RIPM (Le Répertoire International de la Presse Musicale) is widely regarded as the most comprehensive resource offering electronic access to music periodicals from the early Romantic era to the twentieth century. Founded in 1980 by H. Robert Cohen, RIPM is the youngest of the so-called ‘4 R's of International Music Research’; its partner initiatives include RISM (Répertoire International des Sources Musicales), RILM (Répertoire International de Littérature Musicale), and RIdIM (Répertoire International d'Iconographie Musicale).Footnote 1 Cohen's ambitious project was notably visionary in its use of technology: not only did it use computing from the start (beginning with DOS-based indexing systems), but it was also the first of the 4 R's to explore full text searching.Footnote 2 RIPM seeks to address ‘two main problems that have prevented these [historic music] journals from being systematically collected and examined: (1) the limited number of libraries possessing the journals, and (2) the difficulty encountered when one attempts to locate specific information within an available source’.Footnote 3 The project has thus focused on collection building, curation, indexing and accessibility. This international cooperative's accomplishments are impressive: as of July 2020, the database contains 527 music periodicals, 430 available in full text complete runs, totalling 996,000 annotated records and 1.47 million full-text pages of music periodicals.Footnote 4
RIPM's Databases and Interfaces
At present, RIPM consists of two series, each containing two electronic publications. First, the Annotated Series contains the Retrospective Index to Music Periodicals (1760–1966) and the Retrospective Index to Music Periodicals with Full Text (1760–1966). Second, the Preservation Series is projected to contain multiple subject-specific databases in full text. The first two subject-specific collections in the Preservation Series is the European & North American Music Periodicals and, most recently, the RIPM Jazz Periodicals.Footnote 5 Each of these e-publications have their own distinctive purposes and functionality. The RIPM Retrospective Index relies on metadata painstakingly created by scholars; as the website explains, this enables scholars, for example, to search for terms such as harmony even when the term does not appear in the primary source text.Footnote 6 The Retrospective Index with Full Text makes available the full-text of the journals treated in the Index. The RIPM Preservation Series relies on OCR (Optical Character Recognition) to render music journals available which are not part of the Retrospective Index with Full Text.
Moreover, RIPM is available via two distinctive platforms: ripmplus and EBSCOHost, the latter offering a different interface to navigate the same data (theoretically) in RIPM. It appears that the data via the ripmplus platform is more up-to-date than that offered via EBSCOHost; for instance, a search for ‘Der Freischütz’ as keyword in the Retrospective Index with Full Text revealed 8,065 results on ripmplus but only 6,132 results via EBSCO on the same day.Footnote 7 Users and institutional subscribers may wish to keep in mind that the data is not updated asynchronously/in real time from ripmplus to EBSCOHost. RIPM, however, is not the only database affected; RISM, for instance, suffers the same lag on its EBSCO interface. The more pronounced differences between these two platforms lie in functionality. The EBSCO platform has the benefit of feeling more familiar to musicologists who use RILM, boasts an interface available in 29 languages, has the capability of sharing citations through 200+ social media sharing services and offers mobile apps, among other features.Footnote 8 The RIPMplus interface, by contrast, offers a translation module which translates titles and annotations into 52 languages, cross-language search expanders, a drop-down menu for searching individual titles and a virtual keyboard for non-Latin character languages. In short, the RIPMplus interface seems deliberately designed to promote multi-lingual and international use.
Situating RIPM in the Landscape of Digital Periodicals
In the bustling field of digital scholarly resources, databases offering access to periodicals dealing with music have increased sharply in recent decades.Footnote 9 Three major categories exist: first, databases such as RIPM, which focus solely on historic music periodicals – specialist serial publications devoted entirely to music criticism. Second, periodical databases holding titles with a broader coverage of literature, political issues, theatre, and so forth, which nevertheless also include regular reviews of music. Examples here include the Austrian National Library's Open Access resource, ANNO (AustriaN Newspapers Online), offering over 22 million pages of historic newspapers and periodicals from 1689 to 1947.Footnote 10 Finally, musicologists focused on reception studies often use large corpus databases which include periodicals amongst other historic documents: examples include Hathi Trust, an Open Access transnational large corpus database which includes musical documents, including criticism, and Gallica, the Bibliothèque Nationale's Open Access resource for French-language documents.Footnote 11
RIPM's strengths and weaknesses are most effectively assessed within this increasingly complex and saturated marketplace for digital periodical resources. The most obvious strength is RIPM's transnational approach – it aims to bring music periodicals from many different countries together ‘under one roof’. RIPM's decision to focus on specialist music periodicals has allowed it to seek out more obscure publications, journals that appeared only for a short time or with a hyper-local circulation. Its addition of jazz periodicals has also made it a useful tool in diversifying musicological scholarship. If it does eventually add historic journals of Asia and Africa, while continuing to grow its South American content (a few titles are already in place), it has the potential of substantially enhancing studies of global exchange during the nineteenth century.
