erc-dharma / project-documentation

DHARMA Project Documentation
Creative Commons Attribution 4.0 International
3 stars 3 forks source link

IAR in Zotero #290

Closed manufrancis closed 3 months ago

manufrancis commented 4 months ago

@arlogriffiths

How should we zoterise IAR? There are issues with known editor and issues without. And there is also a generic Zotero entry for the journal.

See here the different short titles. Capture d'écran 2024-04-21 105415

arlogriffiths commented 4 months ago

I would probably never cite IAR by the editor, even if known. So I would recommend we first systematize all ZSTs with the pattern "IAR1953-54", "IAR1954-55", etc. (I suspect existing ZSTs not following that pattern have hardly been cited in our XML files so far, but it would be best if @michaelnmmeyer could confirm this.) This will allow citing specific volumes without using the generic entry. Next I would cite by page, e.g. <bibl><ptr target="bib:IAR1965-66"/><citedRange unit="page">114</citedRange></bibl>. This is basically how I proceed for analogous publications like NBG and the anonymous Chronique parts of old BEFEO volumes.

Capture d’écran 2024-04-21 à 18 49 12

The "no name" items are generated by

<bibl><ptr target="bib:NBG14_1876"/><citedRange unit="page">98-100, 106</citedRange></bibl> and <bibl><ptr target="bib:NBG25_1887"/><citedRange unit="page">67</citedRange></bibl>

@michaelnmmeyer: can a rule be formulated whereby in such cases the display shows the abbreviation for the journal in place of "no name"? Since the zotero entries in question do no have an "journal abbreviation" field, some work-around would be necessary. Maybe @manufrancis and @danbalogh have other aesthetically more pleasing solutions to propose.

arlogriffiths commented 4 months ago

@michaelnmmeyer : by the way, in the screenshot just shared, I notice that "bookAuthor" is not adequately represented yet in our display. The Zotero item in question is Brandes1887_01, which exports correctly if I use the personalized CSL:

Brandes, Jan Laurens Andries. 1887. “Beschreven steenen.” In Catalogus der archaeologische verzameling van het Bataviaasch Genootschap van Kunsten en Wetenschappen, by Willem Pieter Groeneveldt, 351–91. Batavia: Albrecht & Co.

michaelnmmeyer commented 4 months ago

(1) We have about 200 citations of the IAR in our texts.

(2) When no author or editor name is available, I propose to fall back to the short title. So we would have e.g.

See NBG14 1876, p. 30.

NBG14 1876. Notulen van de Algemeene en Bestuurs-vergaderingen van het Bataviaasch Genootschap van Kunsten en Wetenschappen: Deel XIV — 1876. Batavia: Bruining, 1877, pp. 98–100, 106.

(3) OK for the display of bookAuthor. We now have:

Brandes, Jan Laurens Andries. 1887. “Beschreven steenen.” In: Catalogus der archaeologische verzameling van het Bataviaasch Genootschap van Kunsten en Wetenschappen. By Willem Pieter Groeneveldt. Batavia: Albrecht & Co., p. 361.

danbalogh commented 4 months ago

I have no constructive suggestions but would like to note that ARIE is also analogous and is a source of annoyance, giving rise to displays like this (all three items of the secondary bibliography are ARIE volumes): image from the code:

