Issue 471: graphical examples

ID: 
471
Starting Date: 
2019-10-25
Working Group: 
4
Status: 
Open
Background: 

Posted by Martin on 28/1/2020

Dear  All,

Attached for your serious consideration the Winkelmann example as spacetime trajectories. I regard this representation as very important for the basic understanding of the introduction.

Firstly, I'd like your opinion about the utility of this graph. 

Secondly, you graphic competition (issue 457

Current Proposal: 

Posted by Carlo on 30/1/2020

The graph is very useful to draw, I’m in fact testing my narrative model by doing the same exercise on the happenings described in the Orlando Furioso, admittedly less relevant from an historical point of view, but decidedly more enjoyable from every other point of view (I must confess I can’t sometimes resist reading it loudly). I’ve extracted the events of the first book, now I am trying to draw them, which is what you are trying to do with Winckelmann. The only difference is that IPosted by Martin on 30/1/2020 have a narration, and I’ll be guided by that.

In my diagrams there are several characters, so I use one line for each character, but then I have to decide whether I use the 2 dimensions of diagrams for space, or I use one of them for time. Unlike you, I’m inclined to go for the first option, so I can use maps. Then I need a separate foil for each time instant, which implies very many foils so I need to decide which instants of time are worth a separate foil. I’ll follow the author, so I will draw a foil (or several foils) every time the author's narration slows down to describe in detail something, such as a place or a duel.

Then I’ll put everything in RDF and will link drawings to the events and to the poetry that describes them. The ontology is the one we have developed for narratives and further enriched in the Mingei project. Everything is mapped to CRM (actually, FRBRoo). So maybe in 20 years I may have something to show to the SIG (but during the coffee breaks only).

Sorry for the long description, as you can guess this is all very exciting to me.

Posted by Martin on 30/1/2020

Hi Carlo,

Great!! Could you provide us with some inspirations from your graphical language?

I prefer the time line, rather that snapshot sequences, didactically to make clear that simultaneity and linearity of causality is an illusion. It does not render the richness of space, but there is a good practice in archaeology to depict the spacetime volumes of periods in graphs with a sing space dimension.

I believe we should refer to such a publication in the introduction.

Posted by Carlo on 30/1/2020

I’ll take photos of my diagrams and send them to you, VERY confidentially. I have them at home. I do not believe they are ground-breaking artefacts, but since you asked. Maybe you can point me to more professional ones.

Posted by Carlo on 31/1/2020

Dear all,

I attach my graphical rendition of the first canto of Orlando Furioso. Every character is a continuous line. Although it incorporates space and time in a single foil, it does not let the reader appreciate the real temporal development of events, i.e., if you draw a vertical line that intersects all 5 characters you do not get the correct representation of who was where at that time. So it’s an approximation.

If you happen to know a professional way of doing this, I’d appreciate the info.

Posted by Carlo on 3/2/2020

Thank you Athanasios, it’s nice to see that there are others who take fiction seriously

Posted by Thanasis on 3/2/2020

It's not doing any better job at delivery correct answers when drawing a line but I could not help thinking of the following:

Andew DeGraff's posters:

http://www.andrewdegraff.com/moviemaps

Something a bit more relevant with time and events:

https://xkcd.com/657/large/

and a machine made version here:

https://www.abc.net.au/news/2015-12-16/star-wars-every-scene/7013826

Github code:

https://github.com/abcnews/d3-layout-narrative

I hope these are relevant.
 

Posted by Thanasis on 3/2/2020

Dear Martin,

I think the time aspect of the graph are clear. I struggled a bit with the spatial:

Why aren't Winckelmann's birth, death and writing of "History of the Art in Antiquity" aligned? Did he not write it in Germany?

Why do we use triangular left and right edges on the event boxes and not use plain rectangles? If they are meant to indicate fuzziness of time they should be on the vertical axis, no?

Also the "seeing the sculpture" event spans both Greece and Rome, shouldn't that be constraint in Rome only?

Posted by Martin on 6/2/2020

All correct!

Posted by Martin on 13/2/2020

Attached a proposal for another figure for the principles, now in a better layout. Please comment!
 

Posted by Martin on 13/2/2020

I'll produce a color proposal(Laokoon) for the whole CRM class hierarchy. It is not good to see only parts. For instance, Persistent Item will never appear in an instance diagram. Therefore its color is relatively irrelevant. 

