September 9, 2002

85 Years of Recorded Jazz (CD-ROM).  By Walter Bruyninckx and Domi Truffandier.  Lange Nieuwstraat 121, B2800 Mechelen, Belgium: Self-published, 2000-2002. $300 for complete alphabet.  Phone: 32-(0)15-41-81-63.  E-mail: lucien.bruyninckx@pandora.be (purchasers in the U.S. will be given the name of a local agent to whom payment can be sent, to avoid the cost of international money transfer).

The Jazz Discography (CD-ROM version 3.3).  By Tom Lord.  1540 Taylor Way, West Vancouver, BC, Canada V7S 1N4: Lord Music Reference, Inc., 2002. $297 ($277 when purchased online through publisher’s website).  Phone: 604-926-9953.  Website:  www.lordisco.com.

A Comparative Review by Tim Brooks

The future of published discography may well lie in those little silver discs called “CD-ROMs.”  Here at last is a way to store and disseminate vast amounts of information at relatively low cost, complete with versatile indexes and even, potentially, pictures and sound.  Pages or whole sections can be printed out as desired, and updates can be issued economically.  Now that CD-ROM drives are being built in to almost all new personal computers they can be used by anyone with access to a PC.

This is a comparative review of the first two major discographies to be issued (at least to my knowledge) on CD-ROM.  Coincidentally they cover almost identical ground–all jazz recordings issued worldwide, from the 1890s to date.  However their approach to CD-ROM publication is quite different, which makes an interesting comparison.  I will first compare loading and navigation (i.e., how you get to the data), and then discuss the quality of the listings in each.

First, a little background on these discographies.  Belgian discographer Walter Bruyninckx first published his magnum opus in the 1960s as 50 Years of Recorded Jazz, on 8,000 looseleaf pages.  This was later updated to cover 60 years, then 75 years.  (The starting year for these counts seems to be 1917, even though the discography includes earlier ragtime and related recordings.)  Facing a new 85-year edition of more than 20,000 A4 pages, he decided it was time to go electronic.  The new edition was announced in 1999, and is being released in four stages, as he works his way through the alphabet.  Each new CD-ROM contains everything on the previous disc, plus additional letters.  So far three discs have been issued, the latest in early 2002 covering 15 letters complete (A-C, F-K, M-R) and two partial (D,S).  When the project is finished in early 2003, his original subscribers will have paid $300 (four discs times $75 each); the cost of the final disc alone to a new subscriber will also be $300.  A “sample” disc is available for $18.

Canadian Tom Lord began publishing his Jazz Discography in hard cover in 1992.  Like Bruyninckx he worked his way through the alphabet, with the 26^th and final volume issued early in 2002.  The cost of all 26 printed volumes is about $1,600, which is prohibitive for most collectors, so he is now offering a CD-ROM with the full contents for under $300.  (Earlier, a partial CD-ROM version was made available to buyers of the books.)  The CD reviewed here is version 3.3; version 4.4, with updates to earlier letters (some of which have not been updated since 1992) is expected in 2003.

The PC I used for this review is a rather old one (c.1998), with a 400 mhz Pentium II processor, 64 mb of SDRAM, and an 8.4 gb hard drive, running Windows 98.  I deliberately wanted to see how these CD-ROM’s performed on a computer that was not the latest high-powered model.  Performance on still older computers might be less, however.  Both The Jazz Discography (henceforth JD) and 85 Years (85YORJ) claim to work on a Mac, but I did not test this.  Remember, you do have to have a CD-ROM drive.

Loading and Navigation

JD requires that you load a minimum of 15 mb of programs on to your hard drive, creating scores of files and roughly a dozen subdirectories to hold them.  This was accomplished in a couple of minutes, with ease.  Once the programs (and data, if desired) are on your hard drive, speed of access is quite fast.  You have to keep the original CD in the CD-ROM drive while using JD, even if the data has been copied to the hard drive.  A small flyer comes with the CD, but more extensive instructions and help screens are available on screen once you start the program.

Both JD and 85YORJ are organized by recording session, laid out in the standard “Rust” format, with matrix numbers at the left, song titles in the middle, and issues at the right, under a standardized header which contains artist, sidemen and recording date.  JD‘s screen colors are attractive and the layout clean, aiding navigation.

JD works by asking the user to specify a leader, session, musician/sideman, or tune title, and then compiling all occurrences of that request into a single on-screen document.  Ask for “Louis Armstrong” as a leader and you get a full discography of his various bands.  Ask for him as a musician and you get all sessions in which he appeared with someone else.  A request for a song title produces all sessions in which that song appears, arrayed in either chronological or alphabetical order as desired.  (I tried “St. Louis Blues”; according to the Index it appears 1,456 times!)  Sessions may also be called up directly, either by date (within a leader/musician), or by session number–a unique number that JD assigns to each session in the discography.  There is also a Multi-Search button, which allows you to look for combinations, for example W.C. Handy playing “The St. Louis Blues.”

