As discussed recently on Markup, the guidelines on textparts and milestones (including <cb/>, <pb/>) need some revision and expansion the better to cover various physical partitions present in the original documents.
Could someone dig out and point to the recent discussion on Markup, please? I'm not sure anyone here remembers what the revisions proposed involve, any more.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Sorry, I should have included a link in the first place. I meant this one: https://lsv.uky.edu/scripts/wa.exe?A2=ind2002&L=MARKUP&P=6445
Nothing particular or urgent, but you did mention that a revision of this topic was in the air and suggest a feature request. The link to the DHARMA Encoding Guide given in my previous message to the Markup list (https://docs.google.com/document/d/1hjWrrwRZQp4hmEqw4jBhhqoXdwJvRlw3EWboJteOPw0/edit?usp=sharing) still works, and I still think it may be a good idea for the wider community to adopt that approach to using <div type="textpart"> and <milestone> in particular situations (as per §3, "Marking up Extrinsic Structure" in that Guide). My <milestone type="pagelike"> can freely be replaced with <cb>, but since many of the breaks I call pagelike have nothing to do with columns, I prefer the former term as a more neutral one.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have revised and expanded the guidelines on textparts and milestones (trans-textpart, trans-nonstructural, trans-foreigntext) following the recent discussion on the Markup list (https://lsv.uky.edu/scripts/wa.exe?A2=ind2002&L=MARKUP&P=6445) and taking into account what suggested by the DHARMA Project Encoding Guide 1.0, in particular at §3.1, §3.3, §3.4, §3.5, §7.2.1, following the basic principle of avoiding <div>s for physical but not semantic breaks.
It would be good to add some more examples from existing projects, e.g. from DHARMA...
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello, thanks for your work on this. I'm happy to provide some actual examples from texts in DHARMA (and some that I initially encoded for Siddham, now revised to DHARMA norms), but most of our partition encoding has more detail than that mentioned in the EpiDoc guidelines. I'm not sure if that is a problem when using these for an illustration; at any rate, feel free to remove any unwanted detail from these examples. I have already manicured them a little to reduce the clutter of irrelevant markup. Note that we do not wrap the whole of an inscription in <ab>, but normally use multiple <p> elements to encode the semantic structure of the text, alternating with <lg> where the text is in verse, and using <ab> only in rare cases such as small chunks of text that are less than a proper sentence. I'm also not sure how best to give you examples, so with apologies for swamping this thread, I'll just dump some right here.
Textparts for physically and semantically separate items (set of inscribed copper plates with an inscribed seal attached to the ring that binds the plates together):
Milestones for a "pagelike" partition: two connected fragments where the break is fortuitously between lines, so that no epigraphic line is interrupted by the fracture. The text on the second fragment is the continuation of that on the first, so textparts were not deemed appropriate.
Another case of milestones for "pagelike" partitions: the text is inscribed on two separate doorjambs. Text continues from the bottom of one to the top of the other just as if a page of paper had been filled and continued on the next page, so again textparts are not appropriate. The inscription begins with a series of verses in Sanskrit, followed by Khmer prose (starting at some point on the second doorjamb), so this is also an illustration of encoding language on the edition div, and encoding a different language separately on specific lg elements.
Milestones for "gridlike" partitions: two fragments with the break cutting across lines, so that the milestone for fragment A is iterated at the beginning of each epigraphic line, and that for fragment B at some point within each epigraphic line.
Thank you very much for the examples Dániel! For the sake of internal coherence in the Guidelines I would replace the<p>s with <ab>s, and perhaps also omit the type="pagelike" attribute: would it be ok for you?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
My pleasure. If there are any other situations for which you would like an example (e.g. one specifically for <pb/>, without the added complication of textparts?), please let me know and I may be able to supply one.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Ah, yes, please: it would be good to provide a specific example of what could be defined as a "page" in an epigraphic context, requirying a <pb/> at its beginning.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Copper plate "inscriptions" are very typical of India, though as far as I know, only occasionally found outside the sphere of Indian cultural influence. Most of these are records of donation (usually of tax-free land to priestly people) and they are more like legal documents than inscriptions in the narrower sense of a public inscription on a monument. Nonetheless, they are routinely considered to be epigraphic records. Some such records, especially among the earliest ones, are single plates inscribed on one side or both, and often with a royal seal soldered on one side. But the typical form is several plates bound together with a ring passed through a hole on each plate, and a seal soldered to that ring. The sides of these plates are inscribed just as the pages of a written book would be, so <pb/> is the natural choice of encoding for them. In DHARMA we make it a rule to use @break="no" on <pb/> elements where applicable, in spite of the same being redundantly present on the adjacent <lb/>. The numbering scheme we use is "1r", "1v", "2r", etc. We have also decided always to encode <pb/> elements for blank pages of such sets, as you can see in the example below at the very beginning and the very end. The outer (first and last) pages of such sets were often left blank because they would have been prone to wear and damage.
If you are willing to include something about this in the Guidelines, here is an image of such a set of plates that may be helpful: https://commons.wikimedia.org/wiki/File:BLL826_INDCH64R_01_L.jpg
There is also a schematic image made by me in "Case study 2A" in Appendix C of the DHARMA Encoding Guide (link above), with images, dummy text and encoding illustration, which you can feel free to lift and use.
Finally, here is a real-life example of an encoded set of copper plates, pared down to just a few bits of text around the page breaks.
