Designs Awaiting Implementation

Handling Original Appearance (EO) Records

Background:

There are a number of cases where the existing approach to handling author or title changes since the first appearance is insufficiently flexible to address the circumstances. For example, there is a two part story in Ghost Stories as follows:

E 12A0~Hichins, Raoul ,(as told to:Mearson, Lyon)~Body and Soul [Part 1 of 2]~ss1927GHSJun~
E 46A0~Hichins, Raoul ,(as told to:Mearson, Lyon)~Body and Soul [Part 2 of 2]~ss1927GHSJul~

The second part (only) was then reprinted in Prize Ghost Stories with a different title and no author credit. The closest we can come to supporting something like this is a construct along the lines of:

E 14A1~Anon.~Woman Who Took off Her Body ["Body and Soul", as by Raoul Hichins and Lyon Mearson]~nv1927GHSJul~The ~
E 14B1~part 2 only~

which is less than ideal and doesn't provide a perfect match anyway.

Original Proposal:

The basic change would be to remove any original title/author information from the EA title field and put it a sub-record, for these examples named EO.

The EO record has the same field structure as an EC record:

E O1~original author(s)~original title~title leading article~sub-title leading article~

ex:

E 14A1~Anon.~Woman Who Took off Her Body ["Body and Soul", as by Lyon Mearson]~nv1927GHSJul~The ~
E 14B1~part 2 only~

would change to

E 14A1~Anon.~Woman Who Took off Her Body~nv1927GHSJul~The ~
E 14O1~Hichins, Raoul ,narr./Mearson, Lyon~Body and Soul [Part 2 of 2]~

or in your format

E 14O1~Hichins, Raoul ,(as told to:Mearson, Lyon)~Body and Soul [Part 2 of 2]~

This way the "as/by" text on the reprint and the dummy original record created by the programs for linkage would exactly match the original.

Current Status:

This was discussed at some length between BC & PSP in May 2016 (e-mail thread "Notes on EO Record Type") but raised sufficient problems and areas of disagreement for it to be shelved for the time being in favour of other (easier) issues that need to be addressed. Some of the issues that remain unresolved include:

1. EC or EO record
   • PSP was unhappy with the proliferation of E record subtypes and with the "distance" between the EA and associated EO record and suggested possibly extending the EC record to handle the new functionality as well.
   • It became clear that extending the EC record would require the inclusion of some form of "flag" to differentiate EC records in the existing sense from the new style as it would not be possible to differentiate the two programmatically based purely on the field contents.
   • There was some disagreement about the placement of such a flag - PSP favoured having it after the title field (which would involve either overloading the existing field or changing all existing EC records) while BC favoured having it after the item additional field (which would involve several field dividers if just an author was specified).
2. To change or not to change
   • BC felt it was critical, to avoid overcomplication of the programs, that any such change, once adopted, must be implemented across the entire database, adjusting all existing records to use the new approach.
   • PSP felt it would be a backward step to force the replacement of the current approach (which works well and is fairly straightforward for the most common places) by the more complex new approach (which was only required in relatively few cases).
   • PSP also felt that such a global change was inherently too dangerous to risk given past failures in making global changes.
3. Uncredited or Indirectly Credited Authors
   • BC felt that the current approach for handling uncredited or indirectly credited authors where the author was known would need to revert to the "old" approach of appending the name as part of a "by:" clause so that the EA record only contained "title and author information for the current publication".
   • PSP could see no advantage in such a move as the EA record would still contain author information not directly related to the current publication, simply moving it from one field to another. In addition, the proposals required the "by:" clause to be attached to the author name on the original appearance and on the EO records for any subsequent uncredited appearances, which was undesirable duplication. PSP suggested as an alternative that the "real name" should be moved either to a separate field in the EA record (overloading the subject field) or specified on a separate type of EO record.
4. Items originally published as part of a larger work
   • The proposals addressed all aspects of Specifying Title or Author Changes since First Appearance except for the case where the item had originally appeared as part of a larger work (e.g. a poem that had originally been published as part of a short story).
   • PSP felt that if the root philosophy was that the EA record should only contain information for the current publication then this also needed to be catered for in the new proposals; BC saw no problem with leaving this part as it is.
5. Partial or Full Specification of details in EO record
   • BC initially suggested that fields should only be specified in the EO record when they differed from the associated field in the EA record. PSP pointed out that this would mean that the EO records associated with different EA reprint records would potentially differ one from another which would make life more difficult.
   • While it would be easy to provide support for both approaches, it would be better to agree on one approach (omit duplicated fields) or the other (include all fields in the EO record) to minimise drift between different versions of the same file.

Handling Translation Detail (ET) Records

Phil S-P (11-Dec-2011):

I've been chatting to Mike Ashley a bit about the BL Index Project and ways I can help him with it, and one topic that came up was that of translations, particularly when the same story is translated by multiple authors. I thought you might be interested in my comments and would welcome any thoughts/views from your end.