Each of these four access points has an accompanying index.  In the case of leader, musician or tune you can browse an alphabetical list of all leaders, musicians or tunes and click on the one you want; in the case of sessions, it’s a chronological list.  At any point you can print out the screen, or up to fifty sessions in your current “document.”  (If that’s not enough, go back and print 50 more!)  The program also allows you to shrink the font size, fitting more on the screen and on the printed page.  All of this is fast and intuitive, and easier to use than it is to explain here in words.  The limitations are mostly those of the underlying data, which I’ll get to shortly.  Overall, this is a highly sophisticated and easy-to-use access system.

85YORJ works on an entirely different principle. Whereas JD is a database, 85YORJ is essentially a long text document residing in a series of Adobe Acrobat files.  To read them you need Adobe Acrobat Reader, a free program which is provided on the CD-ROM and is also available on the Internet.  Once Acrobat has been loaded on your computer–again, an easy task–the discographical files can be read either from the CD, or from the hard drive if you’ve copied them there.  The principal difference from JD is that you are not calling up selected sets of information, but rather scrolling through a book.  All of the Louis Armstrong band recordings will be found in one place, but Louis as a sideman will be found in many different places, listed under various leaders.  Song titles will also be scattered, as they would be in a book.

To keep the size of the Acrobat files manageable each letter is given its own separate file.  You click on the letter you want to scroll.  There is a unified index across letters, which allows you to search for combinations (“strings”) of letters in any file.  Type in “Louis Armstrong” and the index will take you to every place where that name appears.  Type in “St. Louis Blues” and you will find each time the song appears.

This search engine is quite versatile–you can type in not only names and tunes, but anything, a city, “Christmas,” “August 25” or whatever–however it is much more laborious to use than the JD approach.  Bruyninckx himself gives the example of Milt Hinton, perhaps the most-recorded bass player in jazz history, who participated in hundreds of sessions.  To compile his discography the user would have to find each appearance of his name and print out that page, ending with a very large stack of paper (each page would contain much other information).  Alternatively you could cut and paste the Hinton-related sessions found on each page into a separate file, if you know how to do that, and then print that file.  Plan on a couple of hours either way.  This is not nearly as convenient as the JD approach, which compiles the desired discography for you neatly in one place in a matter of seconds.

The black & white 85YORJ screen is quite adequate, if not as pretty as that of JD.  The author does plan to upgrade the appearance on the final CD, adding color; a further enhanced version will follow in 2004.  Note also that an upgraded version of Acrobat is available for a fee that allows you to open these files and add or delete data, thus creating your own personal discography.

Scrolling through the pages is fast, and you can jump quickly to a specified page or to the end of a letter.  (With both discographies I recommend closing other programs that may be open on your computer to maximize speed.)  In some cases, however, the lack of “running heads” makes it difficult to know where you are.  If you are an experienced Acrobat user I suppose you can cut some corners, but for the casual user the 85YORC access system, while serviceable, is less convenient that of JD.

Content

All this technology is of little use if the discographies don’t contain the information you need.  How do JD and 85YORJ compare on content?  In a broad sense, both cover the same ground.  They are primarily giant compilations of previously published research, rather than original scholarship.  Some new information has been added, particularly in the area of identifying LP and CD reissues, but even the authors would admit that their main goal has been to collect in one place what has gone before.  This is no mean task.  Original research on individual artists and labels has appeared in hundreds of widely scattered books and articles (just check any installment of “Current Bibliography” in this Journal!).  No collector or library has them all.  Bringing all this information together in one place is a valuable service indeed.  Most critical is that the compiler filter the information for accuracy, to make sure that mistakes are not propagated, and source it so that users can judge its reliability.

JD has been criticized in some quarters for uncritical verbatim copying from others,  particularly from earlier versions of Bruyninckx.  There does seem to be some truth to this, as not only title conventions but many notes exactly match Bruyninckx’s wording.  A few verbatims don’t even make sense when transplanted into JD, referring to volumes or entries that don’t exist there.  While JD may have copied from Bruyninckx, Bruyninckx himself did plenty of copying, from others.  The matter of who copied what from whom has been covered in some detail in the Dutch magazine Names & Numbers in issues nos. 9 through 20 (1999-2002), which included a running column by Han Enderman entitled “The Discographical Truth.”  It probably doesn’t matter much here, except to rightfully annoy those who aren’t given proper credit for their prior work and to explain why the contents of the two discographies are so similar.  JD (unlike 85YORJ) includes a lengthy bibliography, but it does not mention Walter Bruyninckx!

Despite the overall similarity, a close comparison of individual entries in the two discographies reveals some systematic differences.  Basically, there appears to be somewhat more information in 85YORJ–more issues, more notations, more artists.  I compared the listings for two artists on whom I have done extensive work, Eubie Blake and W.C. Handy.  For starters, 85YORJ begins major entries like these with a short bio of the artist, including birth and death dates.  JD has no bios.  Both discographies include Eubie’s piano rolls (later released on LP), but JD lists only the titles and (for most) a general date span, while 85YORJ gives the original roll number and release date for each title.  It also lists a c.1930 film short released on LP, and two 1960s-70s LP’s, none of which are in JD.   Each discography contains a few modern LP/CD reissues not in the other, but both omit many early 78 rpm issues (for example an original Emerson issue may be shown, but not the reissues on Regal, Symphonola, Puritone, etc.).  This may or may not be intentional, but it devalues both discographies for those who want to know on what 78 rpm labels a given recording can be found.  Such information is important to collectors, and is readily found in Brian Rust’s Jazz Records and other discographies.