Could someone dig out and point to the recent discussion on Markup, please? I'm not sure anyone here remembers what the revisions proposed involve, any more.
Sorry, I should have included a link in the first place. I meant this one: https://lsv.uky.edu/scripts/wa.exe?A2=ind2002&L=MARKUP&P=6445
Nothing particular or urgent, but you did mention that a revision of this topic was in the air and suggest a feature request. The link to the DHARMA Encoding Guide given in my previous message to the Markup list (https://docs.google.com/document/d/1hjWrrwRZQp4hmEqw4jBhhqoXdwJvRlw3EWboJteOPw0/edit?usp=sharing) still works, and I still think it may be a good idea for the wider community to adopt that approach to using
<div type="textpart">and<milestone>in particular situations (as per §3, "Marking up Extrinsic Structure" in that Guide). My<milestone type="pagelike">can freely be replaced with<cb>, but since many of the breaks I call pagelike have nothing to do with columns, I prefer the former term as a more neutral one.I have revised and expanded the guidelines on textparts and milestones (trans-textpart, trans-nonstructural, trans-foreigntext) following the recent discussion on the Markup list (https://lsv.uky.edu/scripts/wa.exe?A2=ind2002&L=MARKUP&P=6445) and taking into account what suggested by the DHARMA Project Encoding Guide 1.0, in particular at §3.1, §3.3, §3.4, §3.5, §7.2.1, following the basic principle of avoiding
<div>s for physical but not semantic breaks.It would be good to add some more examples from existing projects, e.g. from DHARMA...
Hello, thanks for your work on this. I'm happy to provide some actual examples from texts in DHARMA (and some that I initially encoded for Siddham, now revised to DHARMA norms), but most of our partition encoding has more detail than that mentioned in the EpiDoc guidelines. I'm not sure if that is a problem when using these for an illustration; at any rate, feel free to remove any unwanted detail from these examples. I have already manicured them a little to reduce the clutter of irrelevant markup. Note that we do not wrap the whole of an inscription in
<ab>, but normally use multiple<p>elements to encode the semantic structure of the text, alternating with<lg>where the text is in verse, and using<ab>only in rare cases such as small chunks of text that are less than a proper sentence. I'm also not sure how best to give you examples, so with apologies for swamping this thread, I'll just dump some right here.Textparts for physically and semantically separate items (set of inscribed copper plates with an inscribed seal attached to the ring that binds the plates together):
Milestones for a "pagelike" partition: two connected fragments where the break is fortuitously between lines, so that no epigraphic line is interrupted by the fracture. The text on the second fragment is the continuation of that on the first, so textparts were not deemed appropriate.
Another case of milestones for "pagelike" partitions: the text is inscribed on two separate doorjambs. Text continues from the bottom of one to the top of the other just as if a page of paper had been filled and continued on the next page, so again textparts are not appropriate. The inscription begins with a series of verses in Sanskrit, followed by Khmer prose (starting at some point on the second doorjamb), so this is also an illustration of encoding language on the edition div, and encoding a different language separately on specific lg elements.
Milestones for "gridlike" partitions: two fragments with the break cutting across lines, so that the milestone for fragment A is iterated at the beginning of each epigraphic line, and that for fragment B at some point within each epigraphic line.
Inscription text in Sanskrit, with a chunk (semantic paragraph) in Telugu:
Thank you very much for the examples Dániel! For the sake of internal coherence in the Guidelines I would replace the
<p>s with<ab>s, and perhaps also omit thetype="pagelike"attribute: would it be ok for you?Dear Irene, thanks, both changes are perfectly all right.
Great, thank you. I have added all the examples above in the relevant pages of the Guidelines.
My pleasure. If there are any other situations for which you would like an example (e.g. one specifically for
<pb/>, without the added complication of textparts?), please let me know and I may be able to supply one.Ah, yes, please: it would be good to provide a specific example of what could be defined as a "page" in an epigraphic context, requirying a
<pb/>at its beginning.Copper plate "inscriptions" are very typical of India, though as far as I know, only occasionally found outside the sphere of Indian cultural influence. Most of these are records of donation (usually of tax-free land to priestly people) and they are more like legal documents than inscriptions in the narrower sense of a public inscription on a monument. Nonetheless, they are routinely considered to be epigraphic records. Some such records, especially among the earliest ones, are single plates inscribed on one side or both, and often with a royal seal soldered on one side. But the typical form is several plates bound together with a ring passed through a hole on each plate, and a seal soldered to that ring. The sides of these plates are inscribed just as the pages of a written book would be, so
<pb/>is the natural choice of encoding for them. In DHARMA we make it a rule to use@break="no"on<pb/>elements where applicable, in spite of the same being redundantly present on the adjacent<lb/>. The numbering scheme we use is "1r", "1v", "2r", etc. We have also decided always to encode<pb/>elements for blank pages of such sets, as you can see in the example below at the very beginning and the very end. The outer (first and last) pages of such sets were often left blank because they would have been prone to wear and damage.If you are willing to include something about this in the Guidelines, here is an image of such a set of plates that may be helpful: https://commons.wikimedia.org/wiki/File:BLL826_INDCH64R_01_L.jpg
There is also a schematic image made by me in "Case study 2A" in Appendix C of the DHARMA Encoding Guide (link above), with images, dummy text and encoding illustration, which you can feel free to lift and use.
Finally, here is a real-life example of an encoded set of copper plates, pared down to just a few bits of text around the page breaks.
Wonderful, thanks! I have inserted this one too.