********************

Oh, and before I forget, you commented a few emails back on the problem of a story being translated under two different titles by two different translators. That certainly does happen and there's a good example of it. Erckmann-Chatrian's "The Crab Spider" was also translated as, I think, "The Spider of Guyana" (I'm doing this from memory). The latter turned up in THE STRAND whereas the former, which I think was translated by Lafcadio Hearn, had appeared in the US first - probably in SHORT STORIES translated from a French original - and was then picked up by one of the UK magazines and reprinted here. You'll probably find all these variants quite quickly, but how do we link them? THE STRAND's going to be the main culprit in this because it ran so many "Stories for Children" translated from across Europe and some of those translations are also known under other titles elsewhere.

The key question to consider is quite what you have in mind in terms of "linkage". Looking through the assorted FMI (and related files), the handling of this story is rather a mess. We have:

E0489A1~Erckmann, ^E'mile/Chatrian, Alexandre~Spider of Guyana ["The Crab Spider"]~ss1893RMCOct~The
E0489D1~*SNDJan 1899~
E0115A1~Erckmann-Chatrian~Waters of Death ["The Crab Spider"]~ss1893RMCOct~The
E0115D1~*SNDJan 1899~, as "The Spider of Guyana"
E0210A1~Erckmann-Chatrian/Fearnley-Whittingstall, Eithne ,trans.~Crab Spider~ss1893RMCOct~The
E0210D1~*SNDJan 1899~, as "The Spider of Guyana"
E0128A1~Erckmann-Chatrian~Spider of Guyana ["The Crab Spider"]~ss1893RMCOct~The
E0128B1~this translation taken from {The Strand} Jan 1899.
in THE SUPERNATURAL INDEX

E0039A0~Erckmann-Chatrian~Spider of Guyana ["The Crab Spider"]~ss1893RMCOct~The ~
in PIONEER TALES (Miller/Contento Index)

E 39A1~Erckmann-Chatrian~Spider of Guyana~ss1899SNDJan~The
in PIONEER TALES (FMI version)

E0130A1~Erckmann-Chatrian ,(tr:Fearnley-Whittingstall, Eithne)~Crab Spider~ss1981*BstTTEC~The ~
E0130D1~first published in French in <Contes Fantastiques> (1860)
in my fledging index to ASH-TREE books

E0081A0~Erckmann-Chatrian~Spider of Guyana~ss~The ~~~Hardy, Paul~
in THE STRAND

where RMC is {Romance Magazine} and BstTTEC is BEST TALES OF TERROR OF ERCKMANN-CHATRIAN (Millington, 1981).

Now there are multiple confusions and problems here already:

• A quick sniff around the web reveals that "Eithne Fearnley-Whittingstall went back to the original french version and provided Hugh with her translation" (for the Millington 1981 volume) so the entry from the ASH-TREE index is pretty good, but the third one in the Supernatural Index (which claims her translation first appeared in ROMANCE in 1893 and was then reprinted in THE STRAND in 1899) is clearly wrong. As this was another Hugh Lamb anthology, though, it is likely that the translator is correct and hence that the first appearance is actually this anthology (THE TASTE OF FEAR) which appeared five years before the Millington edition.
• The 10/1893 ROMANCE citation comes from (at least) THE ENCYCLOPEDIA OF GOTHIC LITERATURE which implies this was the first appearance of the story (which conflicts with it having appeared in CONTES FANTASTIQUES in 1860).
• Your SFE entry for STRAND MAGAZINE confirms the 1860 original and gives it a title "L'Araigneé crabe"
• We don't have a ToC for that issue of ROMANCE so can't be sure what name it appeared under, but it is likely it was under the pseudonym of "Erckmann-Chatrian", in which case the first entry from the SUPERNATURAL INDEX above gets the author names wrong as well.
• Assuming the STRAND version reprints the ROMANCE translation (which you believe it does) then the note on the final entry in THE SUPERNATURAL INDEX is incorrect as it implies the translation first appeared in THE STRAND.

So what should the entries be? Let's assume for the sake of discussion that the STRAND/ROMANCE versions are translated by Lafcadio Hearn. I'm also dubious about the "Waters of Death" entry (from LOCK AND KEY LIBRARY in 1908) so let's assume, for the sake of discussion, that this is an unknown translation which may, or may not, be by Lafcadio Hearn. IMHO, we should then have:

