Re: [Crm-sig] Modelling bound manuscript copies

2017-10-27 Thread Pierre Choffé
t as far as I
>>> understood, the manuscripts in a codex have been purposely bound together.
>>> There can exist several codices with the same arrangement of manuscripts.
>>> 
>>> 
>>> I think in this context we could see the manuscripts a result of an
>>> industrial production. They are manual copies, hence are not unique in the
>>> way that I understand a F4 Manifestation Singleton to be unique (both
>>> intellectually and physically)
>>> 
>>> 
>>> 
>>> 
>>> Best wishes,
>>> 
>>> 
>>> Florian
>>> 
>>> 
>>> 
>>>> 
>>>> 
>>>> ___ 
>>>> Crm-sig mailing list 
>>>> Crm-sig@ics.forth.gr 
>>>> http://lists.ics.forth.gr/mailman/listinfo/crm-sig 
>>>> 
>>>> 
>>>> 
>>> 
>>> 
>>> On 26. Oct 2017, at 19:29, Christian-Emil Smith Ore < c.e.s@iln.uio.no
>>> > wrote:
>>> 
>>> A small question about a codex containing several manuscripts: Is there
>>> any relationship between the manuscripts (that is, the text they carry) or
>>> is it simply a handy way to handle several manuscripts?  The latter is the
>>> case for some Nordic Medieval codices where the codix is simply a batch of
>>> non related  texts.  
>>> 
>>> 
>>> 
>>> In the recent CRM SIG meeting it was a long dicussion if a manuscript
>>> could be seen as a result of a (production) plan and thus should be an
>>> item of an  F3 Manifestation Product Type.  If so what is the
>>> Manifestation Singleton realising the original expression of the codex
>>> manuscript. Would you claim that the codices are a result of an idustrial
>>> production, mutatis mutandis​?
>>> 
>>> 
>>> 
>>> Best,
>>> 
>>> Christian-Emil
>>> 
>>> *From:* Crm-sig < crm-sig-boun...@ics.forth.gr > on behalf of Florian
>>> Kräutli < fkraeu...@mpiwg-berlin.mpg.de >
>>> *Sent:* 26 October 2017 15:27
>>> *To:* crm-sig@ics.forth.gr
>>> *Subject:* [Crm-sig] Modelling bound manuscript copies
>>>  
>>> Dear all,
>>> 
>>> We're working on a CIDOC-CRM/FRBRoo model to represent a collection of
>>> Islamic manuscripts
>>> 
>>> It is organised into Codices. Further we have the concepts of Witness and
>>> Text. A Witness is a manuscript – a hand produced copy – of a Text. A
>>> Codex contains several Witnesses bound together.
>>> 
>>> A Codex can exist several times, similar to a copy of a book, and appear
>>> in catalogues of other collections. However, the copies of the Codices are
>>> hand-made, binding together several Witnesses.
>>> 
>>> Our difficulty when modelling this comes due to the definition of F5 Item
>>> and F4 Manifestation Singleton in FRBRoo. It would make sense to model our
>>> copy of a Codex as an F5 Item, being an example of F3 Manifestation
>>> Product Type. However, the scope note of F5 states that it is an object
>>> produced through an industrial process, e.g. printing. The physical texts
>>> that are bound together in a codex are however manual transcriptions. The
>>> definition of F4 Manifestation Singletons for the Witnesses is however
>>> also not appropriate, as we know several transcriptions of the same text
>>> exist. F5 Item would be more appropriate for our Witnesses, but does it
>>> apply in our case?
>>> 
>>> Another difficulty is when modelling the Codex as a binding together of
>>> physical manuscripts and the texts they hold. Our direction is to model a
>>> Codex as F15 Complex Work, that is realised in a F24 Publication
>>> Expression carried by an E84 Information Carrier. The Texts are then F14
>>> Individual Work (as members of F15) realised in F22 Self-Contained
>>> Expression (as components of F24). The Witnesses are  E84 Information
>>> Carriers that carry said F22 and P48 compose the E84 Information Carrier
>>> that carries the F24. We did not use F4 or F5 here. Does this make sense?
>>> (See sketch: 
>>> https://oc.rz-berlin.mpg.de/owncloud/index.php/s/AXJLkRmv0E00ecM
>>> )
>>> 
>>> Best wishes,
>>> 
>>> Florian
>>> 
>> 
>> 
> 
>

