Proposal on Sources of Information (RDA 2.2.2)

29 July 2016

Proposal on Sources of Information (RDA 2.2.2)


Submitted by: Renate Behrens, European Regional Representative


Print Friendly, PDF & Email

Tagged , , . Bookmark the permalink.

40 Responses to Proposal on Sources of Information (RDA 2.2.2)

  1. Kathy Glennan says:

    I am especially interested in comments on this proposal from those who are experienced in cataloging:

    1. Moving image resources
    2. Any other resources that have the “booklet seen through the container” situation
    3. Digital resources (both born digital and digitized)

  2. Robert L. Maxwell says:

    Comment on 1.1 Background.

    Paragraph beginning “From the users’ perspective.” Users in a public library searching for a DVD, video, etc., are unlikely to be searching for the exact wording on a particular manifestation (i.e. the wording on a container). They come either (a) wanting to know if the library has a particular movie (e.g. Star wars), or (b) wanting to know what the library has of a particular form/genre (e.g., a French film; an action picture). In neither case is the wording on a particular manifestation relevant to them. It makes more sense to base the title in the description on what will be the same in all the manifestations, that is, the title frames. I admit we don’t always have the ability to play the film. I was just cataloging a Blu-ray today that wouldn’t play in my machine, so I had to go with the container wording. But that was not ideal and there were some aspects of the description I was unsure of because I was unable to look at the title frames.

  3. Robert L. Maxwell says:

    Comment on 1.2, Practical experience.

    The Hansel & Gretel example solidifies my opinion that using the title frames as the first preferred source over other sources such as packaging is the way to go. It’s exactly the same DVD inside the German package as is the DVD in the Italian package. So in some ways transcribing the German title as the title proper (or the Italian) is misleading to the users. If a library happens to have both and a user selects one over the other because he/she thinks one is in German and the other is in Italian, he/she will be disappointed. Other than what’s on the container, neither DVD has either a particularly “Italian” or a “German” character.

    The proposed changes only affect the alternative, but I think there are good reasons why the disc label is preferred currently over the container, so I’m not sure it’s wise to make the container of equal status to the label on the disc. In the Hansel & Gretel example the disc probably would in fact be a better source for the title since the disc is exactly the same in both the German and the Italian packaging.

    • Diane Napert says:

      My comfort level is also, title frames, then disc label, then container
      As an aside, we’ve just set up a video player in our new space as Yale has recently purchased videos

      • Teressa Keenan says:

        I agree that it’s best to stick with title frames as a preferred source. Then use the disc label and then the container in that order.

        While we have also run into the occasional confused patron by the difference between the titles on the container versus that of the title frame, we have solved that fairly simply by adding variant title information to the bibliographic record.

      • Amanda Ros says:

        I agree: title frames, then disc label, then container.

    • Robert Bratton says:

      I was thinking the exact same thing. This example actually proves that preferring the title screens is the best approach. I find it utterly mind boggling that anyone would want a policy to *prefer* taking this kind of information from anywhere other than the screens. I say this as someone who was an AV cataloger many moons ago.

  4. Robert L. Maxwell says:

    1.3. Proposed language

    If the proposal is approved, I recommend much simpler language:


    For a tangible resource, use either

    a) a label with a title that is permanently printed or affixed to the resource
    b) a container issued with the resource.

    The final paragraph “If the resource does not contain a title frame” needs to stay. It’s what you do if you are not applying the alternative.

    I’m fine with adding “a container issued with the resource” to

    I’m also fine with defining (in the glossary) the cover of the booklet as the front of the container.

    • Diane Napert says:

      I like the simpler wording, but would still like a preference for “a” stated
      Of possible interest to the conversation?

      The containers are mentioned in the article (but not in relation to cataloging)

      OK with the glossary addition and will pass along any comments from ARSC members as this mentions CDs

      • Diane Napert says:

        From Jennifer Vaughan ARSC cataloging committee: I’m okay with the change of container definition. I really have no opinion about the revision of the alternative of RDA, since I will continue to use title frames if possible when cataloging moving images. I guess if the resource cannot be played back, it would be good to be able to choose the source with the most information, be it the label or the container

        • Teressa Keenan says:

          I too am fine with adding “a container issued with the resource” to as well as changing the definition in the glossary to include the cover of the booklet as the front of the container.

