Issue 623: Editorial check of CRMsci examples

ID: 
623
Starting Date: 
2022-11-07
Working Group: 
1
Status: 
Done
Background: 

The CRMsci editors have identified a number of examples that need revising. 

"Problematic" examples, where property labels randomly shift from the present to the past tense (f.i. O7, O13), have been marked as "not revised" (anyhting in highlight) or simply "missing".

Classes/ properties whose examples need be revised include: S9, S15, S20, S22, S23, O3, O4, O5, O6, O8, Ο10, Ο11, Ο13, Ο15, Ο17, Ο18, Ο19, Ο20, Ο21, Ο23, Ο25

Post by Thanasis Velios (4 November 2022)

Dear all,

Athina and I are making some progress with version 2.0 of CRMsci and we have identified some inconsistencies with property labels in the past and present tenses. Typically, the labels are in the present tense and the examples using the properties are in the past tense.

These are:

O7 contains or contained
O13 triggers or triggered

Is anyone aware of any relevant guidelines? From our understanding from CRMbase it seems that it should be O13 triggered and O7 confines. Please share any comments.

I think this also relates to the issue of temporal validity of properties.

All the best,

Thanasis
 

Post by Athina Kritsotaki (22 November 2022)

Hello all,

  I reformulate the issue, adding background information regarding a few classes, that would be helpful for the discussion:

Problematic" examples randomly shift from the present to the past tense (f.i. O7, O13), have been marked as "not revised" (anyhting in highlight) or simply "missing".

Classes/ properties whose examples need be revised include:
S9 (S9 Property Type: issue 332: It is postponed, it should be considered together with the issue related to redoing S4 - the review of the definition of this class has been postponed),
S15  (issue 332:It is postponed because the whole entity is under review),
S20 (issue 332: sig accepted the examples but asked Athina to improve the syntax of 4th example),
S22 (issue 332: the sig reviewed the scope note and decided to ask SS and MD to elaborate it further up to the next meeting. The example is rejected. We need an example of a ‘baulk’ from an archaeological record),
S23 (related to the open issue 612),
O3, O4, O5, O6, O8, Ο10, Ο11, Ο13, Ο15, Ο17, Ο18, Ο19, Ο20, Ο21, Ο23, Ο25

Athina 

Post by Christian-Emil Ore (8 December 2022)

ISSUE 623, FOL for O15 (see attached document)

 

I was asked to produce FOL for  the sentence 'This property is equivalent to P156 occupies (is occupied by) with domain E18 Physical Thing and range E53 Place.'  (marked in green). The equivalence is that O15 implies P156 when restricted to E18. The other way round is that P156 implies O15. P156 is a de facto sub property if O15 when seen from CRMsci

 

Christian-Emil

In the 55th joint meeting of the CIDOC CRM and SO/TC46/SC4/WG9; 48th FRBR/LRMoo SIG meeting, TV walked the SIG through some editorial issues that need to be resolved for the release of CRMsci v2.0.

Summary of topics discussed: 

  1. specify the location of an S2 Sampling activity (Introduction section, short description for Figure 3)
  2. Diagram and short description of S19 Encounter Event 
  3. Add/approve/edit examples for
    • S2 Sampling,
    • S14 Fluid Body,
    • S15 Observable Entity,
    • S20 Rigid Physical Feature,
    • S22 Segment of Matter,
    • O3 sampled from,
    • O4 sampled at,
    • O5 removed, 
    • O6 is former or current part of,
    • O7 confines,
    • O8 observed,
    • O10 assigned dimension,
    • O11 described,
    • O13 triggered,
    • O17 generated,
    • O18 altered,
    • O19 encountered object,
    • O20 sampled from typee,
    • O21 encountered at,
    • O23 is defined by,
    • O25 contains,
    • O28 is conceptually greater than
  4. Edit scope note of S3 Measurement by Sampling, O5 removed, O15 occupied   
  5. Update properties of S4 Observation 
  6. Citing papers in scope note formulations

Overall issue for v2.0 that’s still pending: The Present tense marking on O13 triggers (is triggered by). MD to indicate whether this is on purpose or a misnomer. See issue 620.
 

