IEEE Consumer Electronics Magazine - March/April 2020 - 62

Mobile Applications

may save 2-5 h for android
apps and around 3 days in
Apple store in every single
modification.
Other consideration is
that tracking changes can be
done by users with no technical skills by easily and manually editing the JSON, or even
with a graphic interface.
Finally, and considering
Figure 4. Android app performance comparison: CPU use (left) and memory use
the previous results, it can be
(right).
stated that one of the main
innovations of this work is
the application of the AOP
time and not considering the time to market (in
such case, the difference would be even higher), paradigm to a new field, such as mobile app
and considering only 5 changes during the entire tracking, allowing a more efficient mobile apps
life cycle while, in a real case, the number of development and maintenance cycle.
modifications could be bigger.
On this basis, we can confirm that the pro- ACKNOWLEDGMENT
posed solution provides a new method, which
This work was supported in part by the EC
requires less effort in terms of development time H2020-ICT-19-2016-2 76199 EasyTV project and in
for managing the changes needed to dynamically part by the Spanish CDTI-Ministry for Science
track user's activity inside mobile apps.
and Innovation EXP 00064563/ITC-20133062 LPSOn the other hand, left represents the time Bigger project.
spent while developing the app and implementing the changes in percentage. As seen in the
figure, the time spent in tasks that need techni& REFERENCES
cal development skills is significantly smaller in
the AOPþMANTRA solution than in the standard
1. S. Kaur, "The revolution of tablet computers and apps:
one, which also means a total cost reduction.
A look at emerging trends," IEEE Consum. Electron.
To estimate the impact of the library in the
Mag., vol. 2, no. 1, pp. 36-41, Jan. 2013.
performance of the app we used the Android Stu2. S. Van Canneyt, M. Bron, A. Haines, and M. Lalmas,
dio profiler to compare the CPU and Memory
"Describing patterns and disruptions in large scale
usage. As seen in Figure 4, the cost of using the
mobile app usage data," in Proc. 26th Int. Conf. World
libraries is minimal. Concerning disk occupancy,
Wide Web Companion, 2017, pp. 1579-1584.
the app only takes 3.44 extra MB.
3. S. Seneviratne, H. Kolamunna, and A. Seneviratne, "A
measurement study of tracking in paid mobile
applications," in Proc. 8th ACM Conf. Secur. Privacy

CONCLUSIONS
According to the obtained results, the difference on development and maintenance time
(regarding user activity tracking) is significant,
due to the fact that our approach decreases the
time needed to make the changes on the events to
be tracked within the app. Another advantage is
that the app resubmission to the market is no longer needed, which usually adds time on top of the
development time. Considering this, our solution

Wireless Mobile Netw., 2015, Paper 7.
4. D. Contreras, M. Salamo, I. Rodriguez, and A. Puig,
"Shopping decisions made in a virtual world: Defining
a state-based model of collaborative and
conversational user-recommender interactions," IEEE
Consum. Electron. Mag., vol. 7, no. 4, pp. 26-35,
Jul. 2018.
5. X. Li, Z. Zhang, and J. Nummenmaa, "Models for
mobile application maintenance based on update
history," in Proc. 9th Int. Conf. Eval. Novel Approaches
Softw. Eng., 2014, pp. 1-6.

IEEE Consumer Electronics Magazine

IEEE Consumer Electronics Magazine - March/April 2020

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - March/April 2020