        • Diane Napert says:

          From Caitlin Hunter, ARSC Cataloging Committee Co-Chair: Re. tangible resources, my thoughts are that using information from the container was already an allowable alternative option. They could make that their own local practice, if it makes things easier for them, but I don’t know that it will make things easier for others. With published audio, there can be differences between the label and container title which will lead people back to the question of “which is preferred?”

    • Kathy Glennan says:

      Change 1.3 – language of final (non-alternative) paragraph

      I recommend the following revision of this paragraph:

      If the resource does not contain a title frame or title screen, apply the following guidelines for tangible or online resources to choose the preferred source of information [new text follows] by applying these instructions, as applicable:
      tangible resources (see
      online resources (see

      • Kathy Glennan says:

        I recommend the following changes to

        * Keep the 1st paragraph & “a)” through “c)” as they currently stand.
        * Move the alternative from to, and reword it. Place this after the a)-c) list
        * retain the existing final paragraph with no additional changes.

        The reworded alternative would read:

        In preference to the title frame or frames, or title screen or screens, use as the preferred source of information
        a) a label with a title that is permanently printed on or affixed to the resource
        b) a container issued with the resource.

        If none of these sources has a title, use as the preferred source of information another source forming part of the resource itself, giving preference to sources in which the information is formally presented.

        • Kathy Glennan says:

          There has to be a better way to convey the necessary changes to the Glossary.

          For example, the LCRI for 6.0B1 said: “For compact discs and cassettes, consider information which can be read through the closed container (including information on the front cover of a booklet inserted in the container) to be on the container.”

          The MLA Best Practices for RDA say: “Note that a booklet cover which is visible through the front of a container forms part of that container.”

          I suggest the following for the RDA Glossary definition of container:

          [existing 1st sentence, followed by this new sentence]. Consider information that is visible through a closed container to be on the container.

          Or alternatively this as the 2nd sentence:
          Consider information which is visible through the housing to be on the container.

        • Amanda Ros says:

          Passing on comments from an A/V cataloging colleague:

          If there is an alternative for, it makes sense to also have it for “tangible other resources.” I agree that it should be an ALTERNATIVE and be worded as such, instead of the way the original proposal presents it. There is an advantage in choosing the container at times for audio recordings, since they may have “other title information” that the discs don’t have. Under AACR2, catalogers often inserted other title information from the container in square brackets, but RDA state that this element must come from “the same source as the title proper,” and thus, appears to discourage this practice. Having an alternative to choose the container as a preferred source would help resolve this situation when the container has more complete information.

          I agree that related textual information for online resources (landing pages, “jumping off’ pages, etc.) should be considered part of the resource. This is what OLAC task force members and I agreed made sense when we wrote best practice recommendations for streaming media. In our examples, we sometimes obtained title information from the websites when information was missing from the title screen.
          I’m not sure if they would be considered “accompanying material” as a commenter suggested they be described as. I wouldn’t necessarily think of them that way, unless they contained supplemental content besides bibliographic details that would add something to the use of the resource, such as text from a script to accompany a streaming movie. Also, the report from the CC:DA Task Force to Investigate Definitions of and Instructions for Accompanying Material concluded that the term “accompanying material” may no longer be appropriate within an RDA context and recommended exploring the use of other terminology such as “secondary” sources, etc. instead.
          It may only confuse matters to call them accompanying material since RDA already mentions “textual content” as “option a)” if information is not available from sources mentioned in previous rules. It seems that a landing web page would count as “textual content” in this option, and be part of the resource, along with “embedded metadata in textual form,” listed as the second source of information under this rule. RDA also lists “textual content” as the first preferred source for online resources that are not textual or moving images. Again, if one were cataloging a streaming audio file, for instance, it would seem logical to transcribe title information from the web page (which again, would count as “textual content” and be considered part of the resource. So “textual content” seems to suffice in describing such web pages without catalogers needing to consider them accompanying material. In addition, it may also be helpful to include a footnote or reference back to the General Guidelines under RDA and when “textual content” is mentioned

    • Robert L. Maxwell says:

      I see that my proposed language doesn’t include the preference of label over the container. How about this instead:


      For a tangible resource, use, in this order of preference, either

      a) a label with a title that is permanently printed or affixed to the resource
      b) a container issued with the resource.

