![intended audience, system of organization and dissertation or thesis information. And
these are all attributes at work in the current RDA. And they are intended only to be
used in descriptions of works, not in descriptions of other entities, like expressions or
manifestations.
This is the -- a visual showing the attributes of expression in the current RDA, so, in
current RDA, expression has attributes of content type, date of expression, language of
expression, other distinguishing characteristic and identifiers. Those are in chapter 6,
and then in chapter 7, all of these other attributes can be used to describe an
expression, summarization of the content, capture of language, that's a pretty important
one, form of notation, accessibility content, illustrative content, supplementary content,
color content, sound content, aspect ratio and all of these other things for music that
can all be used to describe an expression in the current RDA. Once again, they are
only intended for use in description of expressions, not other entities such as
descriptions of works.
And this next slide is showing the benefit of the reorganization of this, in what I'm
going to call the 2020 text of the RDA, just the new -- the RDA. And in bold are some
new things that are different from the current RDA for the elements of work. So,
category of work is a new element. It is renamed. It used to be called form of work. It's
more or less the same element. New elements include extension plan,
frequency, -- has been moved to work, as is the same with key title. These have all
been moved from the manifestation entity to become attributes of the work entity. Note
on the metadata work is new, and I'm sorry we don't have time to talk about that right
now, because the metadata works are quite interesting -- [ off mic ] used to be an
instruction for a relationship, and it's now an attribute for work. The recording source is
new, as are scope of validity and source consulted.
Missing from the work attributes in the 2020 text are those three music-related
attributes, medium, performance, key, and intended audience. That's because in the
new text, they have become attributes of the expression, as we will see here.
Neither of these lists, by the way, include relationship elements or any subelements or
[ off mic ] elements. So let's talk about what's new in the expression in the 2020 text
attributes. Some new elements include category of expression, the extent of the
expression, intended audience of the expression. This is new to expression, that it
replaces what currently is in the work as the intended audience. It has been moved
from the work level to the expression level. Interactivity mode is a new element. Here
are these three music-related ones, key of expression, medium performance. These
two medium performance attributes have been moved to the expression level. Title of
expression is a brand-new attribute that has never existed, did not exist as a new vision
in the current RDA, which is interesting to me. Expression was the only entity in the
current RDA that did not have a name, and now it does. There is a -- an attribute called
title of expression for the expression entity now.
Deleted from expression are the performer, narrator and/or presenter element and the
artistic [ off mic ] element, which have been hard to remove from the RDA but are still in
the table of contents.
Okay. That's just an overview of what's changing in the new RDA, as far as these
attributes of these two elements, these two entities go, in work and expression. So why
are we even discussing this? What's the problem?](https://image.slidesharecdn.com/021920ala-newconcepts-representativeexpressionsandmanifestationstatements-200220173135/85/New-Concepts-Representative-Expressions-and-Manifestation-Statements-Transcript-February-2020-4-320.jpg)
![I have mentioned already, entity relationship principles require strict boundaries
between the entities, as far as their attributes go, for describing them. So, you can't use
an attribute defined for one entity to describe another entity in the way the principles
work for entity relationship databases. The new toolkit is quite clear about what the
domain of each element is. It's much clearer than the current toolkit, in fact, and this is
necessitated by requirements of link data structures. However, the problem is human
beings don't necessarily think of the bibliographical universe in these terms, and in
particular, database users often associate attributes of expressions with works.
So, for example, many people when they think of Alice's Adventures in Wonderland or
Alice in Wonderland, they think, oh, it's a novel and it's a text and it's in English and it's
for children. Well, the first one, according to RDA, is an attribute. In fact, it's a novel
that is applicable to the work itself, but the other three, the fact that it's a text and it's in
English and it's for children are not attributes of the work, even though normal human
beings think of them in that way, and so they are, in fact -- those three are in fact, in our
model, attributes of the expression. So if we are in an entity relationship database
structure or in linked data, we need to respect those differences.
So, under current RDA, it is not appropriate to include, for example, language or
content type in a description for a work, and the best kind of descriptions that are most
clearly for works are authority records for works. So under the 2020 text, it would be
inappropriate to record these two, intended audience, because none of them are
elements that can be used to describe a work editing. So here we have a work record
for the authorized -- well, a work record for Alice's Adventures in Wonderland, which
includes the authorized access point or [ off mic ] field. It would not be appropriate to
include the fact that it's text or that it's in English there, because those are not
work-related attributes or the fact that it is for children. The fact that it is a novel that is
fantasy fiction or nonsense fiction is perfectly appropriate to record, because those are
work level attributes.
It's clear that cataloguers do want to record this type of information in descriptions of
works, and I don't think they are wrong to want to, although the current RDA doesn't
allow it. So here are some examples from the current name authority file, which are, at
the moment, not correct. So here's an example for a serial called the Ballantonians. It's
true that it is in English, but that isn't an attribute, and since this is a work level record,
the [ off mic ] isn't appropriate.
Here are a couple of more examples, a poem by Paul Verlaine. It is true that it was
originally composed in French, but at the work level, that's not appropriate to record. It
should be recorded in an expression level record. Same for Boy Scouts in an airship,
something that I have not read but which I should read someday, the fact that it's in
English or that the novel is not French. So, clearly people want to record this
information, and I don't think they are wrong to want to do it, but under the current
structure, it's not correct to do this. And it's important, I really don't advocate being
correct just for neatness and correctness sake. It's important for the database structure
to follow rules of the database.
So is there a -- I agree that this is a problem. So is there a solution? There is. It's
called representative expression. This is a new concept introduced with LRM. When
people were talking about it at the beginning, it was initially -- there was initially a
possibility proposed to choose one expression as the, quote, "representative](https://image.slidesharecdn.com/021920ala-newconcepts-representativeexpressionsandmanifestationstatements-200220173135/85/New-Concepts-Representative-Expressions-and-Manifestation-Statements-Transcript-February-2020-5-320.jpg)
![really is a place called Hoople, North Dakota, but since this university is a fictitious
university, I believe, where P.D.Q. Bach supposedly did his work, I wonder if this is
really where this recorded, but it is what the recording says. The duration is an
expression element, and there are two different durations, so I have recorded them
both. One duration for the performance time is found on the score, and another
duration for the actual time that was taken in the recording.
Here's an example of a work that we have not talked about, Claude Debussy's Syrinx.
It has a category of the work, art music, and the date of the work is 1913, but the date of
the representative expression is different because this work was created in 1913. It has
been well researched, and it was created in 1913, but it was not published until 1927, so
here's an example of where we might have a different date of representative expression
from the date of the work that could be recorded in a description of the work.
Medium of performance again is a representative expression. The tabs on this slide
have gotten a little messed up, so the medium performance is flute. Content type is
notated music. Those are all representative expression elements, which right now can't
be recorded in work authority of records but will be once we have implemented the new
RDA.
Here's an example for a film, Wizard of Oz. It is a fantasy film, has a date of work,
and all of these next five or six elements are elements drawn from one of the
expressions of this work, and so I think we can see in this case that these are things
that people think of as being important to the work itself, not just to an expression, the
fact that it's a moving image, that it is both -- this is very famous for being both in black
and white and in color or monochrome and polychrome, as RDA recommends we use
the term. The aspect ratio may be important to some people, when was it captured,
where was it captured, how long it is. These are all drawn from the expression from the
work.
And a map, an example of a map. Here's a map of Eswatini, and it includes some
representative expression elements, cartographic image, the fact that it is color or
polychrome.
So these are examples of descriptions in which we could enhance the work
descriptions by including representative expression elements. Now, one of the reasons
that I haven't given you these examples in MARC is because, as we will see in a few
minutes, there isn't any way to code the representative expression elements in MARC
yet, but that will be coming, I believe. So, actually, here we are. How can I do this in
MARC? The MARC Advisory Committee hasn't yet created any coding for
representative expression elements. Here are some possibilities. They could use
existing coding and just, as most of these elements already have a [ off mic ] coding for
use in the expression authority records. Or another possibility is to use existing field
coding, but define different subfields for the representative expression elements, or just
make up totally different field coding. My opinion is the second possibility would be the
best, so use existing field coding, but define different subfields for the representative
expression elements, simply because I think we need to be able to clearly show in our
records what these elements where, and if they are used both for representative
expression and also for regular expression, it might not be clear.
