Adam P.J. O’Connor, Editor-in-Chief
Size of the volumes. The following statistics may be interesting:
| Rilm Year | Number of Records | Year | No. of Entires | Year | No. of Entires | ||
|---|---|---|---|---|---|---|---|
| 1967 | 2,532 | 1976 | 17,121* | 1985 | 7,639 | ||
| 1968 | 4,311 | 1977 | 6,042 | 1986 | 7,059 | ||
| 1969 | 5,106 | 1978 | 6,459 | 1987 | 9,844 | ||
| 1970 | 4,201 | 1979 | 5,916 | 1988 | 11,753 | ||
| 1971 | 4,639 | 1980 | 5,927 | 1989 | 12,950 | ||
| 1972 | 4,159 | 1981 | 6,641 | 1990 | 12,093 | ||
| 1973 | 4,504 | 1982 | 6,165 | 1991 | 12,802 | ||
| 1974 | 4,235 | 1983 | 6,654 | 1992 | ca. 13,000 | ||
| 1975 | 4,701 | 1984 | 7,681 |
Vol. XXIV (1990): The index portion of vol. 24 was shipped, together with vol. 25, on 17 February. The index has traditionally been delayed, and vol. 24’s was delayed more than usual because some of the effort that would have been directed toward it was directed instead to the 1991 index. This allowed us to produce an integrated volume and at last break the on-again, off-again work cycle that had required the indexes to be delayed by several months. The vol. 24 index was the last that was prepared by a separate indexing staff. This is the last time the indexes will be issued later than the abstracts.
Vol. XXV (1991): Vol. 25 was shipped on 17 February. This volume contains 12,802 entries and 1040 pages. It is the first volume whose indexing was written by the entire editorial staff in the course of editing, the first volume whose indexes appeared with the rest of the book rather than several months later, and the first hard-bound volume; the editorial cycle was about eight months.
Vol. XXVI (1992): Work on this volume has been fully integrated: At the start of the vol. 25 cycle, several staff members were still indexing vol. 24; vol. 26 was the first volume on which the entire staff began work more-or-less en masse. It will also be the first book edited and indexed with the aid of the new 200-page (indexed) editorial and indexing manuals. Vol. 26 is due to be sent to the printer on 31 August.
Editing Manual and Indexing Manual: The Editing manual is an expansion of a long-standing document; the Indexing Manual is the first of its kind at RILM. We anticipate that the manuals will lead to more consistent indexing and editing, and less redoing of work. Copies of the Editing Manual were sent to all national committees. Further copies and copies of the Indexing Manual are available on request.
The editorial process: When records are received in New York, the editorial process (in outline) is as follows: The records are (1) accessed (bibliographic information is keyed into the system); (2) accession proofed (bibliographic information is proofread); (3) edited and indexed (an editor edits and keys the redacted abstract and indexes it); (4) revised (a revision of step 3, requiring more or less effort depending on the difficulty of the material and the skill with which 3 was done); and (5) proofed (any remaining errors or omissions are fixed). When all the records have gone through these stages, various special proofings are done (e.g., all titles and title translations reviewed, all authors regulated, German records proofed by the best German editors, etc.) and machine-based examinations of the data are made (e.g. all symposia being in class 16, Festschriften in 15, headword correctness, uniformity of index strings, etc.); this requires about two weeks. A paper proof of the book is then read twice, and the corrections entered (this requires about two weeks). Not much must be done to perfect reviews because the reviewed book is already entered and subjected to all the above tests, the journal names are defaulted from an authority file, and the review authors regulated with all the other authors. At the end of this process, data can be delivered to the CD vendor and the online services. Typesetting for the printed book requires a further two weeks, and printing and binding the book about a month.
Uploading of committee-keyed data: The Data-transfer protocol gives instructions for tagging data so that it can be typed using a modern word-processing program and is then uploaded directly into the database–i.e., not rekeyed by editors in New York. Committees that have submitted abstracts using this tagging system include Germany, Poland, and Switzerland. We distributed the Protocol and waited to receive test data to complete software for the upload procedure. Although committees are as interested in sending records via e-mail as they are in the tagging itself, the vagaries of e-mail complicated our first trials with the upload. At our request, the German committee sent the same data on a disc, and on 1 June we uploaded approximately 800 records (92 and 93 data) from the committee, and are now evaluating the integrity of the data. Around 15 July we will send committees any amplified instructions for preparing uploadable data. Some committees had found it convenient to submit word-processed data (i.e., bypassing the abstract form) before the advent of the tagging protocol; we will of course continue to accept word-processed data no matter what the limitations of the upload. If the upload system proves successful—as we are confident it will—it will greatly accelerate editorial work in New York, be a convenience to the committees, and enable us all to handle the continual expansion of material. We are grateful to the committees who have pioneered this important work.