For the details of the discussions and the decisions made, plus any pending HW see the attached document

Belval, December 2022

 

Post by Wolfgang Schmilde (8 December 2022)

I would write it like this:

O15(x,y) ⇒ S10(x)
O15(x,y) ⇒ E53(y)
O15(x,y) ∧ E18(x) ⇔ P156(x,y)

Wolfgang

Post by Thanasis Velios (8 December 2022)

Thank you both. I am assuming that Wolfgang's formulation is a shorter equivalent to Christian-Emil's. I will add it to the draft.

Thanasis 

Post by Christian-Emil Ore (9 December 2022)

The compact notation is very good.

Chr-E

Post by Christian-Emil Ore (9 December 2022) --HWL: archaeological examples for S20, S22 (+ references)

 

S20 Rigid Physical Feature

Example:

The top surface of the clay floor A11 (Heterogeneous, yellow to grey silty clay; clear, wavy lower boundary)  (illu p. 1601, Croix et al, 2019)

S22 Segment of Matter  

Example:

the clay floor A11 (Heterogeneous, yellow to grey silty clay; clear, wavy lower boundary)  (illu p. 1601, Croix et al, 2019)

Reference

Croix et al, 2019: Single Context, Metacontext, and High Definition Archaeology: Integrating New Standards of Stratigraphic Excavation and Recording Sarah Croix & Pieterjan Deckers & Claus Feveile & Maria Knudsen & Sarah Skytte Qvistgaard & Søren M. Sindbæk & Barbora Wouters. Journal of Archaeological Method and Theory (2019) 26:1591–1631. https://doi.org/10.1007/s10816-019-09417-x  

Post by Wolfgang Scmidle (9 December 2022) -- FOL of O21 encountered at

Dear All,

I can now formulate more clearly the problem I have with the description of O21 "encountered at". If I understand it correctly, you are saying that the phenomenal place where the encountered things were observed to be present is within the phenomenal place of the S19 Encounter Event. I have no problem with this, and it seems comparable to saying that the phenomenal places of the origin and destinations of an E9 Move are within the phenomenal place of the move. 

But in the scope note you say "the E53 Place at which the S19 Encounter Event took place". I think it should be more clear that "took place at" is not refering to P7 "took place at" but to P161 "had spatial projection“. The FOL axiom is clear about this.

And in the example you are not using the phenomenal place but a larger declarative place, namely "Haifa Bay". According to the FOL axiom O21(x,y) ⇒ (∃z) [E53(z) ∧ P161(x,z) ∧ P89(y,z)], the consequence would be that the phenomenal place of the Encounter Event must contain Haifa Bay. I don't see why this should be the case, and I don't think this is what you want to express. Instead, you should take the examples in P160 as a blueprint. Concretely, the example should be something like
(The Sphaerosyllis levantina specimens) O21 encountered at (the phenomenal place) [P89 falls within "Haifa Bay", or more precisely, P89 falls within "soft sediments of mixed grain sizes in shallow waters in Haifa Bay"]

And finally about the FOL axiom O21(x,y) ⇒ (∃z) [E53(z) ∧ P161(x,z) ∧ P89(y,z)]: To make the axiom future-proof, we should have the shared reference space of y and z in mind. In its present form, the axiom claims that the existence of the phenomenal place y guarantees the existence of the (larger!) phenomenal place z of the Encounter Event in the same reference space. This is an ontological statement that may or may not be true, but neither here nor anywhere else have I seen an argument for it. (Note that this is not the same discussion as the one about P7(x,y) ⇒ (∃z) [E53(z) ∧ P161(x,z) ∧ P89(z,y)], where the inference arrow points from the larger to the smaller place.) So I think the existence of z needs to be assumed rather than inferred:
O21(x,y) ∧ E53(z) ∧ P161(x,z) ⇒ P89(y,z)

Curiously, I see exactly the same three problems with P26 "move to" and P27 "moved from". 

Best,
Wolfgang

Post by Thanasis Velios (5 January 2023) -- evote; examples of CRMsci 

Dear all,

Happy New Year! Please consider the following for an e-vote. This are the pending examples for CRMsci. 

Please vote YES if you are happy to include these examples, or vote NO and explain which ones are not good enough. For minor issues that you picked up please vote YES with a note for the correction.

All the best,

Thanasis

Post by Wolfgang Schmidle (6 January 2023) -- evote; examples of CRMsci

Dear Thanasis,

> Example for S15 Observable Entity:
>
>       • the flight of a male Bearded Vulture observed over Heraklion, Crete in the morning of the 24th of October 2020 (E5) (Claes, 2020)
> Claes, J. (2020) Bearded Vulture - Gypaetus barbatus, Observation.org. Available at:https://observation.org/observation/203043133/ (Accessed: 20 December 2022).

