Not quite two decades ago, western musicology took a notable step forward with the advent in 2001 of GroveMusic, the online version of The New Grove Dictionary of Music and Musicians, 2nd ed. (NGD2).Footnote ¹ That electronic manifestation of the seventh iteration of English-language musicology's most venerable reference work was not without its problems, however, as Lenore Coral's review of GroveMusic trenchantly observed.Footnote ² Chief among the online version's many glitches were a range of issues related to searching for information within the new, amorphous format of a web site.

Most of those early problems were quickly fixed, and within only a few years, the online version of the NGD2 had become a serviceable research tool for anyone who did not want to travel to a library to consult the 29 hard bound volumes – provided one had login privileges at some reference collection or had purchased an individual subscription.Footnote ³ The growing acceptance of Grove Music Online (GMO), as the online version was renamed after Macmillan – publishers of the seven print editions of the dictionary-encyclopedia going back to George Grove's first four-volume edition (1879–89) – transferred the title to Oxford University Press in 2003, came about in no small part because of OUP's expertise in online publishing. In 2008, OUP made GMO part of its own new Oxford Music Online (OMO) platform, along with electronic versions of the multi-volume Encyclopedia of Popular Music and two shorter reference works, The Oxford Dictionary of Music, 2nd ed., and The Oxford Companion to Music.Footnote ⁴ In the years following, GMO added content from the NGD's print ‘spin-off’ encyclopedias, mostly from The New Grove Dictionary of Opera and The New Grove Dictionary of Jazz, 2nd ed., and later and more selectively from The Grove Dictionary of American Music, 2nd ed., and The Grove Dictionary of Musical Instruments.

Thus, in recent years, users of the OMO could hunt for a single topic in multiple music dictionaries and encyclopedias with only one electronic search. Results were displayed in a single list that identified the original published source of each article. These ranged from single-sentence ‘down and dirty’ entries from the Oxford Dictionary and Companion to more extensive general articles from the NGD2 all the way to more specialized essays from one of the spin-off encyclopedias. The ordering of the search results might not always have made immediate sense, but with the identification of the source publications, users quickly learned how to choose from among the various entries, the most relevant of which usually stood near the top of the results list. Students in need of a simple definition of or basic introduction to a topic could read brief entries from the Oxford Dictionary or Companion, while specialists and scholars in need of greater detail could choose lengthier articles. Opera scholars, in particular, now had electronic access to more extensive and focused articles about the theatrical music of composers who worked in multiple genres (and whose primary GMO articles thus surveyed their outputs more generally). Minor historic figures, who might not have merited inclusion or at least much notice in the NGD2, e.g., librettists, singers, conductors, producers, etc., were also brought into the OMO from The New Grove Dictionary of Opera. Entries on individual operas, which combined histories of a work's genesis with a plot summary and commentary on the work's reception, were especially welcome, since no such items had ever been included in the NGD2. Beginning in March 2018, Americanists could likewise find additions to the OMO from The Grove Dictionary of American Music, 2nd ed., although only smaller batches of articles were added at regular intervals, leaving just less than half of the printed contents still off-line.Footnote ⁵ Yet despite the OMO's attempts to serve as a universal ‘one stop’ reference tool, most users (this author included) doubtless thought of the OMO as a single access point for a group of multiple discrete publications.

In fact, the transformation of OMO from a mere electronic copy of NGD2 into ‘a dictionary that is now primarily web-based and global in scope’ has been quietly underway since about 2009, when a new editorial team, headed by Editor-in-Chief Deane Root, was appointed.Footnote ⁶ Among the most important changes noted by Anna-Lise Santella, Publishing Editor of Grove Music Online, were a new search engine, improved links between articles in the OMO, the embedding of audio and visual examples in articles, and the ability to link to various external resources. With these changes came an editorial reorientation to ‘digital-first publication’, which led in turn to a ‘small changes as we go’ approach, as opposed to less frequent ‘batch changes’. As Root wrote in his ‘Letter from the Editor’ (December 2017):

As the OMO's ‘History of Grove Music’ declares: ‘the whole notion of an edition has changed from a concrete model tied to specific historical publication dates to a continually evolving, organic publication’.Footnote ⁸

