Citation Context

...ication will work properly onsall devices without device-specific faults (i.e., faults found onlyson certain devices).sThis problem is well understood by practitioners andsresearchers. Muccini et al. =-=[4]-=- included “new and rapidlysevolving mobile operating systems” and “diversity of phonessand phone makers” in the list of the peculiarities of mobilesapps that have a direct impact on software testing. ...

Citation Context

...ones and tablets) in use at the same time. Thissproblem is especially important for Android devices [2]. As2014 report by OpenSignal claims that 18,796 distinct Androidsdevices were seen in that year =-=[3]-=-. Due to the huge number ofsdevices on the market and the variations in their characteristics,sit is hard to guarantee that an application will work properly onsall devices without device-specific fau...

Citation Context

...s, but such decisions are based on their ownsunderstanding and available resources, rather than on a solidstheoretical basis. Several examples of the practices of differentscompanies are described in =-=[12]-=-, and the number of devicessused for testing varies significantly from 12 (Red Robot Labs),sto 40 (Pocket Gems), to between 30 and 50 (Storm8), and up tos400 (Animoca).sConsidering that research inves...

Citation Context

...” [5]. Commercial tools andsresearch prototypes developed to support different aspects ofsmulti-device mobile testing include Ghostlab [6], Adobe EdgesInspect [7], Silk Mobile [8], AppACTS [9], CTOMS =-=[10, 11]-=-,sand others.sBefore applying any methods and tools for multi-devicestesting, mobile devices should be selected for testing. Thesgeneral approach to this selection, which is quite obvious andscommonly...

Citation Context

...” [5]. Commercial tools andsresearch prototypes developed to support different aspects ofsmulti-device mobile testing include Ghostlab [6], Adobe EdgesInspect [7], Silk Mobile [8], AppACTS [9], CTOMS =-=[10, 11]-=-,sand others.sBefore applying any methods and tools for multi-devicestesting, mobile devices should be selected for testing. Thesgeneral approach to this selection, which is quite obvious andscommonly...

Citation Context

...ific faults, Android,sdevice selection.sI. INTRODUCTIONsAccording to a recent survey, 91% of American adults ownsa cell phone and 50% of them use these devices to downloadsand run mobile applications =-=[1]-=-. Mobile application testing isstherefore important to ensure the quality of such applicationss(apps). One of the reasons that makes mobile testing a criticalsand challenging task is the so-called “fr...

Citation Context

... “fragmentation problem,”sin other words, the large number of distinct mobile devicess(i.e., smartphones and tablets) in use at the same time. Thissproblem is especially important for Android devices =-=[2]-=-. As2014 report by OpenSignal claims that 18,796 distinct Androidsdevices were seen in that year [3]. Due to the huge number ofsdevices on the market and the variations in their characteristics,sit is...

Citation Context

...ice support” [5]. Commercial tools andsresearch prototypes developed to support different aspects ofsmulti-device mobile testing include Ghostlab [6], Adobe EdgesInspect [7], Silk Mobile [8], AppACTS =-=[9]-=-, CTOMS [10, 11],sand others.sBefore applying any methods and tools for multi-devicestesting, mobile devices should be selected for testing. Thesgeneral approach to this selection, which is quite obvi...

Citation Context

...and 50 (Storm8), and up tos400 (Animoca).sConsidering that research investigation of these questionsswould be very useful and promising, we previously took thesinitial step toward filling in this gap =-=[13]-=-.sThe current paperscontinues in this direction by using experimental testing tosevaluate the effectiveness of different approaches to thesselection of mobile devices for testing.s2015 2nd ACM Interna...