About the Free Decimal Correspondence

What is the Free Decimal Correspondence?

The Free Decimal Correspondence, or FDC for short, is a set of decimal numbers ranging from 000 to 999[.9999…], each associated with a particular subject, discipline, or group of subjects and disciplines.  It’s intended to be  reasonably compatible with existing and commonly used library decimal classifications and subject headings, but also as freely usable and adaptable as possible.

You can view or download it from this page.

Why use  a decimal correspondence to subjects?

Many libraries use such a system to arrange their books on a shelf (or their electronic items in a list) in the order given by the decimal numbers, so that they’re organized in a general hierarchy with items on similar subjects located near each other. These numbers, when assigned to particular items, are referred to as “call numbers”. For instance, if you’re interested in political science, you could go to the items with call numbers between 310 and 320, and find lots of political science resources on similar topics presented next to each other.  And you’ll find other social sciences nearby as well.

Decimal systems can also be used to give a language-independent representation of a particular concept. (So, for instance, “mathematics” in English and “matematica” in Italian can both be expressed by the FDC decimal code “510”.)  You can also use FDC to label sets of items you’ve associated with particular decimal numbers and ranges.

There are other possible uses as well, which you’re welcome to invent and try out.  And you’re also welcome to add new terms, or come up with more suitable correspondences and terminology than I’ve used.  This is a free system, after all.

I’ve seen books shelved by call numbers in my local library.  Are they using this correspondence?

Not that I’m aware of, especially since this correspondence was just released in 2009.  They’re probably using another decimal system.  The most commonly used decimal call number system is the Dewey Decimal Classification (DDC).  We’ve tried to make this classification compatible with the present-day Dewey system, so the numbers will in many cases be similar in DDC and FDC for similar subjects.

So wait– is this just the Dewey Decimal Classification under a different name?

No.  Among other things, the FDC is considerably briefer than the DDC, with less detail and almost no editorial apparatus.  It doesn’t include many of the subjects that DDC does.  It associates different terminology in many cases with the numbers than the DDC, and is not guaranteed to be compatible with present-day DDC.  (In particular, we have not consulted the DDC itself when preparing the FDC, except to identify unassigned numbers to skip over in the FDC.) We have made some attempt to be compatible with DDC, however.

What do you mean by compatibility?

We would like to have any number used both by the present-day DDC and the FDC to refer to similar topics in both correspondences (though they may be expressed in different terms).  The goal is that if items classified with DDC are interfiled with items classified with FDC, there shouldn’t be any jarring inconsistencies when looking at assignments in the two systems.  We have not, however, taken all possible steps to ensure compatibility, though, for reasons stated further down, so we can’t guarantee that interfiling will be as seamless as what we hope for.

We also attempt to be compatible with the Library of Congress subject headings (LCSH), in that we attempt where possible to express the subjects we associate with the numbers in the terminology of authorized LCSH headings.  This should hopefully make it easier to move between subject metadata (which often uses LCSH) and decimal classifications such as FDC.

Why not just use DDC, then?

You can, in many cases, and it may well be preferable for you to do so.  But you might need to get permission from DDC’s current owners first, depending on what you plan to do with it.  Recent versions of the Dewey Decimal Classification are copyrighted by OCLC, who also claim trademark on “Dewey Decimal” and a number of related terms.  There are, however, some licenses and inherent limits on the copyrights to OCLC’s decimal classification system.  As Ralph LeVan of OCLC put it recently:

There are absolutely no restrictions on using the Dewey numbers.  You can assign those numbers to your works and then use them to organize your works.  It would be nice if you said something on your site about Dewey being copyrighted.  But otherwise, numbers are numbers and you can use them to your heart’s content.

The problem comes when you try to assign meaning to those numbers; then you’re using the work of the Dewey Editors.  The text associated with those numbers IS copyright.  But, if you can restrict the usage to just browsing up and down the numbers, you’re good.

Do I have to get permission to use FDC as well?

No.  I’ve explicitly placed the FDC in the public domain, so no permissions are required to reuse or adapt it for any purpose you like.  I put no restrictions on your use of the text I’ve associated with my numbers.

Keep in mind a few things, though:

• There is no guarantee for FDC’s continued maintenance or persistence.  So you might want to download and retain a local copy if you plan on using it.
• There’s no warranty on the correspondence, and no guarantee that the system is consistent with itself or any other system. It’s also not as detailed as most other library classification systems.  I don’t recommend throwing out whatever established classification you’re currently using in favor of this.  But you might find it useful for organizing small collections or repositories that don’t need the same level of fine grain cataloging that a full-sized research library needs.
• If you use or modify the FDC and call it a name claimed by someone else, that someone else might get justifiably upset.  So don’t do that.
• The system is subject to change and tweaking, either by me or others.  If you do use the system, it might help minimize possible confusion to explicitly state that you’re using the FDC, to identify the version of  FDC you’re using, and to state whether you’ve modified it from that version.

How can you place FDC in the public domain, if it’s intended to be compatible with the copyrighted DDC?

I don’t use the same “text associated with.. numbers” that DDC does.  In fact, I didn’t even look at the current DDC when preparing the FDC, except to identify unassigned three-digit codes in the most recent edition of the DDC (so I could keep the same codes unassigned in FDC for compatibility’s sake).  The FDC is my own editorial work, done on my own time, drawing on public domain sources and some reverse engineering of the arrangement of some present day library stacks.  While I’m not a lawyer, my understanding is that in the US, this sort of reverse engineering for the purpose of compatibility and interoperability is generally not considered an infringement of copyright.