Okay. So, here is an example where I have kind of made up things. For the Wizard
of Oz, the date of the work is 1939, and so that's perfectly fine. We can have another](https://image.slidesharecdn.com/021920ala-newconcepts-representativeexpressionsandmanifestationstatements-200220173135/85/New-Concepts-Representative-Expressions-and-Manifestation-Statements-Transcript-February-2020-9-320.jpg)
![has to be a cartographic image, but that is, in fact, an element as defined in RDA of
expression. So, yes, of course people are going to be thinking of this map as a
cartographic image, and so that's exactly the point of why it's a good idea to be able to
bring that fact into the description of the work, as we have done here. So I think
that's -- your point is exactly very relevant, and it is a point in favor of implementing the
concept of representative expression.
You could say the same thing about a movie, where your average movie -- I thought [
off mic ] since it is a film could have possibly more than one expression, but in most
cases, the film has only one expression. So all of the elements of expression could be
used to describe that film, but yet in the way that we have parsed out RDA, without the
concept of representative expression, we can't use those elements to describe the work
itself. We have to have expression descriptions. So I think this is a good step forward
to be able to describe representative expression in the work description or authority
record.
>> Colton Ursiny: Thanks, Bob. Someone asked a follow-up question relating to this
one. They said, "A political map could be an RIS file of geocoded boundaries, not an
image?" And they have a question mark there.
>> Robert Maxwell: All I can say is that's probably true. I mean, I suppose that's a
good comment.
>> Colton Ursiny: All right. Thank you, Bob. Let's see. I think time for one or two
more questions before we move on.
Are there guidelines for where to find filming dates and location, which aren't
necessarily described on the manifestations?
>> Robert Maxwell: I am not an expert in that, and so I am not sure there are
guidelines, but if there are, maybe there's a film cataloguer in the audience who could
tell us, but if you are interested in a good place to find this sort of information,
IMDB.com is a good database that includes a lot of this sort of information, if you're
interested in filming dates and locations. And Adam says they are not required. I think
he's talking about the same thing. That's true, they are optional.
>> Colton Ursiny: All right, great. Thank you so much. Last question before we
move on, are there sufficient subfield opportunities in corresponding MARC fields to
accommodate these additions and adjustments?
>> Robert Maxwell: I have seen a discussion paper from one of the members of the
MARC group that's working on RDA, and it is true that we are running out of subfields. I
think in these cases, for this case, I think there are enough subfields for the
representative expressions, but that could be a problem. That's correct.
>> Colton Ursiny: All right, Bob. That's all of the questions for now, so I'll turn it back
to you.
>> Robert Maxwell: Okay. So now we are going to turn to something completely
different, manifestation statements. So, before we start, I would like to say this is -- last
time when I did this, people got -- some people got quite upset at thinking this is what
we are all going to have to do. I want to preface this by saying that this is a new option
being offered in RDA for recording information, and it's not going to force us all to
change if we don't want to. And there are good reasons for not changing what we are
doing right now in recording information, but there are also good reasons for having this
as an option, so I just wanted to preface it that way because we had a little bit of upset](https://image.slidesharecdn.com/021920ala-newconcepts-representativeexpressionsandmanifestationstatements-200220173135/85/New-Concepts-Representative-Expressions-and-Manifestation-Statements-Transcript-February-2020-11-320.jpg)
![last time I gave this.
Okay. So if any of you have the new RDA app, and to get there, you can bring up
your radio RDA and click on -- let's see, what do you have to click on -- I need to look at
this myself, in case any of you want to follow along, click on the explore RDA toolkit site,
and you'll be able to see what I have got in the slides, and you can also explore a little
bit more if you want to.
So to find out more about manifestation statements, let's go to the RDA beta site and
have a look at what it has to say. So, to find this information, go to the guidance tab,
click the are drop-down, and choose manifestation statements.
Manifestation statements are defined as data found on manifestations themselves.
This first section, under principle of representation that tells us that a manifestation can
carry information and describes the manifestation, and it is recorded to follow the
principle of representation, reminds us of the basic cataloguing principle that an entity
should be represented as it represents itself, and we use this principle throughout
cataloguing for many, many things that we do. It's practically unconscious at some
times for us, it's so ingrained in our cataloguing. Now, manifestation statements are, by
nature, unstructured, and so RDA only offers unstructured description as the recording
method. Note there is an option for a normalized transcription under recording and
manifestation statements further down on the page. We'll see the second option talks
about recording and normalized transcription. Normalization can be said to be imposing
some sort of structure to a transcription, but it is not enough in the minds of the creators
of RDA to tip it over into the area of structured description. So, let's briefly look at the
normalized transcription rules.
So, if you're following along in RDA, the beta RDA, click on guidelines on normalized
transcription, and there is really nothing new here. These are all the same guidelines
that are in the current RDA. We have been using normalized transcription rules in
cataloguing from time immemorial. The links are all to instructions that we are quite
familiar with, so let's just look at two, capitalization and punctuation guidelines. So, the
left what you get if you click on capitalization in the previous slide, and the right is what
you get if you click on punctuation. These are all things that we are used to. We
capitalize the first word in a statement. We capitalize proper names. We include
punctuation we have found, but we can omit, modify, or add punctuation for clarity.
The other conventions that we follow for other aspects of transcription are all there in
the revised RDA instructions for normalized transcription, and they will remain available
for our use, and I assume that we will, most of us, will probably continue using these,
these guidelines and conventions. But if you think about it, normalizing data is not
representing the entity as it represents itself, so the new RDA offers a method of more
closely representing the manifestation as it manifests itself, which is called basic
transcription. So, going back to the basic guidelines on manifestation statements, we
are now going to click on guidelines on basic transcription, if you want to follow along.
And this is what you get to, if you do that. So, let's read these. We are told to
preserve the capitalization, punctuation, numerals and diacritics as they appear on the
manifestation. We are to omit symbols, or rather, images, that are to be interpreted as
text that are not available [ off mic ] -- transcription, so, for example, if we are
transcribing a rebus, we might have to omit the symbols in the rebus.
We are to tribe the components of a manifestation statement in the order in which](https://image.slidesharecdn.com/021920ala-newconcepts-representativeexpressionsandmanifestationstatements-200220173135/85/New-Concepts-Representative-Expressions-and-Manifestation-Statements-Transcript-February-2020-12-320.jpg)
![they appear, no transposing, like we do when following the other guidelines in the RDA.
Use these two backward slashes to limit separate components of a statement. This
slide is somewhat confusing and I think needs a little bit more explanation. I have
inquired of the authors of this, and this does not mean dividing between, for instance,
line endings. It means dividing between major components, such as between the title
and responsibility statement and publication statement, for example.
Continuing, we are to collapse -- well, use a mark of omission to indicate the omission
of text within a statement, and as an option, we can use it at the beginning or the end of
the statement. There apparently is no distinction between using a mark of omission to
indicate that the cataloguer omitted something and using a mark of omission to actually
copy what's on the statement that we were transcribing, which may or may not be a
problem. We are to condense all white space, including multiple spaces, new lines,
indentation tabs and block alignment and so forth. This is the alignment that tells us
that we are not going to take into account the fact that we are starting new lines in -- on
the thing we are transcribing. We are to deduplicate a whole bunch of things.
So, you may also note that these guidelines do not take into account type size,
typeface, for example, Fraktur versus Roman, in determining manifestation statement,
type style, such as italic versus Roman, or [ off mic ] or things that appear in the line as
a resource.
Now, returning to the basic page on manifestation statements and scrolling to the
bottom, we learn that there are a number of different categories of manifestation
statements, and we'll begin by clicking through to the general instructions at the top for
manifestation statement. So, the first thing to note is the general manifestation
statement element can be used to record any statement appearing in a manifestation
that is thought to be significant to users. So, it does not need to correspond to one of
the traditional areas or elements of the description, so, a transcription of the entire title
page could be recorded as a manifestation statement, or if you are transcribing
something to include an [ off mic ] note, that is an example of a manifestation statement.
If you are transcribing the form of a name from somewhere in the manifestation to justify
decisions you have made in authority work, that is a manifestation statement. These
are all manifestation statements.
Second, although it's not shown here, if you continue to scroll down in this part of the
new RDA, you will see, as has already been mentioned, that the only recording method
available is unstructured description.
And, finally, although the remainder of my presentation today is going to focus on
basic transcription, basic and normalized description are both legitimate choices and
both count as manifestation statements, so do not get the impression that manifestation
statements is only this exact basic transcription that I'm going to be talking about the
rest of the time. The normalized transcription also counts as a manifestation statement
in the new RDA, which is the kind of transcription that we have been doing.
