Understanding the phenomenology of reading through modelling

Tracking #: 2300-3513

Authors: 
Alessio Antonini
Mari Carmen Suárez-Figueroa
Alessandro Adamou
Francesca Benatti
François Vignale
Guillaume Gravier
Lucia Lupi

Responsible editor: 
Special Issue Cultural Heritage 2019

Submission type: 
Ontology Description
Abstract: 
Large scale cultural heritage datasets and computational methods for the humanities research framework are the two pillars of Digital Humanities, a research field aiming to expand humanities studies beyond specific sources and periods to address macroscope research questions on broad human phenomena. In this regard, the development of machine-readable semantically enriched data models based on a cross-disciplinary "language" of phenomena is critical for achieving the interoperability of research data. This contribution reports, documents, and discusses the development of a model for the study of reading experiences as part of the EU JPI-CH project Reading Europe Advanced Data Investigation Tool (READ-IT). Through the discussion of the READ-IT ontology of reading experience, this contribution will highlight and address three challenges emerging from the development of a conceptual model for the support of research on cultural heritage. Firstly, this contribution addresses modelling for multi-disciplinary research. Secondly, this work addresses the development of an ontology of reading experience, under the light of the experience of previous projects, and of ongoing and future research developments. Lastly, this contribution addresses the validation of a conceptual model in the context of ongoing research, the lack of a consolidated set of theories and of a consensus of domain experts.
Full PDF Version: 
Tags: 
Reviewed

Decision/Status: 
Major Revision

Solicited Reviews:
Click to Expand/Collapse
Review #1
By Roberta Ferrario submitted on 23/Oct/2019
Suggestion:
Major Revision
Review Comment:

I found the paper interesting and the subject definitely in scope of the special issue and my evaluation is overall positive. Nonetheless, there are some things that in my opinion need clarification (some of which are very important from my perspective), so I would recommend some re-working to provide it with a stronger foundation. The ontology is relevant as it addressed a central topic for the cultural heritage domain and it reuses and extends ontologies that are standard. Some concepts should be analysed more carefully though. The paper is well structured, except for the fact that some relevant information about the project are communicated only towards the end, while they could provide more context to the reader if given since the beginning. The paper is overall well written, with some minor typos to be checked. I hope my suggestions could be useful. I group them in general and specific comments below.

GENERAL COMMENTS

- I believe in the introduction some information about the motivations of the READ-IT project are missing, they only appear towards the end of the paper, but they could help the reader in contextualizing better the research were they given since the beginning. Also, I would add a paragraph or two explaining the foreseen applications of the READ-IT ecosystem, possibly with an example.
- Some of the most important concepts introduced are too ambiguous and they shouldn’t, as interoperability is the final aim. More on this in the specific comments below.
- Sometime in the paper a reference to theories about reading, experience and action is made, but no specific theory is ever mentioned, so it is difficult to see how the concepts have been developed, if based on the intuition of the modeler, on previous projects or also on theories. I believe that if theories have been used, these should be explicitly referenced, as they would provide a more solid foundation to the conceptual analysis.

SPECIFIC COMMENTS

Introduction

- p.2 the final paragraph of the section would be more readable if written in a more discursive way, rather than like a list of items with many redundancies between titles and descriptions.

Ontology Development Approach

- p.3 please describe better what the sources are already at this point (2.1).

Ontology Requirements

- p. 5 again, it is not crystal clear whether the listed types of sources are those already used by the project or sources that is possible to integrate in the future.

- p. 6 the authors say: “To be used as training sets, the research data must be integrated by making explicit derivable facts and incorporating validation of annotations by experts to differentiate them from automatic annotations.” This is very interesting and useful, but how is it done?

- p. 6 some details should be provided on how the analysis of the sources is performed. Which are the “units of meaning” under focus? Are they excerpts of descriptions of the experience of reading? If this is the case, better to put it explicitly.

