Paper 61 (In-Use track)

Modeling and Preserving Greek Government Decisions using Semantic Web Technologies and Permissionless Blockchains

Author(s): Themis Beris, Manolis Koubarakis

Full text: submitted version

camera ready version

Decision: accept

Abstract: We present a re-engineering of Diavgeia, the Greek government portal for open and transparent public administration. We study how decisions of Greek government institutions can be modeled using ontologies expressed in OWL and queried using SPARQL. We also discuss how to use the bitcoin blockchain, to enable government decisions to remain immutable. We provide an open source implementation, called DiavgeiaRedefined, that generates and visualizes the decisions inside a web browser, offers a SPARQL endpoint for retrieving and querying these decisions and provides citizens an automated tool for verifying correctness and detecting possible foul play by an adversary. We conclude with experimental results illustrating that our scheme is efficient and feasible.

Keywords: Linked Open Data; Blockchain; Open Government; Semantic Web; Bitcoin; Tamper-proof; Public Services


Review 1 (by Anna Fensel)


The topic and direction of the paper is certainly very trendy and are of large interest to be discussed at the conference. The use case of attaining more transparency in decisions on public sector decisions with the employment of blockchain appears like a direction worth investigating (due to the distributed, decentralized, keeping track of the logs character of the blockchain). Employing Linked Data in this context also appears to be promising, as such data is often produced especially in the government sector.
The paper describes the implementation developments in this direction, which surely are interesting and can deliver important results (e.g. a conclusion on whether or not other countries should be implementing similar systems).
Having said this, there are also a few issues with this work in the current state:
The way how semantic technology has been applied in the addressed scenario, has already been published in detail at the last year’s ESWC conference: Chalkidis, I., Nikolaou, C., Soursos, P., & Koubarakis, M. (2017, May). Modeling and querying Greek legislation using semantic web technologies. In European Semantic Web Conference (pp. 591-606). Springer, Cham.
This makes the level of the contribution of this paper more incremental.
Most of the Greek data appears to be 1-star data (how it is also stated in the paper). It would be necessary to discuss the impact of this limitation, as well as how to overcome it, when it comes to the real-life deployment.
Other contributions (tools) are described in a very compact manner, and their specialties and contributions are vague. For example, when describing the web editor and visualizer, the necessity to design a new tool (instead of using/adapting some existing solutions) for entering the decisions is not motivated, the state of the art in this area is not described either. It is also unclear if such tool would be used and usable at all. 
The conclusions on that part are also well-known in the semantic community. The recommended N3 format has been employed very often in research as well as in practice, when the space is sparse, e.g. in scenarios with mobile applications. For example, see this publication describing such solution (a book chapter with an extended version of SEMANTiCS’08 conference):
While many submissions on this track are industry-driven, or are made in co-operation and co-authorship with industry partners, the level of “industrial” endowment of the work presented here is unclear.
The work appears to be on-going. The evaluation appears to look at the performance time only. This is not the key challenge in this use case in general, especially given the current maturity of semantic technology, and given the fact that there is only a very limited number of government decision taking place daily (actually, a small number, comparing to the volumes of the transactions the blockchain systems are handling in general). And this part is not what makes the use case most interesting, especially taking into account the semantic web in-use angle.
The ”take away message” that could guide further development of the semantic technology, for the semantic web community, is hence also not sufficiently elaborated.
Thank you for the response!
In this paper I have seen very little discussion on the semantic aspect of the technology (in contrary to the paper on the same work that has been published at ESWC a year ago). There were few ongoing developments mentioned (e.g. the editor), but not elaborated and explained in depth. Also the evaluation addressed the scalability aspect, where the connection to the semantic technology is unclear. 
Also I see that the use case has not been deployed even prototypically in reality, neither has involved real-life data.
Note that the papers of this track should be "evaluated on the basis of measurable impact of semantic technologies, and on the extent to which they address real-life problems" - here, I do not see these points as strong enough.
The paper should not go to "In-Use & Industrial Track" in the current state of the work here.
As I see now, other ESWC submissions in similar (and even better) conditions are gretting solid rejects from most reviewers in this track, so I have to update the score and lower my rank here.