How did you create FDC, then?

I started with Melvil Dewey’s Decimal Classification…, 11th edition, published in 1922.  This is public domain just about everywhere, since 1922 copyrights have expired in the US, and Dewey died in 1931, more than 75 years ago.  Dewey’s 1922 classification goes on for hundreds of pages; I mainly considered the high-level classification down to the level of “units” (e.g. “001”, “002”, up to “999”, without any  further decimal elaboration.)

The 11th edition, unrevised, is not very suitable for present-day classification, however.  In order to make it more usable, and compatible with current library practices, I’ve standardized Dewey’s “reformed” spelling, looked through institutional repositories like Penn’s to introduce new subjects not imagined or seen as important in 1922 (like “Computer science”), and looked through present-day library call number listings (like those of the Free Library of Philadelphia) to see what books are currently being filed at different numbers.  (Some subjects have been moved to different numbers since Dewey’s 1922 classification, so I had to decide at each point whether to use Dewey’s 1922 description, revise it, or come up with something  different.)   I’ve also looked at the Library of Congress subject authorities, and tried to make the FDC terminology match authorized terms there, again for compatibility’s sake.

In what forms is FDC available?

I’ve made it available as HTML, for easy human review, and as structured text for easier downloading and machine processing.  The format of the structured text is described at the top of the text file, and hopefully is reasonably self-explanatory.  The HTML is automatically generated from the structured text file.

How much detail does FDC go into?

In the current release, I’ve included correspondences to subject terms for all of the unit values (000, 001… up to 999) that are currently assigned in DDC.  (Note that the 040-049 range and the 921-929 range are completely unassigned, as are some other individual units.)  I’ve gone past the decimal point in some cases, for some locations and subjects that I expect to be common in institutional repositories and small Engligh language collections.

Will there be future revisions?

I released the first version (0.01) on Public Domain Day, January 1, 2009, and have made some other releases since, the latest (0.06) on on August 19, 2010. As noted above, this version gives complete coverage down to the unit level.  There’s still some room for augmentation, though; for example, to include specific subjects that might be common in present-day libraries and institutional repositories but that aren’t defined at the unit level.

I don’t plan to provide long-term maintenance or support for FDC, however.  But since it’s public domain, anyone else is welcome to further revise, adapt, and support it.

Who uses FDC?

Well, it’s just been released recently, and I don’t (yet) know of anyone using it directly in production.  But it’s one of the bases for the Melvil Decimal System recently introduced at LibraryThing. There’s also been some interest expressed in creating an SKOS version.

For The Online Books Page, I use a different call numbering scheme, the Library of Congress classification.  This classification is also in the public domain, at least in the US, and is used in many American research libraries (including Penn’s).  I don’t intend to switch to FDC for this resource.  But some folks prefer numeric classifications, and might find FDC of interest.  (Or DDC, or UDC, for that matter.)  BISAC, used in many bookstores, and a few public libraries, may be worth considering as well.  <!– Folks who want to design a completely new free shelf classification, without concern about compatibility to any previous system, may also be interested in the Open Shelves Classification. –)

Are there better ways of organizing books by subject?

Well, one advantage of online catalogs is that you don’t have to settle for linear orderings of books, or single hierarchies, such as you get in call number systems like FDC and DDC.  Online, you can use ontologies or thesauri that arrange subjects into conceptual networks, where any term can be related to any number of other terms in a wide variety of ways.  I’ve created “subject map” browsing based on such networks for The Online Books Page, using LCSH, and it seems to be more popular and effective than either call number browsing or alphabetic subject browsing, both of which I also provide,

Why did you create FDC, then?

Basically, as a small way to promote free sharing and reuse of library metadata and standards.

FDC didn’t come out of nowhere.  Recently, someone with an institutional repository asked on a DSpace forum whether they could make their items browsable by call number using a decimal call number scheme.  The constraints given for using the Dewey Decimal Classification were not satisfactory to some readers.  So I thought it worthwhile to spend a little time creating a simpler, reasonably compatible system without intellectual property constraints.  While I don’t find decimal classifications the best way of browsing by subject, they do at least require less in the way of programming and data infrastructure than subject maps do.

I’m not that concerned about OCLC claiming ownership of DDC, particularly since I don’t see DDC as an essential part of the future digital library infrastructure, and because I haven’t heard much complaint to date from libraries about OCLC’s conditions of use for DDC.  But if any folks really want to use call numbers but find OCLC’s constraints on DDC unacceptable, they can work around them.  FDC can be seen as one example of a workaround.

I am more concerned about other attempts to lock up library data, some of which are going on right now, such as OCLC’s attempt to claim ownership of any library’s catalog record shared through the WorldCat union catalog.  (The revised policy, which went into effect in Augost of 2010, is somewhat different from the original proposal, but still limits the reusability of records in significant ways.) Policies like WorldCat’s put more substantial and problematic roadblocks for libraries sharing their knowledge about information resources.  But they too can be revised, or worked around, if libraries have the will to do so.  It will take more work to do so than it did to create FDC, but I hope that the same kind of spirit that can bring about free alternatives to DDC can also bring about freer sharing, reuse, and innovation on our libraries’ collective knowledge base.

So, this is FDC, dedicated to the public domain, as is this information page.  I hope you find it useful.

John Mark Ockerbloom

Originally published Public Domain Day, 2009

Last revised for release 0.06, August 19, 2010