• Andrea Omicini
    Andrea Omicini, 16/02/2021 12:33

    Dear Andrea Omicini,

    thank you for having submitted to JELIA 2021. We are happy to inform
    you that your paper

    Lazy Stream Manipulation in Prolog via Backtracking: the Case of 2P-Kt (submission 45)

    has been accepted for presentation at JELIA 2021. Your paper will be
    included in the proceedings and presented at the conference. The
    reviews are appended. If the reviews indicate changes to be made for
    the final version, please make your best effort to improve your paper
    and submit the camera-ready version by

    15th March 2021.

    You will receive more detailed instructions on how to submit the
    final version of your paper in due time.

    Springer communicated the following:

    "Authors should consult Springer's authors' guidelines and use their
    proceedings templates, either for LaTeX or for Word, for the
    preparation of their papers. Springer encourages authors to include
    their ORCIDs in their papers. In addition, the corresponding author of
    each paper, acting on behalf of all of the authors of that paper, must
    complete and sign a Consent-to-Publish form. The corresponding author
    signing the copyright form should match the corresponding author
    marked on the paper."

    Forms, templates, guidelines etc. can be found on Springer's LNCS page:
    https://www.springer.com/gp/computer-science/lncs/conference-proceedings-guidelines

    Springer also wrote:
    "If you or any of the authors contributing to JELIA 2021 are interested
    in Open Access or Open Choice, please refer to our webpage
    https://www.springer.com/gp/computer-science/lncs/open-access-publishing-in-computer-proceedings .
    We would need the address(es) for the invoice(s) and the CC-BY
    Consent-to-Publish form(s) at the same time as the files for the
    publication."

    We are looking forward to see you at JELIA 2021! Due to the pandemic,
    the precise modality of the conference is not yet clear, but we will
    of course announce this as soon as a decision is taken. 

    Wolfgang Faber, Gerhard Friedrich, Martin Gebser
    JELIA 2021 Program Chairs

    SUBMISSION: 45
    TITLE: Lazy Stream Manipulation in Prolog via Backtracking: the Case of 2P-Kt


     REVIEW 1 -
    SUBMISSION: 45
    TITLE: Lazy Stream Manipulation in Prolog via Backtracking: the Case of 2P-Kt
    AUTHORS: Giovanni Ciatto, Roberta Calegari and Andrea Omicini


     Overall evaluation -
    SCORE: 1 (weak accept)


     TEXT:
    The authors describe the integration of streams into Prolog. 

    The paper is easy to read. Some parts could be shortened to be straighter to the point. 

    The paper describes the basic concepts and gives an idea of the design and engineering of the implementation. 

    For potential users of the system, I would suggest focusing more on the application. In particular, the driving example is not what I would expect as a typical example for stream reasoning. I recommend looking at other work in the area of logic programming and stream reasoning and picking up some showcases. Reusing examples would also be helpful for comparison reasons. 

    Extensions to deal with time are greatly appreciated. 

    The most significant question mark is the relation to existing old work in Prolog, which is not mentioned. Stream reasoning in Prolog was extensively investigated in the 1980 and 1990. Why wasn’t this work cited and discussed? 

    Page 11, last sentence before Section 4.2: “… predicates our of lazy …”.


     REVIEW 2 -
    SUBMISSION: 45
    TITLE: Lazy Stream Manipulation in Prolog via Backtracking: the Case of 2P-Kt
    AUTHORS: Giovanni Ciatto, Roberta Calegari and Andrea Omicini


     Overall evaluation -
    SCORE: 1 (weak accept)


     TEXT:
    The paper describes original theoretical and implemented progress in the use of Prolog extensions in order to utilise lazy stream manipulation by means of its backtracking features, worth relating to the community.

    However it does little comparison to other work, and make clear the semantics of the system.

    A issue needs to be answered regarding item 2 in page 8: what if there is more than one stream? in what way are they scheduled relative to one another and why?

    Typos:
    p.8 item 5: consider > considered
    p.9 line 14: retrivable 
    > retrievable
    p.11 line 6: our > out


     REVIEW 3 -
    SUBMISSION: 45
    TITLE: Lazy Stream Manipulation in Prolog via Backtracking: the Case of 2P-Kt
    AUTHORS: Giovanni Ciatto, Roberta Calegari and Andrea Omicini


     Overall evaluation -
    SCORE: -1 (weak reject)


     TEXT:
    This application paper discusses how Prolog may be adapted for lazy stream reasoning. The ideas are rather simple by understanding Prolog solvers as operational state machines that interact using the message passing concept. The implementation is detailed using several code fragments and the paper is easy to read.

    Related work is severely lacking in depth and the authors are apparently unaware of a broad body of research (in the area of Answer-Set Programming) that investigated similar questions as they do here. For example, work on stream reasoning (cf. [1,2,3]) is closely related to the results presented in this submission, but this is never mentioned. Another example is the concept of generators to let the solver query the state of the outside world. Such functionality has been investigated in the logic programming community for over a decade under the term external atoms and HEX programs (cf. [4,5]).

    It appears that the paper follows a purely operational understanding of Prolog, at least I could not make out a clear description of the semantics resulting from the system, i.e., what is the resulting 2P-KT system formally computing? Despite a number of concrete examples that present exemplary implementations, there also is no evaluation of the overall system. What is the throughput of the resulting stream processing system?

    Detailed comments:
    - If figures are referenced, they should be written with upper case "Fig. 1a" instead of "fig. 1a". This similarly holds for sections, it is "this section" but "the Section 2.4" as the latter is taken as a name of something and hence capitalized.
    - Page 4 "can ben modelled" -> can be modelled
    - Page 11 "predicates our of lazy data" -> predicates and of lazy data
    - Page 12 "worth to be highlight" -> worth to highlight
    - Page 13 "this work represents just the first step towards the exploration of LP as a general means for stream processing." The first steps for stream processing using logic programming have been done years ago, please read up on the relevant literature (see pointers below; work on complex event processing may also be related).
    - References consist of 6 self-citations and apart from that only 3 other works. This shows a dangerous lack of knowledge of the research field and should be avoided.

    [1] Beck, Eiter, Folie: Ticker: A System for Incremental ASP-based Stream Reasoning.

    [2] Eiter, Ogris, Schekotihin: A Distributed Approach to LARS Stream Reasoning.

    [3] Beck, Dao-Tran, Eiter: LARS: A Logic-based framework for Analytic Reasoning over Streams.

    [4] Eiter, Germano, Ianni et al.: The DLVHEX System.

    [5] Eiter, Ianni, Schindlauer, Tompits: A Uniform Integration of Higher-Order Reasoning and External Evaluations in Answer Set Programming.


     REVIEW 4 -
    SUBMISSION: 45
    TITLE: Lazy Stream Manipulation in Prolog via Backtracking: the Case of 2P-Kt
    AUTHORS: Giovanni Ciatto, Roberta Calegari and Andrea Omicini


     Overall evaluation -
    SCORE: 1 (weak accept)


     TEXT:
    The main issues found with this paper are two:

    1. There is quite some existing work (old [Prolog] and new [ASP]) that has to be discussed in the final version.

    2. There do not seem to be formal results or experimental results.

    Overall, the main positive aspect is that a working and potentially useful implementation exists, which made the overall evaluation slightly positive.

Partita IVA: 01131710376 - Copyright © 2008-2021 APICe@DISI Research Group - PRIVACY