At the same time, RIPM has limitations in terms of content, and this becomes most evident when searches are conducted. For instance, a search for ‘Fidelio’ in all three databases (RIPM Full Text, RIPM Retrospective Index and RIPM Preservation Series) from 1806–1900 yields 7,124 results on the EBSCO interface. Of these results, 2,877 are German-language sources, 1,572 English, 1,493 French, 483 Italian, 205 Czech and 132 Dutch/Flemish.Footnote 12 By contrast, a search for ‘Fidelio’ from 1806 to 1900 in ANNO yields 17,758 results, many of which are in non-specialist periodicals such as the Wiener Zeitung (one of the central organs for music criticism in nineteenth-century Austria), the Neue Freie Presse or Kladderatsch: Humoristisch-satyrisches Wochenblatt from Berlin, among many others.Footnote 13 This kind of discrepancy in content – that an Open Access resource focused on Austria produces nearly twice as many results for the nineteenth-century reception of a canonic musical work than RIPM across an international search for the same time-frame––has serious implications for scholars and for RIPM.
Most notably, the degree to which music was reviewed in non-specialist periodicals should not be underestimated. To drive the point home, a search for ‘Fidelio’ in journals and newspapers between 1806 and 1900 in HathiTrust yielded 12,398 results, and HathiTrust is not specifically invested in historic periodicals.Footnote 14 It is well worth noting that databases such as ANNO, HathiTrust and Gallica all contain substantial music periodical holdings – as such, there is overlapping content with RIPM that will only increase over time. In an academic climate that celebrates culturally situated studies of music and interdisciplinary scholarship, scholars are increasingly drawn to these non-specialist resources for the sheer scope of their content alone, not to mention that they are Open Access and often boast a vastly superior user experience. Even though some of these non-specialist resources are language specific – Gallica is the best-known example among musicologists working on French materials, for instance – the reality is that most scholars specialize in reception in a specific linguistic area.Footnote 15 For RIPM, the take-away is likely the need to ensure value for this commercial product, and given the changing landscape of online periodicals, to re-evaluate its market differentiation. The most obvious differentiator, at least from my perspective, is the potential for RIPM to assist scholars in large-scale transnational research, offer a superior user experience to their large transnational specialist corpus, and perhaps to continue to offer access to highly local limited-run, obscure music journals as a ‘value-add’.
Moreover, RIPM's stated central goal of curating and making accessible ‘rare’ music periodicals raises an important question in the digital age: what is considered ‘rare’ in 2020? To be sure, what was difficult to access during the 1980s when RIPM was founded might be much more accessible today. Governments around the world have devoted decades’ worth of funding to digitizing their national collections in order to encourage scholars from around the world to study their holdings, thereby promoting their cultural heritage. A quick comparison of the French holdings in RIPM Full text and Gallica suggest that Gallica currently encompasses roughly 63 per cent of the journal holdings available in RIPM; 100 per cent of the Spanish (from Spain) holdings in RIPM Full Text can be found in Hermeroteca Digital;Footnote 16 88 per cent of the Italian holdings in RIPM Full Text can be accessed on Italy's Internet Culturale;Footnote 17 76 per cent of the German titles in RIPM Full Text can be accessed via the Zeitschriften Datenbank at the Bavarian State Library.Footnote 18 Where the adjective ‘rare’ is well deserved are in the Czech, Danish, Dutch, Belgian, Greek, Hungarian, Norwegian, Polish, Portuguese, Russian, Mexican and Swedish sources; these journals, however, are far fewer in number compared to the French, English, German, Italian and Spanish sources. It is well worth noting that the overlap seems to decrease in the RIPM Preservation Series. For now, perhaps, RIPM can still boast offering access to ‘rare’ music periodicals. In the longer term, however, it seems prudent to consider other strategies for market differentiation, particularly in view of the significant aforementioned drawback of not including music reviews from non-specialist journals.