<div type="bibliography">
   <p>First reported in <bibl><ptr target="bib:ARIE1907-1908"/><citedRange unit="page">10</citedRange><citedRange unit="appendix">A/1907-1908</citedRange><citedRange unit="item">2</citedRange></bibl> with a description at <bibl><ptr target="bib:ARIE1907-1908"/><citedRange>48-49</citedRange></bibl>.<!--CHECK: According to bib:Gai1986_01 (Dynastic List), p 33 item 170, these plates were also reported as Copper Plate 60 of 1957. There is no Number 60 among the copper plates in ARIE 1957-58, and there seems to be no ARIE volume for 1956-57, or if there is one, perhaps it was omitted from the reprint.--> Edited from inked impressions by E. Hultzsch (<bibl rend="omitname"><ptr target="bib:Hultzsch1907-1908_07"/></bibl>) with partial estampages<note>Only estampages of 1v and 2r are included in the 1981 reprint of <title>Epigraphia Indica</title>; there is no image of 2v, nor of the seal.</note> and translation. Re-reported in <bibl><ptr target="bib:ARIE1956-1957"/><citedRange unit="page">33</citedRange><citedRange unit="appendix">A/1956-57</citedRange><citedRange unit="item">60</citedRange></bibl>. The present edition by Dániel Balogh is based on photographs taken by myself in February 2023 at the Government Museum, Chennai, collated with Hultzsch's edition and his estampages where available. 
   </p>
    <listBibl type="primary">
    <bibl n="EH"><ptr target="bib:Hultzsch1907-1908_07"/></bibl><!-- one <bibl/> per item of primary bibliography -->
  </listBibl>
   <listBibl type="secondary">
      <bibl><ptr target="bib:ARIE1907-1908"/><citedRange unit="page">10</citedRange><citedRange unit="appendix">A/1907-1908</citedRange><citedRange unit="item">2</citedRange></bibl>
      <bibl><ptr target="bib:ARIE1907-1908"/><citedRange>48-49</citedRange></bibl>
      <bibl><ptr target="bib:ARIE1956-1957"/><citedRange unit="page">33</citedRange><citedRange unit="appendix">A/1956-57</citedRange><citedRange unit="item">60</citedRange></bibl>
<!-- one <bibl/> per item of secondary bibliography -->
   </listBibl>
</div>

Back to the topic at hand, publications of this sort are not journals in my opinion, since they often [though not always] have individual editors and publishing places while typically [but not always] lacking articles by named authors. If it is possible to implement without requiring us to revise too many of our existing Zotero entries, I think it would be nice to display such items (e.g. <ptr target="bib:ARIE1907-1908"/> or <ptr target="bib:IAR1965-66"/>) as follows:

That is to say, I would always show the abbreviation instead of using it just as a fallback option when there is no author or editor available. Author and/or editor, when present, would be shown only in the full bibliography, after the title. (I also think this is what Arlo had in mind above when he said he would never cite IAR with the editor's name.) Place, publisher and publication date may not need to be shown at all, but they could also be added to the full bibliography when available. I would want a space after the journal abbreviation and prefer en dashes instead of hyphens. However, I have no idea how this could be systematised in Zotero to get the desired display. I wonder if the Zotero item type "Report" might not be more suitable for these publications than Book.

manufrancis commented 4 months ago

Thanks both of you!

So I note that ZST for IAR should be of the type "IAR1965-66".

@arlogriffiths I can update the wrong existing ones (IAR1954-01, Ghosh1995_01, etc.) on Zotero. This ZST for IAR should be explained in the next version of the Zotero guide.

@michaelnmmeyer As some target="bib:..." for IAR will have to be updated accordingly to my cleaning of ZST, could you tell me, about the about 200 IAR citations, in which repos they are? And maybe, if easy collected, by which encoders?

manufrancis commented 4 months ago

ARIE is indeed a pain in the ass.

I deal as follows when I need to refer for the same ARIE to a N° in the appendix table and to a § in the detailed notes that follow the appendix tables (see https://dharmalekha.info/texts/INSPallava00518):

<bibl>
        <ptr target="bib:ARIE1938-1939"/>
        <citedRange unit="page">38</citedRange>
        <citedRange unit="appendix">B/1938-1939</citedRange>
        <citedRange unit="item">271</citedRange>
        <citedRange unit="page">91</citedRange>
        <citedRange unit="section">59</citedRange>
</bibl>

Which in display presently gives: image

@michaelnmmeyer

  1. In such a display would it be possible to add something (&, +, or a diamond, ...), rather than a comma, between "p. 38, appendices B/1938-1939, № 271" and "p. 91, § 59"

  2. "appendices" should indeed be "appendix". (which, by the way works fine for another <bibl> in this file in primary bibliography).

  3. Note that after "Government of India" we expect full stop rather than comma.

As for Daniel's suggestions for display of IAR and ARIE in bibliography, it is fine for me if indeed it is possible to implement without requiring us to revise too many of our existing Zotero entries.

michaelnmmeyer commented 4 months ago

@manufrancis

For the IAR, I propose you just modify the Zotero entries in the screenshot. I will update inscriptions accordingly.

For points 1 and 2 of your latest post, I do not see a nice solution. In practice, references vary too much to be expressed in a clear, structured way. In cases like these, it would probably be simpler to use free text.

