blue / papers.hotmobile2015-numerator · Commits

Commit 76b328e5ce6a4384c5f14a2ee7909ae0d31f1f76

Authored by Geoffrey Challen
1 parent 0b387634

Spell check.

Inline Side-by-side
Showing 5 changed files with 18 additions and 19 deletions
figures/tables/tableALL.tex
  View file @76b328e
... ... @@ -142,7 +142,7 @@
142 142  
143 143 \caption{\small \textbf{Evaluating Components of a Value Measure.}
144 144 \PhoneLab{} data is used to weight overall app energy usage in a variety of
145   -different ways. Ommitted results are caused by Android reporting energy
  145 +different ways. Omitted results are caused by Android reporting energy
146 146 consumption for non-apps such as the Android System.}
147 147  
148 148 \label{table-results}
... ...
introduction.tex
  View file @76b328e
... ... @@ -64,13 +64,12 @@ smartphones~\cite{jdpowerbatterylife-url}, we believe this effort is
64 64 worthwhile.
65 65  
66 66 In this paper we motivate the idea of a value measure and describe an early
67   -failure at developing one based on measuring content delivery. We begin in
68   -Section~\ref{sec-usage} by describing how useful such a measure would be and
69   -the results of being able to answer questions like those posed above, as well
70   -as formulating design requirements for the value measure itself.
71   -Section~\ref{sec-measure} presents an overview of possible inputs into such a
72   -measure and discussion of how each could be measured and how useful it might
73   -be. In Section~\ref{sec-results} we present at formulating a value measure
74   -based on content delivered through the video display and audio output---an
75   -attempt that we consider a failure based on the result of a user survey, but
76   -one that we hope sheds light on how difficult this challenge may be.
  67 +failure at developing one. We begin in Section~\ref{sec-usage} by describing
  68 +how useful such a measure would be while also formulating design requirements
  69 +for the value measure itself. Section~\ref{sec-measure} presents an overview
  70 +of possible inputs into such a measure and discussion of how each could be
  71 +measured and how useful it might be. In Section~\ref{sec-results} we present
  72 +at formulating a value measure based on content delivered through the video
  73 +display and audio output---an attempt that we consider a failure based on the
  74 +result of a user survey, but a failure that we hope sheds some light on this
  75 +difficult challenge.
... ...
metric.tex
  View file @76b328e
... ... @@ -26,7 +26,7 @@ foreground information---such as a pedometer app.
26 26  
27 27 Patterns of interaction may also be useful to observe, and inputs such as
28 28 keystrokes and touchscreen events are simple to track. However, there is more
29   -obvious differentation between app interaction patterns between
  29 +obvious differentiation between app interaction patterns between
30 30 categories---users deliver far more keystrokes to a chat client than to a
31 31 video player---so it is clear that interaction statistics will have to be
32 32 used in conjunction with complementary value measure components that offset
... ...
results.tex
  View file @76b328e
... ... @@ -6,7 +6,7 @@ large dataset of energy consumption measurements collected by an IRB-approved
6 6 experiment run on the \PhoneLab{} testbed. \PhoneLab{} is a public smartphone
7 7 platform testbed located at the University at
8 8 Buffalo~\cite{phonelab-sensemine13}. 220~students, faculty, and staff carry
9   -instrumented Android Nexus~5 smartphones and receiv subsidized service in
  9 +instrumented Android Nexus~5 smartphones and receive subsidized service in
10 10 return for willingness to participate in experiments. \PhoneLab{} provides
11 11 access to a representative group of participants balanced between genders and
12 12 across a wide variety of age brackets, making our results more
... ... @@ -112,7 +112,7 @@ redraws.
112 112 \centering
113 113 \includegraphics[width=\textwidth]{./figures/survey.pdf}
114 114  
115   -\caption{\textbf{Participant responses to energy inefficient app sugestions.} The height of each bar
  115 +\caption{\textbf{Participant responses to energy inefficient app suggestions.} The height of each bar
116 116 demonstrates how many of the suggested apps the user is willing to remove for better battery life. }
117 117  
118 118 \label{fig-survey}
... ... @@ -122,7 +122,7 @@ redraws.
122 122 To continue the evaluation of our simple content-based value measure, we
123 123 prepared a survey for the 107~\PhoneLab{} participants who contributed data
124 124 to our experiment. Our goal was to determine if users would be more willing
125   -to remove inefficienct apps, as defined using our content-based measure. As a
  125 +to remove inefficient apps, as defined using our content-based measure. As a
126 126 baseline, we also asked users about the apps that consumed the most energy.
127 127 We used each participants data to generate a custom survey containing
128 128 questions about 9 apps: the 3 least efficient apps as computed by our
... ...
usage.tex
  View file @76b328e
... ... @@ -54,7 +54,7 @@ the second experiment to make it more similar to the first, the adversary
54 54 could remove the apps consuming the least energy up to a given target.
55 55  
56 56 For our value measure we are hopeful that users will prove capable of
57   -assiging cardinal utilities to apps---as in the second experiment---since
  57 +assigning cardinal utilities to apps---as in the second experiment---since
58 58 this matches most directly with our proposed value measure and could provide
59 59 ground truth for a value measure computed automatically. The second
60 60 experiment also engages users directly in the task of allocating energy,
... ... @@ -88,7 +88,7 @@ consumption alone~\cite{carat-sensys13}. Consider attempting to compare a
88 88 chat client and videoconferencing app by only measuring their energy
89 89 consumption. Unless it is terribly written, the chat client will consume less
90 90 energy. But this does not mean that it is efficient, or that the
91   -videconferencing app is not. Ultimately, all the energy consumption
  91 +videoconferencing app is not. Ultimately, all the energy consumption
92 92 comparison truly reveals is that the two apps do different things---which we
93 93 knew.
94 94  
... ... @@ -104,7 +104,7 @@ By computing value and, thus, energy efficiency, we can overcome these
104 104 weaknesses. A value measure should allow us to compare the efficiency of two
105 105 apps in different categories based on how efficiently they use energy to
106 106 deliver user value, making it possible to compare games to email clients to
107   -video players. Comparisons within the same app category shoud allow users to
  107 +video players. Comparisons within the same app category should allow users to
108 108 select the most efficient email client or web browser. Aggregating results
109 109 over all users, differences in app energy efficiency should reflect how well
110 110 the app is written and how well it predicts and adapts to users, not just
... ... @@ -154,7 +154,7 @@ However, all of these previous efforts have ignored the critical question of
154 154 mechanisms are, systems that rely on rates will fail if they provide the same
155 155 rate to Skype and Snapchat, or to a very efficient app and an energy virus.
156 156  
157   -A measure of value can be used alone or in conjuction with energy consumption
  157 +A measure of value can be used alone or in conjunction with energy consumption
158 158 to help prioritize limited energy resources. The simplest approach is to
159 159 attempt to enforce an energy allocation based on the relative value assigned
160 160 to each app. To encourage apps to be more energy efficient, it may also be
... ...