Review 2 (by Sabrina Kirrane)


This paper describes how semantic technologies can be used together with the bitcoin blockchain in order to redesign Diavgeia, the Greek government portal, so that it affords the public greater transparency with respect to public decisions.
Contributions include the development of the Diavgeia ontology, a decision Web editor and Visualizer, a SPARQL endpoint that can be used to query decisions, and blockchain tools for publishing and verifying decisions recorded on the bitcoin blockchain. In addition, the paper includes an evaluation of the scalability of the proposed validator and details of the storage required.  
In terms of suitability of the paper for the in-use track, the paper clearly motivates the impact of Semantic technologies in terms of improved access to governmental decisions in Greece. While, from a reproducibility perspective, the paper includes links to the proposed ontology and source code.
However, although the paper includes an evaluation of the verifier tool and discusses disk space reduction, the paper is missing some important information. 
For example, what compression algorithm was used? Assuming that the N3 files contain the exact same information as the corresponding PDF, how does the N3 compression compare to the standard PDF compression techniques? Is it possible to query the compressed N3 files directly or if the data needs to be decompressed first?  
From an in-use perspective it would be good to know if the Greek government are planning to adopt the proposed system. Also, it would be interesting to include a usability evaluation of the proposed editors and visualizer and a discussion as to the potential challenges moving from the existing system to the proposed system.   
Minor comments:
-The poor contrast in figure 1 makes it difficult to read, especially when the paper is printed in black and white.
-Section 5 “Background on Bitcoin Blockchain” breaks the flow of the paper, consider adding a prelimenaries section before or after the related work or merging with Section 6 “Preserving Decisions using Bitcoin Blockchain”
Many thanks for the information provided in the rebuttal. If the paper, is accepted I would strongly suggest that you include a discussion of the current limitations of the existing compression mechanism and your plans to investigate HDT.


Review 3 (by Anastasia Dimou)


This is a paper which describes a new semantically enhanced version of the Diavgeia portal where all Greek government institutions upload their decisions. The paper introduces Diavgeia, describes the diavgeia ontology which was defined to represent the decisions and DiavgeiaRedefined which is the new portal. DiavgeiaRedifined primarily consists of a blockchain-based back-end and a SPARQL endpoint on its backend and a Web editor and visualizer on its front-end. I believe that the paper is a good match for the In Use track. It shows how Semantic Web technologies are used in a real world setting and to a certain extend compares the semantically enhanced version with the conventional approach.
However, the main concern I have is related to its use within this real world setting. That being said, the paper nicely presents the setup but it does not provide any proof that this is actually used or at least it is under deployment for production use. Within the paper there is no reference to the new portal, but only to the old one. There is only a link to the source code of the implementation but no references to the interface, neither a figure or a screencast. It is a nice use case but is this enough? Of course it is hard to wait until governments adopt such experimental technologies, but I assume that at least a pilot should be in place.
If the paper gets accepted, I'd strongly suggest to include at least an architecture figure and screenshot(s) or a reference to a screencast. I noticed that there is a (complementary?) repository which contains such information.
Minor comment: on page 2, the evaluation is not really a module of the DiavgeiaRefined project. Either the modules are described and the evaluation follows or it is not the modules which are described but the contributions of this paper.


Review 4 (by Tomi Kauppinen)


This is a very strong paper making use of a wide range of semantic technologies together with blockchain tools. Overall the novelty lies in the creative combination of all this tech when applied to support transparency about governmental decisions in Greece. The paper is evaluated in two different ways, both of which provide additional arguments to recommend accepting this paper as it is for ESWC. Great work!


Review 5 (by Anna Tordai)


This is a metareview for the paper that summarizes the opinions of the individual reviewers.
This paper is on a relevant topic for the conference, namely the application of block-chain and semantic technology for sharing data in the public sector. The authors share the ontology and code, linking to a GitHub repository. The main concern of reviewers is regarding the maturity of this paper, in particular in relation to the content of an earlier paper presented at ESWC 2017. Some of the reviewers also note the lack of detail about the portal, in particular lack of architecture figure(s), and missing details with respect to the evaluation and compression algorithm. Reviewers also note that the enhanced government portal isn’t deployed in real life setting.
Laura Hollink & Anna Tordai


Share on

Leave a Reply

Your email address will not be published. Required fields are marked *