For point 3 of your latest post, I am not sure why you want a full stop after the publisher name. Is it because, in this case, page numbers, etc. are given in <citedRange> instead of in the zotero entry? In other words, is it to distinguish page numbers given in the zotero entry (if any) from the ones given with <citedRange>?

@manufrancis @arlogriffiths @danbalogh

For citing entries with an abbreviation instead of the usual author+date scheme, I propose we define a Shorthand field that, when given, would be used in references instead of the usual author+date. This is more-or-less what biblatex does. For the entry Ghosh1961_01, which refers to IAR 1960-61, you would add in the Extra field:

Shorthand: IAR 1960-61

References would look like this:

See IAR 1960-61, p. 10.

And entries would look like this:

IAR 1960-61. Indian Archaeology 1960-61, A Review. Edited by A. Ghosh. New Delhi: Archaeological Survey of India, 1961.

Note that the shorthand includes a date; the date of publication is not automatically pulled from the relevant entry field. This way, it is possible to use a different numbering scheme (series number, etc.), or to use no scheme at all (abbreviations for referencing canonical editions of certain texts, like MBh, etc.).

danbalogh commented 4 months ago

@manufrancis , I'm afraid your solution to 1 is too apocryphal. Using more than one <citedRange> element within a <bibl> is meant to specify a single thing in more than one way (EGD: "a point identified by a combination of entities"), e.g. <citedRange unit="page">38</citedRange><citedRange unit="appendix">B/1938-1939</citedRange><citedRange unit="item">271</citedRange> to mean "Item 271 of Appendix B/1938-1939 on page 38". Since your additional details <citedRange unit="page">91</citedRange><citedRange unit="section">59</citedRange> do not mean that the above is simultaneously also on page 91 and section 59, the citation is strictly speaking uninterpretable. There is also no clean way, as Michaël points out above, for a machine to know that you mean the former three items to specify one cited locus, and the latter two to specify another locus. The method for handling complex references (item X of appendix Z on page N and section K on page M) is that outlined in EGD§10.4.5 in the point starting with "for complex references where multiple points are identified by different (combinations of) entities", i.e. free text, as Michaël said above. Or there is my more verbose way of using two separate <bibl> elements to refer twice to different parts of the same ARIE publication.

Apropos of this, I notice that you use "section" while I use "item" to identify the numbered paragraphs/items in (some issues of) ARIE. I prefer item, since these do not seem to be section numbers to me, but I don't mind changing to "section" if that's your strong preference. I think that a projectwide recommendation/rule should be made for this and included either in the Zotero Guide, which already has a paragraph dedicated to ARIE, or in the EGD (where Example 10.4.5.F is specific to ARIE and already includes a general rule).

Further, if we will be revising the Zotero records for ARIE to add Shorthand fields, I would like to suggest using "ARIE 1938-39" (analogous to the IAR examples above), rather than "ARIE 1938-1939".

@michaelnmmeyer : thanks for your suggestion of Shorthand. From a processing point of view, would this then mean that IF a shorthand is present in the Extra field, then the regular display would be overridden, and the "Desired display" above would be produced instead? Or would the criterion for producing this alternative display be something else, distinct from the presence of a shorthand entry?

michaelnmmeyer commented 4 months ago

@danbalogh A priori, the shorthand would be used by default if present. That said, we can make this opt-in or opt-out if needed. The Extra field in Zotero can store arbitrary key-value pairs, so we could add display options there.

danbalogh commented 4 months ago

Either sounds good to me. I just want to understand what we'd need to do in the Zotero records to achieve what kind of display. This must of course also be clearly written up for the Zotero guide.

manufrancis commented 4 months ago

@danbalogh

  1. Ok. noted for EGD§10.4.5.
  2. I my use of "item" and "section", I am not sure I get your point. I use "item" for an item in a numbered list of inscriptions and the display "N°" fits that. I use "section" for a numbered paragraph and the display "§" fits that.
  3. As for shorthand "ARIE 1938-39" (analogous to the IAR examples above) rather than "ARIE 1938-1939", fine to me. Just to be clear you mean only shorthand? Not Zotero Short title?

