HomeGroupsTalkZeitgeist
  • LibraryThing
  • Book discussions
  • Your LibraryThing
  • Join to start using.

cataloging periodicals

Librarything Series

Join LibraryThing to post.

1rsterling
Edited: Oct 14, 2010, 12:19am Top

I wanted to open up a discussion about how periodicals are cataloged and then listed in CK series.

From what I can tell, there are at least 2 different approaches that LT members use in cataloging periodicals.

1) There are those who catalog each issue as a separate work.

2) There are those who catalog the periodical in general as a work, and then list which issues they hold in the comments - which could include many issues.

I'm in the 2nd camp. In this, I follow the example of actual library catalogs, which typically have a single catalog entry for the periodical, which lists library holdings, rather than separate catalog entries for each issue.* I'm also not the only member who does this.

The series field seems to be used a lot to link cases of approach 1 together, but when that happens, there often seems to be some misunderstanding of cases from approach 2. In some instances, I've noticed they're included in the series, but given a number and sometimes a canonical name of "unknown." That in my view isn't correct, since these are entries for the generic periodical. It seems that either they shouldn't be included in the series at all, or they should be left as generic (unnumbered) entries. I also don't understand why the canonical name would be changed.

* Here's an example of how it's done in one RL library, which gives an idea of the model I and others are using: the catalog entry for the periodical The Economist, in the Los Angeles Public Library:
Title(s) The economist.
Branch Holdings Central Library holdings shown. To check branch holdings go to http://www.lapl.org/catalog/branch_periodicals.html
Uniform Title Economist (London, England : 1843).
Publisher London : Economist Newspaper Ltd., 1843-
{{etc. etc. Some fields omitted from example}}
LIBRARY HAS: 106- (1928-) Microfilm=106-297,338- 1928-1985 1996- Hardcopy=298-337 1986-1995. Current issues held at Reference Desk. Retained in hardcopy until microfilm received.

2keristars
Oct 14, 2010, 1:19am Top

As a stopgap measure for one of the problems, why not use a canonical title of "Periodical title General/Unknown"?

I'm not sure how to resolve the problem of periodicals being entered in different ways - I think that would require changing the way add books works.

3kathrynnd
Edited: Oct 14, 2010, 2:12am Top

I deleted the canonical title for National Geographic which led to one big plus, as the title now comes near the top when using general search.

I kept the magazine in the series as I guess it doesn't do any harm, but I did change the title label so it no longer is listed as 'unknown'.

Edited to say, someone has put it all back again.

I strongly feel that work titles should not be changed for frivolous reasons such as series listings. If the title is good enough for the Library of Congress and other entry sources, (I used ILCSO (Illinois Libraries)) it should be good enough for LT. If no one disagrees strongly I will delete the canonical title once again.

4rsterling
Edited: Oct 14, 2010, 2:13am Top

2 - Just to be clear: I do NOT think that the fact that periodicals are entered in different ways is a problem. Both are valid ways of cataloging. The difficulty is, I think, that people who use one method might assume that everyone does, and make changes to site-wide data (combinations, CK) that reflect their own assumptions about how periodicals are or should be cataloged, but not the actual practice of the people who cataloged the works.

Also, I don't see why using a canonical title at all is helpful in this situation. Why not leave the title as it is, which is, presumably, just the periodical title?

5keristars
Oct 14, 2010, 2:19am Top

4> Right, I was thinking "problem" as in "something that isn't easy to deal with when it gets attention". If you remove the canonical title, someone is going to add it back in. Kathryn is already seeing that. So, instead, add a descriptive one that makes it clear that this particular work is both for a general entry and for unspecific individual editions.

That could (should?) be added to the disambiguation notice, too, but if the CT is going to keep getting added if it's removed, put one in that works for both.

6rsterling
Oct 14, 2010, 2:27am Top

I actually think the case Kathryn mentions IS a generic entry for the periodical as a whole; I looked at a few of the members' catalogs, and that seems to be the case. Several list specific holdings in the comments field, consistent with the approach 2 I describe above. I would suggest adding a disambiguation note to that effect, but that could just get removed. I don't want to get into an edit war. The other option would be to separate out editions and generate a new work page for generic editions, but that seems like unnecessary extra work.

7keristars
Oct 14, 2010, 2:29am Top

Ah...I think I might be confused about what is actually going on with that work, then, and what you're wanting it to do.

8aulsmith
Oct 14, 2010, 8:05am Top

I've been using approach 1, since I eventually want to add the stories and poems inside the issues. I started to add mine and other people's periodicals done in approach 1 to the series entries, since the approach 1 people differ in whether to use the editor or nothing as the author and it's very hard to find which issues have been cataloged. I am not adding approach 2 periodicals to the series entry nor do I see how canonical title helps any of this.

(BTW, in cases where libraries want to keep track of the individual articles in a periodical issue or when individual issues have individual titles, approach 1 would be what the cataloger would do.)

9jjwilson61
Oct 14, 2010, 10:00am Top

Without some other means of dispute resolution, not wanting to get into an edit war just means that the other person wins. Perhaps an edit war is just what is needed to get this other person to come here to discuss the issue instead of unilaterally imposing their point of view.

10KingRat
Edited: Oct 14, 2010, 11:19am Top

The canonical title on that National Geographic entry is "National Geographic". Could someone explain why that's bad?

12rsterling
Edited: Oct 14, 2010, 11:22am Top

I'd rather not make this thread's discussion about any specific case, but rather about how we deal with periodicals in general - for series, canonical title, and combining purposes - given that people have different cataloging approaches.
(edited for typo)

13jjwilson61
Oct 14, 2010, 11:39am Top

10> Someone had set it to National Geographic Unknown on the assumption that since no particular issue was selected then the issue the record is for is unknown. This supposedly made it easier to get all the magazines into their proper spots in the series.

As to whether National Geographic is a good Canonical Name, that depends on what would it be if the Canonical Name were left blank.

14rsterling
Oct 14, 2010, 11:54am Top

10, 13: It is already blank now.

ah - wait, some separating going on... not sure where it went.

15keristars
Oct 14, 2010, 12:33pm Top

13> The CT should be blank now. I deleted it last night to see what it turned into, and it looks like no one has reset it yet. (Unless you're talking about the Author CN, then who knows...)

16alaudacorax
Edited: Oct 14, 2010, 12:57pm Top

#1 - I've never previously catalogued periodicals but for some time I've been contemplating cataloguing my collection of issues of Waterlog, so I've been looking into this.

My reason for cataloguing would be to better keep track of their locations and some of their content (mainly the latter). I can't see any efficient (that is, easy and quick to use) way of doing this other than by cataloguing each issue as a separate work and applying a selection of tags.

Your 'second camp' seems to me to be tailored to the needs of public libraries and to be of doubtful value to the personal libraries of which LT, I imagine, is mainly composed; but, as always, I'm open to the possibility that I've overlooked some efficient ways of doing things.