So, returning to the manifestation statements page, let's now look at a couple of the
narrower elements. The title -- I'm sorry, this is called -- this has got a new name, which
is hard to say, because we are all used to saying title statement and responsibility, and
it is now called the title and responsibility statement. These can serve as examples,
and the other types of manifestation statement work in the same way as these two. So
let's look at manifestation title and responsibility statement. This is a statement that](https://image.slidesharecdn.com/021920ala-newconcepts-representativeexpressionsandmanifestationstatements-200220173135/85/New-Concepts-Representative-Expressions-and-Manifestation-Statements-Transcript-February-2020-13-320.jpg)
![Okay. There isn't currently any place in MARC for a general manifestation of the full
title page, which would be, as I have shown, below the first bulleted point. And there's
not any provision in current MARC for a subelement of a manifestation statement that
doesn't correspond to the expected ISBD order. For example, the manifestation title
and responsibility statement. But the MARC Advisory Committee considered a
discussion paper at ALA Midwinter just last month about manifestation statements that
is expected to return as proposal at Annual. When it comes back as proposal,
hopefully, and it is approved, it will probably be recorded in a newly defined 881 field,
but it is not clear yet what subfields will be approved or how manifestation statements
will be recorded within them.
The discussion paper called for a designated subfield for each type of RDA
manifestation statement. So here are some of the ways that came from the discussion
paper as possible ways to record a manifestation statement. So here's this book called
Crossing the Chasm. The discussion paper suggested this as one possible way of
doing a manifestation statement. This looks like a pretty -- this is without any subfield
codings, which is pretty difficult to take in as a human being, but it is a pretty much top
to bottom manifestation statement, and it includes more than just one source from the
book that's being described. So that's one possibility that was suggested in the
discussion paper.
Here is a second possibility that was discussed -- suggested in the discussion paper,
which includes subfield coding, which looks more useful to me, because it's a little
easier to figure out what's coming from where and what's going on. This probably is not
the final subfield coding that will be approved, eventually if this proposal is approved,
but it's an example of how it might work.
And, finally, a final example gave kind of a label display, which I find, as a human
being, the most useful of all three, because it tells us where we got the information from.
The first [ off mic ] says we got this particular bit of manifestation statement from the title
page, and the second one is information that came from the cover, which isn't shown on
the -- I'm not showing you the back cover on this slide. In any case, don't try this yet. It
hasn't been approved yet, but there are efforts under way to make it possible for us to
record RDA manifestation statements from the RDA in MARC, and so stay tuned for
further developments. At ALA Annual 2020, probably something will be approved, we
hope.
Okay. I can take some questions or comments if anyone has any.
>> Colton Ursiny: All right. Thanks, Bob. One moment. Just pulling up the questions
we have been collecting.
Are there any guidelines in regards to representative expression for work, when the
work is not a well-known work?
>> Robert Maxwell: There were no guidelines at all for representative expression
data at all. So, no, there really aren't any guidelines, so I'm not going to attempt to -- I
guess I will say something that comes to the top of my head. I would think for a work
that is not well-known, you would probably want to use as a representative expression
the first expression of the work, and so the elements I personally would probably
choose from the first expression, some of the language, for instance, would be the
language of the first expression, but that is not any kind of official guideline. That's just
my own opinion. There aren't any guidelines yet allowed for this.](https://image.slidesharecdn.com/021920ala-newconcepts-representativeexpressionsandmanifestationstatements-200220173135/85/New-Concepts-Representative-Expressions-and-Manifestation-Statements-Transcript-February-2020-17-320.jpg)
![>> Colton Ursiny: All right, thanks, Bob.
Is it okay to choose the first manifestation for the representative expression for the
work?
>> Robert Maxwell: Sure, yeah. I mean, that's -- we usually use the first
manifestation to describe the expression, and so that first manifestation would include
elements that we would use to describe the expression, which would go up into the
representative expression element in the work description.
>> Colton Ursiny: All right, thanks.
Has input for manifestation statements been gathered from software developers?
>> Robert Maxwell: I don't know. I do not know the answer to that question. If
there's anybody here who does know the answer, go ahead and tell us, but I don't
know.
>> Colton Ursiny: All right, thanks, Bob.
And let's see, this question here might need some elaboration, but is italic relevant to
the new RDA transcription?
>> Robert Maxwell: No. The RDA manifestation statement instructions do not
distinguish between different types of fonts, so italic versus Roman. I did suggest
during the presentation that it might be useful to allow us to do that, and it might be
good for future development in RDA to distinguish between different type styles or fonts,
but right now, in the guidelines for manifestation statement, that does not take into
account.
>> Colton Ursiny: All right, thanks, Bob.
Let's see. We have one last comment in here. There are lots of historical
bibliographies out there that highly resemble basic transcriptions. It might be viable to
scan a lot of those and give some of those -- historical bibliographical descriptions a
whole new life.
>> I can only say I completely agree with that. Actually, scanning them would be a
good way to ingest the information, and historical bibliographies probably go contain
something very, very close to what RDA defines as a basic transcription of the
manifestation statement. So I think that's a good idea.
>> Colton Ursiny: All right. Well, that's all of the questions about the workshop that
we have right here, Bob. I'll give people a moment to put in any last questions they may
have, and while we do, Bob, I just want to take a moment to thank you and ask if you
have any other final thoughts that you would like to wrap up with.
>> Robert Maxwell: The only thought is I think these are, as probably some of you
may know, from comments I make on -- I don't always agree with everything that goes
on in the development of RDA, but these two things that I have talked about today are
really good developments, and I hope that we have all, you know, are thinking good
things about them, because I think -- [ coughing ] -- excuse me, they will be very helpful
to us in our cataloguing in the future. I really believe that.
>> Colton Ursiny: All right, thank you, Bob. We didn't get any other final questions in,
so I think we can go ahead and wrap up here. Thanks again, Bob, for the great
presentation, and thanks to all of you for joining and for all of your great questions.
We got one last question in though. Since we do have time, I will take a moment here
to ask this.
Why does basic transcription not allow for markers for line breaks?](https://image.slidesharecdn.com/021920ala-newconcepts-representativeexpressionsandmanifestationstatements-200220173135/85/New-Concepts-Representative-Expressions-and-Manifestation-Statements-Transcript-February-2020-18-320.jpg)
New Concepts: Representative Expressions and Manifestation Statements Transcript (February 2020)
- 1. American Library Association New Concepts: Representative Expressions and Manifestation Statements February 19, 2020 CART CAPTIONING PROVIDED BY: ALTERNATIVE COMMUNICATION SERVICES, LLC * * * * * Communication Access Realtime Translation (CART) is provided in order to facilitate communication accessibility. CART captioning and this realtime file may not be a totally verbatim record of the proceedings. * * * * * >> Colton Ursiny: Hi, everyone. This is Colton Ursiny coming on to do another sound check. We are going to get started in about five minutes. Thanks to everyone who has introduced themselves in the chat space. We are happy to hear from you, and if you haven't already, please feel free to chime in. We'd love to hear who you are, where you are from, anything else that you would like to share. I'll do one more sound check before we get started, and in between sound checks, there will be only silence. Thank you. Hi, everyone. This is Colton Ursiny with ALA Publishing eLearning Solutions coming on to do the final sound check for today's workshop. We are going to get started in just under two minutes. If you haven't already, please feel free to introduce yourself or your group in the chat space, and please feel free to continue using that chat space to introduce yourselves, to discuss with one another as we get started. No need to go quiet just because we'll be kicking off in two minutes. So, stand by, and we'll be starting shortly. Thank you. Okay. It's noon Eastern time, so we are going to get started. Thanks for being with us today. I'm Colton Ursiny, with ALA Publishing eLearning Solutions, and we are happy to bring you the first session in this RDA online orientation series on New Concepts: Representative Expressions and Manifestation Statements, with Robert L. Maxwell. I'm going to cover some brief technical things to help everything run smoothly, and then we'll get right to it. The stage event is being live captioned. You can view the captions in the multimedia viewer in the lower right-hand corner of your screen. You just need to click on the arrow next to the text multimedia viewer. If you see an external site message in the viewer, as is seen in this screenshot right now, just click continue, as the captioning site is a safe site. You can use the chat area to interact with the presenter and your fellow attendees. The chat space is on the right side of your screen. You'll see that some chat has already occurred in the area. You can chat in that space at any time. If you do not see the chat window, please click the chat icon in the bottom menu row. If you have technical questions during the event, questions about audio or trouble using WebEx, we ask that you private chat host. We will help you personally without interrupting other discussion in chat. The private chat, simple click on the pull-down window, choose host
