IMOLA: interconnecting Land Registry in Europe

Overall this paper was very difficult to read. Many sections seem more like a loose collection of points and thoughts instead of a coherent description on a topic. It is unclear what the relevance is to this SI on human interactivity with the Semantic Web. This is unfortunate as I believe the main concept - the use of SW technology in an EU project to enable a common understanding of a topic (land registry) - is very interesting and relevant in general.

The abstract is wholly insufficient in describing the objectives, methods, and results of the paper.

The introduction is quite opaque and assumes a knowledge of legal terminology. Avoid references to legal cases without an elaboration on the case and its consequences - this is a CS journal. Parts of the introduction are taken from [https://www.elra.eu/imola-iii/] - on that note, the contents of [https://www.elra.eu/imola/] are much more understandable (I wish the authors had included it instead). After re-reading multiple times, I think the last paragraph starting on the first page is the crux of the matter - an information ecosystem is required that bridges the gap between national property rights and LR systems & EU laws - which necessitates a common semantic model. The rest (including what comes before in the text) seems to flow from that core requirement.

Most of Sections 2 and 3 seems like a summary of the bureaucracy behind the project and is irrelevant to the SW aspect. Sentences as the following make these sections quite opaque as well: "absence by the moment of a European legislative instrument referred to the Land Registry interconnection, such as one for Business Registers, (BRIS Directive), or Insolvency (Regulation on In- solvency Proceedings), bringing about that LRI action by the moment running voluntarily". Both sections seem to talk about the same thing (the need for LRI and IMOLA) and it is unclear how they differ. The first bullet at the end of section 2 is informative and I believe it should be the focus of this section, as it talks about "interoperability of the information and metadata (IMOLA Knowledge Repository) through [..] a common semantic model and controlled vocabularies [...] based on semantic web basis (sic)".

Regarding the rest of the paper, this was an exercise in finding and teasing out useful information. As far as I can determine, pivot points are abstract definitions of LR concepts; which are specialized by national concepts depending on different national legislation (?) There is a related discussion on comparative analysis between different conceptual property right schemes, and how this setup may facilitate it, but it is not elaborated further. ISA core vocabularies represent fundamental characteristics of an entity, and offer cross-domain semantic interoperability. The overall semantic model is called I-KOS. A custom tool called the "Knowledge Manager" was developed adapted to the LR domain and customized to work with I-KOS, but it is not elaborated. There is a KM browser but its relation to the prior KM tool is unclear.

I would recommend the authors to significantly rework the paper, cutting out filler (bureaucratic) parts, restructuring sections around a core topic, and shortening meandering unfocused sentences. I believe the end result could certainly warrant publication.

-- Other issues
There are many unclear and incorrectly formulated statements throughout the paper. Just a small list of examples:
"problems stemming from the increased demand for Land Registers information to the registration of foreign documents and judicial decisions"
"all EU Regulations demand to be effective a wide information system"
"enhancing from it the framework of judicial cooperation on civil and commercial matter"
"is the last stage to close the schema" - what does closing a schema mean?
"IMOLA Semantic Corpus comprises the pri-mary semantic rules and the methodology factors taken into account to determine them"

Words such as "claiming" and "pretend" have a negative connotation that I feel is unintended by the authors. E.g., "that pretends to harmonize the Land Registry information" - to pretend means "speak and act so as to make it appear that something is the case when in fact it is not."

There is actually a section called "5.2.3. Finally" - section titles are supposed to reflect their content!

It is not necessary to explain RDF(S) and OWL in a semantic web journal (p. 10)

IMOLA, LR acronyms should be fully spelled out when first used.

Figures 2 and 3 are dropped in the paper without any explanation.

To be blunt - the first part of Section 5.7 is an exercise in meaningless filler content.

SPARKLE -> SPARQL!

This paper presents the current achievements of the IMOLA project, which aims at providing interoperability among existing Land Registry information systems across Europe. To achieve this goal, the project developed a set of vocabularies, and especially a pivot one to which every IS will be mapped. The paper presents the project incentives, a lot of its achievements, and introduces both the vocabularies and the technical architecture.

However, the paper does not seem to target the research audience of the Semantic Web Journal: i) the author assumes a minimal knowledge about European principles and regulations as well as about LR but uselessly presents the basics of semantic web technologies, and ii) a very little part of it actually deals with semantic web and no particular research aspect in this field is tackled.