- p. 6 I would add an example of cause/effect relations between a moment/episode/reading towards a reader’s condition. Can these three elements act at the same time? Is the effect intended to take place just while reading or also afterwards (or that would rather be called an “outcome”)?

- p. 6 concerning reader’s conditions, does the human/social situation include, besides the contingent situation (being on a train), also a more general one, connected to the “status” of the reader (age, gender, economic situation at the time of reading)? Concerning mental states, should also changes triggered by reading (and while reading) be considered, or also the mental state a reader is in before starting to read and after having read? If these are described in a source, why not?

- p. 6 the concept of “approach to reading” looks too wide, as it includes habits (towards reading in general, reading a type of content, on a specific medium, in a situation), opinions about topics, authors, contents, it appears to me that it is too vague as it is described here.

I understand this section just introduces the requirements, but I believe that a bit more systematic description, suggesting the categories that will be introduced later, would be more understandable for the reader.

Reused Ontologies

- p. 7 the assessment criteria should be described. For instance, what makes a reused ontology better in management of annotations and sources? In which sense an ontology describes better of another the content of sources? My impression is that the reused ontologies have been chosen based on the fact that they are standard, if this is not the case, some characteristics that make them better than others should be mentioned.

- p. 8 “The class socE Intention to Apply, specialization of scoE Mental Attitude, is related to the concepts of E39 Actor and socE Activity Plan as the intention of an actor toward of implementing a plan. In this frame, socE Intention to Apply could be used to encode the intention of reading, as a specialisation of State of Mind preceding and premise of a reading process.” I find this paragraph a bit hard to read, as it presupposes too much knowledge of socE. I would rephrase it at a more conceptual level.

- p. 8 why has FOAF been chosen? Which advantages does it bring?

Reading Experience Ontology

- p. 9 “The reading experiences we are representing with the experience ontology are annotations made on a variety of sources. The reason why we rely on annotation rather than on categorization is to enable researchers to address a wide range of problems in reading studies.” I really don’t understand these two sentences. How can reading experiences be annotations? In which sense? Why relying on annotation allows to address a wider range of problems than relying on categorization? And in which sense? Shouldn’t annotations be based on categorization? Otherwise, how can they bear some semantics?

- p. 9 in “approach to reading” it is said that the gaps have addressed through the analysis of existing theories. Which theories? And how do they help? Are some of the concepts used taken from these theories? Why have they been chosen? I believe this is an important point.

- p. 9 what is the reading in the reader’s mind? Is this a state of mind? Is sense-making a state of mind? How is this related to other states of mind, for instance emotions? Is this aspect treated in the ontology?

- p. 10 the concept of Reading Agent is an important one, if I’m correct, this has to be seen ontologically as a role (it is a description of conditions of a person). First, why necessarily a person? What about artificial agents? It seems to me that the ontology doesn’t need to commit to this, as long as there’s a change in the informational state of the artificial agent. Second, are conditions (ontologically) properties of the agent? Third, in the case of a public reading, how should we consider the listeners? There’s a sense in which we can model them as “Reading Agents”, since their states of mind can be changed by the reading, even if they are not practically those who read.

- p. 10 I’m not sure the distinction between material and immaterial captures the difference between Medium and Content, maybe referring to some ontology of informational objects could help. Also, can we say that all media are material? What about electronic media? I’m aware these can be considered material in a sense, but maybe this point deserves a footnote.

- p. 10 I find problematic defining an alteration as part of a resource, they seem to be ontologically distinct entities. Also, in which sense has a note to be considered an alteration of the content? If the reader adds a note on the margin of a book, they alter their reading experience, but do they really alter the content of the book?

- p. 10 at first I was confused by the claim that experiences are part of sessions, which are part of readings. Maybe the problem is just in the terms used, as experience seems to carry some implications (mental, emotional), that the other two carry less. Is this the case? Is it possible to have a reading experience of a whole session or of a whole reading.

