++++++++++++++++++ FME2001: - Review reports for paper 55 +++++++++++++++++++++ - Dear author(s) : this file was extracted automatically from a very large - mailbox. In case you find any problem concerning mail encodings (or any - other kind of anomaly disturbing your understanding of the reviews) please - email any of the PC cochairs (pamela@research.att.com,jno@di.uminho.pt). ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ PAPER NUMBER: 55 CATEGORY: 2 TITLE: View Updatability Based on the Models of a Formal Specification AUTHOR(S): Michael Johnson Robert Rosebrugh -------------------------------------------------------------------------------- 1. Briefly SUMMARIZE the paper (2-3 lines): A new category theory treatment of the view update problem. It is targetted at formal specification systems and is mainly theoretical. 2. RELEVANCE: Please provide a rating of the paper's relevance to the FME Symposium, using the scale: 0 = Out of scope 1 = Marginal interest 2 = Minority interest 3 = Majority interest 4 = Outstanding interest Numeric Rating: 2.5 Please comment your rating: Predominantly a theoretical paper and requires a certain familiarity with category theory and the vup. 3. OVERALL RECOMMENDATION: Please provide a rating of the paper's acceptability, using the scale: 1 = Strong reject 2 = Weak reject 3 = Could go either way 4 = Weak accept 5 = Strong accept Numeric Rating: 3.5 NB: There should be a correlation between the two rates above. 4. CONFIDENCE LEVEL: Please provide a rating oof your expertise in the area addressed by the paper, using the scale: 1 = Know virtually nothing about this area 2 = Not too knowledgeable, but I know a bit 3 = Know a moderate amount, about average I'd say 4 = Not my main area of work, but I know a lot about it 5 = My area of work, I know it extremely well Numeric Rating: 3 NB: PC members are responsible for ensuring that 1 is not used here. 5. ORIGINALITY. What is NEW and SIGNIFICANT in the work reported here? Comment: I have seen many 'similar' CT papers in different problem areas and development processes. Perhaps the application of CT in the VUP is new but the techniques applied seem familiar. It means that the new is not sufficiently explained and motivated. 6. How WORTHWILE is the goal of the authors? Comment: Hard to tell (see 7) 7. How well is this goal EXPLAINED and JUSTIFIED? Comment: Not very 8. TECHNICAL QUALITY. Are the technical parts (definitions, statements, specifications, proofs, algorithms, etc.) SOUND? Comment: The standard CT terminology can often be hard to follow but everything appears sound (the Proofs ... which seemed a bit rushed... could be tidied up .. I failed to follow some of their reasoning because of textual description) 9. APPLICABILITY. If the work is primarily theoretical or conceptual, are the implications in terms of applicability adequately justified? If the paper is about a new formal technique, are satisfactory arguments presented in favor of the new technique? If a methodology is proposed, is it sound? If experimental results are presented, is their relevance justified? Comment:Not clearly 10. PRESENTATION: Describe the QUALITY of the writing, including suggestions for changes where appropriate. Comment: GOOD 11. Were there any formatting or mechanical problems with this paper?: NO Are the figures and length acceptable?: YES Are the references correct?: YES 12. OTHER COMMENTS you believe would be useful to the author(s), including pointers to missing relevant work: I am not familiar with the view update problem but it seems to be a specific instance of a general problem in software engineering and system evolution. It is not clear that CT offers the best formal model for reasoning about problems of this nature. +++++++++++++++++++++ End of FME 2001 Paper Review Report ++++++++++++++++++++++ PAPER NUMBER: 55 CATEGORY: 2 TITLE: View Updatability Based on the Models of a Formal Specification AUTHOR(S): Michael Johnson Robert Rosebrugh -------------------------------------------------------------------------------- 1. Briefly SUMMARIZE the paper (2-3 lines): In a categorical model, view updatability is expressed through the existence of initial/terminal objects among the objects which express the update. I believe this expresses the intuitive notion of "minimal change". 