The vocabularies mentioned (but not referenced) in the paper are based on SKOS and no formal definition is provided. The paper is rather a project report than a scientific article. The project followed a classical approach regarding semantic interoperability and the paper neither justifies nor details it. Many references are missing and the work is barely evaluated, with no reproducible results given.

Hence, I don't think this paper should have been submitted as a full paper, as per the requirements at http://www.semantic-web-journal.net/authors#types It does not provide theoretical advances nor is the approach original. Instead, it sounds like an application "in real life" at the European scale, and that is the main strength of the paper. IMO, it would better fit in the "Reports on tools and systems" or "Application reports" types. In this case, the authors could shrink it by deleting redundancies and be more straightforward in the use of semantic technologies and their assessment in the project.

Moreover, I this this paper is at too early stage to be published, even as project report. I think the paper lacks:
- project facts (*website URL*, dates, partners, number of people involved...)
- a clear statement (competency questions) of the designed vocabulary
- a state of the art of semantic interoperability approaches and platforms in other European domains (finance and security are cited in the paper)
- a justification of the choices made in the project
- a clearer description of the project methodology and its different "increments" (workplan)
- an assessment of the usage of a FAIR methodology to produce the vocabularies, and their IRIs.

As there is no reference to the project website, I searched on the web and discovered that there were three IMOLA projects on the ELRA website. Which one(s) is this paper talking about?

Readability:
------------

The paper is quite heavy to read and to follow:
- there are many very long sentences,
- the structure of the paper is not straightforward and does not follow that of a research article,
- the titles of the sections are not informative,
- the goals of the project are uselessly repeated at the beginning of many sections,
- many contents in the figures are too small to be readable.

General remarks:
----------------

Keywords are missing

I think that the introduction and sections 2, 3 and 4 say roughly the same thing and could be condensed in a "context" section explaining the IMOLA project objectives.

Intro:
- it would be nice to have references or links to the organizations and projects cited and defined as acronyms.
- The notion of "adaptation principle" seems to be a key that is necessary for understanding the whole picture and IMO should be explained.
- p1 col 2: I am not sure what the "adaptation principle" is, nor why would one want to prevent the European immovables single market.
- Please add also links to the e-justice portal and a reference to the Court decision.
- At the end of the introduction, the reader understands the need of a unified register, but not that of a semantic one. Technically, a common platform or API would do the job.
- Also, the introduction does not meet the usual format: problem-contribution-agenda, but is more a first dive into the problem that is stated in section 2.

Sections 2 & 3:
- p3: "Figure 1 shows a detailed schema according to which the distribution of activities among different actors involved has been distributed on the IMOLA approach.": I could discuss the "detailed" aspect + from my understanding, this figure presents a larger view than the IMOLA approach.
- In Fig. 1, it is not clear to me what activities are in or out the limits of the common platform introduced in section 2 (p.2 col 2). Moreover, this figure seems more to me to gather software modules than human activities.
- Still regarding Fig. 1, I am not sure if all member states have a common, centralized, information system for LR and can provide a single adaptation interface to it with LRI. Is this a European regulation requirement?
- Whereas Fig. 3 is quite clear to me, I think that Fig. 2 needs more explanations than a single sentence, or that the reader could benefit more from a new on-purpose designed schema than from a reused one.
- Section 3 mostly rephrases the introduction and provides no new elements. Especially, at the end of section 3, I still do not have understood the rationale for using semantic technologies in this project.

Section 4:
- This section seems to provide some requirements (XML schemas) / general needs (user-friendly approach), but at a very general level.
- As well, I did not understand the § stating that professors have been involved, since there is no explanation nor reference to the cited methodologies. Hence, I do not understand the link with semantic technologies.
- Section 4 is supposed to deal with an incremental methodology, but I failed to understand this methodology, and especially its incremental aspects. It seems to me that it only presents some requirements and juridical work inspiring the project.
- There are very long sentences with complicated structures and I must admit I was lost at some points, even after reading them several times. For instance, at the end of section 4, not only that I did not understand the "formant methodology", but I am also unable to find out if it has been precised in the end of the sentence (after the semicolons).