Re: [Crm-sig] Modelling bound manuscript copies

2017-10-27 Thread Florian Kräutli
no>> wrote:
> 
> A small question about a codex containing several manuscripts: Is there any 
> relationship between the manuscripts (that is, the text they carry) or is it 
> simply a handy way to handle several manuscripts?  The latter is the case for 
> some Nordic Medieval codices where the codix is simply a batch of non related 
>  texts.  
> 
> In the recent CRM SIG meeting it was a long dicussion if a manuscript could 
> be seen as a result of a (production) plan and thus should be an item of an  
> F3 Manifestation Product Type.  If so what is the Manifestation Singleton 
> realising the original expression of the codex manuscript. Would you claim 
> that the codices are a result of an idustrial production, mutatis mutandis​?
> 
> Best,
> Christian-Emil
> From: Crm-sig  <mailto:crm-sig-boun...@ics.forth.gr>> on behalf of Florian Kräutli 
> mailto:fkraeu...@mpiwg-berlin.mpg.de>>
> Sent: 26 October 2017 15:27
> To: crm-sig@ics.forth.gr <mailto:crm-sig@ics.forth.gr>
> Subject: [Crm-sig] Modelling bound manuscript copies
>  
> Dear all,
> 
> We're working on a CIDOC-CRM/FRBRoo model to represent a collection of 
> Islamic manuscripts
> 
> It is organised into Codices. Further we have the concepts of Witness and 
> Text. A Witness is a manuscript – a hand produced copy – of a Text. A Codex 
> contains several Witnesses bound together.
> 
> A Codex can exist several times, similar to a copy of a book, and appear in 
> catalogues of other collections. However, the copies of the Codices are 
> hand-made, binding together several Witnesses.
> 
> Our difficulty when modelling this comes due to the definition of F5 Item and 
> F4 Manifestation Singleton in FRBRoo. It would make sense to model our copy 
> of a Codex as an F5 Item, being an example of F3 Manifestation Product Type. 
> However, the scope note of F5 states that it is an object produced through an 
> industrial process, e.g. printing. The physical texts that are bound together 
> in a codex are however manual transcriptions. The definition of F4 
> Manifestation Singletons for the Witnesses is however also not appropriate, 
> as we know several transcriptions of the same text exist. F5 Item would be 
> more appropriate for our Witnesses, but does it apply in our case?
> 
> Another difficulty is when modelling the Codex as a binding together of 
> physical manuscripts and the texts they hold. Our direction is to model a 
> Codex as F15 Complex Work, that is realised in a F24 Publication Expression 
> carried by an E84 Information Carrier. The Texts are then F14 Individual Work 
> (as members of F15) realised in F22 Self-Contained Expression (as components 
> of F24). The Witnesses are  E84 Information Carriers that carry said F22 and 
> P48 compose the E84 Information Carrier that carries the F24. We did not use 
> F4 or F5 here. Does this make sense? (See sketch: 
> https://oc.rz-berlin.mpg.de/owncloud/index.php/s/AXJLkRmv0E00ecM 
> <https://oc.rz-berlin.mpg.de/owncloud/index.php/s/AXJLkRmv0E00ecM>)
> 
> Best wishes,
> 
> Florian



