Content area
According to the current (2022) edition of ISO 3297, the standard that governs the International Standard Serial Number (ISSN), a cluster ISSN is "an ISSN assigned to group continuing resources related to each other." Unless the system creating the link and the library catalog receiving the link use the same ISSNs, the links will fail and the end user will experience a dead-end link. ...The goal of the revision to the ISSN standard was to create a "Work-level" (title-level) identifier; one that would supplement the current ISSNs (which are still needed to uniquely define the medium version in support of the supply chain) and allow the overall publication to be identified to support its access.6 To deal with any case where one medium version undergoes a major title change (and is assigned a new ISSN), but the other versions do not, the group does not receive a new ISSN-L until the titles of all the medium versions have changed." A clear conflict between the acquisition supply chain and the goal of linking medium versions in an online environment Solving a problem like this is possible only if all parties to a transaction implement the solution, something that was obvious to Pesch back in 2009: "All parties that may be involved in linking or disseminating information need to accept and distribute the ISSN-L as appropriate for the system to be effective.8 As an added nudge, he noted that "to help bootstrap the various access and linking systems, ISSN mapping files have been provided [by the ISSN International Centre]... free of charge." Fast-forward now to 2017, when TC46/SC9 (Identification and Description), the International Standards Organization group responsible for the maintenance of ISO 3297, undertook one of its periodic reviews of the standard.
First, I want to give an update to my previous column. My column in the September-October issue of Technicalities dealt with the evolution of the British newsmagazine The Economist, including the diverging paths taken by its print and online versions." The following seemed deserving of retroactive inclusion. In 2022, The Economist began issuing a combined issue covering the last week in July and the first week in August, a practice they already applied to the last two weeks in December with their so-called "holiday issue." Two years later, in a novel twist, they began a digital-only issue for the first week in August- the second week of the preceding "combined" two-week issue And just to make things more interesting, they integrated this digital-only issue into the numbering scheme of the print version of which it was not a part. The upshot is that, from 2024 onwards, a library retaining the print version of The Economist may have both a complete run (in that they retain all the issues that were printed) and an incomplete run (in that this does not include all the issues that were produced, as signaled by the gap in numbering). O brave new world that has such creatures in it!
But on to the subject of the column: the cluster ISSN!
And Just What Is a Cluster ISSN?
I am glad you asked! According to the current (2022) edition of ISO 3297, the standard that governs the International Standard Serial Number (ISSN), a cluster ISSN is "an ISSN assigned to group continuing resources related to each other." The ISSN Manual further explains that cluster ISSNs "provide for the identification of specific groups of ISSN, such as the various medium editions of a continuing resource or [spoiler alert] serials related by former and later relationships." To be clear, this is a new concept.
The first cluster ISSN (a cluster ISSN avant la lettre, as the French say) was the ISSN-L or linking ISSN. The ISSN-L was introduced to ISO 3297 in 2007. Its purpose was "to enable collocation or linking among the different media versions of a continuing resource" and more specifically "to enable the use of ISSN in cases where finding or linking to a continuing resource without regard to medium is desired." While the ISSN-L Was a new type of ISSN, the value used as an ISSN-L was not a new one but Was drawn, for any given title, "from among the values assigned to different media versions [of that title]."5 In practice, this has usually turned out to be the ISSN of the print version.