Even without the publication of another physical edition, Root identified 2017 as a genuine moment of change, based in ‘a new approach to the public interface of Grove itself’ and, more importantly, in significant changes to the GMO's taxonomy, that is, the fundamental categories of organization for the entire contents of the GMO, which determines how the search engine looks for information in a database that users no longer explore via an alphabetical scheme. As Santella explains, the new approach to categories involves ‘a much more detailed taxonomy that is more global in scope and … . more closely ties the work of developing content with the work of defining and processing data …, making it easier for researchers to find a meaningful array of resources on a given topic’.Footnote ⁹ Despite the importance attached to this new approach to organization and searching, no explanation of this new taxonomy was forthcoming, beyond the general notion that the GMO reflects recent and current trends in music research of all kinds. A glance at the ‘Advanced Search’ page, easily accessible through a single click next to the simple search window found on any GMO page, demonstrates as much.Footnote ¹⁰

Unlike the relatively simple and straightforward ‘advanced search’ options used by many general library catalogs and other databases,Footnote ¹¹ the GMO's advanced search offers the user 12 broad categories for defining searches: a general heading (author, article titles, heading, bibliography, etc.), Format (article or image), Type of article (biographical, family, place, etc.), Place type (city, country, state, etc.), Life event (with specific dates or ranges), Topic (over 60 choices from audio engineer to voice ranges), Era (the traditional music history divisions with approximate date ranges), Region/Country (choices ranging from continents down to specific countries), Occupation (about 20 different music or music-related fields from Choreography to Scholar), Publication date, and Availability (limited by one's OMO subscription). By checking various boxes (which must be clicked and updated one category at a time), one can construct a previously unimaginable search strategy such as: ‘19th Century, Bohemia (historical)/Czechoslovakia (historical), and Organology’ to generate a list of seven articles about instrument makers active in that time and place. No one who uses this advanced search page will be unimpressed with this feature, and testing it can be positively addictive. Among the minor complaints here are the lack of boxes for a few countries, for example, Scotland and Wales, which are presumably subsumed under United Kingdom, or important states or other geographic subdivisions, such as Bavaria. One can easily work around that modest problem, however, by inserting the relevant term in the general heading of the search. Readers of this journal may be annoyed by temporal divisions that include ‘19th c./Romantic (1800–1900)’, which will force scholars of the ‘long nineteenth century’ to add the eras on either side of that misnomer to ensure they find what they want.

Dazzling as the advanced search function might be, most OMO users will have more traditional, specific targets in mind, and thus the one-window search box located on the crossbar near the top of every page will be the typical starting point for most mundane searches. Here there is less to please the average user. First, the search function is unforgiving of mistyped and misspelled words, and there is no auto completion for even correct partial words, e.g., ‘Beethove’ (absent the final letter) yields nothing but suggestions for fixing an unsuccessful search. The unintentional student error ‘Ludwig von Beethoven’ (recte, ‘van’), when searched without quotation marks, yields 306 hits for articles that contain those three random words anywhere within. More curious is the four entries found when that exact (incorrect) text string is searched, using quotation marks.Footnote ¹² Correctly spelled, nearly any logical bibliographic entry – ‘Ludwig van Beethoven’, ‘Beethoven, Ludwig van’, ‘Ludwig Beethoven’, and ‘Beethoven, Ludwig’ – when searched without quotation marks will put the main article by William Drabkin, et al., at the top of the results list. Oddly, searching some of these formulations inside quotation marks may send the Drabkin article to a much lower spot in the list (104th in one listing, on the sixth page of search results). Even searching just ‘Beethoven’ (with or without quotation marks) puts the Drabkin article fourth in a list that begins with ‘Quartetto Beethoven’, ‘Beethoven Association’, and ‘Beethoven Quartet’. To OMO's credit, one need not worry about diacritical marks – a serious problem in the earliest incarnation of GroveMusic – and thus typing ‘Dvorak’ is sufficient to find the entry for ‘Dvořák, Antonín’. Transliterated names present another challenge, with ‘Tchaikovsky’ and the Germanic ‘Tschaikovsky’ finding the entry for the Russian composer at the tops of their respective search results, while ‘Chaikovski’ lands only two irrelevant hits. Woe betide the student (or scholar) who is unsure of how to spell ‘Wasielewski’ or ‘Reményi’, as OMO's search engine offers nothing for a close miss.