  5. Robert L. Maxwell says:

    3.2. Proposed revision.

    This seems sensible to me. It could be simplified as follows:

    Leave the two paragraphs as is (“When describing …” “When preparing an analytical description …”)

    Add a new paragraph immediately following these two:

    Consider information presented together with or alongside online resources (e.g. landing pages, jump-off pages, metadata pages) to be accompanying material.

    • Teressa Keenan says:

      I agree with the proposed 3.2 revisions.

      I do prefer a simpler text as described by Mr. Maxwell above

    • Kathy Glennan says:

      If this type of change is agreeable, I don’t think this is the right approach. Details regarding describing an online resource should be in &

      • Kathy Glennan says:

        Before we can decide on the ALA response to this proposal, I think we need to determine for #3: what is the resource being described? Is it the e-version, or is it a particular e-version? What if the access to the same scanned copy is available from multiple sources (say Internet Archive & a given library’s own digital repository)? Is this proposal trying to define the resource too broadly, or is this granularity important?

        • kelleym says:

          IMO, the information on the landing page is more like a container than accompanying material. If the German vs. Italian packaging for the same DVD makes a new manifestation (or can make one if you want) then the same should be true for these online pages. Textual content at was meant to refer to this type of web page information and perhaps that should be made more explicit.

          If you catalog the resource with a direct link to the content file then this question is irrelevant, but many resources are embedded in a web page, which you may also want to present to users. In some cases you can’t get at the resource except through the web page. So copies of files (or even the same file, such as a youtube video embedded in multiple different web pages) in the context of different web pages could be considered different manifestations.

          PDF type reproductions could be done as reproductions of the print manifestation contained in the new online resource (PDF + web page), which would connect all the various digital copies. This is less clear with HTML-type reproductions that preserve the text, but not the formatting. These would seem to be reproductions at the expression and not manifestation level (which doesn’t come out with the PN approach)

          You could override this with a PN-type approach, but it seems to me the default should be to describe the resource in its own context. PN is a kluge that we might not need if we had systems that work better.
          I think the reason why the proposal went back to is that the basic question has to be addressed for all the subsections:, .3 and .4 and for them, doesn’t seem to be so much about titles as bracketing. For consistency, perhaps there also needs to be something in the instructions at for online resources lacking a title page saying to look at the landing page.

  6. Diane Napert says: Tangible resources
    This is not an alternative as were the changes to
    My initial reaction is it should be an alternative
    This affects sound recordings so I will ask for opinions from the ARSC cataloging committee
    From the MLA Best Practices: Preferred Source of Information–Other Resources
    This instruction applies to all audio carriers

    • Kathy Glennan says:

      I don’t agree with elevating the container to a) in the list at

      If this change is necessary, and I think it is only if you want to use the container as the preferred source for something other than a comprehensive description, then it should go as the new “c)” — it would be more preferable than accompanying material, but less of a priority than a textual source on the resource itself or an internal source.

      The rationale provided here (the cataloger should be able to decide whether or not to prefer the label to the container) is the same concern raised in RSC/RareWG/6. Why does RDA require a set list of preferred sources? As I noted in my response on the RareWG paper, the bigger question is what other types of resources would benefit from this flexibility — and why didn’t RDA allow for that in the first place?

