Subject: Review of paper C521: A Java Extension Framework for Web-based Simulation From: Judith Bishop Date: Wed, 17 Oct 2001 19:53:40 +0200 To: fox@csit.fsu.edu X-UIDL: c53ccc4c871c0000 X-Mozilla-Status: 0001 X-Mozilla-Status2: 00000000 Received: by mailer.csit.fsu.edu (mbox gcfpc) (with Cubic Circle's cucipop (v1.31 1998/05/13) Sat Oct 20 10:55:41 2001) X-From_: fox@mailer.csit.fsu.edu Sat Oct 20 10:52:10 2001 Return-Path: Delivered-To: gcfpc@csit.fsu.edu Received: from dirac.csit.fsu.edu (dirac.csit.fsu.edu [144.174.128.44]) by mailer.csit.fsu.edu (Postfix) with ESMTP id 017F823A0E for ; Sat, 20 Oct 2001 10:52:10 -0400 (EDT) Received: from localhost by dirac.csit.fsu.edu (AIX4.2/UCB 8.7) id KAA57160; Sat, 20 Oct 2001 10:52:09 -0400 (EDT) Resent-Message-Id: <200110201452.KAA57160@dirac.csit.fsu.edu> Replied: Sat, 20 Oct 2001 10:52:02 -0400 Replied: jbishop@cs.up.ac.za Delivered-To: fox@csit.fsu.edu Received: from kendy.up.ac.za (kendy.up.ac.za [137.215.101.101]) by mailer.csit.fsu.edu (Postfix) with ESMTP id 7E5E423A06 for ; Wed, 17 Oct 2001 13:52:53 -0400 (EDT) Received: from [137.215.18.21] (helo=hades.cs.up.ac.za) by kendy.up.ac.za with esmtp (Exim 3.15 #1) id 15tus6-0006uq-00 for fox@csit.fsu.edu; Wed, 17 Oct 2001 19:52:50 +0200 Received: from indigo.cs.up.ac.za ([137.215.165.66] helo=cs.up.ac.za) by hades.cs.up.ac.za with esmtp (Exim 3.22 #1 (Debian)) id 15tueO-0000q3-00 for ; Wed, 17 Oct 2001 19:38:40 +0200 Message-ID: <3BCDC5A4.232886CD@cs.up.ac.za> Reply-To: jbishop@cs.up.ac.za Organization: University of Pretoria X-Mailer: Mozilla 4.75C-CCK-MCD {C-UDP; EBM-APPLE} (Macintosh; U; PPC) X-Accept-Language: en MIME-Version: 1.0 References: <3B8043CB.80100@csit.fsu.edu> Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 7bit X-Scanner: exiscan *15tus6-0006uq-00*ybQZCHN5MYU* http://duncanthrax.net/exiscan/ Resent-To: Geoffrey Fox Resent-Date: Sat, 20 Oct 2001 10:52:09 -0400 Resent-From: Geoffrey Fox Hi Geoffrey This is number 1, and the fun has just begun. Sorry I'm late. The other two will follow. Judith --------------------- REFEREE'S REPORT Concurrency and Computation:Practice and Experience ********** C: Paper and Referee Metadata Paper Number Cnnn:C521: Date: 17 October 2001 Paper Title:A Java Extension Framework for Web-based Simulation Author(s): Tao Tang and Chu R. Wie Referee: Judith Bishop Address: Department of Computer Science University of Pretoria Pretoria 0002 South Africa Phone: +27 12 460 6683 Fax: +27 12 362 5288 http://www.cs.up.ac.za/~jbishop jbishop@cs.up.ac.za C. Referee Recommendations. Please indicate overall recommendations here, and details in following sections. accepted provided changes suggested are made D: Referee Comments (For Editor Only) I liked this paper (mainly because I have done this same work myself back in 1997). The problem is therefore whether it is new at all. I identify specific issues in this regard below, and make a recommendation for extensions. It is possible that the authors are dividing the thesis up into lots of papers, and this one has landed up as too thin. It is actually rather a weak paper, but innocuous, and would probably be readable and enjoyable. I would just object to giving it prominence, without making them work a little harder. E: Referee Comments (For Author and Editor) This paper presents the framework details for a system which enable simulations to be run across the web. The motivation for such work is clear: accessibility of specialized software and hardware from a variety of platforms. It would seem that the system described here, WebSimMicro, gives access to only one simulator, for microelectronic devices, and the claim is made (p3 para 2) that the framework could be easily extended to other simulation packages. There has been much related work. Page 4 lists many such systems, and I am aware of at least one more (GoWeb, at GMD-First, to be found under the home page of Nikola Serbedzija). It is very important that the reader is made completely aware of the advantages that the WebSimMicro approach has over the earlier work. Page 5 is not adequate in this regard. At the end of section three, four desirable qualities are ascribed to WebSimMicro, but one needs to be more convincing. What does it mean to say that "the end user's attention [is moved] away from the the syntax ... to the physical design"? Could you give an example here? Ease of use is always a desirable trait, but how is it measured? can you show that your system is easier to sue than some other? What are the criteria? The paper needs an evaluation section, which can be pointed to from section 2. Page 6 starts off with an exposition of the standard three tier architecture model, and you should acknowledge that this is a standard model, and not your own invention. Clearly, the components you select for the tiers are tailored to your simulation, but you could make more of the genericity of the design here. It did seem to me to be quite good. The example starting on page 7 needs to be emphasized more, and you need a screen dump of the presentation tier at least, if not of some monitor screen showing how things are happening behind the scenes. The sequence diagram is useful, bu not everyone can read these, so screens are helpful. I found section 3.2 problematic, because it is written in the future tense - "components with operations like" "this tier can achieve" "the UI update can be made the responsibility". It is better to be definite about what you have done. One of the more significant parts of the work is the Engine Container and the design of the different engines. (pages 12-14) However, the requirements specified here are too vague for a reader to believe without an example. Specifically, consider the description of the Interactive Engine. "It is required that such applications must provide an API for manipulating the command interpreter" etc. This is quite a tall order, and it is necessary to illustrate what such an API would look like, its size and complexity. In IEEE Software May.June 1998 which was a special issue on web-based access to programs, there is a description of a geo package for accessing GIS software. Is yours similar? You should give a listing of it's main components/methods. As mentioned above, I do not think the reader is convinced that your framework is significantly different or better than any that has gone before. You need a comparison table (always a good idea) or some metrics for evaluation. The design and use of such frameworks is no longer breakthrough science, so you need to be very clear as to what contribution you are making to the field. F: Presentation Changes The structure of the paper is fine. There are some minor grammar errors which should be picked up by a careful read of the paper by an English speaker. The references are clear and complete enough. the authors may want to consider the GoWeb work mentioned above. -- .