Re: [Crm-sig] Modelling bound manuscript copies

2017-10-27 Thread Pierre Choffé
 wrote:
>> 
>> A small question about a codex containing several manuscripts: Is there
>> any relationship between the manuscripts (that is, the text they carry) or
>> is it simply a handy way to handle several manuscripts?  The latter is the
>> case for some Nordic Medieval codices where the codix is simply a batch of
>> non related  texts.  
>> 
>> 
>> 
>> In the recent CRM SIG meeting it was a long dicussion if a manuscript
>> could be seen as a result of a (production) plan and thus should be an
>> item of an  F3 Manifestation Product Type.  If so what is the
>> Manifestation Singleton realising the original expression of the codex
>> manuscript. Would you claim that the codices are a result of an idustrial
>> production, mutatis mutandis​?
>> 
>> 
>> 
>> Best,
>> 
>> Christian-Emil
>> 
>> *From:* Crm-sig < crm-sig-boun...@ics.forth.gr > on behalf of Florian
>> Kräutli < fkraeu...@mpiwg-berlin.mpg.de >
>> *Sent:* 26 October 2017 15:27
>> *To:* crm-sig@ics.forth.gr
>> *Subject:* [Crm-sig] Modelling bound manuscript copies
>>  
>> Dear all,
>> 
>> We're working on a CIDOC-CRM/FRBRoo model to represent a collection of
>> Islamic manuscripts
>> 
>> It is organised into Codices. Further we have the concepts of Witness and
>> Text. A Witness is a manuscript – a hand produced copy – of a Text. A
>> Codex contains several Witnesses bound together.
>> 
>> A Codex can exist several times, similar to a copy of a book, and appear
>> in catalogues of other collections. However, the copies of the Codices are
>> hand-made, binding together several Witnesses.
>> 
>> Our difficulty when modelling this comes due to the definition of F5 Item
>> and F4 Manifestation Singleton in FRBRoo. It would make sense to model our
>> copy of a Codex as an F5 Item, being an example of F3 Manifestation
>> Product Type. However, the scope note of F5 states that it is an object
>> produced through an industrial process, e.g. printing. The physical texts
>> that are bound together in a codex are however manual transcriptions. The
>> definition of F4 Manifestation Singletons for the Witnesses is however
>> also not appropriate, as we know several transcriptions of the same text
>> exist. F5 Item would be more appropriate for our Witnesses, but does it
>> apply in our case?
>> 
>> Another difficulty is when modelling the Codex as a binding together of
>> physical manuscripts and the texts they hold. Our direction is to model a
>> Codex as F15 Complex Work, that is realised in a F24 Publication
>> Expression carried by an E84 Information Carrier. The Texts are then F14
>> Individual Work (as members of F15) realised in F22 Self-Contained
>> Expression (as components of F24). The Witnesses are  E84 Information
>> Carriers that carry said F22 and P48 compose the E84 Information Carrier
>> that carries the F24. We did not use F4 or F5 here. Does this make sense?
>> (See sketch: https://oc.rz-berlin.mpg.de/owncloud/index.php/s/AXJLkRmv0E00ecM
>> )
>> 
>> Best wishes,
>> 
>> Florian
>> 
> 
>

Re: [Crm-sig] Modelling bound manuscript copies

2017-10-27 Thread Florian Kräutli
 If so what is the Manifestation Singleton 
> realising the original expression of the codex manuscript. Would you claim 
> that the codices are a result of an idustrial production, mutatis mutandis​?
> 
> Best,
> Christian-Emil
> From: Crm-sig  <mailto:crm-sig-boun...@ics.forth.gr>> on behalf of Florian Kräutli 
> mailto:fkraeu...@mpiwg-berlin.mpg.de>>
> Sent: 26 October 2017 15:27
> To: crm-sig@ics.forth.gr <mailto:crm-sig@ics.forth.gr>
> Subject: [Crm-sig] Modelling bound manuscript copies
>  
> Dear all,
> 
> We're working on a CIDOC-CRM/FRBRoo model to represent a collection of 
> Islamic manuscripts
> 
> It is organised into Codices. Further we have the concepts of Witness and 
> Text. A Witness is a manuscript – a hand produced copy – of a Text. A Codex 
> contains several Witnesses bound together.
> 
> A Codex can exist several times, similar to a copy of a book, and appear in 
> catalogues of other collections. However, the copies of the Codices are 
> hand-made, binding together several Witnesses.
> 
> Our difficulty when modelling this comes due to the definition of F5 Item and 
> F4 Manifestation Singleton in FRBRoo. It would make sense to model our copy 
> of a Codex as an F5 Item, being an example of F3 Manifestation Product Type. 
> However, the scope note of F5 states that it is an object produced through an 
> industrial process, e.g. printing. The physical texts that are bound together 
> in a codex are however manual transcriptions. The definition of F4 
> Manifestation Singletons for the Witnesses is however also not appropriate, 
> as we know several transcriptions of the same text exist. F5 Item would be 
> more appropriate for our Witnesses, but does it apply in our case?
> 
> Another difficulty is when modelling the Codex as a binding together of 
> physical manuscripts and the texts they hold. Our direction is to model a 
> Codex as F15 Complex Work, that is realised in a F24 Publication Expression 
> carried by an E84 Information Carrier. The Texts are then F14 Individual Work 
> (as members of F15) realised in F22 Self-Contained Expression (as components 
> of F24). The Witnesses are  E84 Information Carriers that carry said F22 and 
> P48 compose the E84 Information Carrier that carries the F24. We did not use 
> F4 or F5 here. Does this make sense? (See sketch: 
> https://oc.rz-berlin.mpg.de/owncloud/index.php/s/AXJLkRmv0E00ecM 
> <https://oc.rz-berlin.mpg.de/owncloud/index.php/s/AXJLkRmv0E00ecM>)
> 
> Best wishes,
> 
> Florian