But what was the practical purpose of such a number? Well, we have a very detailed explanation of the very real problems that the ISSN-L was intended to address in a 2009 article by Oliver Pesch, chief product strategist at EBSCO Information Services:
ISSNs are currently assigned to each medium version of a continuing resource. For example, a print version of a journal will have one ISSN and the online version of the same journal will have its own ISSN. This makes perfect sense for the acquisition supply chain where the ISSN is often used as a product identifier for ordering. Clearly the print and online versions of a publication need to be distinguished when ordering a journal; however, in the increasingly networked world of online information, the ISSN is often used in links as the unique identifier for the title and having different ISSNs can be problematic ... Unless the system creating the link and the library catalog receiving the link use the same ISSNs, the links will fail and the end user will experience a dead-end link. ...The goal of the revision to the ISSN standard was to create a "Work-level" (title-level) identifier; one that would supplement the current ISSNs (which are still needed to uniquely define the medium version in support of the supply chain) and allow the overall publication to be identified to support its access.6
To deal with any case where one medium version undergoes a major title change (and is assigned a new ISSN), but the other versions do not, the group does not receive a new ISSN-L until the titles of all the medium versions have changed." In such cases, some version(s) would presumably carry two linking ISSNs-the old one retained while the other versions caught up and the new one assigned once the other versions did, in fact, catch up. I have never actually encountered an example of this, and I hope I never will.
I have, however, encountered the problem that Pesch described in his 2009 article. This happened when our library's link resolver sent an OpenURL query to the journal database of a major publisher only to be told that the desired article was not there when in fact it was. Such "404" errors can occur for a variety of reasons, but in this case our link resolver had sent the print version's ISSN to the publisher while the publisher just had the ISSN of the online version stored in its metadata because the journal was no longer published in print. A clear conflict between the acquisition supply chain and the goal of linking medium versions in an online environment
Solving a problem like this is possible only if all parties to a transaction implement the solution, something that was obvious to Pesch back in 2009: "All parties that may be involved in linking or disseminating information need to accept and distribute the ISSN-L as appropriate for the system to be effective.8 As an added nudge, he noted that "to help bootstrap the various access and linking systems, ISSN mapping files have been provided [by the ISSN International Centre]... free of charge."
They can be used by online databases to add the ISSN-L to their content systems. Link resolvers can use these tables to create a comprehensive list of alternate ISSNs to be used during the look-up process as well as to determine the ISSN-L for creating links to other systems. Library catalogs can use this list to add the ISSN-L to bibliographic records (022 $1) as well as include and index the additional ISSNs related to a title to allow more links to succeed.9
So, more than a decade later we are still experiencing problems implementing the ISSN-L as it was intended. This will elicit an unsurprised nod of recognition from many librarians.
Fast-forward now to 2017, when TC46/SC9 (Identification and Description), the International Standards Organization group responsible for the maintenance of ISO 3297, undertook one of its periodic reviews of the standard. As part of that review, users of the ISSN were surveyed on various questions, including users" interestin what the survey called "family" ISSNs.
Users were asked to rate the priority of different options for a family ISSN including overall interest on a scale of High to Low with an option of Zero interest. The results suggest that overall there is interest/support for the notion of family ISSNs with a total of almost 77% registering either High or Medium interest ... There is even higher combined interest in title history, reaching 89%, and slightly less interest in either geographic families or language families but still reporting high/medium interest of 63% and 67%, respectively.10
The group had expected the title history family to be popular among survey respondents, so it included further questions asking them to identify the relationships that might define such a title history family. The winning combination was one that included not just the more straightforward one-to-one continuations but also the more elaborate splits and mergers, basically all the relationships encoded in MARC 21 fields 780 (Preceding Entry) and 785 (Succeeding Entry). Users also preferred reusing an existing ISSN (as had been done with the ISSN-L) rather than assigning a new ISSN.11
But TC46/SC9 chose not to designate a new "title history" family and an associated ISSN type in ISO 3297. Rather, they decided to extract and generalize the model implicit in the ISSN-L. They created the cluster ISSN.
As noted earlier, they defined the cluster ISSN as "an ISSN assigned to group continuing resources related to each other." They devolved responsibility for designating specific cluster ISSNs to the Registration Authority for ISO 3297, the ISSN International Centre, and the ISSN-L was removed from the standard proper, retained only as an example and reframed as a "cluster ISSN designated by the ISO 3297 Registration Authority. "12
But to return to that looming "title history" family, something like it had been around in the past, if in a more limited context.
