to

synonyms. This is an article I have bookmarked on my laptop for future reference, when I will

no longer be in the positio
n to say, “I am interested in ONIX, let’s use that one.”


After reading Miller an
d Zeng this week, t
he steps to designing and documenting a

metadata schema are similar to choosing one for a particular digital project, in that it’s important

to

analyze context, content, and users to determine functional requirements. Determining the

purpose and intent of a schema, whether it’s established or still in process, is the most important

step. Selecting and developing an element set does sound daun
ting, and there are many domain
-

specific elements sets that already exist, such as VRA, CDWA, and TEI. For now, is designing

and documenting a new metadata schema even necessary, due to all the existing schemas that are

available? I personally don’t
th
ink it is.


I do think that the Miller readings are much easier to understand and absorb than Zeng, as

Miller writes in a more accessible language than Zeng and Qin. Miller also makes developing

and designing a metadata schema sound less daunting than
it really is. In 10.1.4 of
Metadata for

Digital Collections
, I like his simple instructions on establishing element and database

S
pecifications
, even though the process itself is much more complicated and arduous
. In terms

of obligation, should an ele
ment or field be required? Mandatory or required if applicable? Or

optional?
The sections on developing content guidelines, establishing controlled vocabularies

and encoding schemes, and documenting the scheme are helpful to learn about the process and


kind of work that goes into the design of a new schema, and I like how Miller doesn’t talk down

to the reader while writing in a more accessible tone and language. For a library student, the

thought of designing a metadata schema can sound very intimi
dating. This chapter shows the

kind of work that goes into it, and makes it sound like something that could be done in the distant

future.



March 13, 2012 : week 12



Miller, Steven J.
Metadata for Digital Collections.
Chapter 11, “Metadata, Linked
Data,
and the Semantic Web,” p. 303
-
324.



Berners
-
Lee, Tim, Dan Connolly, and Ralph R. Swick. Web Architecture: Describing
and Exchanging Data, W3C Note 7 (June 1999).
http://www.w3.org/1999/04/WebData



Ber
ners
-
Lee, Tim. (2007). Linked Data.
http://www.w3.org/DesignIssues/LinkedData.html


I enjoyed today’s set of presentations and was happy to see my classmates show us what they

learned about th
eir schemas and apply them to an electronic resource.
Also, I will not complain

about
how ONIX has over 200 data elements, because CDWA has even more than that.
I now

understand why Assignment 5 was given more priority over the final group project, as
it is a

culmination of what we’ve learned this semester, along with us applying the concept of “learning

to learn.”
I’ve noticed that once someone is familiar with Dublin Core and creates a crosswalk

from that sche
ma to a different
one, it will not be t
hat difficult to learn other schemas, as even

though the number of data elements may vary, there are similar in elements in each schema, such

as some form of title, creator, etc.
That is probably one of the most important things I ended up

learning in
this class. During week one, I remember how intimated I was when everyone else

seemed to be familiar with the schemas. Even in 703, the term “metadata” confused me, as all I

knew was that it was data about data, and the closest I ever experienced it wa
s on iTunes, when I

would get frustrated from having to edit the data from music I got over the Internet. Worse yet, I

had no idea what Dublin Core actually was until this semester.


One of the other concepts that once intimidated me was linked data, especially the

prospect of it taking over MARC in the future. Now that the semester is almost over, I am more

comfortable with the idea of “using the web to connect related data that wa
sn’t previous linked,

or using the Web to lower barriers to linking data currently linked using other methods” (Miller,

2011, p. 303). Dublin Core and ONIX both have data elements for related works, and FRBR

brings

together related works, so in the future, linked data will be playing a significant role in the

organization of knowledge. Now, Figure 11.1 on page 304 of Miller makes a lot more sense,

especially after we went over DBpedia this semester and all the ot
her data points back at it.


I also agree that with metadata, “
statements
are the basic unit of metadata rather than

records”
(
Miller, 2011, p. 231). With legacy cataloging, the opposite feels mostly true, as there

is a focus on creating and maintainin
g records so users can access items as quickly and as easily

as possible. With metadata
, each statement consists of a property
-
value pair that resides in

database records in a Linked Data environment that may be harvested and processed separately

one d
ay, and not into a record structure.
Also, MARC is not compatible with Linked Data and

the Semantic Web, the way RDA is. This isn’t change for change’s sake, but looking ahead and

keeping up with the times. I understand that there are catalogers out th
ere who don’t

even want to convert to RDA, let alone utilize the semantic web, but since the beginning of this

semester, I realized that I belong to the other camp, which is the one who looks forward to RDA,

and likes the possibility of Linked Data and
the Semantic Web.




.