- p. 11 I believe that the relation between Temporal Entity and Activity should be made more explicit for the reader who’s not too familiar with CIDOC, but more importantly, the relation between Reading_Process and reading should be made explicit, as the terminology doesn’t help.

- p. 11 the difference between engagement and transportation is not very clear. It seems that transportation is an engagement towards the immaterial component (the content). But this would imply that engagement can be also towards the material (the medium). Is this the case? If yes, I found it a bit counterintuitive. In other parts of the paper, transportation seems to involve emotions, while engagement not necessarily. For instance, I can engage in the interpretation of a difficult argument without been transported. Am I correct?

- p. 11 it is said that the experience is a change of the reader’s mental state. But I assume this is not the case for session and reading, so maybe this can be used as a specific property of experience distinguishing it from other parts of session or reading.

- p. 11 State_of_Mind represents a revision of the mental state of the reader. But how are called mental states that are not changes, for instance the fact that the reader was sad when they started the session?

- p. 11 it is not clear what the “Undefined” orientation is used for. Maybe an example could help.

- p. 11 if one can have State_of_Mind that precedes a Reading_Process, why are they represented as changes?

- p. 11 why is Reading_Frame defined as the union of Reading and Session? Wasn’t Session just a continuous and active part of Reading?

- p. 11 “In the first case, accordingly with the definition of Reading, Session and Experience, a State_of_Mind is evidence of experiences occurring during the Reading_Process.” This sentence makes me suspect that the word “experience” refers to two different concepts. If this is the case, it should be clarified.

- p. 12 it seems strange to define the aim of reading as an expectation. Should the reader always anticipate some effects of reading?

- p. 12 is remembrance necessarily about other readings or resources? Couldn’t it be about something external (like an event in the past), just triggered by the reading?

- p. 12 which is the ontological relation between events and situations?

- p. 12 is Reading Resource necessarily a material entity? So here the reference is more to the medium than to the content, is it intentional?

- p. 13 why it is said that there is no material interaction involving reader and reading? How can this happen without the medium (which can be material)?

- p. 13 which is the criterion to decide whether two events are comparable or not? BetterThan and WorseThan seem too vague, how can you model that a reading is better than another under one aspect, but worse under another?

Validation

- p. 14 it seems strange to me that the questions that guided the modeling are answered by the modelers themselves and used as validation. Maybe this methodological point can be clarified.

- p. 15 are posts, web novels, comics, comments, posters, example of content? Some of these (posts, posters) look more like media, others (comics, web novels) more like types of resources.

- p. 15 if suggestions and reviews are considered contents about other contents, rather than part of the content they are suggestions and reviews of, why are notes treated differently?

- p. 18 Fig. 15 p1 is two different readers, r1 and r2. This would support the idea that reader is a role. So, do the same person is a different reader with respect to different resources? Or to different readings (possibly of the same resource, or the same content carried by different media)? This is interesting, but should be clarified and put explicitly.

To conclude my review, I believe the paper is interesting and that it deals with an extremely complex topic, which brings many open issues. For this reason, I recommend acceptance with major revision, as I believe that the paper could benefit a lot from a more thorough conceptual analysis.

Review #2
By Carmen Brando submitted on 31/Oct/2019
Suggestion:
Minor Revision
Review Comment:

The present paper describes the design of the ontology READ-IT which models the reading phenomenon, i.e. the human’s experience of reading some content available on a physical medium. The topic is very original and seems to be of interest as it gathers many scholars from several disciplines in an european project. This ontology will be used as the backbone of a digital ecosystem which will provide scholars whose research topic is reading with a mean to collaborative- and semantically annotate their sources of study, both cultural heritage sources and online content (blogs, social media, ..). READ-IT reuses CIDOC-CRM, an event-based model, which is the standard for semantically enriching cultural heritage sources and also because the reading experience can be modeled in fact as an event. It also reuses WADM which is a W3C standard and is used for handling the source annotation aspects. The ontology has been implemented in OWL and validated, and is available online in a GitHub repository. The paper also describes the process of creating the ontology, i.e. the ontology engineering aspects.
The content of the paper is well structured so it's very comprehensible.