Posted by George on 15/2/2020

I object to the colour proposal because there is already a community of use that has gone the other way and this unnecessarily diverges. It will mean people have to work 5 hats on to use CRM. Colour attribution is arbitrary and has already happened. Starting a change now requires a cascade down the chain of people who use CRM and is in practice non monotonic. 

Posted by Martin on 15/2/2020

Dear George,

Your arguments well taken, I still want to point out that there was no decision of CRM-SIG yet, that we have agreed on a competition and that concerns about the quality of published documents should be answered.

We talk about publication of the authoritative text, didactic material about the schema, and graphical representation of instances. we need to decide if we follow one or two styles.

We do also have a large set of graphics in the style presented in authoritative document. I would like to understand better, what the amount of material is in the community of use you talk about, and who is involved and what sort of material it is. We always do respect  communities of use. Whatever the decision, one has to adapt.

I do not agree that color attribution is arbitrary. If it were arbitrary, no standard would be needed.

So, please inform as better...

Posted by Martin on 15/2/2020

On 2/15/2020 3:54 PM, George Bruseker wrote:
> I object to the colour proposal because there is already a community of use that has gone the other way and this unnecessarily diverges. It will mean people have to work 5 hats on to use CRM. Colour attribution is arbitrary and has already happened. Starting a change now requires a cascade down the chain of people who use CRM and is in practice non monotonic.

Please also comment on the style to present instances, regardless color!

Martin

Posted by George on 16/2/2020

Dear Martin,

>
> Dear George,
>
> Your arguments well taken, I still want to point out that there was no decision of CRM-SIG yet, that we have agreed on a competition and that concerns about the quality of published documents should be answered.

It was my recollection that at the last meeting we actually agreed the colours and that this changed in our email thread. However, the competition is fine; I trust the process.

>
> We talk about publication of the authoritative text, didactic material about the schema, and graphical representation of instances. we need to decide if we follow one or two styles.
>
> We do also have a large set of graphics in the style presented in authoritative document. I would like to understand better, what the amount of material is in the community of use you talk about, and who is involved and what sort of material it is. We always do respect  communities of use. Whatever the decision, one has to adapt.

Groups that I know of that use this already are: LinkedArt, SARI, Pharos, CHIN, CIDOC conferences for the last three years, MASA in France. Documentation is not just done at the model level but at the mapping level, so we are talking about hundreds of diagrams with thousands and thousands of representations.

>
> I do not agree that color attribution is arbitrary. If it were arbitrary, no standard would be needed.

On colour, Goethe believed the following:

> Goethe also included aesthetic qualities in his colour wheel, under the title of "allegorical, symbolic, mystic use of colour" (Allegorischer, symbolischer, mystischer Gebrauch der Farbe), establishing a kind of color psychology. He associated red with the "beautiful", orange with the "noble", yellow to the "good", green to the "useful", blue to the "common", and violet to the "unnecessary". These six qualities were assigned to four categories of human cognition, the rational (Vernunft) to the beautiful and the noble (red and orange), the intellectual (Verstand) to the good and the useful (yellow and green), the sensual (Sinnlichkeit) to the useful and the common (green and blue) and, closing the circle, imagination (Phantasie) to both the unnecessary and the beautiful (purple and red).[26]

https://en.wikipedia.org/wiki/Theory_of_Colours#Goethe's_colour_wheel

whereas in Chakra theory the following is true

> It is believed that each Chakra has a particular colour. The base or Root Chakra is red, it makes you more alert, increases circulation to all tissues, warms and awakens. The second Chakra above, the Sacral Chakra, is orange, the color of joy and wisdom, energy and internal connection. By applying orange to you sacrum and choosing Yoga poses for this region may balance your digestion.
> The next above is the Solar Plexus Chakra and it is yellow corresponding to self worth, personality and ego. Green is for the Heart chakra, it is healing and loving by nature, making you grateful for giving and receiving. The colour of the Throat Chakra is blue used for matters relating to speech and communication. The Brow (sometimes known as the 3rd eye) Chakra and it is Indigo. Concentrating on the “3rd eye” point during Yoga practice and Meditation will help enhance self responsibility and intuition. The last is the Crown Chakra and it is bright white, although colors of violet and gold are also seen in this area. Violet energy brings about increased feelings of spiritual connection, intuition and mental clarity, often helpful in emotional problems.
https://powerofnature.wordpress.com/2007/08/09/colour-therapy-for-yoga/

My favourite philosopher had the following ideas on colour

> For by convention colour exists, by convention bitter, by convention sweet, but in reality atoms and the void

Here is a pop culture reference on the meanings of colours, pink means ‘girlish’ etc.

https://www.empower-yourself-with-color-psychology.com/meaning-of-colors...