Re: [Crm-sig] Modelling bound manuscript copies

2017-10-27 Thread Pierre Choffé
Dear Florian, dear Christian-Emil,

I hope this finds you well. Just a few words about this interesting discussion 
if I may. If I understand well: 
* the codices we are talking about are collections  of manuscripts

* there can be multiple versions of the same codex
* there can exist other codices incorporating either part of, or all of the 
manuscripts plus other manuscripts, thus differentiating themselves from the 
above
am I wrong if I say we have 2 “industrial" production processes, one of 
manuscripts on the one hand, and one of codices on the other hand? These should 
be described separately.

It starts off simple: we have an author creating a Work (F15) realising an 
Expression (F22), and an original manuscript (F4 Manifestation Singleton) 
carrying the Expression.

Then at some point we have copies of the original manuscript. This is the first 
industrial process, resulting in the production of F5 Items (new manuscripts) 
R6 carrying a F24 Publication Expression, itself P165 incorporating the 
original F22 Self-Contained Expression (or not, or not exactly, but this is 
another discussion).

The second one is more complex. Being a “collection”, I would suggest that the 
codex is an F17 Aggregation Work (subclass of F14 and F16), which realises an 
F22 Self-Contained Expression (the expression of the collection) which itself 
P165_incorporates as many F22 as there are manuscripts. Note that a manuscript 
is a carrier, so it can carry one or multiple Expressions (e.g. poems).

There we have a second industrial process resulting in the production of F5 
Items (the codices) R6 carrying a F24 Publication Expression, itself P165 
incorporating the original F22 Self-Contained Expression (the expression of the 
collection).

If I have time I will try to make a schema, but I hope this is clear (and 
correct!).

I am sure there are complex cases, where the same manuscript can be found in 
completely different codices, or where manuscripts differ, etc. but I think 
this modelling allows for describing all sorts of situations. Or not?

Have a nice day, 

Pierre

On Fri 27 Oct 2017 at 09:45 "Florian Kräutli" < ">"Florian Kräutli" > wrote:

> 
> Dear Christian-Emil,
> 
> 
> Thanks for your reply. I will check back on this, but as far as I
> understood, the manuscripts in a codex have been purposely bound together.
> There can exist several codices with the same arrangement of manuscripts.
> 
> 
> I think in this context we could see the manuscripts a result of an
> industrial production. They are manual copies, hence are not unique in the
> way that I understand a F4 Manifestation Singleton to be unique (both
> intellectually and physically)
> 
> 
> 
> 
> Best wishes,
> 
> 
> Florian
> 
> 
> 
>> 
>> 
>> ___
>> Crm-sig mailing list
>> Crm-sig@ics.forth.gr
>> http://lists.ics.forth.gr/mailman/listinfo/crm-sig
>> 
>> 
>> 
> 
> 
> On 26. Oct 2017, at 19:29, Christian-Emil Smith Ore < c.e.s@iln.uio.no
> > wrote:
> 
> A small question about a codex containing several manuscripts: Is there
> any relationship between the manuscripts (that is, the text they carry) or
> is it simply a handy way to handle several manuscripts?  The latter is the
> case for some Nordic Medieval codices where the codix is simply a batch of
> non related  texts.  
> 
> 
> 
> In the recent CRM SIG meeting it was a long dicussion if a manuscript
> could be seen as a result of a (production) plan and thus should be an
> item of an  F3 Manifestation Product Type.  If so what is the
> Manifestation Singleton realising the original expression of the codex
> manuscript. Would you claim that the codices are a result of an idustrial
> production, mutatis mutandis​?
> 
> 
> 
> Best,
> 
> Christian-Emil
> 
> *From:* Crm-sig < crm-sig-boun...@ics.forth.gr > on behalf of Florian
> Kräutli < fkraeu...@mpiwg-berlin.mpg.de >
> *Sent:* 26 October 2017 15:27
> *To:* crm-sig@ics.forth.gr
> *Subject:* [Crm-sig] Modelling bound manuscript copies
>  
> Dear all,
> 
> We're working on a CIDOC-CRM/FRBRoo model to represent a collection of
> Islamic manuscripts
> 
> It is organised into Codices. Further we have the concepts of Witness and
> Text. A Witness is a manuscript – a hand produced copy – of a Text. A
> Codex contains several Witnesses bound together.
> 
> A Codex can exist several times, similar to a copy of a book, and appear
> in catalogues of other collections. However, the copies of the Codices are
> hand-made, binding together several Witnesses.
> 
> Our difficulty when modelling this comes due to the definition of F5 Item
> and F4 Manifestation Singleton in FRBRoo. It would make s