An Antecedent
What had been around was the XISSN Title History tool (Figure 1), an online tool that had been introduced by OCLC some years earlier as part of its XISSN web service. This tool took as input an ISSN and produced as output a diagram showing the related earlier and later titles and ISSNs as well as the ISSNs of any related medium versions, so long as these were present in the OCLC database.13
OCLC introduced its XISSN Title History tool in late 2007, and it lasted about a decade. I confess that I remember it fondly. As you can see from Figure 1, it gave a very nice presentation of the history of a serial, at least in more straightforward cases. I used it in the first edition of my book, RDA and Serials Cataloging, to produce an image illustrating a not-so-straightforward case, the title history of The Saturday Review.14 It performed its duty admirably.
Sadly, however, the xISSN Title History tool is no more. OCLC retired it on March 15, 2016, along with all of its xID services, due to low usage.15 Consequently, it was no longer available when it came time for me to produce a new edition of RDA and Serials Cataloging, and I had to cobble together on my own what can only be described as a poor substitute. 10
But to return to our story...
A Family ISSN?
The ISSN International Centre's strategic plan for 2020-2024 listed as Objective 3 "Implement the sixth version of the ISSN standard by creating the family ISSN and other cluster ISSNs."
Wait a minute. Where did this "family ISSN" come from?
A survey of our user community in 2019 revealed an interest in what could be called an ISSN-F grouping together titles identified by ISSNs and representing the same family. Discussions within the ISSN International Centre determined that an ISSN-F could be embodied as a URI providing access to a visual representation of a publication's history. This representation could be easily created and managed centrally without additional work for the National ISSN Centres as it is already available in the ISSN Portal Without yet having a URI or being identified via an ISSN-F.17
So the family ISSN was our old friend, the "title history" family ISSN.
I have to admit there is something appealing in viewing the titles involved in a serial's history as a family. In this sense, I suppose the "representation" referred to in Objective 3 might be akin to a family portrait.
But to realize this ambition, there was work to do. A family needs a home, and that meant MARC. On May 27, 2021, the ISSN Review Group submitted MARC Discussion Paper no. 2021-DP07, "Recording Cluster ISSNs in MARC 21 Bibliographic, Authority, and Holdings Formats," to the MARC Advisory Committee (MAC).' It proposed a new MARC field 023 for cluster ISSNs, modeled on the 022 field then used for both plain vanilla ISSNs and ISSN-Ls. The new field would be used for ISSN-L (which would migrate from its dedicated subfields in field 022) and for other cluster ISSNs "to be defined in future such as a possible ISSN-F or "family ISSN." In its review of the discussion paper, MAC proposed a few clarifications, and noted that "[t]he paper will likely return as a proposal."
As indeed it did, though it took its time doing so. It was two years later, on June 28, 2023, that MAC took up MARC Proposal no. 2023-06 defining a new repeatable field for Cluster ISSNs. Deanna White of the ISSN International Centre presented the proposal, which was discussed and ultimately approved by MAC (and subsequently by the MARC Steering Committee).' I should note that, for the incurably curious, a recording is available. As are recordings of all subsequent MAC meetings, a wonderful historical record."
By the time the proposal came before MAC in 2023, the name "Family ISSN" used in the original "Discussion Paper" had been abandoned in favor of the more descriptive, if more prosaic, "History ISSN"
The ISSN International Center has now developed the History ISSN, abbreviated ISSN-H, which ... has the structure of an 8-digit unique number butis prefixed by ISSN-H and does not repeat any of the ISSNs which are in the cluster. It clusters the ISSNs of sequential titles of a continuing resource over time. The resultis a unique ISSN and accompanying URI for the title history of a continuing resource.
The proposal noted that the ISSN International Centre would retroactively assign ISSN-Hs to "clusters linked by preceding and succeeding titles including all corresponding medium versions" in the ISSN database. (You may have noticed this mirrors the old xISSN Title History tool.) Prospectively, an ISSN-H would be assigned to any title to which a preceding (780) or succeeding (785) title is linked, as well as the related title(s) and any associated medium version(s)
In the MAC discussion, Regina Reynolds, head of the US ISSN Center, observed that the proposal
supports Linked Data going into the future, and it also, via the ISSN-H, really helps the dilemma of people, users, publishers, or databases that have a need to give the whole history of a serial that has changed titles many times, and having an identifier for this title history supports those who think of a serial that, even if it's changed title, as one work, one publication."
