Title: Application and device...
Relevance to magazine:
Presentation and clarity:
Depth of contribution:
Detailed comments for the authors
I did not think that the first study was very exciting. It seems rather obvious that some general applications could be ported across several devices (isn't that the entire point of the J2ME and visual tools used to construct the applications?). I was also curious about what interface elements were used in the applications. For example, a PDA with a touch screen can use buttons, but a cellphone interface usually cannot. I get the sense that only list items were used in the two applications. This does not prove that any application can be written and easily ported to different mobile devices. Some of the detail used in the first experiment seemed unnecessary, such as the list of task orders. Since the null hypothesis was taken in every case, the list of hypotheses could have been quickly summed up rather than listed explicitly. The pool of users selected seemed to bias the experiment; all users had used mobile phones but only 3 had used PDAs.
The second study about configuring a desktop application to work on a PDA also was not very exciting. Again, it seems obvious that an application can be made mobile and there did not seem to be any significant results or conclusions drawn from this exercise, only general observations. The methods used in the two experiments contrasted so much it seems that they were written by two different authors and just stuck together. While it is indicated that five users interacted with the applications, no experiment was conducted and only five opionions were taken. This is a pilot study and does not seem to be at a stage where it can make a contribution. It is stated that all users had difficulity with the transcriber, however the expressed problem of not understanding where to write could have easily been avoided by a quick, general training session with the device. The authors also state that they cannont control the pen software from VB, so it seems obvious that to fix this problem they should incorporate another tool or write some native code to control the pen.
Overall, the results of this paper do not make a significant contribution. The fact that the authors reference that these are pilot studies shows that much work needs to be done before results from this study are ready to publish. There are many sources of related work left out of this paper. The authors mention "seperating functionality from the interface" but fail to cite any systems such as Pebbles or XWeb that do this. There are also tons of commericial applications that have been ported to mobile devices that would prove the point of their second study (such as devices for mobile delivery applications). A simple google search on inventory management using a PDA brought up over 10 pages of commercial applications. I also quickly found many papers about interface design on PDAs, which very likely site many of the observations of this paper. There are also quite a few typos and errors in this paper.
Comments for EIC and AEICs only:
This almost looks like a quickly mocked up class project. The authors did not do much research into their topic and I would
not be suprised if this work had been done before.