17brightcopy
Oct 14, 2010, 12:57pm Top

I think the "second camp" takes into account the labor shortage when it comes to trying to catalog every issue of a periodical. It's just not worth it to pay someone to do that. We don't have so much of a labor shortage here.

18lorax
Oct 14, 2010, 1:05pm Top

16>

Your 'second camp' seems to me to be tailored to the needs of public libraries and to be of doubtful value to the personal libraries of which LT, I imagine, is mainly composed

If I cataloged periodicals, I'd fall into the second camp. Here's why.

I wouldn't be interested in tracking individual articles on LT; it's not the right solution for that. Let's say I had a complete run of National Geographic from 1985 on, except that I lost May 1994 somewhere along the line. I could either add a couple hundred items, one per issue, or I could add "National Geographic Magazine", and note in the comments that I have issues xxx through zzz except yyy. Far more efficient, and more useful for the "which issues do I have" question. If you aren't interested in finer granularity than the issue, and have a large number of issues, this approach makes more sense. (On the other hand, if I only have a few issues, then it makes more sense to catalog them individually. Both have their merits.)

19brightcopy
Oct 14, 2010, 1:21pm Top

18> I think the real answer is that they'll be plenty of instances of people wanting to do it BOTH ways, and that for any camp it will make more sense to them. LT just doesn't handle this well at the moment, and it shows. Perhaps the contained in/by functionality will eventually come to the rescue. People could catalog the general "omnibus" work that is ALL National Geographic (noting in some other field which issues they actually have), or they could catalog the individual issues.

20alaudacorax
Oct 14, 2010, 1:57pm Top

#18 - I wouldn't be interested in tracking individual articles on LT; it's not the right solution for that.

Why not, and what would you suggest as a better solution?

21lorax
Oct 14, 2010, 2:17pm Top

20>

I'm not going to try to talk you out of doing it your way; I'm explaining why an individual person might want to do it the other way. If cataloging individual articles here works well for you, great. I find it adds too much clutter for me. I don't think I'd bother to catalog individual articles for leisure reading like Nat Geo; for scholarly articles, I'd use something like CiteULike. (I've not heard of "Waterlog" and don't know where it falls on this spectrum.)

22brightcopy
Edited: Oct 14, 2010, 2:22pm Top

21> I'm curious, as you've done this a few times. Why do you keep saying "individual articles" rather than "individual issues"? Is it a case of library jargon versus non-library definition? Each time you say it, I get the image of people cataloging each article that's in an issue separately, much like people individually catalog short stories that make up an anthology.

23jjwilson61
Oct 14, 2010, 2:33pm Top

I imagine because the reason that some people have given for listing individual issues is for the purpose of tracking articles. And if you're adding each issue its only one more step to add each article.

24rsterling
Oct 14, 2010, 3:01pm Top

I actually think the LT system handles it just fine whichever approach you want to use. The problems arise when one camp assumes there's only a single way to do it, and then makes changes that affect site-wide data.

It would be problematic if people with approach 2 decided to combine together all individually-cataloged issues into one work for the whole journal, just as it would be problematic if people from approach 1 decided to treat instances of approach 2 (generic work for the journal) as if they were incorrect, unclear or deficient instances of approach 1.

25brightcopy
Oct 14, 2010, 3:20pm Top

23> Yes, but it's a completely different topic than the general topic of this thread. So I'm trying to figure out if when lorax says articles she means what most of us would think of as entire issues.

24> Except, people with approach 2 couldn't really legitimately do that, could they? When you combine together different editions into a work, those different editions are supposed to each have the same content as one another. This would be a pretty clear cut instance of LT working as it is designed. Individual issues of periodicals ARE individual works. The place where it gets fuzzy is when you basically make custom boxed sets. That's where you start getting into fuzzy undefined area in terms of how the LT framework applies.

26lorax
Oct 14, 2010, 3:31pm Top

22>

Rankamateur mentioned wanting to track content of issues -- not just which issues he has, but what's inside the issues. I've been referring to the finer-granularity-than-issues content as "articles". (I guess it could be stories or poetry, instead, depending on the nature of the periodicals.)

27alaudacorax
Edited: Oct 14, 2010, 6:10pm Top

#21, #26 - Yes, I wanted to track individual articles, but not to catalogue them individually, thus I was thinking of using tags to do so.

#24 - This worry is the reason I haven't entered my periodicals yet.

(#21 - It's simply a fishing magazine - I don't know why I used the title rather than generalised - there's no significance to it.)

28brightcopy
Oct 14, 2010, 3:56pm Top

26> Ah, okay. Just seems like it's yet another topic for suggestions.

Collection->Issue->Article

More and more, it seems like the contained in/by ideas would work so much better for periodicals than Series do.

29rsterling
Oct 14, 2010, 4:24pm Top

24. It would be illegitimate in both cases. That's the point I'm trying to make.

Individual issues of periodicals ARE individual works.

In your view. Not in mine. I consider the periodical in general a work. I'm ok with leaving the issues, which others consider and have cataloged as individual works, as works. I would just like others to leave my work - the generic periodical - as a work, and not second guess it and assume that I'm trying to catalog an individual issue as a work and not succeeding.

What you are defining as "custom box sets" may well be something I consider to be a single work - the periodical in general - of which I am specifying my holdings. I don't want my entry to be considered as a custom box set and tried to fit into someone else's schema of what counts as a work, and manipulated just to fit a series listing, or grouped together with various unclear or unknown copies and presumed to be some kind of misfit.

30brightcopy
Oct 14, 2010, 4:40pm Top

29> I don't mean in my view or your view, I mean by the LT definition of a work.

What are works?

The purpose of works is social. Books that a library catalog considers distinct can nevertheless be a single LibraryThing "work." A work brings together all different copies of a book, regardless of edition, title variation, or language. This works system will provide improved shared cataloging, recommendations and more. For example, if you wanted to discuss M. I. Findley's The Ancient Economy, you wouldn't really care whether someone else had the US or the British edition, the first edition or the second.

What is a book then?

The "This Book" data on the book information page is particular to YOUR copy—the distinct edition that's sitting on your bookshelf. None of the combining and separating of works will change anyone's personal book data.


http://www.librarything.com/concepts

I'm not sure how you go from this to "I put three different copies of National Geographic magazine together on my shelf, and now that is a 'work.'" If you go that route, there really IS no such objective concept as a work any longer, as it can be whatever you want it to be.

But I think there's a difference here between what really is a work and what is permitted as a work. Tim permits movies, cds, bobcats and perfumes to be works. He doesn't embrace it, though, and it's one of these "off-label" uses. I'm not taking a stand against custom boxed sets such as yours, because LT simply doesn't have a good way to handle them and they are legitimate book-ish things (far more than movies, cds, bobcats and perfumes).

So don't take me as calling you a misfit. Far from it. You're trying to solve a very real shortcoming of LT. That doesn't change the fact that LT has certain concepts that are defined in such a way that they don't fit what you're doing.

