|
4 | 4 | <meta charset="UTF-8"> |
5 | 5 | <meta name="viewport" content="width=device-width,initial-scale=1"> |
6 | 6 | <meta name="turbo-root" content="/CodeGRITS"> |
7 | | - <meta name="turbo-cache-control" content="no-cache" data-turbo-track="reload" data-track-token="3.5.0.757102960125"> |
| 7 | + <meta name="turbo-cache-control" content="no-cache" data-turbo-track="reload" data-track-token="3.5.0.757226652152"> |
8 | 8 |
|
9 | 9 | <!-- See retype.com --> |
10 | 10 | <meta name="generator" content="Retype 3.5.0"> |
|
32 | 32 | <script data-cfasync="false">(function () { var el = document.documentElement, m = localStorage.getItem("doc_theme"), wm = window.matchMedia; if (m === "dark" || (!m && wm && wm("(prefers-color-scheme: dark)").matches)) { el.classList.add("dark") } else { el.classList.remove("dark") } })();</script> |
33 | 33 |
|
34 | 34 | <link href="../favicon.png" rel="icon"> |
35 | | - <link href="../resources/css/retype.css?v=3.5.0.757102960125" rel="stylesheet"> |
| 35 | + <link href="../resources/css/retype.css?v=3.5.0.757226652152" rel="stylesheet"> |
36 | 36 |
|
37 | | - <script data-cfasync="false" src="../resources/js/config.js?v=3.5.0.757102960125" data-turbo-eval="false" defer></script> |
| 37 | + <script data-cfasync="false" src="../resources/js/config.js?v=3.5.0.757226652152" data-turbo-eval="false" defer></script> |
38 | 38 | <script data-cfasync="false" src="../resources/js/retype.js?v=3.5.0" data-turbo-eval="false" defer></script> |
39 | | - <script id="lunr-js" data-cfasync="false" src="../resources/js/lunr.js?v=3.5.0.757102960125" data-turbo-eval="false" defer></script> |
| 39 | + <script id="lunr-js" data-cfasync="false" src="../resources/js/lunr.js?v=3.5.0.757226652152" data-turbo-eval="false" defer></script> |
40 | 40 | </head> |
41 | 41 | <body> |
42 | 42 | <div id="docs-app" class="relative text-base antialiased text-gray-700 bg-white font-body dark:bg-dark-850 dark:text-dark-300"> |
@@ -282,6 +282,31 @@ <h4> |
282 | 282 | </h4> |
283 | 283 | </doc-anchor-target> |
284 | 284 | <p>See <a href="../developer/#accommodating-new-ides">Accommodating New IDEs</a>.</p> |
| 285 | +<doc-anchor-target id="q5-how-efficient-is-the-processing-of-raw-eye-gaze-data"> |
| 286 | + <h4> |
| 287 | + <doc-anchor-trigger class="header-anchor-trigger" to="#q5-how-efficient-is-the-processing-of-raw-eye-gaze-data">#</doc-anchor-trigger> |
| 288 | + <span>Q5. How efficient is the processing of raw eye gaze data?</span> |
| 289 | + </h4> |
| 290 | +</doc-anchor-target> |
| 291 | +<p>In CodeGRITS, the efficiency of the processing of gaze data is negligible. For each gaze, we calculate the average time |
| 292 | +from the timestamp in the raw data to the timestamp after all processing is complete. This processing primarily involves |
| 293 | +location mapping and upward traversal of the AST. The average delay is 4.32 ms, equating to delays of approximately |
| 294 | +12.98% for 30Hz, 25.96% for 60Hz, and 51.92% for 120Hz eye gaze data. With such |
| 295 | +a high sampling frequency, meaningful changes in the content of the code editor's page are extremely rare within this |
| 296 | +short time frame, which ensures the accuracy of gaze processing. Moreover, compared |
| 297 | +to <a href="https://www.i-trace.org/">iTrace</a>, our method of receiving data from the |
| 298 | +eye-tracking device is more efficient, ensuring that all sampled gazes are mapped without any loss.</p> |
| 299 | +<doc-anchor-target id="q6-how-much-storage-space-does-codegrits-require"> |
| 300 | + <h4> |
| 301 | + <doc-anchor-trigger class="header-anchor-trigger" to="#q6-how-much-storage-space-does-codegrits-require">#</doc-anchor-trigger> |
| 302 | + <span>Q6. How much storage space does CodeGRITS require?</span> |
| 303 | + </h4> |
| 304 | +</doc-anchor-target> |
| 305 | +<p>Generally speaking, a high sample frequency generates a large amount of gaze data. To conserve storage space, we |
| 306 | +only perform upward traversal of the AST of the first gaze and record the hierarchy structure, and mark the rest |
| 307 | +as <code v-pre>same</code>. This approach significantly reduces storage space. In a previous debugging study, we set the eye-tracking |
| 308 | +device's sample frequency to 60Hz, and during the 20-minute experiment, the eye-tracking data amounted to only about |
| 309 | +40MB.</p> |
285 | 310 |
|
286 | 311 |
|
287 | 312 | <!-- Required only on API pages --> |
|
0 commit comments