- 2. and type your chat and it will go directly to us. We will have a Q&A session halfway through and again at the end of the workshop. To ask questions during the workshop, type your questions into the chat box and make sure that the "to" box is set to all participants. You can chat your questions at any time. We'll keep track of them when we pause for Q&A. Please keep in mind that we have got close to 200 people here, so we might not get to every single question. If your audio breaks up or drops out during the presentation, you can hang up and click communicate at the top of your screen or three dots at the bottom, then select either use computer for audio or call in. If you're listening through audio broadcast and you notice an echo, make sure that you don't have two audio broadcast windows open simultaneously. If you do, simply close one. Please note internet audio quality can be affected by any number of factors in network speed or traffic. If you're having trouble, try clicking disconnect and then reconnecting. We are recording this presentation. Within a day, you'll receive an email with a link to the slides and to the archive recording, which includes the full audio, video, and chat record. ALA Publishing eLearning Solutions offers an array of workshops and online courses throughout the year. You can view additional eLearning opportunities on the ALA store. And we are happy to have Robert Maxwell with us for today's workshop. Bob is a senior librarian at the Harold B. Lee Library at Brigham Young University. He has taught cataloguing at Brigham Young University and the University of Arizona school of resources and library information science. He holds a JD from Brigham Young University and a Ph.D. in classical languages and literatures from the University of Toronto. And as I am sure many of you know, Bob is the author of Maxwell's Handbook for RDA from ALA editions. With all of that out of the way, I'm going to turn things over to Bob. Let's just get you off mute here. Welcome, Bob. >> Robert Maxwell: Thank you, Colton. Welcome everyone to this webinar in the RDA New Concept series. We'll be discussing two separate topics today, representative expressions and manifestation statements. I just want to note that this is a repeat of the webinar that was given last summer, although it's been updated. That's the same for all of the sessions in this series. I want to apologize to begin with for my voice. I am coming down from a rather severe cold, but I think I am on the mend, so I hope I will last through the whole hour. Okay. Let's start by talking about the new RDA. Where does it come from? Sometime ago, the RDA steering committee agreed that RDA would be based on current international cataloguing models, and so the revised RDA, which is likely to become official, although not immediately implemented in December of this year, is based on the new IFLA Library Reference Model or LRM. The -- restructuring of RDA that has been going on for the last two years. LRM is a consolidation of functional requirements for bibliographic records, Functional Requirements for Authority Data, and Functional Requirements for Subject Authority Data that most of us have become quite familiar with over the last decade or so. The original authors took on the task of analyzing the bibliographic universe to see if it could be divided into discrete entities that could be used in an entity-relationship model. They found four bibliographic entities which were work, expression, manifestation, and
- 3. item, as well as a number of other entities that could have a relationship to these four entities in one way or another. The diagram was taken from the original FRBR, but the four entities remain in the consolidation of the documents which is known as the library reference model, as I mentioned, or the LRM. In link data terms, which we'll talk about more extensively in a couple of weeks after the webinar, one of the entities can be diagrammed using a series of RDF triples. In short, an RDF triple is a statement describing an entity that has a subject, which is called a domain in RDA, a predicate, which can be the alternate label in RDA, and an object, which is called the range in RDA. I have an -- if you go into the new RDA and click on element reference in any of the -- in the ID of the elements, you can see these things that I have been talking about, the domain, the range, and the alternate label. The domain is defined as the RDA entity that is described by an element. This element can be used -- this means that the element can be used to describe the RDA entity, which is called -- described as its domain. So, for instance, expression -- well, in link domain terms, the subject is an RDF. Range means the RDA element that is the value of the element. That means that the element can be related to the RDA element and described in terms of its range. The alternate label can be used as the predicate on RDF triple. Here are some examples in fairly high terms of these triples. In the first triple, work is the domain, expression is the range. So we can say work as an expression of a work, and that's the expression. That's an RDF triple. That same expression is now the domain in the second RDF triple, and manifestation is the range, so we can say that expression has a manifestation of the expression, and that's a manifestation. This all seems pretty obvious at this level, but this is how the structure works, and similarly, that manifestation becomes the domain in its own RDF triple with either message range. So let's take a more concrete example. Alice in Wonderland, that's a work. Lewis Carroll's English language text is an expression of the work, and when this expression was published by McMillan, it became an expression. The two RDFs are Alice in Wonderland as an expression, that is Carroll's English language text, and that same expression -- on the screen, the English language text has a manifestation, and that's an 1865 publication by Macmillan. So these are examples which are important in the link data environment. So, in the model, the framers who made up this model were very careful in defining the entities to ensure that they were completely distinct from each other. Attributes of one entity were not used to describe any other entity, so this is a basic thing about entity relationship modeling. Entities have attributes that be used to describe instances of the entity and distinguish one entity from another, and at the model level, these entities have to be defined so that they do not overlap, meaning that tributes used to describe one entity can't be used to describe another entity. Now, in the current text of RDA, these are the attributes of the entity called work. You can find them in the RDA tab, under six and seven, and they are title of work, form of work, date of work, place of origin of work, other distinguishing characteristics, history of work, identifier for work, and then for musical works, they are medium performance and numeric designation and key are attributes that can be used to describe a work, and then under chapter 7, we have several other attributes that can be used to describe a work, the nature of the content, coverage, coordinates for a map, an equinox, epochs,
- 4. intended audience, system of organization and dissertation or thesis information. And these are all attributes at work in the current RDA. And they are intended only to be used in descriptions of works, not in descriptions of other entities, like expressions or manifestations. This is the -- a visual showing the attributes of expression in the current RDA, so, in current RDA, expression has attributes of content type, date of expression, language of expression, other distinguishing characteristic and identifiers. Those are in chapter 6, and then in chapter 7, all of these other attributes can be used to describe an expression, summarization of the content, capture of language, that's a pretty important one, form of notation, accessibility content, illustrative content, supplementary content, color content, sound content, aspect ratio and all of these other things for music that can all be used to describe an expression in the current RDA. Once again, they are only intended for use in description of expressions, not other entities such as descriptions of works. And this next slide is showing the benefit of the reorganization of this, in what I'm going to call the 2020 text of the RDA, just the new -- the RDA. And in bold are some new things that are different from the current RDA for the elements of work. So, category of work is a new element. It is renamed. It used to be called form of work. It's more or less the same element. New elements include extension plan, frequency, -- has been moved to work, as is the same with key title. These have all been moved from the manifestation entity to become attributes of the work entity. Note on the metadata work is new, and I'm sorry we don't have time to talk about that right now, because the metadata works are quite interesting -- [ off mic ] used to be an instruction for a relationship, and it's now an attribute for work. The recording source is new, as are scope of validity and source consulted. Missing from the work attributes in the 2020 text are those three music-related attributes, medium, performance, key, and intended audience. That's because in the new text, they have become attributes of the expression, as we will see here. Neither of these lists, by the way, include relationship elements or any subelements or [ off mic ] elements. So let's talk about what's new in the expression in the 2020 text attributes. Some new elements include category of expression, the extent of the expression, intended audience of the expression. This is new to expression, that it replaces what currently is in the work as the intended audience. It has been moved from the work level to the expression level. Interactivity mode is a new element. Here are these three music-related ones, key of expression, medium performance. These two medium performance attributes have been moved to the expression level. Title of expression is a brand-new attribute that has never existed, did not exist as a new vision in the current RDA, which is interesting to me. Expression was the only entity in the current RDA that did not have a name, and now it does. There is a -- an attribute called title of expression for the expression entity now. Deleted from expression are the performer, narrator and/or presenter element and the artistic [ off mic ] element, which have been hard to remove from the RDA but are still in the table of contents. Okay. That's just an overview of what's changing in the new RDA, as far as these attributes of these two elements, these two entities go, in work and expression. So why are we even discussing this? What's the problem?