31rsterling
Edited: Oct 14, 2010, 4:51pm Top

Sorry, but we're going to have to agree to disagree here. I am following the example of numerous actually existing libraries in their cataloging practice. They use one entry - i.e. one work - for the periodical, and list their holdings. LT wasn't designed to list periodicals, sure, but I am using the ISSN as the distinctive mark of a "work" in the periodical world. To repeat: I am not the only one who catalogs periodicals in general as works on LT. To me this is much more intuitive and natural than listing each individual issue as a work, for which I've never seen precedent in any library catalog -- and library cataloging is my model for what we're doing here and how to list my own books. All I'm asking is that people not enforce their own assumptions about what counts as a work in the cases of periodicals on my cataloging.

PS - I meant the work was being considered as a misfit (and thus various cludgy solutions are being devised to make it "fit" a model it wasn't meant to fit), not that members are being considered misfits.

32brightcopy
Oct 14, 2010, 4:50pm Top

31> I'm not really sure what following the examples of libraries has to do with it. Do libraries have a concept of work vs book? Not to my knowledge, but I'm no librarian. If they have no concept of work vs book, how do you think their cataloging is really applicable to LT?

33rsterling
Edited: Oct 14, 2010, 5:00pm Top

This is exactly why I started this thread. I would like people to be aware that there is more than one approach actually being used on LT to catalog periodicals. In my view, both approaches are legitimate. Some people, however, seem to assume that that theirs is the (only) "correct" one, and thus try to fit other people's data into that model. I'm happy to let those who do approach 1 continue to do so. Just leave my data alone.

ETA - sorry if I sound/am grumpy. Need more sleep. But there is no LT "rule" about how periodicals are supposed to be cataloged. LT wasn't designed for them. A periodical issue is not a book. Therefore the application of the work/book distinction does not work neatly here anyway. I am just trying to make people aware that the way they may assume things are done or supposed to be done is not the only way they could be - or are being - done, nor the only legitimate way.

34brightcopy
Oct 14, 2010, 5:02pm Top

33> But that's the thing, you ARE using the LT model. You can't use LT and not use the model. By "the LT model", I mean a model of works and books, like the above. And as it says above, works are shared social concept and books are your own data that is sacrosanct. So really, LT is built around people not leaving works alone.

It's like if I said "Hey, I'm going to be cataloging all my books as a single work per author per series. So for example for an Fleming I'm going to make a 'James Bond' work. And I want everyone else to leave it alone and not combine it into any of the other James Bond books and not touch it at all." If you decide LT works don't have to simply be abstract forms of a single published work, then this is as legit as anything else.

35rsterling
Edited: Oct 14, 2010, 5:13pm Top

I think you misunderstand what I said; I didn't use the words LT model.

Anyway, someone who uses approach 2 could take the social "work" concept and say that what's interesting is whether or not anyone else shares a subscription to The Economist, not whether they share a particular issue. If e.e. cummings subscribed to the same journals as me, that's a point of social interest, despite the fact that our lifetimes, and thus our holdings, didn't overlap. Personally, I'm not going to make that argument or act on it: I'm not going combine all instances of The Economist - all generic entries and all issues - together into one big The Economist work. That's because I respect the fact that others want to treat their individual periodical issues as books (and moreover, as separate books, i.e. as works).

Let's leave aside the work/book distinction for a minute, since that brings in another set of issues, and just look at it at the "book" level. Approach 1 people are cataloging individual issues of periodicals as if they were "books." They're not, but fine. Approach 2 people are cataloging the ISSN-identified periodical as if it were a "book." These 2 "books" are not the same and should not be combined. Nor should my instance of the "book" (the generic journal) be renamed because it doesn't fit someone else's idea of what belongs.

37brightcopy
Edited: Oct 14, 2010, 6:11pm Top

36> But if you did combine two different issues of the Economist, you'd be very clearly violating the LT rules. You'd be combining a real, physical book with another real, physical book that doesn't have the same content. You'd be combining "Lord of the Rings" with "The Hobbit." No one would even have to question if you'd be wrong.

On the other hand, the whole existence of this thread shows that the other case (combining random bundles of issues) is a much more debatable one.

And as far as the social aspect goes, this is exactly why I say the contained in/by seems a better approach. It could still let these social aspects interact in a useful way.

Approach 1 people are cataloging individual issues of periodicals as if they were "books." They're not, but fine.

How in the word are they not? Just because they're part of a greater whole? How is this different than the serialized form of novels, which ARE considered books in their own right? Or a trade paperback of a graphic novel, which IS considered a book, even though it may only be a few issues from an entire run?

I think we have this disagreement on a fundamental level. If we can't even agree that a thing that's part of a whole can be a "book", I'm not sure what we can agree on.

38rsterling
Edited: Oct 14, 2010, 6:26pm Top

A book and a magazine are different things. I'm fine with you cataloging individual issues of a magazine as if they were books. I'm using books here in the everyday terminology, not the specialized LT concept. LT was designed for cataloging books, in the everyday sense of the word. Both ways of cataloging periodicals - approach 1 and approach 2 - were not really what LT was designed to do, because it was designed for books. But be that as it may, I have no problem with people cataloging periodicals.

They're not books in the everyday sense of the word book. I'm ok with them being cataloged as if they were books and as "books" in the LT-specific sense/concept of book.

If we can't even agree that a thing that's part of a whole can be a "book", I'm not sure what we can agree on.

This is not what I'm saying. This is going off into a different point.

Please look at the examples of the ways that other people have cataloged periodicals in general as a work. Clearly you disagree that this is a legitimate way of cataloging. Others think that it is. I understand your point of view. I don't want to catalog the way you prefer for periodicals because - as lorax explains very well above - I'm not interested in cataloging periodicals that way, or distinguishing that level of detail. What I really disagree with is any dogmatic view that there is one single correct way to catalog periodicals on LT - which as you rightly point out was not designed to catalog periodicals in the first place.

39brightcopy
Edited: Oct 14, 2010, 6:32pm Top

38> I don't know how I can say this again clearly enough, as you keep thinking I'm saying the opposite:

I am not against people cataloging periodicals however they want on LT.

I am not saying there is one single correct way to catalog periodicals on LT.


What I keep trying to say is that you're swimming upstream. It's why we wind up having a long thread like this. Things won't work as well as they will with boooks. People will combine in ways you don't like, etc. I find that approach 1 is less swimming upstream than approach 2 because it more closely fits the LT work/book model. That's the simple point I've been talking about, not about some dogmatic approach that there's a "correct" way.

40rsterling
Oct 14, 2010, 6:37pm Top

Thanks for the clarification. I think we've been talking past each other. It may be that what I'm trying to explain is a hard point to get across clearly or to understand, and thus people are liable to misunderstand what those of us who catalog generic periodicals are doing. That's why I started the thread: to open a discussion; to make people aware that there's more than one way that this is done.