I am in favour of the publication of the paper after revising minor issues that I list below.

The idea of doing research on reading, for those who are not involved in this project, remains very abstract. Real examples about what the reading phenomenon is as well as the challenges and why scholars are interested in this topic arrive too late in the paper, at page 6. To further motivate the interest of this work, I would recommend to describe some examples in the introduction.

Previous (and simpler) ontologies on the subject have been developed and are currently being used (e.g. UK-RED), nevertheless the authors decide not to reuse RED nor LED terms. READ-IT must thus ensure compatibility with the existing developments, I believe the issue of change management should be addressed in the paper.

The ontology does not seem to be deployed for concrete use by the researchers of the projects, this may be the main weaknesses of this proposal.

Also, I would like to see a short subsection about the reasoning capabilities of this ontology.

The project’s aim is to address macroscope questions regarding the evolution of reading in Europe during the last three centuries, could you develop on these questions?.

Also, it would be useful to further develop in subsection 2.1 (analysis of sources) with examples of specific sources that have actually been studied during the design of the ontology..

In subsection 6.3.1, what does imply the following statement: Reader is a subclass of cdc:E3_Condition_State and therefore of cdc:E2_Temporal Entity?

I wonder, what about skimming (i.e. reading a text quickly to get a general idea of meaning), is this kind of reading modeled in the ontology?

The article can benefit of English proofreading.

Minor issues:
“The engagement of technical partners involved...”: what does technical implies in this context? Is it necessary to add “technical’ to partners? Do you mean ‘developers’?
“which are relevant for the model”: “which are relevant to the model”
The following phrase repeats itself, it’s redundant: “The sources are provided by researchers involved in READ-IT and represent a significative set of the different types of sources and reading experiences.”
“.. informed by the- ory”: avoid splitting the word
“.. integrates the perspectives of the different research groups”: what groups? this term must be defined.
“... the simple schema encoding of? the research data limits …”: the whole phrase needs to be reformulated..
”As musical performances and works .. after that ontology, combined with BIBO.”: to rephrase, otherwise, phrase too complicated
“This model to connects a.. ”: the verb is missing

Review #3
By Pietro Liuzzo submitted on 04/Nov/2019
Suggestion:
Major Revision
Review Comment:

While the quality and relevance of the described ontology are extensively documented with facts and methodologies, the illustration, clarity and readability of the describing paper are lacking and some major re-organisation and harmonisation of the content, together with few corrections need to be made.

My overall Reading Experience results in the impression that quite a lot of work needs still to be done by the authors on the style and text of this contribution, eventually to the point of deciding if really all of it should be in there or it may be better addressed to two or three different venues for example.

In the Introduction it is hard to follow the argument and for example how the common language and the specific question on the evolution of reading fit in relation to one another. It sounds like this is a summarised or excerpted version of some other type of formal document where keywords are more important than arguments. At such an important place in the paper structure as the introduction, we are offered generic statements about DH, which are cursory to the actual material in the paper, as they are nowhere else discussed or reused. The 9 (!!) sections of the paper are listed but a useful discourse to join them to one another is missing here as it is missing between one section and the other, leaving the reader only with this would-be-"self-explanatory" list, which is also somehow inconsistent in its structure, as the only guide through the paper. It is not self explanatory and I would suggest, if the authors decide to keep everything (which I would advise against), either here in the introduction or both here and in the text, to join the parts with simple sentences to help the reader transition from one point to the other and follow the description being made. Another useful thing which has been entirely omitted is a brief definition of the problem at stake, which emerges page 8. I did not even know that there was somebody studying "experiences of reading" until reading this paper and I honestly struggled to figure out what the authors where speaking about. The references provided are all ontology oriented (understandably) but this does not help framing the issue at hand.

