A Post-Mortem on Google Glass: Researchers' Autopsy into Gadget's Bust Discovers How to Improve 'Underpowered' Mobile Devices
By Lori Cameron

As the Web gets bigger and mobile devices get smaller, power and access erode dramatically.

Researchers now look to a most unlikely source—the marketing flop Google Glass—for answers.

These so-called “underpowered devices”—smaller and smaller tech gadgets such as Google Glass—are suffering as the internet becomes more complex and dynamic. These demands strain underpowered devices.

In spite of the first-iteration woes of Google Glass, researchers continue to study the beleaguered device to glean information about how to create a better user experience with the next generation of underpowered mobile devices.

At stake could be a revolutionary new small device—remember how transformative was the tiny iPod?

Such a future, however, could be a long journey—according to researchers who authored “The Web for Underpowered Mobile Devices: Lessons Learned from Google Glass.”

“In general, the browsing performance on Glass is worse than a smartphone, primarily because the same content is being delivered to Glass and smartphones regardless of the device type. Glass is an underpowered mobile device with smaller processing power and smaller battery than smartphones,” the authors wrote.

Here are their findings on what went wrong—and recommendations for a better future. While a mass market disappointment, Google Glass is now being re-purposed for the manufacturing industry.

How the experiments panned out

A profiler monitored Google Glass and a smartphone through six experiments for four performance metrics—power consumption, temperature variation, downloaded bytes, and webpage load time using WiFi.

The results demonstrate that, in order to increase the functionality of underpowered devices, web developers should reduce web content, use simpler image formats, rely more on the cloud, and use better security protocols.

Important results from the study

The more web objects, servers accessed, and JavaScripts running on a page, the worse Glass performed than the smartphone. Furthermore, it took Google Glass about twice as long to load popular websites while consuming one-and-a-half times as much power.

Comparison of web-browsing performance
Comparison of web-browsing performance. (Upper-left) correlation analysis, (upper-right) total power consumption (mW), (bottom-left) webpage load time, and (bottom-right) temperature rise on Glass.

JavaScript is the most resource-intensive web component. The Glass browser proved to be about three to eight times slower than the Google Chrome browser on a Nexus 5 smartphone, while executing the same JavaScript benchmarks. The execution time for third-party analytics and ad scripts on Glass is about twice that of a smartphone.

Time to execute JavaScripts
Time to execute JavaScripts.

WebP image format is more energy-efficient than JPEG and PNG on Glass. For example, using WebP instead of JPEG on m.wikihow.com results in 45-percent savings in power consumption and 50 percent lower webpage load time.

Power consumption for image formats on Glass
Power consumption for image formats on Glass.

The cost of accessing a website using HTTPS on Glass compared to a smartphone increases with increasing webpage size, number of web objects, and number of servers accessed through the webpage.

Glass power consumption is 27 percent lower than that of a smartphone for webpages with less than 64 web objects. However, power consumption on Glass becomes 17 percent higher than that of a smartphone when loading webpages having more than 64 web objects.

Comparison of HTTP versus HTTPS performance
Comparison of HTTP versus HTTPS performance. (Upper-left) correlation analysis (HTTPS/HTTP), (upper-right) total power consumption (mW), (lower-left) webpage load time, and (lower-right) temperature rise on Glass.

Additionally, the number of websites tailored for Glass was dismally low, demonstrating a need for web developers to build better sites for a growing market of lightweight devices.

Only seven out of 50 studied websites are optimized for content delivery to Glass. For example, m.espn.go.com optimizes by serving images according to Glass screen dimensions.

The takeaways in a nutshell

The authors came up with four takeaways for any industry seeking to capitalize on underpowered devices:

  • Reduce web content by using fewer JavaScripts. Loading fewer CSS, images, and ads and adjusting to the display capability of the requesting device can also reduce the content.
  • Use efficient image formats. WebP instead of JPEG or PNG can be used on webpages to save power and lower the webpage load time.
  • Use cloud-based acceleration rather than a full-fledged browser as Glass uses. Also, use a data-compression proxy on the cloud, thereby reducing network bandwidth, power consumption, and webpage load time.
  • Use new protocols. SPDY can be used in place of HTTPS to improve browser performance.

The authors of this study are Jagmohan Chauhan of Aalto University, Mohamed Ali Kaafar of Macquarie University, and Anirban Mahanti of Motley Fool.


Research related to mobile devices in the Computer Society Digital Library:



About Lori Cameron

Lori Cameron is a Senior Writer for the IEEE Computer Society and currently writes regular features for Computer magazine, Computing Edge, and the Computing Now and Magazine Roundup websites. Contact her at l.cameron@computer.org. Follow her on LinkedIn.