- 5. I have mentioned already, entity relationship principles require strict boundaries between the entities, as far as their attributes go, for describing them. So, you can't use an attribute defined for one entity to describe another entity in the way the principles work for entity relationship databases. The new toolkit is quite clear about what the domain of each element is. It's much clearer than the current toolkit, in fact, and this is necessitated by requirements of link data structures. However, the problem is human beings don't necessarily think of the bibliographical universe in these terms, and in particular, database users often associate attributes of expressions with works. So, for example, many people when they think of Alice's Adventures in Wonderland or Alice in Wonderland, they think, oh, it's a novel and it's a text and it's in English and it's for children. Well, the first one, according to RDA, is an attribute. In fact, it's a novel that is applicable to the work itself, but the other three, the fact that it's a text and it's in English and it's for children are not attributes of the work, even though normal human beings think of them in that way, and so they are, in fact -- those three are in fact, in our model, attributes of the expression. So if we are in an entity relationship database structure or in linked data, we need to respect those differences. So, under current RDA, it is not appropriate to include, for example, language or content type in a description for a work, and the best kind of descriptions that are most clearly for works are authority records for works. So under the 2020 text, it would be inappropriate to record these two, intended audience, because none of them are elements that can be used to describe a work editing. So here we have a work record for the authorized -- well, a work record for Alice's Adventures in Wonderland, which includes the authorized access point or [ off mic ] field. It would not be appropriate to include the fact that it's text or that it's in English there, because those are not work-related attributes or the fact that it is for children. The fact that it is a novel that is fantasy fiction or nonsense fiction is perfectly appropriate to record, because those are work level attributes. It's clear that cataloguers do want to record this type of information in descriptions of works, and I don't think they are wrong to want to, although the current RDA doesn't allow it. So here are some examples from the current name authority file, which are, at the moment, not correct. So here's an example for a serial called the Ballantonians. It's true that it is in English, but that isn't an attribute, and since this is a work level record, the [ off mic ] isn't appropriate. Here are a couple of more examples, a poem by Paul Verlaine. It is true that it was originally composed in French, but at the work level, that's not appropriate to record. It should be recorded in an expression level record. Same for Boy Scouts in an airship, something that I have not read but which I should read someday, the fact that it's in English or that the novel is not French. So, clearly people want to record this information, and I don't think they are wrong to want to do it, but under the current structure, it's not correct to do this. And it's important, I really don't advocate being correct just for neatness and correctness sake. It's important for the database structure to follow rules of the database. So is there a -- I agree that this is a problem. So is there a solution? There is. It's called representative expression. This is a new concept introduced with LRM. When people were talking about it at the beginning, it was initially -- there was initially a possibility proposed to choose one expression as the, quote, "representative
- 6. expression," unquote, of work, in which you would describe a work, and then you would pick one of the expressions as the representative expression. However, when it was finally published, LRM chose instead to design an attribute of the work, which is called the representative expression attribute, and its definition is an attribute which is deemed essential in characterizing the work and whose values are taken from a representative or canonical expression of the work, so it's an attribute that people think of as characterizing the work that are actually values of an expression of the work. RDA has adopted both the concept of representative expression as a relationship as well as the representative expression attributes, which are work-level elements. In other words, back to the previous slide, RDA accepted both of these possibilities, either choosing an expression as the representative expression or recording attributes of the work as representative expression attributes. Here's what it looks like in the new RDA. There is an element called representative expression, defined as an expression that's considered the canonical source for identifying a work. So, this is how we could do it as a relationship. We could say Alice in Wonderland has a representative expression, the 1865 English language text. I think if we had to pick one expression of Alice in Wonderland as the representative expression that represents all of the expressions in the work, I think most people would agree that the 1865 English language text is that representative expression. So we could actually record that starting from day one with no need for new work accounting of any kind. We could record in the work record for Lewis Carroll's Alice in Wonderland, here's the same work record that we saw earlier, we could record the 500 field to record a relationship to the English language 1865 expression and the relationship designator would be representative expression probably, something like that. This has never been established, so I made up a hypothetical authorized access point for that expression. Now, it seems reasonable for a work like Alice in Wonderland to be able to pick a representative expression, in this case, the first expression. But what about another work, like The Odyssey by Homer? This is an ancient Greek poem, an adventure tale, but what would its representative expression be? What you might call the original expression, in that it came from the mouth of the original narrator, does not exist. There are dozens of scholarly editions, each purporting to be as close as possible to the original text, each different, so each a different expression. So is it possible to choose one of those as the representative expression of The Odyssey? Or what about the -- what is the representative expression of that? The score? Which score. A recording? Which recording? So that's the first problem that I see with picking a representative expression and making a relationship like this. Second, who has the authority to declare one out of all possible expressions in a given database as the representative expression. What if there is disagreement? So, in my opinion, a much more practical solution to the problem is that which was adopted by LRM and is available in the 2020 RDA, to use the new representative expression attributes and record them in the work description, whereas the new toolkit suggested we call this the metadata description set for the work. Before we go there, I need to point out that RDA's implementation of representative expression does not require that one expression be chosen as the representative expression. A relationship can be created from the work entity to more than one expression, all of which might be considered representative in one way or another. So, for example, here we have the
- 7. original score and the original cast recording of Wicked, which are both expressions, and they were both created in 2003. In the revised RDA, each could be considered a representative expression, and so the box to the upper left is an authority record for the work, and it could have two separate 500 fields relating to two different expressions as representative expressions, and then the other two boxes are hypothetical authority records for these two expressions. Note that subfield H has not yet, looking at the authority records, has not yet been incorporated for importing content type, notated music note here, but it is just on the verge of being implemented, and we will soon be able to use H. At the moment, these are being recorded, and they have been recorded ever since we have been doing RDA as subfield S, but I am just explaining in the MARC coding, we will soon be able to use subfield H. Okay. Here's another example, P.D.Q. Bach's Little Pickle Book. Here we have the original recording and the score. These are both expressions created in 2003, and in the revised RDA, each could be considered a representative expression, so just like in the example of Wicked, we could have the authority record for the work up in the upper left with the little pickle workbook. It is a musical parody, and it can have two separate relationships two to two different expressions as representative expressions. So, RDA does not require us to pick just one representative expression. I still think in many cases, it would be difficult to even pick a multitude of these as relationships. But to return to the point we were making a few slides ago, possibly or probable a more practical solution that was adopted by LRM and is available in the 2020 RDA is to use the new representative expression attributes and record them in the work description, the metadata description set for the work. These are the available representative expressions elements in the 2020 RDA for a work with textual aspects, so in the work record, we could record, if we thought it appropriate, the color content of a representative expression, the content type of a representative expression, the date of a representative expression, the extent of a representative expression, the intended audience, language, and the script. So, in terms of Alice in Wonderland, for instance, probably the color content isn't very relevant, the content type is text, so that might be useful to recording the work description, the date. The representative expression probably coincides with the date of the work. The language is something people find relevant, and the intended audience. These could all be recorded under the new RDA in an authority record or a metadata description set for a work. Works with moving image aspects, these are the possible representative expression elements, aspect ratio, color content, content type, date of capture, date of the expression, duration, extent, intended audience, language, media performance of choreographic content, media performance of musical content, place of capture and sound content. All of these could be recorded in a work level description for a moving image work. Sound recording, works with sound recording aspects can -- will be able to include as elements content type, date of capture, date of the representative expression, duration extent, intended audience, key language, media performance. These two media performance elements, place of capture and sound content. Notated music will include content type, representative expression, duration, extent, media performance and script.