2. RELEVANCE: Please provide a rating of the paper's relevance to the FME Symposium, using the scale: 0 = Out of scope 1 = Marginal interest 2 = Minority interest 3 = Majority interest 4 = Outstanding interest Numeric Rating: 2 Please comment your rating: I am a little curious as to why the authors chose this forum, a databases conference with a formal slant seems more obvious. However, the view update problem and data modelling are very relevant issues in the FM community, too, so I don't mind seeing a paper on this subject at FME, although I'd like to see the FM link covered a little better (see below). 3. OVERALL RECOMMENDATION: Please provide a rating of the paper's acceptability, using the scale: 1 = Strong reject 2 = Weak reject 3 = Could go either way 4 = Weak accept 5 = Strong accept Numeric Rating: 4 NB: There should be a correlation between the two rates above. 4. CONFIDENCE LEVEL: Please provide a rating oof your expertise in the area addressed by the paper, using the scale: 1 = Know virtually nothing about this area 2 = Not too knowledgeable, but I know a bit 3 = Know a moderate amount, about average I'd say 4 = Not my main area of work, but I know a lot about it 5 = My area of work, I know it extremely well Numeric Rating: 4 NB: PC members are responsible for ensuring that 1 is not used here. 5. ORIGINALITY. What is NEW and SIGNIFICANT in the work reported here? Comment: There is evidence (at least in the references section) of the practical use of this categorical method of data modelling, which I am glad to see. I don't think the "solution" to the view update problem is very exciting: it is obvious that view updates are propagatable if the updated databases which could have provided the updated view can be ordered in such a way that one of them represents a "minimal" change. I would have liked to see whether the categorical model adds anything beyond the intuitively obvious, e.g. through an implicit "up to isomorphism", or through variation of the ordering underlying the category. 6. How WORTHWILE is the goal of the authors? Comment: Very - giving data modelling diagrams a usable formal meaning, and solving the view update problem. 7. How well is this goal EXPLAINED and JUSTIFIED? Comment: Reasonably, though with little reference to non-database applications. 8. TECHNICAL QUALITY. Are the technical parts (definitions, statements, specifications, proofs, algorithms, etc.) SOUND? Comment: I believe so, but the level of category theory is quite high, I could have done with a definition of fibration, e.g. 9. APPLICABILITY. If the work is primarily theoretical or conceptual, are the implications in terms of applicability adequately justified? If the paper is about a new formal technique, are satisfactory arguments presented in favor of the new technique? If a methodology is proposed, is it sound? If experimental results are presented, is their relevance justified? Comment: Justified except for the missing link to FM. 10. PRESENTATION: Describe the QUALITY of the writing, including suggestions for changes where appropriate. Comment: Very well written overall. In the conclusion, advantage (1) seems very relevant, could it be refered to earlier in the paper? Reference [16]: I know of a Rom La*n*gerak. 11. Were there any formatting or mechanical problems with this paper?: No Are the figures and length acceptable?: Yes Are the references correct?: Yes 12. OTHER COMMENTS you believe would be useful to the author(s), including pointers to missing relevant work: I believe the update problem is quite closely related to the framing problem. The solution to the update problem is to impose minimal change, and one approach to the framing problem is to insist that very little else changes apart from that which is explicitly required to change. The framing problem could even be seen as an instance of the view update problem, where the view is a projection on the variables explicitly required to change. Exhibiting this link in the paper would make it a more worthwhile contribution to FME, I would suggest looking at the following papers for discussion of the framing problem: @Article{Jackson95, title = "Structuring {Z} Specifications with Views", author = "D. Jackson", journal = {ACM Transactions on Software Engineering and Methodology}, year= 1995, pages= {365--389}, volume= 4, number = 4, month= oct, } @CONFERENCE{ Borgida93, Author= {A. Borgida and J. Mylopoulos and R. Reiter}, Title= {And Nothing Else Changes: The Frame Problem in Procedure Specifications}, Booktitle= {Proc. 15th International Conference on Software Engineering}, Address= {Baltimore, Maryland}, Publisher= {IEEE Computer Society Press}, Month= May, Year= 1993} @ARTICLE{ Pitt94, Author= {D. Pitt and P. Byers}, Title= {The Rest Stays Unchanged (Concurrency and State-Based Specification)}, Journal= {Formal Aspects of Computing}, Volume= 6, Pages= {471--494}, Year= 1994} +++++++++++++++++++++ End of FME 2001 Paper Review Report ++++++++++++++++++++++ PAPER NUMBER: 55 CATEGORY: 2 TITLE: View Updatability Based on the Models of a Formal Specification AUTHOR(S): Michael Johnson Robert Rosebrugh -------------------------------------------------------------------------------- 1. Briefly SUMMARIZE the paper (2-3 lines): This paper presents a category-theoretic treatment of information system modelling. This could be seen as an extension of E-R modelling, but allows a neat characterisation of updatability. 