UX (User Experience)
In software development, UX (User Experience) refers quite simply to the following: how well does the product assist the user in achieving their goals? Clarifying ‘who’ the user is, is a critical first step, and users should not be confused with buyers. In the case of RIPM, users likely include scholars, students and teachers with their specific research questions, whereas the buyer likely refers to institutional libraries, who may have different priorities (acquiring the maximum number of volumes, for instance). UX includes everything from onboarding the user, to providing a UI (User Interface) that is intuitive and pleasurable to use, and offering support to ensure that users are successful in achieving their research goals. As a means of evaluating the RIPM UX, I shall refer to some case studies – research goals taken from or at least inspired by scholarship on music criticism featured in this issue – covering each of the RIPMplus and EBSCO interfaces.
Our first case study is inspired by Lesley A. Wright's article on Berlioz's and Reyer's reviews of Bizet in the Journal des débats.Footnote 19 This is a non-specialist journal not available via RIPM. However, scholars have shown considerable interest in composers who also served as critics, and a study on Berlioz's criticism more broadly lies well within RIPM's target use case. Our imaginary scholar seeking to study Berlioz's music criticism (within the RIPM corpus) thus proceeds to the RIPMplus interface and notices that the simple search only offers ‘keyword’ and ‘language’ search fields. Clicking on ‘Advanced Search’, they then proceed to type ‘Berlioz’ in the ‘keywords’ search bar as a first step of exploring a broad range of search results and getting a feel for the tool. While the ‘Advanced Search’ offers a fuller range of search fields: ‘keyword(s)’, ‘languages’, ‘publication place’ (with a drop-down menu), ‘periodical’ (with a drop-down menu), and ‘Year from … to’, our researcher is faced with an interesting conundrum: there appears to be no way to differentiate between Berlioz the composer and Berlioz the critic on this interface. Our researcher is forced to conclude that generating a list of results containing the music criticism of Hector Berlioz is simply impossible via RIPMplus interface.
If they have their wits about them, they might proceed to the EBSCO-interface, which boasts a more nuanced search functionality.Footnote 20 Typing ‘Berlioz’ into the author field of the ‘Advanced Search’ with a date limit of 1828 to 1910 yields 2,002 results. The faceted search, especially with respect to language, is extremely useful to the researcher as it breaks down the number of sources by language. Thrilled by these results, our researcher becomes fascinated by the broad dissemination of Berlioz's music criticism – his reviews were translated into at least 12 languages. A range of follow-up research questions could follow: which of Berlioz's reviews were widely disseminated? Were particular works or reviews from a specific journal (perhaps La Revue musicale which occupies the first two results) of interest beyond France? Might it be possible to discern geographic patterns of dissemination of Berlioz's criticism? Clicking through some of the results reveals that details of the ‘original’ periodical review are rarely included in the metadata, which certainly complicates the investigation. In this instance, the researcher will need to manually cross-reference Berlioz's reviews with the translations that appeared outside of France, and the original sources appear to be unknown.
Most scholars at this stage would wish to download the data into some usable format for further analysis and to flesh it out. Downloading the results into a spreadsheet, perhaps excel or airtable, seems a logical first step. Here, there is another roadblock: it is impossible to select all the results for download; 50 is the maximum number for an export via the EBSCO interface. Assuming that our imaginary researcher is willing to tediously click through 2,002 results to export them 50 results at a time into a format that could be useful – for an excel spreadsheet, likely csv (comma separated value) format, it is worth examining precisely what fields one is able to export. Having placed some items in a folder, the researcher can now select ‘export’, and choose their desired .csv format (other popular options include Refworks and Endnote). The following metadata is available for export: article title, author, journal title, publication date, volume, issue, first page, page count, accession number, document type, and permanent link. This should serve our researcher reasonably well in starting their spreadsheet of Berlioz's music criticism and its dissemination. Having the place of publication available as well as more cross-referencing of translations of reviews – a widespread yet vastly understudied practice in nineteenth-century music criticism – would greatly facilitate this research. Incidentally, a search for ‘Berlioz, Journal des débats’ yields a wonderful list of 265 translated reviews illustrating the potential power of this kind of metadata.Footnote 21 Having fleshed out their spreadsheet with some additional data, our imaginary researcher is now in a position to analyse the metadata more meaningfully, to see if there are clusters of sources of particular interest.