      • Tina Shrader says:

        I personally like the idea of having RDA specify a preferred source, and then leaving the choice of secondary preferred sources up to catalogers judgement, AS LONG AS there’s a requirement to include information about the source of title (e.g. in a note currently, but possibly as a separate data element in future linked data applications). This preserves consistency where it’s possible, but has enough flexibility to cover a variety of circumstances. And as long as other catalogers can see where the information came from, they know how to deal with it in terms of identifying matching resources for copy cataloging purposes, etc.

        • Teressa Keenan says:

          I also like the idea of specifying a preferred source and beyond that leaving the choice of secondary preferences up to the cataloger. I also agree that this should be accompanied by a note indicating where the data came from. This provides a nice balance between consistency and flexibility.

  7. Tina Shrader says:

    Passing along a couple of comments from my A/V cataloging colleagues at NLM:

    1. “… I would not feel comfortable just using container and label information and not viewing the title screen(s) of the resource. Often time, containers and labels can contain misleading or actually erroneous information. I have also found that the title screens don’t match the container/label titles at all, though admittedly those cases are rare. Also, other languages present on the resource are not always immediately apparent from containers and labels. The presence of these other languages (either on the soundtrack (dubbing) or in subtitling) are important to record from a FRBR standpoint if they represent other expressions of the work.

    Moreover, the document mentions that finding the title screen over advertisement and other extraneous material on the resource is labor-intensive. I haven’t found that to be the case in my experience. It mentions that playback devices are not always available, though I find that statement a bit strange except for the smallest of libraries on the tiniest of budgets.

    So to sum it up, I guess I don’t see the problem in broadening the definition of a moving-image preferred source of information to include containers and labels on an equal par with the title screen, but I personally still prefer the title screen.”

    2. “I’m not an advocate for elevating the role of the container in describing a tangible moving image. … The title screen and the opening credits usually offer the most accurate information about a film. These credits cannot be manipulated by a third party. Sometimes the container information is typed in by a contractor to the library. Mostly, the container information is reliable but very brief. I don’t believe “the container normally gives more information” from page 3 of the proposal is true. I rarely see a source of responsibility on a container. I also don’t buy into the statement that “users are not familiar with the wording of the title screens” (page 1).”

  8. Amanda Ros says:

    As a cataloger who has spent more time with non-print tangible materials than with print materials, I agree with the consensus. Title frames are preferred and disc label and container are equal as alternative preferred sources of information.

  9. Amanda Ros says:

    Passing along comments from one of my A/V cataloging colleagues:

    1.2 and 1.3
    I also feel that title screen should still be given preference over disc label or container as the preferred source for moving images for all the reasons people have stated. Contrary to what the proposal says, I don’t think that doing so would violate RDA’s principle of representation (RDA, which says that one should prefer the title most frequently found in the resources embodying the work in its original language. The title screen is where you are most likely to get that information and as others have stated, it has the most accurate information.

    But I recognize the problem where some libraries don’t have equipment or sufficient time or administrative support to always view videos for cataloging. I also understand the point the writers of the proposal make for foreign films where the container may have the translated title in one’s own native language that users would look under, and that they may not be familiar with the original title of a foreign film that appears on the title frames. Personally, I am not convinced that the latter is a reason to prefer the container over the title frames, since one can always transcribe original title and include a parallel title for the translated title from any source, including the container. But apparently the authors of the proposal don’t agree that parallel titles are an adequate solution and think it may still confuse patrons. I can only assume that this is based on their own actual experiences with patron behavior in Germany.

    If they want an alternative to treat the container as a preferred source in addition to the disc label, I don’t have an objection to it. Most of the time, the title on the container will match what is on the title frames so it will often not be an issue. When they don’t match and someone has “cataloged from the container,” I have not had a problem detecting whether the record matches the item in hand as long as there is a note in the record stating that the title is from the container and it matches what I have. As there is a rule in RDA ( requiring such a note¬ when title is taken from a source other than the title frames for moving images, this appears to be covered. If this proposal is approved, it would really help to add an instruction near this alternative reiterating this and referring readers back to this rule the way the “Early printed resources” proposal does in its proposed wording.