Re: [Crm-sig] Modelling bound manuscript copies

2017-10-27 Thread Florian Kräutli
Dear Christian-Emil,

Thanks for your reply. I will check back on this, but as far as I understood, 
the manuscripts in a codex have been purposely bound together. There can exist 
several codices with the same arrangement of manuscripts.

I think in this context we could see the manuscripts a result of an industrial 
production. They are manual copies, hence are not unique in the way that I 
understand a F4 Manifestation Singleton to be unique (both intellectually and 
physically)


Best wishes,

Florian

> On 26. Oct 2017, at 19:29, Christian-Emil Smith Ore  
> wrote:
> 
> A small question about a codex containing several manuscripts: Is there any 
> relationship between the manuscripts (that is, the text they carry) or is it 
> simply a handy way to handle several manuscripts?  The latter is the case for 
> some Nordic Medieval codices where the codix is simply a batch of non related 
>  texts.  
> 
> In the recent CRM SIG meeting it was a long dicussion if a manuscript could 
> be seen as a result of a (production) plan and thus should be an item of an  
> F3 Manifestation Product Type.  If so what is the Manifestation Singleton 
> realising the original expression of the codex manuscript. Would you claim 
> that the codices are a result of an idustrial production, mutatis mutandis​?
> 
> Best,
> Christian-Emil
> From: Crm-sig  <mailto:crm-sig-boun...@ics.forth.gr>> on behalf of Florian Kräutli 
> mailto:fkraeu...@mpiwg-berlin.mpg.de>>
> Sent: 26 October 2017 15:27
> To: crm-sig@ics.forth.gr <mailto:crm-sig@ics.forth.gr>
> Subject: [Crm-sig] Modelling bound manuscript copies
>  
> Dear all,
> 
> We're working on a CIDOC-CRM/FRBRoo model to represent a collection of 
> Islamic manuscripts
> 
> It is organised into Codices. Further we have the concepts of Witness and 
> Text. A Witness is a manuscript – a hand produced copy – of a Text. A Codex 
> contains several Witnesses bound together.
> 
> A Codex can exist several times, similar to a copy of a book, and appear in 
> catalogues of other collections. However, the copies of the Codices are 
> hand-made, binding together several Witnesses.
> 
> Our difficulty when modelling this comes due to the definition of F5 Item and 
> F4 Manifestation Singleton in FRBRoo. It would make sense to model our copy 
> of a Codex as an F5 Item, being an example of F3 Manifestation Product Type. 
> However, the scope note of F5 states that it is an object produced through an 
> industrial process, e.g. printing. The physical texts that are bound together 
> in a codex are however manual transcriptions. The definition of F4 
> Manifestation Singletons for the Witnesses is however also not appropriate, 
> as we know several transcriptions of the same text exist. F5 Item would be 
> more appropriate for our Witnesses, but does it apply in our case?
> 
> Another difficulty is when modelling the Codex as a binding together of 
> physical manuscripts and the texts they hold. Our direction is to model a 
> Codex as F15 Complex Work, that is realised in a F24 Publication Expression 
> carried by an E84 Information Carrier. The Texts are then F14 Individual Work 
> (as members of F15) realised in F22 Self-Contained Expression (as components 
> of F24). The Witnesses are  E84 Information Carriers that carry said F22 and 
> P48 compose the E84 Information Carrier that carries the F24. We did not use 
> F4 or F5 here. Does this make sense? (See sketch: 
> https://oc.rz-berlin.mpg.de/owncloud/index.php/s/AXJLkRmv0E00ecM 
> <https://oc.rz-berlin.mpg.de/owncloud/index.php/s/AXJLkRmv0E00ecM>)
> 
> Best wishes,
> 
> Florian



Re: [Crm-sig] Modelling bound manuscript copies

2017-10-26 Thread Christian-Emil Smith Ore
A small question about a codex containing several manuscripts: Is there any 
relationship between the manuscripts (that is, the text they carry) or is it 
simply a handy way to handle several manuscripts?  The latter is the case for 
some Nordic Medieval codices where the codix is simply a batch of non related  
texts.