At this point, our imaginary researcher embarks on ‘close reading’ of source material, which, is currently the most widely used technique in studies of music criticism (four of five contributors to Nineteenth-Century Music Review's issue ‘Critical Responses to Nineteenth-Century French Music’ employ close reading of music reviews at some point in their article). RIPM should be a good fit for ‘close readings’, as one is able to click through to the PDF of the review in question via the list of search results. At the same time, it is most useful for studies which require only a very small selection of sources; Ralph Locke's examination of reliability of opera criticism related to Félicien David's Le Désert and Herculanum is one example of this kind of project (though, Locke, like all other contributors in the issue, draws heavily on music reviews in non-music journals, the content of which is beyond the scope of RIPM).Footnote 22 Content aside, Locke is dealing with a small number of reviews and within a narrowly defined scope (two operas), which is the kind of research that RIPM seems to be designed for. That said, while it is possible to download a PDF of a single article, there are severe copyright limitations on downloading materials, including that researchers ‘agree to not download a complete issue of a journal, multiple copies of any article(s) or a substantial portion of a journal’.Footnote 23 This is a barrier and stands in stark contrast to major periodical sites such as Gallica and ANNO who maintain much more generous user terms and have gone to great lengths to provide scholars with .pdf or .txt download options to facilitate scholarship of their digital collections. For scholars with investigations as broad as that of Christopher Moore, who studied constructions of Fauré in France from roughly 1920 to 1950, there are many more sources involved.Footnote 24 Indeed, a RIPM search via the EBSCO interface for ‘Fauré’ in the ‘All Text’ search field, limited to publications in France within the aforementioned date range yields 3,724 results. Limiting the results with an additional search term ‘classique’ still yields 1,631 results.Footnote 25 Moreover, scholars who work music criticism would not necessarily forgo looking at the first set of results as reviewers might describe a composer who is considered ‘classic’ or canonic without using the term ‘classique’. It is at this stage of research that institutions such as HathiTrust hold a real advantage, as scholars are able to create their own collection, carrying out various kinds of fine-grained searches within their collection. HathiTrust even provides an API (application program interface) and dedicated data analysis supports for those interested in text mining.Footnote 26
Evaluating RIPM's UX, therefore, is intimately connected to considering ‘who’ the ideal user is. It is at least curious that of the five contributors who independently generated an article on French music criticism for this special issue, it would seem that none made use of RIPM. Part of the issue lies with limitations in content – all scholars involved, except me,Footnote 27 were engaged in cultural issues related to French music criticism and therefore necessarily engaged extensively with reviews in non-music periodicals. But I believe that the UX in both platforms in RIPM plays a significant role as well. More specifically, one issue that seems to be at play represents a common pitfall in software development: that the functionality facilitates activities that our user does not need, and much more crucially, that are fundamentally disconnected from their overall goals. Put another way: it is not enough to build (or in the case of EBSCO, overlay) generic search functions around a corpus and assume that this will enable users to meet their research goals. On the contrary, extensive user research needs to be conducted in order to gain a nuanced understanding of the broad goals for scholars and students of music criticism and their inextricably connected research activities.
In its current state, there are cases that RIPM would serve well. For example, a scholar or student sees a short excerpt of a review quoted in a book or article, with the full citation easily available, and wishes to read the full review. This is cognitively equivalent to using a card catalogue to locate a specific item in a collection. Or a scholar is interested in highly technical musical issues (i.e. the diminished seventh chord) that would only be discussed in a music-specific journal and wishes to search the corpus for mentions of this. But there exists a large category of scholars – those conducting culturally based close readings of issues in music criticism and a growing minority who, like me, are interested in newer digital approaches such as text mining, for whom RIPM does not appear to (and in the case of text mining cannot) serve as the ‘go to’ tool at present. And yet, I believe that it should be our ‘go to’ tool. There is tremendous unrealized opportunity to capture these scholars of music criticism as more engaged users and to fully unleash the potential of RIPM.
RIPM 2.0?
Given the already high and ever-increasing percentage of specialist music journals available on Open Access platforms, combined with the fact that non-specialist journals also contain rich and plentiful holdings of music criticism, one might sense an opportunity to re-envision a tool such as RIPM. One obvious (and technologically possible) solution for a ‘RIPM 2.0’ is to collaborate with all the Open Access platforms, many of which are academic or national libraries, to leverage their (comparatively abundant) resources for ongoing digitization, and to create a new interface, tailor-made for musicologists, capable of providing powerful searching across these sources. As I mentioned earlier, RIPM's strongest asset is the potential for transnational research and concomitant contributions to global music history. However, extensive user research and user testing is required to build an interface that would assist scholars in answering research questions that occupy them today. This new interface could also be designed from the outset to facilitate future areas of digital research: big-data research including text-mining, network visualization and the like. Cohen was without doubt a technological visionary in the 1980s; re-positioning RIPM at the forefront of digital scholarship would further honour his legacy of technological innovation in studies of music criticism.