With the search results displayed, the next challenge is sorting out the results, a task that used to be guided in large part by the designations of the original publications for all articles. Beginning in December 2017, however, those designations were removed, which raised a modest kerfuffle on the now defunct AMS-L email list. Most of the scholarly concerns were with the inability to identify easily the primary entry on a topic, usually that of a composer, especially when obvious candidate articles did not display at the top of a results list. Experienced scholars might recognize the provenance of an article by its ‘published in print’ and ‘published online’ dates, which replaced the titles of the source publications previously found at the top of each article, but students and other novice users, who generally lack this knowledge, can easily be led astray by OMO.

For example, searching just ‘Weber’ (as a student might do) for the father of German Romantic opera, yields a long random list of articles about individuals with that surname. ‘Weber, Carl Maria (Friedrich Ernst) von’ by Clive Brown appears as the twenty-seventh item (seventh on the second page of results).Footnote ¹³ Astute individuals might think to check the third item on the first page, ‘Weber family’ by Paul Corneilson, et al., even though there is no indication that C.M. von Weber is included in that entry.Footnote ¹⁴ Alternately, searching ‘Carl Maria von Weber’ yields a more helpful list, with Clive Brown's article now first in the list of results, and Michael C. Tusa's essay on the same composer, now drawn out of the ‘Weber family’ article and identified as ‘(9) Carl Maria (Friedrich Ernst) von Weber / In Weber family’, in the second slot. One might assume that pride of place marks Brown's article as the primary one on Weber, but in fact, that piece was written in 1992 for The New Grove Dictionary of Opera. While its nearly 4,000 words offer a good overview of Weber's life and work in the theater, it pales in comparison to Tusa's 23,000 words covering all aspects of Weber's life and music. A clarinetist interested in Weber's contributions to his instrument might wonder why Brown makes no mention of Weber's two concerti and other works, which Tusa does discuss. Adding some additional confusion to the mix, the ‘Weber family’ article is marked as ‘revised and updated on 1 July 2014’, which might suggest that Tusa's essay is fairly new. In fact, Tusa's essay is identical with the ‘previous version’ of 2001, which is also available via a link above the article, suggesting that only other portions of the ‘Weber family’ entry have been updated.Footnote ¹⁵

The entries for Saverio Mercadante, a significant secondary figure in the history of nineteenth-century Italian opera and also a composer of instrumental music, provide another example of problems in the OMO's search methodology. Searching for ‘Mercadante’ or other formulations of his name finds the article ‘Mercadante, (Giuseppe) Saverio (Raffaele)’ by Michael Whittmann, usually at the top of a relatively short list of hits.Footnote ¹⁶ Written in 2001 for the NDG2 and running to over 7,400 words, the article is an excellent overview of its topic. The List of Works includes all genres, although its information on the operas is rather barebones. Still, Whittmann's article seems sufficient, unless one notices the box at the very bottom of the page labeled ‘More on this topic’ with a link to ‘Mercadante, (Giuseppe) Saverio (opera)’ by Michael Rose. That ‘hidden’ entry – it cannot be found through any normal search and even the advanced search cannot find it – was written in 1992 for The New Grove Dictionary of Opera.Footnote ¹⁷ In fact, the article's roughly 2,900 words have been very lightly edited from Rose's essay for the 1980 NGD and thus might seem to offer less than Whittmann's essay. Nevertheless, Rose's entry has a much more thorough and easier to use list of Mercadante's operas (but no other works). Moreover, a ‘see also’ at the end of Rose's text (immediately above the works list) links to articles about five of Mercadante's more important operas, and in turn, each of those five articles includes a link to an additional article about the opera's librettist. While links to the same five operas appear in Whittmann's essay (but now after the works list and bibliography), along with links to two other potentially useful articles, many readers might miss these items due to their placement at the very bottom of the page.