- 8. Cartographic types of work can include color content, content type, date, extent, intended audience, language, protection, scale, and script. Now, I have mentioned the word metadata description a couple of times. This is new vocabulary for what we in the past have called records, for example, authority records for persons. I think it's a good new way of talking about these things, so I recommend that we try to start using this new vocabulary. The glossary definition of a metadata description set is one or more metadata statements that describe and relate individual instances of one or more RDA entity. And a metadata statement is defined as a piece of metadata that assigns a value to an RDA element that describes an individual instance of an RDA entity. Now, these two definitions are kind of tough to understand without an example, so, here we go. Here's an example. An RDF triple is an example of a metadata statement, so as a subject, predicate and object, so a person is an RDF element, and the object is a time span. Here's an example of a metadata statement. Terry Pratchett has a date of birth of April 28th, 1948. So that is one metadata statement, and when you have a set of these triples and combine them to describe an entity, that's called a metadata description set, and when you think about it, that's what our authority records certainly -- I'm pretty sure we could call them metadata description sets. Here is an example of a description of a work but not as a metadata description set, and it could be converted into MARC, but at the moment we are going to look at it as a metadata description set. So, this particular work has a creator, Carroll, comma Lewis. It has a preferred title, Alice's Adventures in Wonderland. It has a category of work, which we are calling form of work, novels, fantasy fiction, nonsense fiction. Its date of work is 1865, and now we can start in 2020, we are going to be able to include these representative expression elements. So we can include in the description of the work, it has a language of representative expression, and that's English. It has a content type of representative expression, which is text. It has an intended audience of representative expression, which is children, and then it has an authorized access point for the work, Carroll, Lewis, 1832-1898 period, Alice's Adventures in Wonderland. So, this is going to be possible in the new RDA, and it's also going to be possible to translate this into a format that we are more familiar with. Here is the metadata description set for Wicked, so it has a creator, Schwartz, Stephen, a preferred title, Wicked, and it is a musical. It was created in 2003. It is performed music, and it is notated music, so that's the content type, and right now it's not appropriate to put that in the authority record for the work, but it will be appropriate in the future, once we have implemented the new RDA. It has a language of representative expression, English. It was captured on November -- this is a recording. It was captured on November 10th, 2003. It has a place of capture for the representative expression, and it has a duration for the representative expression, and then, finally, the authorized access point for the work is Schwartz, Stephen. So, just a reminder, this is a future description of the work, not of the expression, and the work will be able to include elements that are drawn from the expressions, which are called representative expression elements. Here is the same for Little Pickle Book, and I'm just going to point out the representative expression elements are the content type for notated music, the language, English. The date of capture, the place of capture. I want to point out there
- 9. really is a place called Hoople, North Dakota, but since this university is a fictitious university, I believe, where P.D.Q. Bach supposedly did his work, I wonder if this is really where this recorded, but it is what the recording says. The duration is an expression element, and there are two different durations, so I have recorded them both. One duration for the performance time is found on the score, and another duration for the actual time that was taken in the recording. Here's an example of a work that we have not talked about, Claude Debussy's Syrinx. It has a category of the work, art music, and the date of the work is 1913, but the date of the representative expression is different because this work was created in 1913. It has been well researched, and it was created in 1913, but it was not published until 1927, so here's an example of where we might have a different date of representative expression from the date of the work that could be recorded in a description of the work. Medium of performance again is a representative expression. The tabs on this slide have gotten a little messed up, so the medium performance is flute. Content type is notated music. Those are all representative expression elements, which right now can't be recorded in work authority of records but will be once we have implemented the new RDA. Here's an example for a film, Wizard of Oz. It is a fantasy film, has a date of work, and all of these next five or six elements are elements drawn from one of the expressions of this work, and so I think we can see in this case that these are things that people think of as being important to the work itself, not just to an expression, the fact that it's a moving image, that it is both -- this is very famous for being both in black and white and in color or monochrome and polychrome, as RDA recommends we use the term. The aspect ratio may be important to some people, when was it captured, where was it captured, how long it is. These are all drawn from the expression from the work. And a map, an example of a map. Here's a map of Eswatini, and it includes some representative expression elements, cartographic image, the fact that it is color or polychrome. So these are examples of descriptions in which we could enhance the work descriptions by including representative expression elements. Now, one of the reasons that I haven't given you these examples in MARC is because, as we will see in a few minutes, there isn't any way to code the representative expression elements in MARC yet, but that will be coming, I believe. So, actually, here we are. How can I do this in MARC? The MARC Advisory Committee hasn't yet created any coding for representative expression elements. Here are some possibilities. They could use existing coding and just, as most of these elements already have a [ off mic ] coding for use in the expression authority records. Or another possibility is to use existing field coding, but define different subfields for the representative expression elements, or just make up totally different field coding. My opinion is the second possibility would be the best, so use existing field coding, but define different subfields for the representative expression elements, simply because I think we need to be able to clearly show in our records what these elements where, and if they are used both for representative expression and also for regular expression, it might not be clear. Okay. So, here is an example where I have kind of made up things. For the Wizard of Oz, the date of the work is 1939, and so that's perfectly fine. We can have another
- 10. 046, the date of capture of representative expression. I have put dollar sign X in there to show that it's not defined yet, which we can do. 336 we can include with a subfield of some sort for a two-dimensional moving image. The fact that it was captured in Culver City, California is subfield. The language of representative expression is 377 for the subfield, and so on. With the other three elements at the bottom, there isn't currently a coding that could take them, I don't think, in a MARC -- but I think we need to be able to record those things. Okay. We can pause it now, if we have any questions, which I haven't been looking at the chat, so you'll have -- somebody will have to ask me. >> Colton Ursiny: Yep, not a problem. We have been keeping track of questions as they have been coming in. First here, we currently use authority records for different translations of works. For example, Defoe, Daniel, 1661?-1723, Robinson Crusoe, Spanish. How would that work with the current system? >> Robert Maxwell: This isn't going to change anything about what kind of authority records and how many we make. For instance, under current RDA, if we have more than one Spanish translation of Robinson Crusoe, it is possible to create more than one authority record for each individual expression, but that wouldn't make really any difference in using -- for recording representative expression elements in the work level. This all would take place in my mind in authority records at the work level, that is using representative expression elements. They wouldn't really change, I don't think, what we are doing at the representative expression level, like a Spanish translation of Robinson Crusoe. >> Is this a metadata set for Wicked the musical as performed, or recorded music and in the recorded cast album. >> Robert Maxwell: So slide 42 was it? >> Colton Ursiny: Yeah. >> Robert Maxwell: This is not a metadata description for -- it's for the work. It shows a good point for the reason for structure. The performance and the score are two different expressions, so they need to be described on their own expression authority records, which would be related to this work record. The point of this work record is that it is bringing in some elements from all of those expressions that the person who wrote the record, the metadata description set, thought were important to describing the work itself. So, this can include elements from both the expression for performed music and the expression for notated music, but I just wanted to make clear this is not a description of those expressions. This is a description of the work, which brings in elements from the expressions to help describe the work itself. So, this is not going to -- in my opinion, this is not going to obviate the need for separate expression authority records or set description sets for expressions. They would exist separately and be related to this record for the work. >> Colton Ursiny: All right. Thanks, Bob. This person asked, just curious, how can a political map, a work, not be a cartographic image in expression? >> Robert Maxwell: Well, it is. I mean, you're probably asking why isn't that an element of the work itself. Some works, some elements can only be -- there's only certain elements they can be, so you're right. In fact, that's sort of coming to the heart of why representative expression is a useful concept, because a political map obviously
- 11. has to be a cartographic image, but that is, in fact, an element as defined in RDA of expression. So, yes, of course people are going to be thinking of this map as a cartographic image, and so that's exactly the point of why it's a good idea to be able to bring that fact into the description of the work, as we have done here. So I think that's -- your point is exactly very relevant, and it is a point in favor of implementing the concept of representative expression. You could say the same thing about a movie, where your average movie -- I thought [ off mic ] since it is a film could have possibly more than one expression, but in most cases, the film has only one expression. So all of the elements of expression could be used to describe that film, but yet in the way that we have parsed out RDA, without the concept of representative expression, we can't use those elements to describe the work itself. We have to have expression descriptions. So I think this is a good step forward to be able to describe representative expression in the work description or authority record. >> Colton Ursiny: Thanks, Bob. Someone asked a follow-up question relating to this one. They said, "A political map could be an RIS file of geocoded boundaries, not an image?" And they have a question mark there. >> Robert Maxwell: All I can say is that's probably true. I mean, I suppose that's a good comment. >> Colton Ursiny: All right. Thank you, Bob. Let's see. I think time for one or two more questions before we move on. Are there guidelines for where to find filming dates and location, which aren't necessarily described on the manifestations? >> Robert Maxwell: I am not an expert in that, and so I am not sure there are guidelines, but if there are, maybe there's a film cataloguer in the audience who could tell us, but if you are interested in a good place to find this sort of information, IMDB.com is a good database that includes a lot of this sort of information, if you're interested in filming dates and locations. And Adam says they are not required. I think he's talking about the same thing. That's true, they are optional. >> Colton Ursiny: All right, great. Thank you so much. Last question before we move on, are there sufficient subfield opportunities in corresponding MARC fields to accommodate these additions and adjustments? >> Robert Maxwell: I have seen a discussion paper from one of the members of the MARC group that's working on RDA, and it is true that we are running out of subfields. I think in these cases, for this case, I think there are enough subfields for the representative expressions, but that could be a problem. That's correct. >> Colton Ursiny: All right, Bob. That's all of the questions for now, so I'll turn it back to you. >> Robert Maxwell: Okay. So now we are going to turn to something completely different, manifestation statements. So, before we start, I would like to say this is -- last time when I did this, people got -- some people got quite upset at thinking this is what we are all going to have to do. I want to preface this by saying that this is a new option being offered in RDA for recording information, and it's not going to force us all to change if we don't want to. And there are good reasons for not changing what we are doing right now in recording information, but there are also good reasons for having this as an option, so I just wanted to preface it that way because we had a little bit of upset