Anyhow, there are strategic reasons for choosing the arbitrary colours we chose given prevailing cultural prejudice in the West towards what a colour might mean, which is what we originally based the extant colour division on. The parts of the model that are different in ontological nature should have sufficiently different colours in the spectrum so that they are not confused with one another. Different part of colour spectrum = different kind of thing. Red was not used at all because it is the colour of STOP, violence, do not go further, whereas blue was chosen as a central colour as it is easy to look at and promote tranquility. When staring at CRM graphs, we ought to facilitate such calm. Brown is an earthy colour connoting things of a more material nature. etc. etc.

Anyhow, I guess we should just save the colour conversation for the in person meeting since I think we will just end up having to vote. 

 

Posted by George on 16/2/2020

Dear Martin et al.,

 

Regarding the Laoocon example, I drew up the attached diagram. I couldn’t find a means of representing everything in a square formation, as it seemed that the progression of time as linear is important, so I took the freedom to go rectangular. Another small difference, is that I find it confusing to mix representations of things with things themselves, so I made the images instances of visual item so that this distinction was clear. Otherwise, it should bear the same content and the basic strategy for representing the instances remains the same (ie they are glued to the classes instead of getting nodes in themselves as is often done). For consideration.

 

 

Posted by Martin on 17/2/2020

Dear George,

Nice graphics. I would like to see the color choices actually in the same layout. Otherwise we confuse layout and color. The Laookon example does not have the additional representation links because I wanted to focus on the ones introduced in the text.

By the way, may I remind everybody that our SIG rule is, that no proposal can be decided in the same meeting it was proposed. It can only be accepted as a proposal. Decisions can only be taken on issues that have been formulated in a decidable form before the meeting, originally 14 days. No decision can be undone in the same meeting. If this is not fixed somewhere in text, it should be.

Posted by George on 17/2/2020

Dear Martin,

>
> Dear George,
>
> Nice graphics. I would like to see the color choices actually in the same layout. Otherwise we confuse layout and color. The Laookon example does not have the additional representation links because I wanted to focus on the ones introduced in the text.

Thanks! I attach another variant below which is less linear as another option with the idea to focus the encounter event. 

 

Here is the link to the file for all the graphs, it is open for people to experiment with colours and so on:

 

https://drive.google.com/file/d/1afVfdWWU3UMADkS_OkAcifJBxWMlVRem/view?usp=sharing

 

Ideally, if performing an experiment, it would be good to copy the particular tab that is being worked on to keep the different versions available.

 

 

Posted by Martin on 17/2/2020

Dear All,

For true comparison, please see attached George's colors and two proposals of mine. We need to compare exactly the same layout. Layout rules are a different issue...:-)

I have been referring to psychological experiments of the 1970's, studying reactions to colors. Indeed, blue lowers blood pressure and red raises. Red is a color of activity and alert. Not "stop", but attention. Red stands out most from all colors. Whatever hue of red. My point is the event orientation of the CRM. In a knowledge graph, does it immediately render the message? Is the contrast good enough? Is the difference worth discussing it?

To be discussed:-) 

Posted by Martin on 19/2/2020

Hi George,

Here is the pdf, and odp format of Laokoon graph

Posted by Nicola Carboni on 28/2/2020

Attached the RDF version of the Winkelmann example in TTL and RDF/XML.
Feel free to let me know if something has to be changed.
Best,
Nicola

In the 46th joint meeting of the CIDOC CRM SIG and ISO/TC46/SC4/WG9; 39th FRBR - CIDOC CRM Harmonization meeting; It was discussed together with issue 457.  The sig decided to pause this issue until after the release of CIDOC CRM v.7.0 (so 7.1 probably)

Athens, February 2020