In saying so often that the approach 2 goes against the work system, it sounded to me like you were saying it is not a legitimate way of doing things. Hence my confusion/misunderstanding. I see now that that's not what you were saying.

41brightcopy
Oct 14, 2010, 6:44pm Top

41> No, the only thing I ever said was illegitimate was your example of combining together different works that had completely different content.

I totally understand what you're trying to say, I think we just have some fundamental disagreements on what things should be considered LT works. Perhaps it's that sometimes you interpreted my use of the term "work" to not be talking about LT works. But when you say things like "individual issues of periodicals are not individual works", I just can't agree with anything you derive from such a position. Maybe it's because you didn't mean LT works. But LT works are an extremely fine-grained thing. If you're claiming that a magazine that I can go up to a newstand and buy and put on my bookshelf doesn't count as a work, I am going to be a bit dogmatic on that one.

If that's what you think, I really do think you are wrong in an objective way, not just a subjective one. A short story can be considered an LT work, for crying out loud. A poem can be considered an LT work. I just do not agree with the logic that somehow an issue of a periodical (apparently, not a periodical that is a comic book, as you didn't disagree with that) is not an LT work. That's not the same as not understanding the logic.

42rsterling
Oct 14, 2010, 7:00pm Top

For *my* catalog, and my use, I do not consider individual issues of periodicals to be works. There's a distinction here between "I consider" and "can be considered" or "should be considered." I am fine with the fact that other people do consider them works, in their own catalogs. I am not making any claims about what can be considered a work on LT. I'm talking about what I consider to be a work, for my purposes, as someone who uses the second approach. Similarly, I do not consider a poem to be a work for LT cataloging purposes unless it was separately published in book form; I would not include that as a work in my own catalog. However, I see that others do, and I would not keep them from doing so. (It would be nice, of course, if LT had a better way of dealing with differing formats; I would prefer, for instance, if individual scholarly articles or short stories didn't show up as equivalent to books (in everyday use) on author pages, or if there were some good way of distinguishing these. But that's another issue.)

The problem I'm trying to highlight is that sometimes people see works of type 2 and assume that they are works of type 1 with some data missing, and combine them, rename them, do other things with them, in order to fit them into some idea of what they think the data should look like. This doesn't affect anyone's core catalog data, sure, but it does affect how my entry shows up in various places, for me: it can mean that my entry gets wrongly combined with others with which it does not belong, or that my entry starts displaying the wrong title in the "work title" column, because someone else decided the canonical name should reflect some status that doesn't apply to my data.

43brightcopy
Oct 14, 2010, 7:08pm Top

42> But can you see my point about how even though you can see that individual issues of a periodical DO fit the concept of an LT work (even if that's not the way you want to organize yours), that a random collection of issues does not naturally fit the concept of an LT work? Do you see my analogy with deciding to catalog all my James Bond books as one single "James Bond" "work"? Do you see how that's fighting against the way LT works fit, as part of a community with people editing the data all over the place? Can you see how since every person's "National Geographic" work will contain different content depending on which issues they mean it to include, how that makes it not fit into the system in a problem free way as approach 1, regardless of your own personal preference of approaches?

These are simply the points I'm trying to make, which isn't about trying to force you to do things one way or the other.

44lorax
Oct 14, 2010, 7:14pm Top

34>

It's like if I said "Hey, I'm going to be cataloging all my books as a single work per author per series. So for example for an Fleming I'm going to make a 'James Bond' work. And I want everyone else to leave it alone and not combine it into any of the other James Bond books and not touch it at all."

People in fact have been doing this for years, either to duck the 200-book limit, out of laziness, or both. Leaving them alone and uncombined is all we can do -- and, unlike rsterling's case, there's no library precedent for this practice.

45rsterling
Edited: Oct 14, 2010, 7:15pm Top

43 - I see what you're saying, but I disagree. You are saying that what I consider to be a work, and catalog as such, isn't really a work. I understand that you don't want to force others to change the way they use their catalog. But you can see why I got confused and thought you were saying it wasn't legitimate to catalog periodicals that way, because they aren't really works. With all this discussion, it's still hard not to read what you're saying as "you're doing it the wrong way." There is no right way. There are no rules on how to handle periodicals. There are different practices. Mine has precedent, and logic to it. Yours does as well. Fine. Let's let both live. That is all I'm saying.

I would like to get some other people to weigh in on this, because I don't think I'm doing a very good job explaining the logic of cataloging generic journals - using the ISSN - as single works: why either library catalogs or individual personal catalogs might want to do so. I don't think the analogy to RL books applies very well here because a periodical is a different beast.

46lorax
Oct 14, 2010, 7:21pm Top

45>

For what it's worth, your approach, and your explanation, make perfect sense to me. I think I'm thinking of them as "all issues of this magazine within a timeframe", and brightcopy seems to be thinking of them as "an arbitrary random set of issues", which may be part of the disconnect. The unified whole is also how I think of my own periodicals, which I don't catalog here; I'd say that I have say, Birding magazine from 2008 onwards, not "I have January 2008, and February 2008, and March 2008..." The latter approach seems like listing individual chapters in a book to me.

47brightcopy
Oct 14, 2010, 7:24pm Top

45> With all this discussion, it's still hard not to read what you're saying as "you're doing it the wrong way."

I'm sorry it's difficult, but that's not what I'm saying. What I'm saying is, "You're doing it the way that will most likely cause you difficulties because it maps to LT structures poorly."

Approach #1 isn't the "right" way to do it, either. It's the way that causes less grief to the person using it. I know, I know, to you that's just not acceptable because the grief of having individual issues in your catalog outweighs the grief of having other people edit "your" data.

What I think is the RIGHT way to do it is only with an expansion of LT. I think that's one thing maybe we can agree on.

As a semi-related topic, there's a work that came up recently in discussion in Bug Collectors. You posted on the thread, so I'm sure you remember it. But I'll recap for the record. Basically, two (or possibly three) different books named "Italy" were written by George Kish. They didn't have ISBNs. They got combined together (either automatically by the system or manually by a user) and there's no way to separate them.

This is the kind of thing I'm talking about when I say you're swimming upstream. It's not about right or wrong, but about how much LT is going to fight you on the different approaches.

48brightcopy
Edited: Oct 14, 2010, 7:27pm Top

46> No, I believe you'll find the "entire run" (for a year, etc.) is only a subset of what rsterling is talking about. Look at the examples he posted. They're almost all scattered collections of issues.

49rsterling
Oct 14, 2010, 7:29pm Top

Ok, so I'm swimming upstream. Obviously I'm not the only one. I'm posting this so that others can perhaps understand that there's a method of cataloging periodicals that hadn't occurred to them, and won't fight the tendency quite so much.

My way doesn't cause me any grief, unless LT members - not LT as a data structure - start doing things with my entry that aren't appropriate for my entry.