(What constitutes a serial "Work" will have to wait for a later column.)
A diagram giving an abstract representation of a simple ISSN-H cluster (Figure 2) accompanied the MARC proposal. After extensive discussion, MAC accepted the proposal as submitted.
You might reasonably then have expected a similar outcome for the ISSN International Centre's analogous proposal to update the UNIMARC format. After all, they are both MARC formats. But you would be wrong.
The ISSN-His a cluster identifier distinct from the ISSN and ISSN-L Itis automatically assigned from a special block of ISSNs starting with 9... The ISSN-H groups all titles linked in the ISSN+ tool through MARC?21 fields 776, 780, and 785.3
This includes what the ISSN Manual calls "simple" ISSN-L clusters (two medium versions linked by one 776 field pair but with no earlier/later relationships). Consequently, the set of continuing resources brought together in ISSN-L clusters can, in terms of cluster membership, be viewed as a subset of those brought together in ISSN-H clusters.
So Where Do Matters Stand Now?
Atthe time of writing-mid-September 2025-the ISSN-H remains timidly off-stage, currently enabled only in the ISSN Network's ISSN+ tool, its wider deployment dependent on a still-pending deployment of the new ISSN Portal. Likewise, at the time of writing, MARC field 023, while officially part of the MARC21 formats, had yet to be authorized for use in CONSER records, and the expanded use of UNIMARC 011 has yet to make its way into the ISSN Manual.
So, at this point we must be left to speculate. At the very least, the ISSN-H will be a unique element and value present in all continuing resource descriptions linked by earlier/later relationships and/or medium version relationships. How any given agency decides to exploit this new element remains to be seen. The possibilities are many.
Perhaps we will see a display akin to what OCLC's xISSN service used to provide, though one can imagine cases where such a display could become quite elaborate. The mind boggles when contemplating, for example, how it would handle the Bulletin signalé-tique, an abstracting journal published decades ago by the French Institut de l'information scientifique et technique (INIST), which started out fairly simple but rapidly ramified into a complex of intermingling and entangling descendants that resembles nothing so much as a plate of spaghetti, a structure that would challenge the sense-making abilities of the most sophisticated image renderer.24
On the other hand, an implementor might choose no display at all, but just one very long run-on title history note, such as can be found in the Notice historique de périodique under Bulletin analytique. Philosophie in the catalog of the Bibliotheque nationale de France (BnF) at https: //catalogue.bnf. fr/ark:/12148/cb327168628. (I will not reproduce the note here.)
Or perhaps something in between. Or perhaps something yet unimagined. Perhaps the wisest thing one can do at this point is simply anticipate the arrival of the ISSN-H, keeping the shape of the thing arrivingin a suspended state of "something to be anticipated," as the character Tony does here in the musical West Side Story:
Couldit be? Yes, it could
Something's coming, something
good,
If I can wait.
Something's coming, I don't know
What it is,
Butitis
Gonna be great!
We shall leave it at that for now.
Postscript (AI Doom Watch)
After completing this column, I did a search in my smartphone's web browser for "issn-h" and "issn+" and got back a Google AI Overview exploring the intersection of the two topics that was both comprehensive and linked to authoritative sources. I find this both encouraging - highly reliable information presented at the top of a search result - and personally dispiriting: Who needs me then?
References and Notes
1. Ed Jones, "Serial Evolution: À Case Study," Technicalities 45, no. 5 (Sept./Oct. 2025): 1, 8-11
2 International Standards Organization, Information and Documentation -International Standard Serial Number, ISO 3297:2022, 7% ed. (Geneva: ISO, 2022), 3.4.1.