- 12. last time I gave this. Okay. So if any of you have the new RDA app, and to get there, you can bring up your radio RDA and click on -- let's see, what do you have to click on -- I need to look at this myself, in case any of you want to follow along, click on the explore RDA toolkit site, and you'll be able to see what I have got in the slides, and you can also explore a little bit more if you want to. So to find out more about manifestation statements, let's go to the RDA beta site and have a look at what it has to say. So, to find this information, go to the guidance tab, click the are drop-down, and choose manifestation statements. Manifestation statements are defined as data found on manifestations themselves. This first section, under principle of representation that tells us that a manifestation can carry information and describes the manifestation, and it is recorded to follow the principle of representation, reminds us of the basic cataloguing principle that an entity should be represented as it represents itself, and we use this principle throughout cataloguing for many, many things that we do. It's practically unconscious at some times for us, it's so ingrained in our cataloguing. Now, manifestation statements are, by nature, unstructured, and so RDA only offers unstructured description as the recording method. Note there is an option for a normalized transcription under recording and manifestation statements further down on the page. We'll see the second option talks about recording and normalized transcription. Normalization can be said to be imposing some sort of structure to a transcription, but it is not enough in the minds of the creators of RDA to tip it over into the area of structured description. So, let's briefly look at the normalized transcription rules. So, if you're following along in RDA, the beta RDA, click on guidelines on normalized transcription, and there is really nothing new here. These are all the same guidelines that are in the current RDA. We have been using normalized transcription rules in cataloguing from time immemorial. The links are all to instructions that we are quite familiar with, so let's just look at two, capitalization and punctuation guidelines. So, the left what you get if you click on capitalization in the previous slide, and the right is what you get if you click on punctuation. These are all things that we are used to. We capitalize the first word in a statement. We capitalize proper names. We include punctuation we have found, but we can omit, modify, or add punctuation for clarity. The other conventions that we follow for other aspects of transcription are all there in the revised RDA instructions for normalized transcription, and they will remain available for our use, and I assume that we will, most of us, will probably continue using these, these guidelines and conventions. But if you think about it, normalizing data is not representing the entity as it represents itself, so the new RDA offers a method of more closely representing the manifestation as it manifests itself, which is called basic transcription. So, going back to the basic guidelines on manifestation statements, we are now going to click on guidelines on basic transcription, if you want to follow along. And this is what you get to, if you do that. So, let's read these. We are told to preserve the capitalization, punctuation, numerals and diacritics as they appear on the manifestation. We are to omit symbols, or rather, images, that are to be interpreted as text that are not available [ off mic ] -- transcription, so, for example, if we are transcribing a rebus, we might have to omit the symbols in the rebus. We are to tribe the components of a manifestation statement in the order in which
- 13. they appear, no transposing, like we do when following the other guidelines in the RDA. Use these two backward slashes to limit separate components of a statement. This slide is somewhat confusing and I think needs a little bit more explanation. I have inquired of the authors of this, and this does not mean dividing between, for instance, line endings. It means dividing between major components, such as between the title and responsibility statement and publication statement, for example. Continuing, we are to collapse -- well, use a mark of omission to indicate the omission of text within a statement, and as an option, we can use it at the beginning or the end of the statement. There apparently is no distinction between using a mark of omission to indicate that the cataloguer omitted something and using a mark of omission to actually copy what's on the statement that we were transcribing, which may or may not be a problem. We are to condense all white space, including multiple spaces, new lines, indentation tabs and block alignment and so forth. This is the alignment that tells us that we are not going to take into account the fact that we are starting new lines in -- on the thing we are transcribing. We are to deduplicate a whole bunch of things. So, you may also note that these guidelines do not take into account type size, typeface, for example, Fraktur versus Roman, in determining manifestation statement, type style, such as italic versus Roman, or [ off mic ] or things that appear in the line as a resource. Now, returning to the basic page on manifestation statements and scrolling to the bottom, we learn that there are a number of different categories of manifestation statements, and we'll begin by clicking through to the general instructions at the top for manifestation statement. So, the first thing to note is the general manifestation statement element can be used to record any statement appearing in a manifestation that is thought to be significant to users. So, it does not need to correspond to one of the traditional areas or elements of the description, so, a transcription of the entire title page could be recorded as a manifestation statement, or if you are transcribing something to include an [ off mic ] note, that is an example of a manifestation statement. If you are transcribing the form of a name from somewhere in the manifestation to justify decisions you have made in authority work, that is a manifestation statement. These are all manifestation statements. Second, although it's not shown here, if you continue to scroll down in this part of the new RDA, you will see, as has already been mentioned, that the only recording method available is unstructured description. And, finally, although the remainder of my presentation today is going to focus on basic transcription, basic and normalized description are both legitimate choices and both count as manifestation statements, so do not get the impression that manifestation statements is only this exact basic transcription that I'm going to be talking about the rest of the time. The normalized transcription also counts as a manifestation statement in the new RDA, which is the kind of transcription that we have been doing. So, returning to the manifestation statements page, let's now look at a couple of the narrower elements. The title -- I'm sorry, this is called -- this has got a new name, which is hard to say, because we are all used to saying title statement and responsibility, and it is now called the title and responsibility statement. These can serve as examples, and the other types of manifestation statement work in the same way as these two. So let's look at manifestation title and responsibility statement. This is a statement that
- 14. includes titles and names of agents responsible for the content of the expression, and there are, surprise, conditions for the presence of title proper, other title information and statement of responsibility. And the basic instruction is simply to record the information. Thereon, we are reminded that we can use basic or normalized transcription, and then we can only use unstructured description. To summarize, the manifestation publication statement is a statement that includes information about the publication, release, or issuing of a manifestation. And there are conditions for the presence of the name of publisher, date of publication, and place of publication, and the basic instruction, again, is simply to record the information. Thereon, we are reminded, such as we were with the title and responsibility statement, that we can use basic or normalized transcription and that it is an unstructured description, in terms of the types of description in the new RDA. So, keeping this in mind, as an aside, if you click through to the publication statement element in the first condition, you might see that there are some options there for recording the publication statement in traditional order, place of publication, name of publication, name of publisher and date of publication. I draw your attention to the fact that this is under recording a structured -- manifestation statements are always unstructured descriptions, so a manifestation publication statement, as opposed to a publication statement, which are two separate elements in the new RDA, the manifestation publication statement will always be recorded in exactly the order found. So, in the new RDA, we have to be more careful than ever to make sure that we understand exactly where we are and what we are looking at when we are trying to understand what the guidelines are asking us to do. So it's going to take some getting used to, but we will all get used to it, I assure you all. But we are now in the elements that are called manifestation something or other, so this is manifestation publication statement. It's not the same as the element called just plain publication statement. Okay. So let's see how this might work with a real example. Here we have a novel by Terry Pratchett, and so the manifestation title and responsibility statement would read exactly what we found in the title page, so with all of that same capitalization, and the manifestation publication statement reads exactly as all of the stuff at the bottom of the title page, Harper, an imprint of HarperCollinsPublishers. Note that there is not a provision for that line between the word Harper and the phrase an imprint of HarperCollinsPublishers, so I was not able to include it in my manifestation statement. If we had a full manifestation statement for the entire title page, we would put the two backward slashes between the title and responsibility statement and the manifestation publication statement. This is sort of transcription that I think is most likely to be used when we ingest produced metadata, for example, OCR. I'm not thinking it's going to be that useful for us in our day-to-day work, because at least when I did this, it took me a lot longer to do it this way than to just use the normalized transcription, but it is an option that is available to us to use. Here's an example of the possible application of the manifestation statement concept. For ingesting digital information, this is a -- as a matter of fact, there is a cataloguing program that takes this digital information and transforms into normal-looking data for CIP cataloguers, but if I were in a hurry, I could adjust the information for using RDA by simply copying and pasting the information with no further information, which is exactly what I did to create this slide. Best practices have not been worked out for