50brightcopy
Oct 14, 2010, 7:35pm Top

49> Considering your library is private, wouldn't it be pretty difficult for other LT members to tell what you are cataloging by looking in your comments?

51rsterling
Edited: Oct 14, 2010, 7:48pm Top

I'm talking about the general issue. I'm talking about the existence of 2 different approaches. I'm asking people who use one approach to be aware of and respectful of the other approach.

I'm definitely swimming upstream in this thread.

ETA in my own case, I've had miscombinations happen when I was public, and when I had specifications like those in the examples in my comments field. I separated them out and put a disambiguation notice on. But I'm not talking about my own library per se or privacy status - which comes with its own, separate set of issues - but rather the general issue, which concerns not only me but other people who adopt approach 2.

52brightcopy
Oct 14, 2010, 8:06pm Top

51> The general issue is always made up of individual cases. As such, I think it's pretty relevant.

So, basically, if anyone wants to combine/separate editions that just say "National Geographic"/no author/no ISBN, they need to go look at the publication data for each individual book or the comments. One thing that would come in handy there would be being able to jump to specific books that roll up into editions which roll up into works. Right now it's a right pain in the butt to get from the editions list to individual book data. And I suppose it would be impossible for the private users.

Or is the better solution that nobody should ever stick their nose into periodicals and try to keep them cleaned up like they do all the other works on the site?

53TineOliver
Oct 14, 2010, 8:13pm Top

Isn't there a more simple way to deal with this?

I actually can't find which the 'current' version of the National Geographic 'work' is but what about:
(a) Separating out your edition from the current work (you could add in a dummy ISBN if necessary).

(b) For the newly separated work -

(i) Set a disambiguation notice something along the lines of "This work is for the National Geographic periodical as a single work. Individual issues should not be combined here"

(ii) Set the canonical title to something like "National Geographi Periodical" (you could leave it blank, but I think this might help prevent future combining/editing)

(iii) Set the series to something like (3000|omnibus)

(iv) Any other known camp (2) entries can be combined here.

(c) On the old work, set a disambiguation notice along the lines of "This work is for unknown/unspecified individual issues of the National Geographic magazine. If you consider the entire periodical as a single work, please separate out your edition and combine it with the relevant omnibus work"

(d) Leave a note in the combiners! group.

54brightcopy
Oct 14, 2010, 8:22pm Top

53> (ii) Set the canonical title to something like "National Geographi Periodical" (you could leave it blank, but I think this might help prevent future combining/editing)

That was actually the first response to the thread. I believe the "Approach 2" posters have a feeling that this is somehow an imposition on them. From messages 3 and 4, they appear to be agin' it.

55aulsmith
Oct 14, 2010, 8:36pm Top

Wow, I'm overwhelmed. I thought the original post was really clear and the only problem was how to tell people using approach 1 and using series to group things not to try and group the approach 2 works. I think it's pretty clear to most of us who have been around a while that cataloging periodicals on LT is not what it was designed for, but the fact is, that because each of our libraries is our own, we can each do what is useful and meaningful for us. It's in the common areas that we need to understand the different ways people are doing things.

32: Actually librarians invented the concept of work (fairly recently) to help deal with a number of thorny cataloging issues. Tim borrowed the concept and gave it a social rather than a bibliographical meaning.

In the library concept of work an entire run of a periodical could be a work. So could an individual issue. So could an article/story/poem in an issue. It's clear why this is so, since different people on LT have had reasons for defining all these different things as works depending on the information they're trying to collect in their catalogs.

56TineOliver
Oct 14, 2010, 8:38pm Top

I think people need to figure out what's more annoying - being combined with various individual issues and them showing up as unknown or having a canonical title that maybe isn't perfect but makes it clearer what the work actually is?

I can see the problem with 'unknown' in the canonical title, because people using approach (2) know which copies they are, but I can't see how having 'periodical' at the end would pose such a problem?

So, for those people actually using approach (2) what is the part of how things are know that bugs you most? What can you live with?

I'm just trying to think whether there's not some sort of logical way we can separate the works so that those using approach (2) have their own 'work' that should not have any copies held by those using approach (1) and it's clear what the work is for so people don't combine them again.

Then those using approach (2) can do what they like with the rest of the CK.

57TineOliver
Oct 14, 2010, 8:42pm Top

55: I think the only way you can 'tell' people that you're using an alternative approach is to make it clear from the work what it is and I think that involves a good disambiguation notice and probably some sort of canonical title.

Not everyone who combines, separates or edits CK looks at the Series group (or talk at all for that matter), so I think the work needs to make your point on its own.

58vpfluke
Oct 14, 2010, 8:46pm Top

I have been under the impression that some people list each issue of their periodicals separately in order to show each cover separately. I am not sure that this has been discussed. If this person has 150 issues of a periodical and it is considered to be one work, then they have a mass of duplicates.

Most peridoicals only have ISSN's, books have ISBN's. If an item has both an an ISSN and an ISBN, what is the thought of people in this thread. I have a 900 page volume whose ISSN is 0084-1382, hardcover ISBN is 0-88687-659-1 and softcover ISBN is 0-88687-658-3 (LCCN is 4-3781).

59jjwilson61
Oct 14, 2010, 9:08pm Top

I read up to about msg #50 and I have to leave so I hope this comment is still relevant.

Perhaps it would help to think of entering a magazine with missing issues as one "book" as equivalent to entering a book with missing pages with which pages are missing in the comments. I don't think anyone would say that a book with missing pages is a different work than the main work.

60rsterling
Edited: Oct 14, 2010, 9:43pm Top

I think the disambiguation notice is the best way to go. However, sometimes these are ignored or disregarded, and things get combined anyway. I don't see any need to change the Canonical Title unless it's confusing (ETA by which I mean if a weird version of the title is coming up as the dominant one, or in other words for the same reasons we'd always use Canonical Title). The periodical title on its own -- Annals of Ancient Greek Basketweaving -- seems fine as a title, no?

I would just be happy if people combining and renaming would not automatically presume that every instance of a periodical is intended by the cataloger to be for a specific issue - that people be aware that some entries might be intended as generic entries. Sometimes that means looking at members' catalogs to try to gauge if the person meant to catalog a single issue, or the periodical as a whole. Obviously that doesn't mean that people shouldn't try to clean up periodicals - that doesn't follow at all - but rather that in cleaning them up, some care and caution is taken not to try to fit square pegs in round holes.

55 - "the only problem was how to tell people using approach 1 and using series to group things not to try and group the approach 2 works." "It's in the common areas that we need to understand the different ways people are doing things."
Yes, those are good ways to put it.

54 - I believe the "Approach 2" posters have a feeling that this is somehow an imposition on them
?? Where did you get that impression? Personally, I'm happy to separate things when they get mistakenly combined and to put disambiguation notices on, either retrospectively, or pre-emtively, where necessary. It's not an imposition. I just don't think Canonical Title is the way to do disambiguation in this case; no need to add artificial status information to the title, particularly when we have a disambiguation field specifically for that purpose.

