Should epistolary numbers for letters be encoded as IDNOs so that they list with the Reference Numbers?

[davidamichelson]

See https://syriaca.org/work/3004

Where The Letters of Severus, II.12 is the same as Sixth Book of the Select Letters I.1.

There are other famous letters that have multiple numbers.

[dlschwartz]

This letter is part of two different ancient collections. The "Select Letters" collection, which has its own numbering, also mention the number of the same letter in the no-longer-extant ancient comprehensive collection. The multiple numbering systems note the different numbering of this same letter in the two different collections.

[wlpotter]

Yes but I think @davidamichelson was saying it might be a nice feature to add these as idno elements, so they show in the Reference Numbers section as they are a kind of standard identifier

[dlschwartz]

I don't think they quite count as idnos and both are already listed as relations. Moreover, there is a lot of work to do on these. It's been on my rader screen but I figure Phil's letters project will supersede whatever I'm able to do in the short run anyway. We can keep this issue open though.

[davidamichelson]

I think it would be useful to list letter numbers either as idnos or maybe as titles, certainly for other letter collections the letter numbers are the way scholars often know the letters. Such as Basil's Letter 2 to Gregory.

…

On Thu, Jul 2, 2026 at 3:23 PM Daniel Schwartz ***@***.***> wrote: *dlschwartz* left a comment (srophe/syriaca-data#1386) <#1386 (comment)> I don't think they quite count as idnos and both are already listed as relations. Moreover, there is a lot of work to do on these. It's been on my rader screen but I figure Phil's letters project will supersede whatever I'm able to do in the short run anyway. We can keep this issue open though. — Reply to this email directly, view it on GitHub <#1386?email_source=notifications&email_token=AAXQ37IJKPL735Z65FFZWLT5C3ACNA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTIOBXGAYTMMBRGAY2M4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJLDGN5XXIZLSL5RWY2LDNM#issuecomment-4870160101>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAXQ37LAGDRLZKBFMRYUHJL5C3ACNAVCNFSNUABEKJSXA33TNF2G64TZHMZTINBXGE2DOMZ3JFZXG5LFHM2DOOJXG4YDEOJXG2QXMAQ> . Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS <https://github.com/notifications/mobile/ios/AAXQ37PQYM7CMQU5YKQQN235C3ACNA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTIOBXGAYTMMBRGAY2M4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJKTGN5XXIZLSL5UW64Y> and Android <https://github.com/notifications/mobile/android/AAXQ37LLEULK6EL574AQYT35C3ACNA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTIOBXGAYTMMBRGAY2M4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJLTGN5XXIZLSL5QW4ZDSN5UWI>. Download it today! You are receiving this because you were mentioned.Message ID: ***@***.***>

[dlschwartz]

We should discuss. I think this would require a policy for which dcterms:isPartOf relationships also require an idno. Would this be a one-off for these letters? Would we have an idno for any such relationship for which the parts are numbers: Aphrahat's demonstrations, Ephrem's homilies on the Nativity, the Lives of the Eastern Saints, etc.?

Also, re these letters, we would probably only want one of the two ancient numbering systems, the Select Letters system. The other ancient system is little used and we often don't know this number for a letter. And unlike our current use of idno, these would need text to make them meaningful. It would need to indicate the source. I don't think an idno like "I.23" is all that helpful. We would need something like "Select Letters I.23" or "Comprehensive Collection III.17" or "PO 14 letter 29". This might be useful but it's a lot more text that we typically have in our idnos.

The PO letters present another problem. They are a grab bag of letters from various manuscripts (unlike the Select Letters which are an ancient collection). The numbering system we have was invented by Brooks in a haphazard fashion. Two pieces of one letter appear in two different parts of Brooks because he found more of one letter published in one volume after that volume was printed and published it in the second. That one letter has two Brooks PO numbers. Do we actually want to consider this an idno?

I trust Phil will be creating canonical numbering systems to sort all of this stuff out which is why I was wondering about waiting. Anyway, we can discuss.