In the recent CRM SIG meeting it was a long dicussion if a manuscript could be 
seen as a result of a (production) plan and thus should be an item of an  F3 
Manifestation Product Type.  If so what is the Manifestation Singleton 
realising the original expression of the codex manuscript. Would you claim that 
the codices are a result of an idustrial production, mutatis mutandis??


Best,

Christian-Emil


From: Crm-sig  on behalf of Florian Kräutli 

Sent: 26 October 2017 15:27
To: crm-sig@ics.forth.gr
Subject: [Crm-sig] Modelling bound manuscript copies

Dear all,

We're working on a CIDOC-CRM/FRBRoo model to represent a collection of Islamic 
manuscripts

It is organised into Codices. Further we have the concepts of Witness and Text. 
A Witness is a manuscript - a hand produced copy - of a Text. A Codex contains 
several Witnesses bound together.

A Codex can exist several times, similar to a copy of a book, and appear in 
catalogues of other collections. However, the copies of the Codices are 
hand-made, binding together several Witnesses.

Our difficulty when modelling this comes due to the definition of F5 Item and 
F4 Manifestation Singleton in FRBRoo. It would make sense to model our copy of 
a Codex as an F5 Item, being an example of F3 Manifestation Product Type. 
However, the scope note of F5 states that it is an object produced through an 
industrial process, e.g. printing. The physical texts that are bound together 
in a codex are however manual transcriptions. The definition of F4 
Manifestation Singletons for the Witnesses is however also not appropriate, as 
we know several transcriptions of the same text exist. F5 Item would be more 
appropriate for our Witnesses, but does it apply in our case?

Another difficulty is when modelling the Codex as a binding together of 
physical manuscripts and the texts they hold. Our direction is to model a Codex 
as F15 Complex Work, that is realised in a F24 Publication Expression carried 
by an E84 Information Carrier. The Texts are then F14 Individual Work (as 
members of F15) realised in F22 Self-Contained Expression (as components of 
F24). The Witnesses are  E84 Information Carriers that carry said F22 and P48 
compose the E84 Information Carrier that carries the F24. We did not use F4 or 
F5 here. Does this make sense? (See sketch: 
https://oc.rz-berlin.mpg.de/owncloud/index.php/s/AXJLkRmv0E00ecM)

Best wishes,

Florian


[Crm-sig] Modelling bound manuscript copies

2017-10-26 Thread Florian Kräutli
Dear all,

We're working on a CIDOC-CRM/FRBRoo model to represent a collection of Islamic 
manuscripts

It is organised into Codices. Further we have the concepts of Witness and Text. 
A Witness is a manuscript – a hand produced copy – of a Text. A Codex contains 
several Witnesses bound together.

A Codex can exist several times, similar to a copy of a book, and appear in 
catalogues of other collections. However, the copies of the Codices are 
hand-made, binding together several Witnesses.

Our difficulty when modelling this comes due to the definition of F5 Item and 
F4 Manifestation Singleton in FRBRoo. It would make sense to model our copy of 
a Codex as an F5 Item, being an example of F3 Manifestation Product Type. 
However, the scope note of F5 states that it is an object produced through an 
industrial process, e.g. printing. The physical texts that are bound together 
in a codex are however manual transcriptions. The definition of F4 
Manifestation Singletons for the Witnesses is however also not appropriate, as 
we know several transcriptions of the same text exist. F5 Item would be more 
appropriate for our Witnesses, but does it apply in our case?

Another difficulty is when modelling the Codex as a binding together of 
physical manuscripts and the texts they hold. Our direction is to model a Codex 
as F15 Complex Work, that is realised in a F24 Publication Expression carried 
by an E84 Information Carrier. The Texts are then F14 Individual Work (as 
members of F15) realised in F22 Self-Contained Expression (as components of 
F24). The Witnesses are  E84 Information Carriers that carry said F22 and P48 
compose the E84 Information Carrier that carries the F24. We did not use F4 or 
F5 here. Does this make sense? (See sketch: 
https://oc.rz-berlin.mpg.de/owncloud/index.php/s/AXJLkRmv0E00ecM)

Best wishes,

Florian