According to the linked webpage, the Bearded Vulture was observed near the village of Loukia, in the south of the regional unit of Iraklio.

> Example for S19 Encounter Event:
>       • the encounter of the marble floor of the Villa of the Papyri in Herculaneum during the digging of a well in 1750 (S19) (Koekoe, 2017)
> Koekoe, J. (2017) ‘Herculaneum: Villa of the Papyri – World History et cetera’, 17 January. Available at: https://etc.worldhistory.org/education/villa-papyri/ (Accessed: 20 December 2022).

The linked webpage doesn't mention a marble floor.

>       • the encounter of oak planks from a ship during a dig in a mound at the farm Lille Oseberg in Norway in 1908 (S19) (‘Oseberg Ship’, Wikipedia, 2022)
> ‘Oseberg Ship’ (2022) Wikipedia. Available at: https://en.wikipedia.org/w/index.php?title=Oseberg_Ship&oldid=1127078631 (Accessed: 20 December 2022).

According to the linked Wikipedia page it was excavated 1904–1905, so the encounter of oak planks was likely in 1904? And I assume from the example that these oak planks were the first part of the ship that was found, but the article doesn't say this.

> Example for O19 encountered object:
>       • The encounter of a marble floor during the digging of a well in 1750 (S19) encountered object the Villa of the Papyri in Herculaneum (E18). (Koekoe, 2017)
> Koekoe, J. (2017) ‘Herculaneum: Villa of the Papyri – World History et cetera’, 17 January. Available at: https://etc.worldhistory.org/education/villa-papyri/ (Accessed: 20 December 2022).

see above

>       • The encounter of oak planks from a ship during a dig in a mound at the farm Lille Oseberg in Norway, in 1908 (S19) encountered object the Oseberg Ship (E18). (‘Oseberg Ship’, Wikipedia, 2022)
> ‘Oseberg Ship’ (2022) Wikipedia. Available at: https://en.wikipedia.org/w/index.php?title=Oseberg_Ship&oldid=1127078631 (Accessed: 20 December 2022).

see above

> Example for O21 encountered at:
>
>       • The encounter of the Oseberg Ship in 1908 (S19) encountered at the farm Lille Oseberg in Norway (E53). (‘Oseberg Ship’, Wikipedia, 2022)
> ‘Oseberg Ship’ (2022) Wikipedia. Available at: https://en.wikipedia.org/w/index.php?title=Oseberg_Ship&oldid=1127078631 (Accessed: 20 December 2022).

I think the example should be like this:
The encounter of the Oseberg Ship in 1908 (S19) encountered at The phenomenal place of the encounter of the Oseberg Ship (E53) [P89 falls within: the farm Lille Oseberg in Norway]

The argument is the same as in the example of the Sphaerosyllis levantina specimens in Haifa Bay: Otherwise you would state that the phenomenal place of the Encounter Event contains the whole farm, which I don’t think is true.

> Examples for O31 has validity time-span:
>
>       • The measurement of the position of the Titanic by captain Smith after hitting an iceberg (S23) has validity time-span from 15 April 1912 23:40 to 15 April 1912 00:15 (E52) [This was the time-span between hitting the iceberg and ordering for a distress signal (time-span A). Captain Smith measured the position during a time-span B within time-span A. [The two time-spans can be related with property 'P86 falls within'] (Tikkanen, 2022)

The O31 scope note, version in the email from 24.11.2022:

> This property associates an instance of S23 Position Measurement with the instance of E53 Time-Span for which the measurement is valid. No inferences can be made in relation to the validity of the measurement outside this time-span despite the fact that some measured entities are relatively stable and their positions may remain the same after the measurement. The time-span of validity should fall within (P86 falls within (contains)) the overall time-span (P4 has time-span (is time-span of) of the process of measurement.

Did this change since then? Because in your example it seems to be the other way round, the time-span of validity (A) contains the time-span of the measurement (B).

In the same email, O31 is declared as a subproperty of P4 has time-span. But P4 denotes the phenomenal time-span of a temporal entity. This doesn’t seem to fit with either version? Or what would be the temporal entity whose phenomenal time-span is the validity time-span of the measurement?