In section 2, we are introduced to the ontology development. The image and its explanation help in a very limited way the understanding of the ontology without some minimal but strongly needed short examples. I would suggest to add one in each of the subsections of 2 so that for example we get to know which sorts of gaps are filled (see first sentence of 2.2). 2.3 results very vague, and still undecided between philosophy and project application.

Section 3 needs to be complemented with section 6.3 and section 9, where we get to read finally what the limitations are. This aspect of the relation with the RED ontology is unclear throughout the paper. It is said in the conclusion that this is a challenge in some respects. I would actually invite the authors to consider to move to this section some of the considerations left for the conclusion in the current version. And why is LED also discussed here?

Section 4 moves on to the ontology requirements and apparently ignores section 3, so that the projects we have just learned about are recalled again in a footnote to "Digital Humanities Research on cultural experiences". There are also strange beasts around like "cultural heritage sources" which is a concept I have never heard of mixing textual sources, cultural heritage objects in a strange way. The minuscule subsections should be perhaps downgraded to points in a list and made less elliptic, see the last sentence of 4.1.2 . The list of functional requirements additionally adds a bag full of concepts which are without any reference. These lists are hardly referred to in other parts of the article, and perhaps they may be omitted here if they cannot be properly connected to the descriptive argument.

In Section 5 page 8 col 1 the second paragraph starts with an incomplete sentence, in the second column there is a doubtful use of "toward of implementing a plan" and then some formatting caos between the caption of the image and beginning of section 6. Fig. 4 is very important for the rest of the argument in section 6 and should be recalled in more occasions and explained in its design. Where is for example the property connecting the reader to the reading process?

In section 6.3. the prefix used for CIDOC (cdc) should be introduced in the same way as FOAF.
:r1 is a Reader in Fig 7 but in Fig 9 is a Reading, while in Fig 10 :read1 is a reader. Perhaps it is worth sticking to the use of :r1 for the reader and perhaps :read1 for the process, if I have correctly understood? Fig. 10 is introduced as an example of :causeOfAlteration, but does not seem to me to exemplify that at all. Additionally in Fig. 10, I would have expected a property :involving to link to :m1 which would have been linked to :c1 by :providingAccessTo as in Fig. 8, instead I am presented with two (sub?) properties :involvingAMedium and :involvingAContent. On page 11, col 2, we meet an "eral" whose definition I failed to find. There is also an uncompleted paragraph starting with "Following the definition" and a strange "accordingly with" and later on page 12 "preceeding to" and "reading to 'killing time'".

The section on validation is interesting and instructive but still disjoint from the other sections and including some incomplete sentences like the third point of the third list in section 7.2.3 and the second paragraph of 7.3.2.

I find quite mysterious the usefulness of section 8 in general and its placement at this point secondly. section 8.1. should be part of 7 and a state of the art, although not so titled was already represented in my understanding by section 3, where also the paragraphs on related ontologies could be integrated. This section starts quoting a definition of reading in the LAWD ontology, which, however, has little or nothing to do with in my understanding to the conceptualisation being done in the ontology.

Some additional punctual remarks:

page 1 col 2, "quantitative and qualitative" should perhaps be "quantitative, and qualitative"
page 2 col 1, "humanistic research" I doubt the use of this adjective, although I am not sure what is meant. perhaps "research in the humanities" ?
page 2 col 1, "macroscope" should perhaps be "macroscopic"
page 2 col 2, "section 2 - 2." should be "Section 2 - " hyphens and en-dashes are mixed in this list
page 3 col 1 low, section 2.2 seems to start with a sentence which misses something, or its punctuation is misleading.
page 4 col 1, event (event) is the only one in the list without a footnote with a link.
page 9 col 1, "model to connects" is probably wrong
page 19 col 1, looks like the reference to Brewster et al. needs to be reformatted to [22].