@michaelnmmeyer

  1. I will update the Zotero Short titles for IAR.
  2. As for the full stop after the publisher name. Yes, this is because, in this case, page numbers, etc. are given in instead of in the zotero entry. And yes this is it to distinguish page numbers given in the zotero entry (if any) from the ones given with <citedRange>. See what Axelle had for a <citedRange> image

And here two displays of the same Zotero reference, the first with a <citedRange> image

Note that your actual display does not show the page quotations of the said article. image

manufrancis commented 4 months ago

I have now updated the existing IAR entries in Zotero

Capture d'écran 2024-04-23 091839

I have followed for the ZST the current model for ARIE (ARIE0000-0000), thus IAR0000-0000. Should we decide to change this model to IAR0000-00, the ZST should thus be updated in Zotero and in the edition XMLs.

danbalogh commented 4 months ago

@manufrancis , thanks for cleaning up the Zotero entries. On your above points,

2: I have no objection to your use of "item" and "section" for these things, but in my own editions I use "item" for both (entries in the appendix and items of discussion). The numbered parts of the ARIE text descriptions don't really look like sections to me. But as I said above, I don't mind changing to "section" if that's your strong preference. The point is that it would be best if you and I used the same unit when referring to the ARIA, and even better if everyone in the project did that. So if you are sure you want to keep using section for the discussion, let me know, and I'll change to section in my editions.

3: Yes, I meant only the shorthand; I would not want to change the ZST.

manufrancis commented 4 months ago

To your point 2

In secondary literature I find as a kind of traditional reference "part n, para. n"

Capture d'écran 2024-04-23 094607 [note: 272 is the page number]

In my oxygen I find:

Capture d'écran 2024-04-23 094744

and

Capture d'écran 2024-04-23 094809

For the moment, relying on EGD§10.4.5 I would encode a complex ARIE ref as follows:

<bibl>
        <ptr target="bib:ARIE1938-1939"/>
        <citedRange unit="page">38, 91 (§59)</citedRange>
        <citedRange unit="appendix">B/1938-1939</citedRange>
        <citedRange unit="item">271</citedRange>
</bibl>

But would I adopt double , yes I would use "item" for a number in appendix tables and "section" for paragraph in detailed discussion

To your point 3

If we thus agree on using shorthand, am I right that it means, e.g. for ARIE 1955-1956, of which the ZST is "ARIE1955-1956", we should add on Zotero "ARIE1955-56" in the field "Extra" ???

@michaelnmmeyer

If we adopt Daniel' encoding

   <listBibl type="secondary">
      <bibl><ptr target="bib:ARIE1907-1908"/><citedRange unit="page">10</citedRange><citedRange unit="appendix">A/1907-1908</citedRange><citedRange unit="item">2</citedRange></bibl>
      <bibl><ptr target="bib:ARIE1907-1908"/><citedRange>48-49</citedRange></bibl>
      <bibl><ptr target="bib:ARIE1956-1957"/><citedRange unit="page">33</citedRange><citedRange unit="appendix">A/1956-57</citedRange><citedRange unit="item">60</citedRange></bibl>
   </listBibl>

would you easily be able to have a display where both <ptr target="bib:ARIE1907-1908"/> result in only one bibliographical Venkayya, followed by two pages quotations, one to p. 2 and one to pp. 48-49?

danbalogh commented 4 months ago

@manufrancis Re point 2: I understand your reasons for preferring "section" and was not asking for further details of why. I just want our practice to be homogeneous. Since my practice so far has been to use "item" for those numbered bits of prose, my question was simply whether you are certain you want to keep using "section" for the same. I'll take this as a yes and change my practice to follow suit; I'll also make a note for myself to recommend this explicitly in the next EGD.

On the complex EGD reference, I repeat with emphasis that the encoding you show is wrong, as its meaning is ONE locus which is simultaneously on page 38 AND on page 91 AND in section 59 AND in Appendix B/1938-1939 AND in item 271 - which is impossible. To create a single reference to two items, one specified as (page 38, Appendix B/1938-1938, No. 271) and the other specified as (page 91, section 59), you can only use the non-rigorous free-text mode of referring for which I cited the EGD above: <citedRange>p. 38, Appendix B/1938-1938, No. 271; p. 91, §59</citedRange> - where you can write anything in the contents of citedRange and it's up to you to make sure that it will be interpretable for readers and if possible, consistent across your editions.