E0489A1~Erckmann, ^E'mile/Chatrian, Alexandre ,(tr:Hearn, Lafcadio)~Spider of Guyana ["The Crab Spider", as by Erckmann-Chatrian]~ss1893RMCOct~The
E0489D1~translated from the French ("L'Araigne^e' Crabe", <Contes Fantastiques>, Hachetter 1860)~
E0115A1~Erckmann-Chatrian~Waters of Death~ss~The
E0115D1~translated from the French ("L'Araigne^e' Crabe", <Contes Fantastiques>, Hachetter 1860)~
E0210A1~Erckmann-Chatrian/Fearnley-Whittingstall, Eithne ,trans.~Crab Spider~ss1976*TastoFr~The
E0210D1~translated from the French ("L'Araigne^e' Crabe", <Contes Fantastiques>, Hachetter 1860)~
E0128A1~Erckmann-Chatrian ,(tr:Hearn, Lafcadio)~Spider of Guyana ["The Crab Spider"]~ss1893RMCOct~The
E0128D1~translated from the French ("L'Araigne^e' Crabe", <Contes Fantastiques>, Hachetter 1860)~
in THE SUPERNATURAL INDEX

E0039A1~Erckmann-Chatrian ,(tr:Hearn, Lafcadio)~Spider of Guyana ["The Crab Spider"]~ss1893RMCOct~The ~
E0039D1~translated from the French ("L'Araigne^e' Crabe", <Contes Fantastiques>, Hachetter 1860)~
in PIONEER TALES

E0130A1~Erckmann-Chatrian ,(tr:Fearnley-Whittingstall, Eithne)~Crab Spider~ss1976*TastoFr~The ~
E0128D1~translated from the French ("L'Araigne^e' Crabe", <Contes Fantastiques>, Hachetter 1860)~
in the index to ASH-TREE books

E0081A1~Erckmann-Chatrian ,(tr:Hearn, Lafcadio)~Spider of Guyana ["The Crab Spider"]~ss1893RMCOct~The ~
E0081D1~translated from the French ("L'Araigne^e' Crabe", <Contes Fantastiques>, Hachetter 1860)~
in THE STRAND

Now "The Spider of Guyana" and "The Crab Spider" would be linked because they are alternate titles for the same story but currently the only link to "The Waters of Death" would be indirect via the notes. The use of the "E0489D1~*SNDJan 1899~" records currently in place might sound attractive but actually cause problems for Bill's programs _and_ would not be valid anyway as they only apply to the same story (i.e. translation).

What I would _like_ to do in the next version of the software is to add much more formal support for things like translations. Thus, thinking on my feet, we might have a new ET record type such that the STRAND entry became instead:

E0081A1~Erckmann-Chatrian ,(tr:Hearn, Lafcadio)~Spider of Guyana ["The Crab Spider"]~ss1893RMCOct~The ~
E0081T1~French~Araigne^e' Crabe~1860*CntsFnt~L'~

which would be easier to type, much easier to validate, would guarantee a consistent format of output (which could, itself, be varied from index to index) _and_ would allow for an entry in the author index (and maybe the title index) for L'Araigne^e' Crabe listing all the translations.

Bill C. (15-Dec-2011):

An ET type record would be useful for tying different translations together. I see the Author List looking like this:

ERCKMANN-CHATRIAN; pseudonym of Alexandre Chatrian & Émile Erckmann (chron.)
The Crab Spider, (ss) Romance Magazine Oct 1893; also as “The Waters of Death”.
The Crab Spider, (ss) The Taste of Fear, ed. Hugh Lamb, London: W.H. Allen 1976; tr. by Eithne Fearnley-Whittingstall; translated from French; "L'Araigne^e' Crabe", (ss) <Contes Fantastiques>, Hachetter 1860
L'Araigne^e' Crabe, (ss) <Contes Fantastiques>, Hachetter 1860
as "The Crab Spider", (ss) Romance Magazine Oct 1893
as "The Waters of Death", (ss) Romance Magazine Oct 1893, as "The Crab Spider"
as "The Spider of Guyana", (ss) The Strand Jan 1899
as "The Crab Spider", (ss) The Taste of Fear, ed. Hugh Lamb, London: W.H. Allen 1976; tr. by Eithne Fearnley-Whittingstall
The Spider of Guyana, (ss) The Strand Jan 1899; translated from French; "L'Araigne^e' Crabe", (ss) <Contes Fantastiques>, Hachetter 1860
Ghostly by Gaslight, ed. Sam Moskowitz & Alden H. Norton, New York: Pyramid 1971
The Waters of Death ["The Crab Spider"], (ss) Romance Magazine Oct 1893; translated from French; "L'Araigne^e' Crabe", (ss) <Contes Fantastiques>, Hachetter 1860
Library of the World’s Best Mystery and Detective Stories: French: Italian: Spanish: Latin, ed. Julian Hawthorne, The Review of Reviews 1907
The Lock and Key Library: Modern French, ed. Julian Hawthorne, The Review of Reviews 1909

In addition to the English titles there'd be an entry for the original title, followed by the various English titles and different translations. The ET record would be similar to an EA record with the Author field listing the original language.

Named Serial Parts

While a format has been agreed ([{Part m of n{, serial part title};} {"Original title",} {as by original name}]) to allow the capture of the names of individual serial parts, this is not yet supported by the programs.