Section 5:
- This section presents the whole contribution of the project. It is more than 12 pages long (paper overall length is 19p). I think it could be split in several sections, as it deals with theoretical (semantic), technical, user-centered and evaluation aspects of the project.
- Section 5.1 misses a lot of references to understand previous work, as well as references to the produced vocabularies. Are they issued using a FAIR methodology?
- As there are several vocabularies, I think a figure relating these vocabularies to one another (and if necessary other existing ones in the domain) would help the reader get the big picture.
- In Section 5.2, please do reference EIF, Insofar, ISA2 and TOOP.
- Using a pivot language is not innovative at all in the field of semantic interoperability, as it is known to reduce the number of mappings by one order of magnitude. Please remove this statement.
- In section 5.2.2, if ISA2 is not a vocabulary issued by the project, it should not be detailed in this section but in a state of the art one. Otherwise, please state how it is specific to the LR domain.
- Reading Section 5.3, I think that the author does not have the skills in semantic technologies to publish in this journal and is not aware of its audience. The overview of the semantic web is at too low level and reinventing the RDF acronym as "RDS" is not tolerable.
- I did not understand where the KM software comes from: paper reference 24 does not exist, and ref. 25 is a foundational paper from 1998, so none of them gives a clue about this system. From what I see on Fig. 9, it is the Protégé interface.
- Section 5.4 contains an unnecessary presentation of semantic technologies, but gets to the point of the article. The vocabularies defined earlier could have benefit from that to be more clearly defined.
- In Section 5.4:
- It was not necessary to define an "equivalence relation", as OWL already provides an owl:equivalentClasses property, which can be used in SKOS (see https://www.w3.org/TR/skos-reference/#L4858).
- As the author claims using RDFS, I cannot figure why, on Fig. 13, imola:07869504 is not modeled as an individual of a class named something like IndependentUnit? This also could be inferred automatically and would not require storing an extra boolean value.
- Given the title of Section 5.5, I expected to read a technical definition / architecture, but most of it presents the tool's user interface.
- Section 5.7:
- I think the 4 § at the beginning (before "Two levels of tests") are useless and could be removed.
- Given its title, I first thought this section would present an evaluation of the methodology (vocabs and/or platform), but as far as I understand, it presents a part of the process of publishing data in the platform. It is indeed a crucial phase, which I think should be stated at the beginning of the section and not at its end.
- There are plural terms (that do not conform to the quality criteria) like Encumbrances and Burdens in section 5.4.
- Section 5.8:
- Fig. 20 is useless as unreadable.
- I do not understand the link between the two sentences mentioning the XML Schema instance (please correct as well) file and the electronic template. BTW, what exactly does this template template? Hasn't this been done already in a previous step?
- I think Section 6 could be shrinked and added to the conclusion as a "perspectives" subsection.

Minor remarks:
--------------

p1: ELRA acronym should be between parentheses
p1 col2: the sentence starting with IMOLA is too long and barely understandable. In the next sentence, what does "they" stand for?
p2: "increasing increasingly": is that on purpose?
p3: "To accomplish these burdens is necessary a task distribution among different actors". Please correct syntax.
p3: the texts in the boxes in figure 1 are too small.
Fig. 2 & 3 are hardly readable, and some text items are even not readable at all given the image definitions.

p4: The very long sentence in §2 of section 4 (starting with "Due to") is hardly understandable and grammatically incorrect. As a general rule, kindly use bullet or numbered lists for enumerations rather than long sentences when talking to a computer science audience.
p5: what does ELRN CP stand for?
p7: please watch titles accuracy. For instance, section 5.2.3. is titled "Finally", which is not very informative in the middle of the paper.
p7: Fig. 5 caption states "based on XSD 1.0". XSD is an XML grammar; I guess the document is based on a schema instance, not on the grammar itself.
p10: please provide a legend for Fig. 10, explaining which terms come from which vocabulary.
p11: I don't understand the structure of the text in the 2 last § of column 1 (starting with "It helps").
p12: please do correct the spelling of "SPARKLE" (unless you designed a new query language)!
p15: 5.7, a.
- "semantical" -> "semantic"
- "because of" -> "according to"?
p16: Remove the extra quote at the end of the sentence beginning with "Mapping existing data models"... Or remove the whole sentence, as this is the goal of using a pivot language.