Without belaboring the point, clearly there are problems with OMO's taxonomy and search engine, and what is most frustrating is the effort one must expend in some cases to locate basic information that is generally among the best and most reliable in the musicological world. At this point, the OMO has fallen short of its avowed goal to ‘mak[e] it easier for researchers to find a meaningful array of resources on a given topic and also to quickly identify the best resources for the task at hand’.Footnote ¹⁸ For now, professionals will have to bring their A-game research skills to bear when trying to extract much of what lies buried in the lower reaches of the OMO's taxonomy, but how can students and novices be expected to navigate these same uncharted waters?

Among the resources in the OMO designed for less experienced users, are two single-volume reference works, The Oxford Dictionary of Music, 2nd ed., and The Oxford Companion to Music.Footnote ¹⁹ Either can be opened as a separate item via a link under ‘What Would You Like to Do?’ on the ‘About’ page, followed by another link (‘Browse all content from this title’) on the new page. Both of these dictionaries display as alphabetical lists that can be searched by letter or even a single word, and the relevant article opened with a click. For example, the Oxford Dictionary's unsigned entry on Verdi runs to 650 words, and is followed by a list of his operas, each of which is a clickable link to a scant sentence describing the opera's basic information. The Companion's entry on Verdi, signed by Roger Parker, is a more substantial 2,700 words and includes a brief bibliography of six items. There are no additional links in the entry, but most of Verdi's operas have separate entries similar to those of the Dictionary elsewhere in the Companion. All of this seems more than sufficient for students and novice users.

Other items of interest for less experienced researchers can be found under ‘Tools and Resources’ (via the link at the very top of all pages).Footnote ²⁰ These include practical tools like the ‘Grove Music Online User's Manual’, and information on various abbreviations, both of which should be read (or at least skimmed) by any serious user. Still other content-directed resources may be found via a link on the ‘Tools and Resources’ page under ‘Subject Guides and Research Resources’.Footnote ²¹ These include a set of three Opera Indexes (title, role, and composer), Research Resources, Timelines (for 100 Important Opera Premieres, Contemporary Music [beginning in 1880], and Women in Music), and a series of Topical Guides including ones for ‘Romanticism’ and for ‘Women in Music’ (sortable by either name or date).Footnote ²² Again, readers of this journal will probably be disappointed by a collection of articles that lumps the long nineteenth century under the rubric of ‘Romanticism’, while ignoring composers like Rossini and Richard Strauss (in a list that runs from Beethoven to Mahler), and performers like Thalberg and Patti (with only Paganini and Joachim in the list). Again, these are minor issues. The ‘Research Resources’, which are apparently free to any user, registered or not, allow access to 13 major OMO articles on ‘Congress Reports’, ‘Dictionaries and Encyclopedias’, ‘Editions’, ‘Periodicals’, ‘Sources, Manuscript’, ‘Sources of Instrumental Ensemble Music to 1630’, ‘Sources of Keyboard Music to 1660’, ‘Sources of Lute Music’, ‘Collections [of music, letters, and other primary sources], Private’, ‘Festival’, ‘Instrument, Collections of’, ‘Libraries’, and ‘Sound Archives’, most of which should become required reading in graduate music bibliography and research courses.

At its center, the OMO remains a remarkable object, with over 33 million words encompassing 60,000 articles by 8,900 authors. For every complaint one might raise, there is plenty elsewhere for which to be grateful, starting with the extensive works lists (usually indexing the major sets of collected works) for almost every significant composer. One might wish for more frequent updates of articles and especially the bibliographies, but much of the current editorial work is dedicated to expanding the content. With many articles still holding up well in their third decade, updates and revisions can take a backseat to the ‘Southeastern and East Central Europe update project’ and similar content additions, at least for a few years.Footnote ²³

At present, the primary question seems to be, how does one navigate such an expansive database? The sheer volume of information in the OMO can make online music research something akin to spotting icebergs, with the first easily found items masking an even greater abundance of worthy scholarship lurking below the surface. While OUP's software engineers work to improve the search engine and the OMO editors wrestle with the database's taxonomy, OMO users will need both to embrace the opportunities and to be aware of the challenges this new electronic world presents.