Original-language abstracts: Dr. Leibnitz of the Austrian committee has asked if the publication of original-language abstracts would be feasible. Our answer has always been that it would mean considerably more work for the New York office (which is already on a five-year forced-march to currency), we could not insure that so much foreign-language material would be presented error-free. There was some discussion of this in Ottawa, Maria Calderisi suggesting that at least French and German be considered as languages for the abstracts. Presentation of the original-language abstract met with some approval, though it was agreed that English was the most widely understood language. The Guidelines for Preparing Abstracts printed on the verso of the abstract form states: “Abstracts will be published in English although publication in more than one language is not precluded.” The delivery of machine-readable data to the International Center raises the possibility that the original language abstract could be presented along with the English translation. This is, of course, wholly dependent on the success of the upload operation. A considerable amount of work is done on most abstracts, and little editorial work would be done on the original-language version; therefore some divergence between the text of the original and the edited English version is inevitable. To reiterate: If it is decided that original-language abstracts are to be included, the editors would at most reduce the original version by some deletion of material; they would be unable to edit the originals properly. The text supplied by the committee would be presented more or less as it arrives. Therefore, for the presentation of the original version to succeed, the submitting committee would have to follow the Editorial Manual closely, and feel confident of the perfection of its keying. The English versions of the abstracts are and would continue to be gone over several times, but the original-language material would be merely checked for basic data integrity, and presented to the reader. Some committees might elect to present material not in their own language but in French or German (e.g., an abstract might appear in English and German from a non-English, non-German source); in this case, of course, there is a somewhat greater exposure to error, the writers of the abstracts not being native speakers, and the editors in New York not able to check the language. But committees will doubtless know whether or not to certify their keying as perfect. This matter will have to be approached with great sobriety. The editors would reserve the right to withhold non-English versions that would not reflect well on the publication or the submitting committee. It would perhaps be best to begin this in a small way. Such an additional data burden will increase the expense of the printed book, but it might also improve subscriptions in some countries. As this would be a fundamental change in editorial procedure, it is a matter for the Commission Mixte. If the decision is taken to follow this course, the New York office will notify the national committees.
Cumulative Index 5: Vol. 25 was the final volume in a five-volume cumulative period, and we are beginning to plan for another Cumulative Index. We will treat the work on Cum 5 as a wholly autonomous project, to preserve the progress of the annual volumes. Software development will begin in the next couple of weeks, and editorial work is projected to begin in the late summer.
Thesaurus: A faceted-access Thesaurus is underway. A fifth printed thesaurus will be prepared in tandem with Cum 5.
NISC: The New York office immediately discovered flaws in the latest NISC CD (followed by more discoveries by a few other users). At NISC’s suggestion, a call was sent out on the MLA and IAML listserves for reports of any further problems. NISC advised us that, depending on the nature of the problems, the disc could be fixed with either a patch diskette or a newly pressed CD. NISC ultimately elected to press a new CD. At Alan Green’s suggestion, we asked NISC to wait for us to provide them with the 1991 data, as this was nearly complete. Unfortunately, difficulties with a new file server in New York kept spoiling our sorting and downloading of data. After a delay of a few weeks, we sent the 1991 data and improved 1990 and 1989 data to NISC. A long series of unrelated delays followed, and the New York office field-tested discs for NISC, with three more aborted releases. This is where the situation remains.
OCLC: After some negotiation, aided considerably by Alan Green, we signed a contract with OCLC in November. According to OCLC, RILM will be available on FirstSearch and EPIC by fall 1995. We believe OCLC will be more convenient to users than DIALOG. We agreed to send OCLC data on a quarterly basis, though they seem to have revised this to a semiannual system. This will change our work flow in New York somewhat.
Adam P.J. O’Connor
Editor-in-Chief