I generally do not think it desirable to start adding idiosyncratic display solutions for specific combinations of bibl elements. But if Michaël says it is indeed easy to create a merged display for the structured bibliographies, so that if two consecutive <bibl> elements contain a <ptr> with the same @target, then the bibliodata of the second are not shown, but instead the citation details are appended to the first, separated, say with a semicolon, instead of a line break (or indented, if a line break is inevitable), then fine by me.

But what I was speaking about above when I noted that two separate bibl elements can do the job, I meant the kind of unstructured context in my own encoding cited above:

First reported in <bibl><ptr target="bib:ARIE1907-1908"/><citedRange unit="page">10</citedRange><citedRange unit="appendix">A/1907-1908</citedRange><citedRange unit="item">2</citedRange></bibl> with a description at <bibl><ptr target="bib:ARIE1907-1908"/><citedRange>48-49</citedRange></bibl>.

In the (structured) secondary bibliography, these would still have to be listed as two separate references.

manufrancis commented 4 months ago
michaelnmmeyer commented 4 months ago

@manufrancis It is feasible, but I agree with Dániel that a free text reference would be preferable. If you have several cases like this, we can find a solution together.

danbalogh commented 4 months ago

@manufrancis: I must correct myself on one detail. It had escaped my memory that for this kind of non-rigorous reference, we have agreed to no longer use <citedRange> without any attribute; instead, it has to involve <citedRange unit="mixed">. This way, the computer can know that if there are no attributes, then a page number is meant. So if you use the free-text format, please make it <citedRange unit="mixed">p. 38, Appendix B/1938-1938, No. 271; p. 91, §59</citedRange>. I can add it as an Example to the next EGD.

manufrancis commented 4 months ago

@danbalogh Noted. And yes, please, an example in next EGD would be useful.

@michaelnmmeyer Yes, I have several cases of double ref pointers in the same ARIE. Nothing urgent though.

arlogriffiths commented 4 months ago

I didn't have time to follow all of this discussion, but I did see that @danbalogh requested an addition to ZG. Since Axelle's departure, I no longer know whom I am supposed to be collaborating with for maintenance of the ZG.

@manufrancis : this is a subject for our next AG. @danbalogh : could you kindly insert a comment at a relevant place in the ZG gdoc?

arlogriffiths commented 4 months ago

The shorthand solution doesn't work yet, right? I have just made Zotero entry JA1908_01 and encoded a bibliography entry as follows:

<bibl><ptr target="bib:JA1908_01"/><citedRange unit="page">331</citedRange></bibl>

with "Shorthand: JA 1908" in Extra.

display:

Capture d’écran 2024-04-25 à 07 41 11
danbalogh commented 4 months ago

I've added a comment in the ZG with reference to this thread. I cannot suggest even a draft of what should be added, because this Shorthand will need integration with the present ZG recommendation/rule of including the short title in the extra field (to be deprecated, I guess), and because this discussion does not seem to be finished.

michaelnmmeyer commented 4 months ago

@arlogriffiths The "shorthand" thing is not implemented yet. I will annotate the Zotero guide, but I need some time.

arlogriffiths commented 4 months ago

thanks, take your time.

michaelnmmeyer commented 4 months ago

I added support for shorthands and modified the IAR entries accordingly. Now we have e.g.:

image

image

In the second case, I find it undesirable to have the editor name at the beginning of the field. The following looks more natural:

[IAR] IAR 1957-58. Indian Archaeology 1957-58: A Review. Ed. by A. Ghosh. New Delhi: Department of Archaeology, Government of India, 1958. Page 56, plate 19.

But it is not possible to determine automatically which editors/authors names should be pushed forward in the entry, we would need an extra option for this.

danbalogh commented 4 months ago

Earlier you said,

References would look like this:

See IAR 1960-61, p. 10.

And entries would look like this:

IAR 1960-61. Indian Archaeology 1960-61, A Review. Edited by A. Ghosh. New Delhi: Archaeological Survey of India, 1961.

