Gunilla Jonsson wrote:
I find the proposed treatment better, both from a principal point of view and a practical one. The recording of the title is descriptive information that primarily shall support identification. For the sake of global record exchange, it is also better to follow the instruction proposed in 1.6.8 -- you cannot be sure that all cataloguers in all countries will detect misprints in all languages. It's also better for the sake of consistency in the cataloguing of materials from different time periods. In cataloguing hand press manifestations you would not apply the procedure you propose.
I take your point; but I'm not yet convinced. This is indeed a troublesome issue, an instance of a conflict between transcription (which is an application of the principle of accuracy) and the user's expectations; we do well to remember that the user's convenience ranks above purely technical considerations. I'm inclined to think that literal accuracy, in the case of orthographical variation (including typographical errors), is primarily a technical consideration. It's true that literal transcription of the title proper is significant in making a match between a document and a variety of records which may or may not be correct; this impinges on selecting the correct record from a utility or other database (something that is part of my daily activity), but also on database management, where control of duplicate records loaded from different sources is an ongoing problem. May I pose a question: suppose that a typo occurs, on the title page of a printed book, or the title screen of an electronic resource. Subsequently, the publisher of the book issues a reprint with a corrected title page, or maybe issues a corrected title page with instructions to paste it over the original (I have actually received such pages); or the issuing agency of the electronic document silently corrects it. To my mind, this is a change in the label information, not the substance, not a significant change: I see this as the *same manifestation*. How is this best handled?
This said, it is of course necessary to provide an access point even for the correct spelling, but that is an *access* point, not descriptive data. I think it is useful to observe that distinction. I also have some hesitations about discarding the "[sic]" after the typo. It is an efficient way of telling the user that this is not a cataloguer's typo (of which there are so many more!).
On the other hand, interpolation of "[sic]" into the text disturbs the transcription. I would prefer that something like "[sic, as printed]" be added to the end of the title data, before the statement of responsibility. One of the matters discussed here has been the requirements for content of minimal descriptions (corresponding to AACR2 Level One). An added title access point can perhaps be categorized as "mandatory" for higher-level descriptions, but I doubt that staff constructing minimal descriptions would defer to that. Finally, while the closest possible correspondence between document and record has to be achieved, we mustn't overlook what it is that users bring to the catalogue. Apart from "resource discovery", which may be by a variety of search criteria, users bring citations to their searching. Those citations may be perfect or imperfect. While, as text-oriented people, librarians generally take it for granted that users have good citations which they obtain in written form, in fact users receive citations also in speech: conversation, academic lectures, conference presentations, radio programmes, and so on. Citations from such sources will seldom convey inaccuracies in the labels of the documents referred to. Access (both by searching indexed strings and by keyword search) has to provide for such unconsciously corrected data. Moreover, even in printed or electronic reference lists, it's not unlikely that apparent errors will be editorially corrected. Where does the principle of accuracy place us in such cases? Hal Cain Joint Theological Library Parkville, Victoria, Australia [EMAIL PROTECTED] -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.371 / Virus Database: 267.14.17/228 - Release Date: 12/01/2006