The comparison of Handy discographies produced similar findings.  85YORJ includes a take number, performance detail on the 1938 Library of Congress recordings (which JD mistakenly credits to “Capt. John Handy”!), and two 1950s LPs that are not in JD.  Both discographies miss several other Handy LPs that I have identified, as well as numerous early 78rpm reissues.  This latter point is especially annoying in the case of Handy’s 1922 Paramounts. Those masters appeared on at least ten other labels at the time, but 85YORJ and JD each list only the Paramount number and one Banner release, nothing else.

In order to compare the number of artists covered, I looked at the first 50 artists listed in JD under the letter “R” (the last full letter covered by both discographies).  All 50 names were found in 85YORJ, plus 15 more.  A few of the 15 additional names were blues or gospel artists, which JD does not cover, but most were legitimate, post-World War II jazz-related artists.  In all, JD claims to include 24,000 leaders, 136,000 sessions and 238,000 tunes.  85YORJ gives no totals, but it must have even more.

There are a few other idiosyncracies.  Neither discography is very good at listing the contents of medleys, so for example you won’t find any early recording of Eubie Blake playing his famous song, “I’m Just Wild About Harry.” (In fact there was a lengthy interpolation of the tune in his 1921 Victor recording of “Bandana Days,” but that is not indicated here).  Also, deviant spellings and missing or unreliable dates can wreak havoc with JD‘s elegant indexing system.  You won’t find Gershwin’s “‘S Wonderful” at all unless you know exactly how JD spells it (“‘S-space-Wonderful,” not “‘Swonderful”) or “Star Dust” unless you know it’s now “Stardust.”  “Green Dolphin Street”?  Nope, it’s “On Green Dolphin Street.”  More cross-references or a more forgiving system would have helped.  Sessions by an artist in a given year won’t be found if they are not individually and exactly dated, which is sometimes the case.

Both discographies, being copied somewhat uncritically from other sources, perpetuate numerous errors.  For example both omit the leader on the historic 1910s Ciro’s Club Coon Orchestra recordings, Dan Kildare.  Both perpetuate the misdating of Gus Haenschen’s personal recording of “Maple Leaf Rag” on Columbia 61070 as 1910; the correct date is easily ascertainable since that numerical series was not begun until 1917.  (To add insult to injury, both also mistype Haenschen’s banjo orchestra as “banko” orchestra!)  And 85YORJ dutifully includes as jazz or blues a mysterious recording of “Jim Jackson’s Affinity” on a Document CD; this is actually a Len Spencer-Ada Jones vaudeville sketch, which just happened to use the same name as that of the famous bluesman.  Has anyone listened to this record?

Finally, to see where these two discographies stood in terms of current jazz scholarship, I tested them against five examples.  These are all instances where information has come to light, and been published, that seriously questions certain listings.  The compiler does not have to accept the new information, of course, but in my opinion it is his obligation to point out that there are questions, and not to uncritically print old, and quite possibly wrong, information with no warning at all.

Issue #1: Blake’s Jazzone Orchestra, “The Jazz Dance” (1917): this single Pathe recording was originally attributed to Eubie Blake, but Eubie said he didn’t remember making it and research as early as the 1960s indicated that it might be someone else entirely.  There is still no evidence who this is, and to say flatly that it is Eubie, without qualification, is in my opinion irresponsible.  RESULT: both discographies flunk this one.  Each lists it in two different places, identifying it in one place as Eubie and in the other as “unknown” (!)  In one listing they also misspell the name as “Jazztone,” give the wrong date (1918) and get the matrix and record numbers wrong (Pathe 20340/mx 65787 instead of the correct Pathe 20430/66787).

Issue #2: Ciro’s Club Coon Orchestra, “St. Louis Blues” (1917): this wild recording is significant as the first vocal version of the famous song.  RESULT: neither discography lists a vocal.

Issue #3: W.C. Handy’s 1919 recordings for Lyric: four very interesting titles by “Handy’s Memphis Blues Band,” possibly with Johnny Dunn on cornet, were announced in Talking Machine World in 1919, and as a result have found their way into many modern discographies.  But no one has ever found copies of any of them, and recent research strongly suggests they were never issued (the numbers were immediately reassigned to other recordings).  RESULT: both discographies list them, neither acknowledges they probably don’t exist.

Issue #4: The Original Dixieland Jazz Band’s first recordings for Columbia (1917): the date of the first “jazz” record is surely of some significance.  The date is often given as “January 30, 1917,” even though an analysis of the Columbia files shows that date to be most likely a figment of cornetist Nick LaRocca’s imagination.  In fact the band’s first recording was for Victor in February, with the Columbia session occurring in May.  This information was published fifteen years ago in an article by Brian Rust, in his magazine Needle Time, and has been reprinted since then.  RESULT: both discographies give the old date of January 30^th for Columbia, with no comment.