linking separate documents on the same papyrus across DDB+HGV/DCLP divide

[paregorios]

A question for @rla2118 @jcowey @jds15 @rogerbagnall @hcayless @ryanfb ...

Consider the case of BGU 2 597. This papyrus seems to carry two documents, one presumably a case of reused whitespace:

• First there's a letter, which has been assigned TM 9253 and has been treated by HGV+DDB: http://litpap.info/ddbdp/bgu;2;597
• Then there's a hexameter, which has been assigned TM 100219 and has been treated by DCLP: http://litpap.info/dclp/100219

It's my naive expectation that, as a user of a DCLP-enabled papyri.info, the HTML display for one of these would alert me to the existence of the other. But that's not what I see happening at litpap.info.

1. Is that a reasonable expectation?

2. Are there examples now in papyri.info or litpap.info where such a linkage is made and, if so, how does that look?

There does not seem to be any specific content in either the HGV or the DCLP metadata for these documents that spells out the relationship. I note two potentially relevant aspects:

• In the HGV metadata, the value of <idno type="ddb-hybrid"> is "bgu;2;597." The same value is encoded in the DCLP metadata's <idno type="dclp-hybrid">.
• Trismegistos seems to know about the relationship, since the two entries are cross-linked sub "reuse."

3. Are there established metadata patterns for effecting the sort of semantically clear linkage I'm imagining?

[rogerbagnall]

This is indeed a reasonable expectation. I haven't been in any discussions of how to handle this, but I think that the TM knowledge of the relationship is the key to a solution, as it has been pretty attentive to the relationships between texts and objects. That's not to claim that it will always have picked it up correctly, but I think in principle it tries to. If I'm mistaken, the others can correct me. So I would look to drawing on TM information to deal with this, but I can't advise technically how one would do that.

[jcowey]

Just so that we are clear about what is on the ONE piece of papyrus. There is a private letter on the recto with its accompanying address on the verso (http://litpap.info/hgv/9253 = http://papyri.info/hgv/9253: see also http://berlpap.smb.museum/01860/ for the distribution of the lines of these two components). Also on the verso of the ONE piece of papyrus is one hexameter. It is written upside down in relation to the address of the letter. This "text" received the TM Number 100219 (note the LDAB number is 10835:: the fact that it differs from the TM number is an indication that this "text" was included fairly early in the life of LDAB). The DCLP metadata for it is to be found at http://litpap.info/dclp/100219. For the placement of this piece see http://berlpap.smb.museum/16469/. Note that the text of the hexameter WAS entered into DDbDP all those years ago (http://papyri.info/hgv/9253 = http://litpap.info/hgv/9253). There is no text in http://litpap.info/dclp/100219.

[jcowey]

@paregorios, you brought up the fact that Trismegistos knows of the "reuse" of the above mentioned papyrus to write the hexameter. That brings this issue into the area of the considerations raised by #259.

[ryanfb]

So is the question whether pn-mapping should know/use the TM "reuse" metadata, if it has access to it, vs. using something in the XML itself as in #259? Maybe something could write the TM "reuse" metadata into our XML as a one-time (or periodic) process, and then pn-mapping could make use of that normally?

[paregorios]

blocked by #259 per note from @jcowey , above

[paregorios]

The ability to do this has arisen as a key desideratum of participants in the user workshop in NYC (August 2017).