61prosfilaes
Edited: Oct 14, 2010, 11:10pm Top

#60: Well, yeah, if I took La Faraono and broke it into three pieces, I'd be back to La Faraono, volume 1, volume 2 and volume 3, not three copies of the La Faraono work. Which is a touch more accurate comparison, since just having National Geographic doesn't mean that your series intersect at all.

62brightcopy
Oct 15, 2010, 3:20am Top

60> Where did you get that impression?

You don't think it's a correct summation? It's something that people are imposing on you, and you don't think it's necessary yourself. Especially when people stick on thinks like "Unknown" or otherwise make your title something less than you'd like (e.g. just "National Geographic" instead of something less pleasing - "National Geographic (Periodical)").

63BarkingMatt
Oct 15, 2010, 6:07am Top

Would like to mention a third option:

3.) Catalog periodicals by volume / year.

This is used much (here in Holland at least) because people don't necessarily want to consult every issue when they want to consult a periodical. That could get huge after all.

Not saying people should adopt this method, of course.

64kathrynnd
Oct 15, 2010, 3:44pm Top

63 >> This sounds like a pretty logical way to enter periodicals to me, and I've noticed several members are doing so, but rsterling (msg 1) would like to discuss how periodicals are cataloged and then listed in CK series.

What has been happening in some cases recently is that canonical titles have been added to CK to change periodicals entered by volume / year to year / volume. There is absolutely no reason for doing this. Series order can be set by the CK series statement.

65fdholt
Oct 15, 2010, 6:40pm Top

I'm late weighing in on this and I am going to give you all a warning. This is a catalogers view and when a cataloger talks, eyes glaze over. Even other librarians have been known to fall asleep when we talk about cataloging, MARC, and associated topics.

First what #1 is referring to is a serial (magazine or journal), mean by the publisher to be put out in successive parts, has numeric or chronological designations and is to be continued indefinitely. Of course there are a lot of serials who start and we only see one published issue. I don't know if #1 was referring to annuals (or yearbooks) which are also serials. A series is kind of serial since it bears a collective title as well as it's own title proper. It does not need a numeric designation and there is usually an ending date. (from AACR2R)

Libraries handle serials in several ways. First it costs money to catalog anything and we don't have lots of it. So we use a "serial workform" and do one record as the surrogate for the complete run of the serial, whether current or dead. However due to CONSER rules, name changes now require a new record. (And if you think I am going have 8 or 9 records in my library's catalog for Atlantic because the publisher can't make up his mind what the name of the serial is, think again. The main rule is to do what is best for your users. And remeber it costs money!)

However libraries may, if it helps the user, make in analytics for issues that have a theme or other reasons. An analytical entry just means you've made a record for something that you have a complete entry. (If you look at my library catalog you will see lots of analytics for Clinical infectious diseases and Journal of infectious diseases among others. Just remeber that most of this stuff had to be originally cataloged and tagged by me - money angle again.)

#58's work is a serial type, an annual. Some publishers assign an ISBN in addition to an ISSn for the complete run. It helps with inventory control and books sales when you have this type of item. Since the content does not change in format, just the facts, it is the same work. (Which is why I wondered as a newbie why some editions of my works were combined when they were very different and others which were similar were separate but that's another issue.)

When someone uses a library for their data, they will get a serial record in most cases. If they don't indicate that they are cataloging a single issue, we can't know whether it is the complete run. Hence we don't know if all of the members listed under that work have an individual issue or the serial as a whole. (And as an LT user, I don't like the idea of making dummy titles as place holders - like a $h in a title field, a general material designation - but I could do that.)

The problem with National Geographic is the famous "Atlantic" problem. The publisher changed it's title in the 50's from The National Geographic magazine to National Geographic. So CONSER says this is 2 different serials. Is it? My patron wants National Geographic, its common name and doesn't care. There is nor should be a canonical title. It is one or the other depending on what issue you have and it is wrong to impose one. If I have an issue from 1902, I don't want someone saying the title is National Geographic when it is not.

On to another issue, how we catalog in LT. I have a cross stitch magazine from 1984 to the present and it is published 6 times a year. That's over 140 issues so far and stiill coming. So I chose to have one LT work and indicate in the comments field the ISSN and what issues I have. I chose one representative title cover and use it (which causes another set of problems with combining). Even if I only own one or 2 issues, I still use the serial approach. By using the series page, it avoids a title search although the results should be the same.

By putting a canonical title on a serial run (which can be the former or the newer title), it defeats the prupose of the title proper and I would have to oppose doing that for any reason.

I do want to repeat that any approach for the user which works is not wrong - the whole run, the individual issue, the volume (however many issues that would be) or anything else that works.

That's the beauty of using LibraryThing.

66jjwilson61
Edited: Oct 15, 2010, 7:30pm Top

It is one or the other depending on what issue you have and it is wrong to impose one. If I have an issue from 1902, I don't want someone saying the title is National Geographic when it is not.

You're right for individual libraries in LT but not for works. It's like saying that The Philosopher's Stone and The Sorcerer's Stone is one book or the other depending on whether it was printed in England or America. LT rules say that it's one work and it has to have one title or the other with the more popular winning.

67aulsmith
Oct 15, 2010, 7:32pm Top

65: Why would you need/want to use the series title for OP's approach #2 (the serial title as a work (in cataloger's speak) approach)? I've only been using series titles for OP's approach #2 (the issue of a serial is a work approach). And the only reason I started doing that is because so many people were entering their approach #1 records with such varyiing forms of authors (editors, serial uniform title, "various", blank) that there was no good way I could figure out how to find out what issues people had except to use series.

68fdholt
Oct 16, 2010, 9:49am Top

#66

When we have a work for each issue, I don't see that having the correct title for the title on the issue would be wrong. The 1943 issue 4 of National Geographic Magazine should reflect its title and not be changed to National Geographic, especially if the user has entered it that way. (And I would guess that combining can be tricky if someone doesn't enter a title correctly but that's another issue entirely.) But I do agree on a uniform title (or canonical title as LT calls it) to avoid the CONSER approach when LT has works that reflect the whole series, especially if the title change is trivial. (Definition: addition of the word Magazine to the title, change in subtitle, change in title after the first 5? - I don't have my CONSER manual at home so this is a guess - words.) If the title change is radically different, I would keep it on its own record. Maybe someday we will have linking fields that would take care of this problem but I don't think it's high on my list.

And #66, I do agree with you on the Rowling book, the differing languages of works, and editions in many cases being combined. These are not serials with the complexities involved.

69fdholt
Oct 16, 2010, 10:02am Top

#67

Personally I wouldn't use a series for the title as a whole but only for the individual issues, especially when someone uses the current editor for the main entry. After posting yesterday, I tried to find some serials by using the search in title. It's hard to see the whole picture so I do think now that making a series page is the easiest answer for now in the first approach.