- 15. manifestation statements yet, since nobody has any extended experience with them, so a question arises on this particular transcription, for which there currently isn't an answer: Should the vertical bar between the parts of the publication statement be included? It's not a mark of punctuation. On the other hand, it is a symbol that is available in my character set. Further, when I directly ingested the data using copy and paste, it pasted right into my manifestation publication statement, so I left it. Future best practice might be to omit this sort of thing. If so, however, I point out that having to go in and delete stuff defeats one of the purposes, at least if one of the purposes is to achieve greater efficiencies for cataloguers. Anyway, there are some comments on that particular example. Here's an example where a manifestation statement concept could be very useful for identification, using publications, especially publications of sheet music, like this, are notorious for absence of or misleading information about aspects of the publication, especially the date. Here are two pieces of music, each with the copyright date 1978 and no formal publication date, but the one on the right is actually a different manifestation from about 1995. There are a few other differences that could be used to distinguish, but one obvious distinction is aspects of the typography, which could be brought up by using basic transcription. This could be useful, for example, for a copycat logger coming to a set of records in OCLC and who might be confused about which record represents the piece at hand. So here are two manifestation statements, using basic transcription, based on these two separate manifestations. So, I think in this case, the manifestation title and responsibility statement does a very good job here of distinguishing between the two manifestations, although there are also a few other elements that distinguish. Another case where the manifestation statement concept could be very welcome is with the description of early printed books. Basic transcription lends itself to this type of description. As an aside, before we continue, books from this period include letter forms that aren't used nowadays, including the long S, which you can see in the words reason and dissolution on this title page, as well as several other places in the publication statement. These look a little bit like an F, but they are not Fs and they should never be transcribed as Fs. They are simply a form of the letter S with certain positions in the word. Even in basic transcription, they should be transcribed as an S, because that's what they are. Bibliographers -- this is possible because when such a book was produced, sometimes the printer had to go back and reset a page. Perhaps he got interrupted or discovered a mistake when the job was partway through. Because it was expensive to trash what had already been done and started over, printers often simply made the correction and then continued the production of the book, so there wound up being more than one version of the book. Since RDA doesn't recognize the concept of issue and state, which is what we are talking about here, which is something that different versions might be called, in the RDA LRM model, we would probably have to say that these were different manifestations. But how do you distinguish between them? Basic transcription of one or more manifestation statements can help to distinguish between these different manifestations. As a matter of fact, this particular book does exist in two different versions, as evidenced by the title page. If you look carefully, you'll see when the printer reset the
- 16. title page, for whatever reason, the new version was not quite the same as the first. I don't know the exact order of these two title pages from 1863, but let's assume that the one on the left was the first. When it was reset, the word counsel was spelled with one less L on the sixth line and the -- and dissolution is capitalized in the first but not the second. The typographical elements are different. Looking further at the publication statement, the one on the left has five lines, and the one on the right has four. Otherwise, the wording is identical, except the spelling with -- and the spelling of the word sign. Capitalization throughout is identical. So in this case the application of the new RDA in the manifestation statement is helpful, but not as much as it could be because it does not take into account line breaks, which is the main thing that distinguishes these two title pages. Here's another example of an early printed book that exists in nearly two identical manifestations, the distinguishment of the layout on the title page. I direct your attention to the differences between those two title pages, and these would be able to be brought out using the RDA manifestation statement basic transcription. The word English on the seventh line is all caps, and it is italicized on the left and it is in large and small caps and not italicized on the right. Three words down, the word dispute -- Zachary Coke is in all caps on the left. On the right, it is capitalized but translated in minuscule characters. This has yet to be developed. On the left, the word -- is italicized, but on the right -- the publication statement, unless I missed something, is identical between the two words, except the word sign in the second to last is capitalized in the version to the right. So, using basic transcription in the manifestation would be quite useful to a rare materials cataloguer, and it would partially between these two title pages. It fails to distinguish between them as well as it could, because some of the only differences would be in the word Zachary Coke, which is in all caps on the left. It might be useful with the notation of the line breaks and type styles, which would really be of interest in these two if -- surrogates. It has also been pointed out that the inclusion of a picture on the title pages would suffice for this purpose, and that is true, but it isn't something that we are generally able to do at the moment in our catalogue groups. The introduction of the concepts of manifestations omitted in basic transcription will be a great benefit, I believe, to many aspects of our description of resources, and I remind you again, though, that no one is going to be forced to do these basic transcriptions. The standardized transcription that we have been using will remain as an option for us, and most of us probably will continue to use that. So the question is how do I do this in MARC. Well, the cataloguer can choose to apply basic transcription in a MARC record right now. Even though it's not spelled out, it's allowed in the current RDA, but it is limited to the subelements in MARC, so it might be appropriate to do this 245, where it says A Novel of Discworld, Terry Pratchett, having inverted them, but perhaps not in the full manifestation statement without manipulating it in our current MARC environment. The second example is probably not appropriate. By the way, I'm aware that the registered trademark is not according to current policy transcribed, but there isn't any reason to drop it, aside from that policy and using the manifestation statement guidelines, we would probably transcribe that.
- 17. Okay. There isn't currently any place in MARC for a general manifestation of the full title page, which would be, as I have shown, below the first bulleted point. And there's not any provision in current MARC for a subelement of a manifestation statement that doesn't correspond to the expected ISBD order. For example, the manifestation title and responsibility statement. But the MARC Advisory Committee considered a discussion paper at ALA Midwinter just last month about manifestation statements that is expected to return as proposal at Annual. When it comes back as proposal, hopefully, and it is approved, it will probably be recorded in a newly defined 881 field, but it is not clear yet what subfields will be approved or how manifestation statements will be recorded within them. The discussion paper called for a designated subfield for each type of RDA manifestation statement. So here are some of the ways that came from the discussion paper as possible ways to record a manifestation statement. So here's this book called Crossing the Chasm. The discussion paper suggested this as one possible way of doing a manifestation statement. This looks like a pretty -- this is without any subfield codings, which is pretty difficult to take in as a human being, but it is a pretty much top to bottom manifestation statement, and it includes more than just one source from the book that's being described. So that's one possibility that was suggested in the discussion paper. Here is a second possibility that was discussed -- suggested in the discussion paper, which includes subfield coding, which looks more useful to me, because it's a little easier to figure out what's coming from where and what's going on. This probably is not the final subfield coding that will be approved, eventually if this proposal is approved, but it's an example of how it might work. And, finally, a final example gave kind of a label display, which I find, as a human being, the most useful of all three, because it tells us where we got the information from. The first [ off mic ] says we got this particular bit of manifestation statement from the title page, and the second one is information that came from the cover, which isn't shown on the -- I'm not showing you the back cover on this slide. In any case, don't try this yet. It hasn't been approved yet, but there are efforts under way to make it possible for us to record RDA manifestation statements from the RDA in MARC, and so stay tuned for further developments. At ALA Annual 2020, probably something will be approved, we hope. Okay. I can take some questions or comments if anyone has any. >> Colton Ursiny: All right. Thanks, Bob. One moment. Just pulling up the questions we have been collecting. Are there any guidelines in regards to representative expression for work, when the work is not a well-known work? >> Robert Maxwell: There were no guidelines at all for representative expression data at all. So, no, there really aren't any guidelines, so I'm not going to attempt to -- I guess I will say something that comes to the top of my head. I would think for a work that is not well-known, you would probably want to use as a representative expression the first expression of the work, and so the elements I personally would probably choose from the first expression, some of the language, for instance, would be the language of the first expression, but that is not any kind of official guideline. That's just my own opinion. There aren't any guidelines yet allowed for this.
- 18. >> Colton Ursiny: All right, thanks, Bob. Is it okay to choose the first manifestation for the representative expression for the work? >> Robert Maxwell: Sure, yeah. I mean, that's -- we usually use the first manifestation to describe the expression, and so that first manifestation would include elements that we would use to describe the expression, which would go up into the representative expression element in the work description. >> Colton Ursiny: All right, thanks. Has input for manifestation statements been gathered from software developers? >> Robert Maxwell: I don't know. I do not know the answer to that question. If there's anybody here who does know the answer, go ahead and tell us, but I don't know. >> Colton Ursiny: All right, thanks, Bob. And let's see, this question here might need some elaboration, but is italic relevant to the new RDA transcription? >> Robert Maxwell: No. The RDA manifestation statement instructions do not distinguish between different types of fonts, so italic versus Roman. I did suggest during the presentation that it might be useful to allow us to do that, and it might be good for future development in RDA to distinguish between different type styles or fonts, but right now, in the guidelines for manifestation statement, that does not take into account. >> Colton Ursiny: All right, thanks, Bob. Let's see. We have one last comment in here. There are lots of historical bibliographies out there that highly resemble basic transcriptions. It might be viable to scan a lot of those and give some of those -- historical bibliographical descriptions a whole new life. >> I can only say I completely agree with that. Actually, scanning them would be a good way to ingest the information, and historical bibliographies probably go contain something very, very close to what RDA defines as a basic transcription of the manifestation statement. So I think that's a good idea. >> Colton Ursiny: All right. Well, that's all of the questions about the workshop that we have right here, Bob. I'll give people a moment to put in any last questions they may have, and while we do, Bob, I just want to take a moment to thank you and ask if you have any other final thoughts that you would like to wrap up with. >> Robert Maxwell: The only thought is I think these are, as probably some of you may know, from comments I make on -- I don't always agree with everything that goes on in the development of RDA, but these two things that I have talked about today are really good developments, and I hope that we have all, you know, are thinking good things about them, because I think -- [ coughing ] -- excuse me, they will be very helpful to us in our cataloguing in the future. I really believe that. >> Colton Ursiny: All right, thank you, Bob. We didn't get any other final questions in, so I think we can go ahead and wrap up here. Thanks again, Bob, for the great presentation, and thanks to all of you for joining and for all of your great questions. We got one last question in though. Since we do have time, I will take a moment here to ask this. Why does basic transcription not allow for markers for line breaks?
- 19. >> Robert Maxwell: Well, I agree with the question. I don't know why. I didn't -- I'm not -- I was not one of the writers of RDA, and I think if I had been doing these guidelines, I would have included the possibility of marked as line breaks, and I do recommend that as a possible future development for RDA, but it's not there right now. >> Colton Ursiny: All right, thanks, Bob. And thanks again to all of you, and we hope to see you at the next session in this series. Take care, everyone. Bye, Bob. >> Robert Maxwell: Thank you.