3. ISSN International Centre, ISSN Manual, MARC 21 Version, Edition: December 2024, Section 7 "Clustering ISSNs," www.issn.org/ wp-content/uploads/2025/05/ISSN_ Manual_pdf_en html (accessed Sept. 17, 2025).
4. International Standards Organization, Information and Documentation- International Standard Serial Number, ISO 3297:2007 (Geneva: ISO, 2007).
5. Ibid, 2.9, 6.
6. Oliver Pesch, "ISSN-L: A New Standard Means Better Links," The Serials Librarian 57, no. 1-2 (2009): 41-42, DOI: https://doi.org/10.1080/03615260802669052.
7. ISSN International Centre. ISSN Manual, MARC 21 Version.
8. Pesch, "ISSN-L," 44.
9. Ibid, 46.
10. Gaélle Béquet and John Akeroyd, "Irreplaceable ISSN: From 200 to the Future," The Serials Librarian 75: no. 1-4 (2019): 24. DOI: https: //doi.org/10.1080/0361 526X.2019.1592795.
11.1bid.; 25:
12. International Standards Organization, Information and Documentation -International Standard Serial Number, ISO 3297:2020 (Geneva: ISO, 2020), 3.4.1, 3.4.4.
13. Mike Beccaria, Roy Tennant, and Adam Traub, "Can't We Write a Little Script for This? Managing Serials Data and XISSN," The Serials Librarian 60, no. 1-4 (2011): 181-85. DOI: https://doi.org/10.108 0/0361526X.2011.556031
14. Ed Jones, RDA and Serials
Cataloging (Chicago: ALA Editions, 2013).
15. OCLC Developer Network, "Change to xID Services" (Dec. 10, 2015), archived on Internet Archive Wayback Machine, https://web. archive.org/web/20151214053251/ http://www.oclc.org/developer/ news/2015/change-to-xid-services. en.html (accessed Sept. 17, 2025).
16. Ed Jones, RDA and Serials Cataloging, 2nd ed. (Chicago: ALA Editions, 2025).
17.ISSN International Centre, Strategy 2020-2024 (Paris: ISSN International Centre, 2020), Objective 4, www. issn.org/wp-content/ uploads/2020/07/Strategy-ENG.pdf (accessed Sept. 17, 2025)
18. Library of Congress, MARC Standards, ISSN Review Group, "MARC Discussion Paper No. 2021-DP07 - Recording Cluster ISSNs inthe MARC 21 Bibliographic, Authority, and Holdings Formats," May 27, 2021, https://loc.gov/marc/ mac/2021/2021-dp07.html (accessed Sept. 17, 2025).
19. Library of Congress, MARC Standards, ISSN Review Group, "MARC Proposal no. 2023-06 - Defining a New Field for Cluster ISSNS in the MARC 21 Bibliographic, Authorities and Holding Formats, May 25, 2023, www.loc.gov/marc/ mac/2023/2023-06.html (accessed Sept. 17, 2025).
20. Library of Congress, MARC Standards, "MARC Advisory Committee Meeting Recordings, 2023-2025," www.loc.gov/marc/mac/recordings. html (accessed Sept. 17, 2025)
21. Library of Congress, MARC Standards, ISSN Review Group, "MARC Proposal no. 2023-06."
22.Ibid., Annual 2023, Meeting #1 (June 28) Recording, at 00:50.
23. ISSN International Centre, ISSN Manual, MARC 21 Version, Edition: December 2024 (compiled Sept. 16, 2025), 7.1, www.issn.org/ wp-content/uploads/2025/05/ISSN_ Manual_pdf_en html (accessed Sept. 17, 2025).
24.The Bulletin signalétique survives in an archived database provided free by Inist-CNRS and combining the content of its online analogs - PASCAL and FRANCIS -from 1972 to 2015: https://pascal francis. inist.fr/cms/?lang=en (accessed Sept. 17, 2025).
Copyright Media Periodicals Division, The Kansas City Gardener, Inc. 2025