2. RELEVANCE: Please provide a rating of the paper's relevance to the FME Symposium, using the scale: 0 = Out of scope 1 = Marginal interest 2 = Minority interest 3 = Majority interest 4 = Outstanding interest Numeric Rating: 2 Please comment your rating: The paper is a good theoretical contribution, but will be inaccessible to most FME delegates. 3. OVERALL RECOMMENDATION: Please provide a rating of the paper's acceptability, using the scale: 1 = Strong reject 2 = Weak reject 3 = Could go either way 4 = Weak accept 5 = Strong accept Numeric Rating: 4 NB: There should be a correlation between the two rates above. 4. CONFIDENCE LEVEL: Please provide a rating oof your expertise in the area addressed by the paper, using the scale: 1 = Know virtually nothing about this area 2 = Not too knowledgeable, but I know a bit 3 = Know a moderate amount, about average I'd say 4 = Not my main area of work, but I know a lot about it 5 = My area of work, I know it extremely well Numeric Rating: 3 NB: PC members are responsible for ensuring that 1 is not used here. 5. ORIGINALITY. What is NEW and SIGNIFICANT in the work reported here? Comment: Through the use of category theory, the authors have a succinct and precise description of the view update problem, and a means to characterise views which exhibit (or do not exhibit) the problem. 6. How WORTHWILE is the goal of the authors? Comment: Very worthwhile. The problem is significant and likley to become more and more widespread as databases are linked and interact. A good solution will render such interactions much more modular and updateable. 7. How well is this goal EXPLAINED and JUSTIFIED? Comment: The goal is adequately explained, but the paper is light on examples. The reader needs to be quite familiar with information systems to understand the problem domain (and with category theory to understand the solution). 8. TECHNICAL QUALITY. Are the technical parts (definitions, statements, specifications, proofs, algorithms, etc.) SOUND? Comment: The technical parts appear carefully constructed and sound. This referee is not competent to check the proofs. 9. APPLICABILITY. If the work is primarily theoretical or conceptual, are the implications in terms of applicability adequately justified? If the paper is about a new formal technique, are satisfactory arguments presented in favor of the new technique? If a methodology is proposed, is it sound? If experimental results are presented, is their relevance justified? Comment: The applicablility of the technique is not discussed. This may be seen as an early piece of theoretical groundwork. Whilst, therefore, a fully worked-out methodology would be inappropriate, a continuation of the example from the initial description through to the final results would have been useful. 10. PRESENTATION: Describe the QUALITY of the writing, including suggestions for changes where appropriate. Comment: The paper is well-written, but assumes a great deal. Additional hints (in terms of sets and functions, say) for this audience would be particularly welcome --- e.g. a 4-word reminder of what `monic' means when we first encounter it. I detected no typos or spelling mistakes. 11. Were there any formatting or mechanical problems with this paper?: Are the figures and length acceptable?: Yes Are the references correct?: Yes 12. OTHER COMMENTS you believe would be useful to the author(s), including pointers to missing relevant work: I have just one concern with the approach, as described on p6: On the one hand, (end of second bullet point) you tell us that this approach is able to capture constraints usually missing from informational modelling; thus this is a more formal technique. On the other hand, the first bullet point tells us that the names on the nodes and arrows have no formal significance. It therefore is hard to know in what sense this particular constraint (as distinct from any other) has indeed been captured --- I have trouble reconciling the two statements. +++++++++++++++++++++ End of FME 2001 Paper Review Report ++++++++++++++++++++++