So why not stick with that? Does somebody need the editor/author name to be "pushed forward"? Or why did the alternative in youd screenshot (with Ghosh before the title) arise? I think extra options should be avoided unless absolutely necessary, and the format with "Edited by" (preferable imo to "Ed. by") after the title could be used in all cases where a shorthand is present in the Extra field.

michaelnmmeyer commented 4 months ago

I have in mind journal articles, books, etc., that have authors, but no editors. In this case, it might seem more natural to have the author name at the beginning. Compare e.g.:

RAC. Nilakanta Sastri, K. A. and M. Venkataramayya. “Telugu Chola records from Anantapur and Cuddapah.” EI 27, (1947-1948), pp. 220–251.

RAC. “Telugu Chola records from Anantapur and Cuddapah.” By K. A.Nilakanta Sastri and M. Venkataramayya. EI 27 (1947-1948), pp. 220–251.

But OK for your solution.

danbalogh commented 4 months ago

It was not "my solution" but quoted from what you had proposed earlier on. In my opinion, shorthand should definitely not be applicable to journal articles, and to books only for a given definition of "book". I thought the shorthand was introduced to handle journal-like serial publications. I have no idea what RAC means, but I am strongly against using any kind of shorthand for an article in EI. If this is the explicit wish of a PI, then I wash my hands of it. It's all the same to me.

arlogriffiths commented 4 months ago

I may have muddied the waters with my attempt to use Shorthand to handle an authorless item from the Journal asiatique:

“Chronique : XVe Congrès international des Orientalistes. — IIIe Congrès de l’histoire des religions. — Varia.” 1908. Journal Asiatique 10e série, 12: 312–38. [https://gallica.bnf.fr/ark:/12148/bpt6k932739].

But I don't think Shorthand should come into question for any itemType:journalArticle item that has an explicitly identified author and date, as is the case in Michaël's examples

RAC. Nilakanta Sastri, K. A. and M. Venkataramayya. “Telugu Chola records from Anantapur and Cuddapah.” EI 27, (1947-1948), pp. 220–251.

RAC. “Telugu Chola records from Anantapur and Cuddapah.” By K. A.Nilakanta Sastri and M. Venkataramayya. EI 27 (1947-1948), pp. 220–251.
michaelnmmeyer commented 4 months ago

This is implemented now. See e.g. http://dharmalekha.info/texts/INSPallava00402. There is no documentation yet.

arlogriffiths commented 4 months ago

I see there:

Capture d’écran 2024-05-17 à 18 24 19

To my mind the display (with "By Anonymous") is not yet satisfactory. Now I see that it must be due to "anonymous" having been entered in the Zotero entry itself:

Capture d’écran 2024-05-17 à 18 28 21

There is nothing in ZG to sanction this use of "anonymous".

Do I have everyone's permission to remove any such cases from our Zotero groub library?

@michaelnmmeyer : will the bit with "By ..." disappear in display if there is no author/editor etc. field with any contents?

michaelnmmeyer commented 4 months ago

@arlogriffiths Yes, the "By ..." will disappear.

Generally speaking, the "Author" field should be empty whenever none is given. It is the job of the bibliography processor to add "Anonymous" or "No author", etc.

michaelnmmeyer commented 4 months ago

The shorthand feature is now documented, under ZG §4.15 Extra.

arlogriffiths commented 4 months ago

Thanks a lot Michael.

I have searched for "anonymous" in our Zotero Group library and cleaned up a number of cases.

@danbalogh and @manufrancis : can I leave it to you to make any necessary modifications in the following three items?

Anonymous. 1847. “Translation of the Inscription in the Nagarjuni Cave, given in Plate X. of the Present Volume.” Journal of the Asiatic Society of Bengal 16: 594–96. = Siddham_Anonymous_1847

———. 1894. “Obituary Notices: Major-General Sir Alexander Cunningham.” Journal of the Royal Asiatic Society, New Series, , 166–77. = Anonymous1894_01, apparently not from Siddham but maybe Dan can deal with it anyhow

Rajavelu, S. 2008. “Recent Discoveries near Mamallapuram.” In Airāvati: Felicitation Volume in Honour of Iravatham Mahadevan, edited by anonymous, 177–90. Chennai: Varalaaru.com. = Rajavelu2008_01

manufrancis commented 4 months ago

Done for Rajavelu, S. 2008.

danbalogh commented 3 months ago

Done for Nagarjuni Cave and the Cunningham Obituary.

arlogriffiths commented 3 months ago

Thanks. So I presume this can now be closed. Feel free to reopen if I am wrong.