    In the meantime, the writers of the proposal suggest that the same manifestation may be packaged with different containers for different countries. I’ve not experienced this situation, but if a DVD is packaged in another language for a different country, and a cataloger in Spain decided to base description on the container, it would probably not be a problem, since we wouldn’t be using their records anyway (according to the “parallel record” policy in OCLC). I imagine a German library would also prefer records created in their own native language than one created in Spain where the non-transcribed elements of the record are in Spanish. Thus, Germans can continue to create parallel record based on the German language container while Spanish-speaking countries can base theirs on the Spanish language container for the same manifestation and they won’t have to re-use each other’s records.
    The wording that Kathy Glennan proposed for looks good to me. I also agree with her wording for “container” in the glossary, but feel that a few selective examples can be added to illustrate the definition. The particular example in the original proposal (e.g., for a jewel CD case…) can be retained, since it is a useful illustration in my opinion.

  10. Mary Anne Dyer says:

    Passing along comments from a colleague who catalogs videos and films:

    I understand that we do need to consider the title frame as the preferred source of information. I do support however the proposal to use cataloger’s judgment to use either the label on the tangible resource or container. I also support the idea of redefining what a container is for a tangible resource like a DVD.

  11. kelleym says:

    Here are my thoughts on the first few practical questions. I’ll post my more general comments separately.

    I prefer to retain the preference for an integral label over the container. There is a lot to go through in all these comments and I’m not quite sure: Are we trying to rewrite the alternative so that the label and container are equal while keeping the list at the same? Although, I do prefer the current hierarchy, given that it is not widely enforced and how commonly the alternative seems to be used, I am not sure how much practical impact prescribing a priority list of preferred sources for moving images actually has. The cynical part of me wonders if any of this actually matters.

    I do see why they want to move the alternative since the final sentence for when a resource doesn’t have a title frame only applies when you are actually try to look at the title frame. However, it also makes sense in its current location as an explicit alternative to preferring the title frame. I am not sure that it makes sense to move the alternative to since it really is an alternative to what is stated in It’s true that the alternative only applies to tangible resources, but if that has to be made explicit, maybe it would be better to begin the alternative with “for tangible resources, use a label…” What is clearly missing from the alternative is an explicit instruction about what to do when there is no label.

    I agree that information seen through jewel cases, etc. should be considered part of the container. The new definition just makes explicit what people have been doing anyway. Either of Kathy’s proposed wordings is fine with me and better than what is in the proposal.

  12. kelleym says:

    It seems like they are trying to finagle the rules so that they can always make the preferred title of the manifestation be the title in the language of the catalog. We don’t do this for other materials, including music, which is similar to moving images in that the language that the manifestation uses to describe itself is disconnected from the language abilities that the user needs, However, I am sympathetic to the aim. Most North American users are not helped by a title in Romanized Chinese. And moving images do differ from books in that they are often accessible to users in languages other than the one that the preferred source uses to describe itself. On the other hand, most multilingual DVDs do not come with alternate containers as described in the proposal. Many DVDs provide dubbed or subtitle tracks in multiple languages without varying the container so it is unclear to me that it is realistic to always provide a manifestation title in the language that the user intends to use the DVD in. This also wouldn’t be very helpful for international record exchange. As someone pointed out, right now we have parallel records in WorldCat, but there may be better ways to integrate multilingual records in the future.

    I do think it is important to continue to prioritize the title frames. The title frame is the title most integral to the manifestation. It is also the title most likely to stay the same in all manifestations. IMO, the most important reason to record the title frame title is that it is useful, sometimes even essential, for identifying the work. For that reason, I think it is important to record the title on the title frames even if it is not used as the preferred title of the manifestation. This is not so much a problem with widely-available major motion pictures where information about title variations is likely available in external reference sources. However, there are situations in which it is vital to know the title on the title frames. Sometimes moving image works sold to the educational market are distributed by more than one distributor (say FFH and Insight Media) and each distributor has used a different container/label title. Without the title frames title to bring them together, there’s no way to know that they are duplicates. I have also recently had a need to track down info about some Chinese-language films where the manifestation has had an English language parallel title, but I am unable to find any info in English-language online sources (using either the English or Romanized title). I have had to look up the original title in Chinese-language sources and use Google translate. Without the original title, I would have been stuck.