And out of curiosity: How does the concept of a validity time-span relate to E3 Condition State ("the time-span for which a certain condition can be asserted may be shorter than the real time-span, for which this condition held“)?

Best,
Wolfgang

Post by Christian-Emil Ore (6 January 2023) -- evote; examples of CRMsci 

Dear all,

I am the one responsible for the Oseberg example. I made a typo, the planks where found in 1903 and of course not in 1908. Thank you, Wolfgang! I choose the Wikipedia as source. All texts with the information about how  the tenant (not the owner) at the farm Nedre Oseberg, found the oak planks are in Norwegian.

Phenomenal place: It is ok, but makes the example harder to read. I am not sure if we need to specify that it is a phenomenal place. I can of course give the geo-coordinates for the finding with an accuracy of 20 meters since the mound is easily identifiable and not very big.

Best,

Christian-Emil

Post by Wolfgang Schmidle (6 January 2023) -- evote; examples of CRMsci

Dear Christian-Emil,

> Am 06.01.2023 um 13:45 schrieb Christian-Emil Smith Ore <c.e.s.ore@iln.uio.no>:
>
> Dear all,
> I am the one responsible for the Oseberg example. I made a typeo, the planks where found in 1903 and of course not in 1908. Thank you, Wolfgang! I choose the Wikipedia as source. All texts with the information about how  the tenant (not the owner) at the farm Nedre Oseberg, found the oak planks are in Norwegian.

I really like the example, and I fondly remember the tour to the museum. But I think a cited source should contain the information given in the example, so I would suggest to replace either the source or the example. Similarly for the Villa of the Papyri: The example mentions the marble floor and the source doesn't, so I would suggest here as well to change either the source or the example.

> Phenomenal place: It is ok, but makes the example harder to read. I am not sure if we need to specify that it is a phenomenal place. I can of course give the geo-coordinates for the finding with an accuracy of 20 meters since the mound is easily identifiable and not very big.

We can shorten "phenomenal place" to "place" if that helps. And after P89 you can specify any place attestation you like, for example the burial mound or its geo-coordinates or the farm.

In my opinion this is part of a wider issue of cutting corners because "we all know what is meant", until someone doesn't. It's the same in the examples for E9 Move / P26 moved to / P27 moved from. And in the case of O21, mixing up the phenomenal place with a place attestation really leads to an incorrect statement about the phenomenal place of the encounter event.

Or take P179 had sales price, where the example "The purchase of 10 okka of nails by the captain A. Syrmas on 18th September 1895 (E96) had sales price 20 piastre (grosi) (E97)" suggests that "20 piastre" is directly the monetary amount, as opposed to "had sales price: Sales price of the purchase (E97) [20 piastre (grosi)]", if monetary amounts are supposed to behave like all the examples in the superclass E54 Dimension of E96 Monetary Amount.

If you have a string "20 piastre (grosi)" in a database field and are not interested in splitting it up into currency and value, then by all means just say "had sales price 20 piastre (grosi)". Or just say "the Tutankhamun exhibition moved to the Saatchi Gallery" if you don’t know or don’t care care about the fact that this place attestation is not the phenomenal place. But shouldn't at least the official examples be correct?

Best,
Wolfgang
 

Post by Martin Doerr (7 January 2023) -- evote; examples of CRMsci

I vote YES,

small correction of the first example (the rare bird was observed in the southern mountains of the district of Heraklion, not over the city):

"the flight of a male Bearded Vulture observed near Loukia, Heraklion, Crete in the morning of the 24th of October 2020 (E5) [Gypaetus barbatus, a threatened species in Crete] (Claes, 2020)"

         

Post by Christian-Emil Ore (8 January 2023) -- evote; examples of CRMsci 

I also vote YES.

The small corrections (like Martin's, my typo, Wolfgang's comments) are at 

an text editorial level and is left to the editors. If not we will never finish our work.

Best,

Christian-Emil

Post by Wolfgang Schmidle (8 January 2023) -- evote; examples of CRMsci

I agree with Christian-Emil and also vote YES.

Wolfgang

Outcome: 

Note from the Editors:

By the decision of Issue 637 the examples of S23 Position Measurement, O30 determined position, O31 has validity time-span, and O32 measured position of have been approved by the SIG and will form part of CRMsci v2.0. This is also the case for the proposed relabelling of S23 to Position Determination). 

Issue closed

8 March 2023

Meetings discussed: