Re: [Indic-texts] question about use of @source in stanza + paraphrase translation dyads

I always use a note element as Andrew suggested but include directly in it bibl elements with corresp attributes for linking. Other attributees are also available. Multiple bibl elements can go in the note element. I would put just the text reference in the bibl with a corresp linking to its bibliography in a listbibl element in a listbiblfile. I would put the citation within that text in the biblScope element as a child of the bibl element. I’d use s-elements for text withing the p-element:

<l>kāntāravanadurgeṣu kr̥cchreṣv āpatsu sambhrame |</l> <l>udyateṣu ca śastreṣu nāsti dharmavatāṁ bhayam ॥28॥</l> </lg> <note><bibl corresp=“mbhgretil”>MBh.<biblScope corresp=“05,039.053”>5.39.53</biblScope></note> </quote> <s>lavan ta vaneh, riṅ hlət, riṅ alas, riṅ priṅga, riṅ laya, salvirniṅ duhkhahetu, ri papraṅan kunəṅ, tar tka juga ikaṅ bhaya, ri saṅ dhārmika, apan ikaṅ śubhakarma rumakṣa sira.</s> </p> I’d use the ana attribute instead of the met attribute for the meter name and save the met for the meter definition. But if this is a running commentary on the “basetext”, I would not use quote at all. I would put the base text in a different file coordinated with the commentary by identical xml:id attributes, and use quote only for your second type of quote. We hope to discuss commentary encoding at the WSC in Canberra. Yours, Peter ****************************** Peter M. Scharf, President The Sanskrit Library scharf@sanskritlibrary.org https://sanskritlibrary.org ******************************

On 2 May 2020, at 3:38 AM, Patrick McAllister wrote:

On Fri, May 01 2020, Andrew Ollett wrote:

...

...

...

It is more cumbersome, but I might recommend a <note> element, perhaps with a dedicated @type, embedded in the <quote> element, and this <note> element can provide bibliographic references and parallels (say with the <ptr> element). The advantage to this is that you can include multiple references, say to Indische Sprüche, various editions of the Mahābhārata, and so on, and you can also comment on the closeness of the quotation to each source (as far as I know <quote> only provides the @cert attribute, which would be interpreted to mean the certainty that the text in question is a quotation, which is not exactly what you mean).

Andrew is here suggesting a format that would easily lend itself to display in a PDF or HTML Format.

If you want a more “machine readable” approach, you could also consider using tei:linkGrp and tei:link elements (section 16.1 in the TEI Guidelines, https://www.tei-c.org/release/doc/tei-p5-doc/en/html/SA.html#SAPT)

That would allow you to formalize your connections and keep them apart from the base text (in an appendix, or a separate file). It could look like this:

<div xmlns="http://www.tei-c.org/ns/1.0"> <!-- this is just from your example, minus the @source and plus an @xml:id --> <p n="29"> <quote xml:id="id__1" type="basetext" xml:lang="san-Latn"> <lg n="28" met="anuṣṭubh"> <l>kāntāravanadurgeṣu kr̥cchreṣv āpatsu sambhrame |</l> <l>udyateṣu ca śastreṣu nāsti dharmavatāṁ bhayam ॥28॥</l> </lg> </quote> lavan ta vaneh, riṅ hlət, riṅ alas, riṅ priṅga, riṅ laya, salvirniṅ duhkhahetu, ri papraṅan kunəṅ, tar tka juga ikaṅ bhaya, ri saṅ dhārmika, apan ikaṅ śubhakarma rumakṣa sira. </p>

<p n="32"> <quote xml:id="id__2" type="basetext" xml:lang="san-Latn"> <lg n="31" met="anuṣṭubh"> <l>arjayej jñānam arthāṁś ca vidvān amaravat sthitaḥ |</l> <l>keśeṣv iva gr̥hītaḥ san mr̥tyunā dharmam ācaret ॥31॥</l> </lg> </quote> mataṅnya deyanika saṅ meṅət, apagəh kadi tan kneṅ pāti, lviraniran paṅarjana jñāna, artha, kunaṅ yan paṅarjana dharma, kadi katona rumaṅgut mastakanira, ta pva ikaṅ mr̥tyu denira, ahosanā palayvana juga sira.</p> <!-- This is what’s new --> <div> <linkGrp type="quotes-from-historical-texts"> <link target="#id__1 http://gretil/some/path/mbh.html#MBh05,039.053" ana="#arlo-thinks-it-is-a-quote #pada-c-differs"/> </linkGrp> <linkGrp type="general-parallels"> <link target="#id__2 bibl:Boehtlingk1870_01-94/32" ana="#todo-link-to-the-e-text"/> </linkGrp> </div> </div>

You wish to encode these features:

...

On Fri, May 1, 2020 at 1:39 AM Arlo Griffiths wrote:

...

Now to the specific problem on which I need your help: the correspondence with the reading of the @source is often not precise. In other words, even though a relationship with the @source is affirmed, we do not affirm that the reading of @source is identical to that of our text. In fact, I’d like to have a means of encoding that the correspondence is precise, which is rare, and maybe also that the correspondence is remote, which is also rare — the remoter the correspondence, the less likely that I’d want to cite the include a reference to the that source at all.

The example I provided uses the @type on the tei:linkGrp elements for the two broad categories you wish to distinguish. More details are supplied with the @ana on the individual links, which point to a list of useful interpretations that you supply somewhere else. You could also use another @type on the tei:link element for a tighter control on the possibilities (define a list of permitted types, for example).

In practice, I often find myself switching from a rather messy group of @ana attributes to a better regulated @type (and @subtype) attribute after I’ve encoded a big enough number of passages to get a feeling for the possibilities.

Unlike Andrew’s suggestion, this approach would require you to write a program to process those groups of tei:link elements, probably adding (where you think it’s right) the @source attribute to a tei:quote.

I would like to mention one other possibility, which I must add is a horrible hack that avoids the point of the TEI:

Instead of

you could add parameters to the @source attribute, which, after all, is just a (list of) URIs:

Best wishes,

-- Patrick McAllister long-term email: pma@rdorte.org _______________________________________________ Indic-texts mailing list Indic-texts@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/indic-texts