Remember Ranganathan's 4th law - save the time of the reader. The series approach for loose issues does that.

For those who are interested, a quick summary via Wilipedia
http://en.wikipedia.org/wiki/Five_laws_of_library_science

Where using series gets interesting is the gathering approach (when someone catalogs volumes that are bound together as a unit) or the complete run of the serial (how much of the run does someone have and does it matter). With the former, how do you anticipate the hundreds of combinations that may occur with just National Geographic? And how do you account for them? I don't know the answer here because omnibus isn't quite right. This does need further discussion. And once decided, carried through by everyone doing serial combining/separating.

70aulsmith
Oct 16, 2010, 10:05am Top

68: Since experienced CONSER catalogers have trouble with major/minor title changes, I don't think we're going to get Lters to do them. ;)

Maybe less combining/linking of serials (since it's a non-standard use of LT) should be the norm??

71fdholt
Oct 16, 2010, 10:14am Top

#70

Serials catalogers are special people and deserve thanks. I try to avoid serials cataloging and wait for LC or a CONSER library to do it.

And it's important to note that catalog librarians live by rules (AACR2R and soon RDA) and LCRIs (Library of Congress rule interpretations) but the reason always wins with me - we need to serve our users in the best way possible.

I don't know how to respond to less (or more) combining since I haven't done anything except getting my serials on the correct work records.

72jjwilson61
Oct 16, 2010, 10:19am Top

68> But I do agree on a uniform title (or canonical title as LT calls it)...

I don't agree that uniform title and canonical title are the same concept and if you think they are that could be leading to some misunderstandings. The canonical title is supposed to be the title under which the work is most commonly known and in the vast majority of cases it shouldn't be set. Only when LT picks a title that is clearly wrong by anyone's standards should it be used.

73fdholt
Oct 16, 2010, 1:08pm Top

#72

From what I have seen the uniform title and canonical are essentially the same. Granted Hamlet's uniform title is:

Shakespeare, William, 1564-1616. Hamlet.

Instead of Hamlet.

But we all do agree to not use the "official" uniform title since this play is known as Hamlet no matter whether the title proper is Hamlet; Hamlet, Prince of Denmark; The tragedy of Hamlet, etc.

I do not ever assign canonical titles since I have had the bad experience of a stitching chart called Americana combining with Americana by Don DeLillo. No author so I had to put in a publisher as author (grit my teeth!) to get it not to combine. And I finally discovered that a canonical title was the problem. The title doesn't seem to need it but I left it alone.

Tell me what the rules on LT are and, as a cataloger, I will follow them. (And thank you to all of the wonderful members who have helped me fix all my mistakes and have offered advice on porblems. You know who you are.)

74eromsted
Oct 19, 2010, 10:37am Top

I'm not sure I've digested all of the issues above but here are a few comments.

In LT it is standard for works with the same author/title/ISBN to be combined. Considering comments is a lot of work and may be impossible (private libraries), therefor they should not be part of the equation. LT will auto-combine works with identical author/title/ISBN so some book records will be lumped regardless of what the combiners do (e.g. this Nation Geographic work).

To me a disambiguation notice is designed to alert users not to combine works that are in fact different by author/title/ISBN but are often combined regardless. It seems to me strange to use a disambiguation to try to differentiate your particular copy of a generically titled work. Again, as in the example above, your copy may be combined automatically despite the disambiguation notice.

So in the example of National Geographic I would combine every work that does not give more specific information in the title. This would result in a work that contains book records that are known to be different (by the owner, though not by users consulting works and editions nor by the automatic system). I don't see this as a problem. It accomplishes the social function of linking users who list a periodical generically and this strikes me as more interesting than not linking them at all. It also allows for a simpler rule for combiners to follow. And, as always, it doesn't impact anyone's data in her/his own catalog.

I don't see the need for a canonical title in this case, but if one is used for clarity it should at a minimum not be wrong as a description of the work, although it may not fully describe the underlying book records.

75Collectorator
Oct 19, 2010, 12:05pm Top

To The Person(s) To Which This Pertains:
Your distaste and abhorrence of the Evil Canonical Title would be so much more plausible if you didn't have them all over your own works. Plank, eye comes to mind.

761dragones
Oct 19, 2010, 2:02pm Top

Currently, I do not catalog periodicals for various reasons mentioned throughout this thread. If I do decide to catalog my periodicals (of which I retain issues from several different publications), I would be most likely to follow the practices of camp 1.

My reasoning for the choice of camp 1 practices is that, in the comments field of the appropriate record, I would want to list the individual articles contained within each issue. That table of contents data would quickly become unwieldy if I followed the practices of camp 2... not to mention that either the article listings or the issue listings would probably need to be bumped into the private comments field for space reasons if camp 2 practices were followed. Putting the listing of individual articles or issues in the private comments field would mean that the other comments I sometimes place there would be more difficult to locate the the quantity of data from either a listing of individual issues or articles included in the same field.

In no case would I ever catalog the articles separate from the magazine issues in which those articles are contained. That practice would lead to several other difficulties with which I am not prepared to cope. I am hoping that the contains/contained in feature will soon be introduced and will handle the complications inherent in listing individual articles and stories, which is a record I would like to have for reference purposes.

77Proclus
Oct 19, 2010, 8:23pm Top

I'm firmly in camp #2.
I would have close to a thousand new "works" if I cataloged each serial issue separately, and I'm not about to do that.

I deal with the combining issue by simply ignoring whether, or with what, any of my periodicals are combined. Most of the info I want to see is in my list view, so I have little reason ever to look at the work pages.

78rsterling
Jul 9, 2011, 1:37am Top

Bumping, to keep these considerations on people's radar, since I'm seeing some work on periodicals recently that doesn't seem aware of this.

79aulsmith
Jul 9, 2011, 7:07am Top

78: Perhaps you could summarize the points you think people need to remember? It's a longish thread with a lot of back and forth.

80theapparatus
Jul 9, 2011, 12:05pm Top

+1 on 79 there. :)

I;m more concerned/annoyed with the folks who are separating out copies of certain knitting magazine when they're the same issue. *sigh*

81rsterling
Jul 9, 2011, 12:16pm Top

Yeah, when I read the whole thing again last night I was surprised.

I think the main point of the original post was this:

People combining, separating, entering series, etc. should be aware that not everyone catalogs their periodicals issue by issue. Some people catalog just the whole periodical, and then list their holdings, like libraries do (I call those generic editions, or, above, approach 2). Please do not assume that everyone lists periodical issues individually, and please do not combine, separate, or add CK on that assumption. Please do not combine generic editions with individual-issue editions.

82rsterling
Jul 9, 2011, 12:16pm Top

80 - That's another issue. I have seen this happening with other series as well.

83theapparatus
Jul 9, 2011, 2:04pm Top

I do hope no one ever combines all of the Dragon magazines.

I personally think that they should be individual records where each issue is under their own record.