    Presumably the reason for preferring the label on the resource itself over the container is that it is more integral to the resource itself. Does a different container necessarily make a new manifestation? If a book comes in a container or case and the publisher switches it out for a new design, would you consider it a new manifestation? Would most bibliographic utilities want you to? OCLC recommends ignoring packaging dates on video containers for determining whether something is a match so they are clearly encouraging catalogers to put videos with different containers on the same record (I am not sure this is always a good idea as the packaging date often does best reflect the publishing/release date, but, on the other hand, there are publishers like FFH who just put whatever the current year is on their container without making any other changes). RDA doesn’t define the boundaries between manifestations this specifically and maybe some groups would consider each container (and book cover) to be a separate manifestation. It does seem like you would want to represent the information from the German and Italian containers in their example in some way. Then again, if it’s the same disc, for most purposes of selecting a manifestation, what container it came in is irrelevant. Where I work, for better or worse, they remove DVDs from their containers for circulation. Most patrons never see the containers and this does not seem to have generated widespread complaints. Another reason to play the resource is that containers sometimes lie.

    Perhaps the basic question is, what is the purpose of having a preferred title for a manifestation?

    Is it that different catalogers should consistently arrive at the same title to make it easier to identify when something is the same manifestation? This makes it easier for a cataloger to select the correct manifestation from a list and consistency in title identification would seem to be important for de-duping records in bibliographic utilities.

    Or is it to choose the best title to display to the user? There probably are cases where the spine title of a book would be more useful than a title page title to users and the spine title is what users will see when scanning the shelf.

    Often these two goals don’t conflict, but if we really want to do both things, perhaps it would be better to have two elements: (1) the title identified using a standardized prioritization of sources for identification and comparison; (2) the title deemed most useful to display to users in a given catalog.

    In fact, rather than basing the title to display to users on the preferred source, it should be possible to mark up each title with information that would enable an algorithm to pick the title to display to the user. If you could give the source (or sources if the same title appears on multiple sources) and the language of the title, it would be possible to say this is the German title and it appears on the container and the label while this is the English title and it appears on the container and the label and the disc menu and the title frames. The algorithm could then choose to display the title in the language of the catalog, preferring certain sources if there is more than one title in the language of the catalog. You could have some way to mark a title so that the algorithm is overridden and a manually-selected title is chosen for display.

    I actually think the rule where you’re not required to say that you got the preferred title from the title frames is a mistake and that catalogers should be required to always give a source of title. Implied data is a bugaboo of mine. You never know if they really took the title from the title frames or just forgot (or chose not to bother) to say where they got the title from.

    If you have a cluster of titles that are all marked with where they came from in a machine-actionable way, it seems less important to me to have to pick a single preferred one. This is true for other data where the source could be made explicit, as well.

  13. kelleym says:

    Comment from an OLAC CAPC member: I feel strongly that the title frame should stay as the preferred source. Next in terms of preference should be the disc label, and then finally the container (which would include the accompanying sheet with cover art and information about the film. I also feel it’s important to add note to the bibliographic record regarding whether the title is coming from the disc label or the container. I read through the many comments posted regarding the proposal and I like the simpler wording proposed by some. There were several of these and I’m not partial to any particular one.

  14. kelleym says:

    From an OLAC CAPC member:

    For the source of information proposal, Tina Shrader expressed my thought exactly:

    “I personally like the idea of having RDA specify a preferred source, and then leaving the choice of secondary preferred sources up to catalogers judgement, AS LONG AS there’s a requirement to include information about the source of title (e.g. in a note currently, but possibly as a separate data element in future linked data applications). This preserves consistency where it’s possible, but has enough flexibility to cover a variety of circumstances. And as long as other catalogers can see where the information came from, they know how to deal with it in terms of identifying matching resources for copy cataloging purposes, etc.”

Leave a Reply