84fdholt
Jul 9, 2011, 4:18pm Top

#83 Except for those of us who have lots of issues. Haven't done this title yet but plan to have only one record and I wouldn't want it combined with a single issue record.

85theapparatus
Jul 9, 2011, 4:46pm Top

Well with 200+ Dragon magazines in existence (Or whatever the count it, I'm drawing a blank and I've hacked a catalog terminal to type this so I'm typing fast) I'd rather have 200+ in my record account than 200+ on my dupe item report.

But yes I can see why one would rather have everything under one record.

Another issue with placing all of those items under a single record, it makes it harder to combine with other records. I know I;ve had more than one occurrence where I've put together a series and had to list a record as Item #0 because someone has Items 3,4, and 9 (for example) as a single record. Makes a mess.

86fdholt
Jul 9, 2011, 6:08pm Top

#85 In the record I do not say how much of the run I own. I use my comments field for holdings as a library uses the MARC holdings field. The assumption, like a library, is the complete run.

87Collectorator
Jul 9, 2011, 6:27pm Top

This is so simple. Either make it clear that it is the entire set of magazines, or separate it when it gets combined. eos.

88rsterling
Edited: Jul 9, 2011, 6:34pm Top

Either make it clear that it is the entire set of magazines

This can be done by disambiguation notices, and has been, though sometimes people disregard them.

There's a distinction here between "entire set" and "generic entry" though. I'm talking about the latter, and about an approach to cataloging that fdholt describes in 86, and that several people on LT use.

My hope is that if more people are aware that there are different approaches to cataloging periodicals, it might help prevent things from being wrongly combined and wrongly separated.

89fdholt
Jul 9, 2011, 7:09pm Top

And I do try to write a disambiguation notice saying that this record is for the complete run of the serial and that it should not be combined with single issues. As #88 states, the notice can be ignored.

90Collectorator
Jul 9, 2011, 7:09pm Top

Combinations and separations are easy to accomplish. Simple, even. I think this is a waste of time and effort.

91vpfluke
Jul 9, 2011, 11:37pm Top

Combinations are easy to do, but separations can take time. However, I usually only work with serials that are not periodicals.

92theapparatus
Jul 12, 2011, 5:18pm Top

fdholt, didn't say it was you. :)

Another example is the Battlestar Galactica series. Had a few of "Season #, Episodes 1 3 7 8-11" Those were fun.

93ErkDemon
Mar 15, 2012, 6:16pm Top

Firstly, yes, LibraryThing wasn't designed to handle periodicals ... and it shows. And unfortunately, if you have a collection that includes periodicals, that's often a reason not to use LibraryThing.

I'm currently doing some volunteer work at a small museum, and I'd like them to use LibraryThing for the books in their resource room, but if LibraryThing can't cope properly with the periodicals, then we'd need two systems, which isn't a supportable option. I think they'd rather wait unto someone has a single system that's designed to deal with both types of publication.

They need to catalogue "items" that are individually movable, because they need to know where things are, and who has them. So they need the option of cataloguing by volume when they have a bound volume of issues, but also the option of cataloguing individual loose issues, and right now, I don't see any obvious structure built into Librarything that is designed to cope with that.

A suggestion:

1) include a switch or checkbox for periodicals. If it's a periodical, you get an "ISSN number" field instead of an ISBN.

Include specific fields for:
(a) Periodical Name,
(b) Periodical ISSN
(c) Volume
(d) Issue Number (yearly)
(e) Issue Number (global)
(f) optional date as printed on the cover (dd/mm / mm / freeform),
(g) Additional

If "book" names need to be unique, then the name would be automatically constructed by Librarything from the contents of those fields. So, if there's a volume number but no issue number or sub-date, then we assume that the item being catalogued is an entire volume. If an issue number or sub-date is included, we assume that it's an individual issue. If there's no volume or year or issue number or date, we assume that it's an entire collection. If we have a bound volume that's missing an issue, we can use the additional field to enter an item-specific qualifier, so that we can distinguish between two sets, "Nature Vol.50" and "Nature Vol.50 (incomplete).

For comic collectors, the "Additional" field could distinguish between "SuperThing #572" and "SuperThing #572 (Collector's Special Edition, silver cover)"
The "Additional" field could also be used to label special issues of a serial that existed outside the number sequence - special yearly indexes for science journals, Christmas specials, bonus issues, Scientific American "Specials", and so on.

If the system needs a unique ID number (taking the place of the ISBN for single books), then we could use the ISSN number with selected additional data appended (eg ISSN-YYYY, or ISSN-YYYY-Mo).

If people still want to make up their own free-form systems, then fine, but LibraryThing ought to be creating and supporting a generic system for periodicals for those that want it. If someone owns "Spiderman issue whatever", and someone else owns a copy and fills in the description that it's the famous issue that marks the first appearance of "SuperBaddyX", then we want that information to propagate to the individual listings of everyone else who has that same issue listed individually, without relying on people magically arriving at the same way of typing out all the issue identification details into a title. We need a centralised and organised core system that people can use and tap into, and which can be changed centrally without people having to go back and reformat the titles of hundreds of issues manually. For speed, we need to be able to click on an issue, copy it, modify the issue number of the copy and resave, and have the identifying title of the copy change automatically to reflect the change without being manually retyped.

If we want to get really advanced, we could have a customisation string for each ISSN number, that defines how the name of a single issue of a serial is constructed from the field data. For some serials it'd be preferable to use Volume/Issue, for others, date/year, for most comicbooks, I’d guess that an issue’s title would normally be the name of the serial plus a hash-prefixed issue number (“The Amazing Spiderman #100”). The default title for individual issues could then be changed and updated globally for a whole series without damaging any of the underlying metadata.

Field “g” is the special-case field for any "manual" qualifier text that goes onto the end of the auto-generated title, to distinguish between something like "Batman #300 (Catwoman cover)" and "Batman #300 (Joker cover)" (same volume, issue and date, but still different items), or between "Physical Review Vol 40" (a bound volume) and "Physical Review Vol 40 (yearly index)" (an cumulative index issued every year, or every few years)

Once your "periodicals" system can cope with comicbooks and science journals, then it should be able to deal with almost anything. The comicbook community are probably a good place to start, since they take cataloguing very, /VERY/ seriously, they've probably already put a fair amount of thought into most of the problems that you might come across, and they won't be shy about telling you when you've missed out a critical feature.

94AdocentynLibrary
Jun 16, 12:18am Top

So ErkDemon, #93, did you find an alternative to Library Thing that handles periodical as well as books? We are at the stage where we need to figure out how to handle the journals, magazines and periodicals we hold and are having a hard time figuring out what to do.

Group: Librarything Series

127 members

3,916 messages

About

This topic is not marked as primarily about any work, author or other topic.

Help/FAQs | About | Privacy/Terms | Blog | Contact | APIs | WikiThing | Common Knowledge | Legacy Libraries | Early Reviewers | 92,361,820 books! | Top bar: Always visible