This is a review that I have been making of the Google Glass device from a User Experience perspective.
This was an unusual experience because I had no clear usage agenda but was interested in discovering what could be done with Google Glass and how it was to used.
Hardware Experience: Although it is not the reason for using a device, the first interaction with one is through the hardware. The first impression is that it is easy to wear and fits well. It’s not heavy or uncomfortable.
The bulk of the left arm holds the CPU and the micro-USB port for plugging in the additional ear piece or the USB cable used for charging and data transfer. The end of the arm contains the battery. The CPU used is a Texas Instruments OMAP System on a Chip (SOC) based on an ARM processor with additional GPU and memory interface functions, and is the same generation of chips as used in the Samsung Galaxy S2 phone from 2011.
Interaction with the device is by tapping or swiping the arm, or by speech recognition. Output from the device is by sound or visual output on the display. The viewfinder is positioned above the line of vision of the right eye and produces a 640 x 360 pixel image. Some interaction could be made by tipping the head or winking.
This Glass came with 16GB of FLASH memory, 12GB of which is usable for storing photos and videos or for adding apps. It has a camera capable of capturing 5 megapixel images or 720p video.
Communication Media Experience: Judging the quality of images and video depends entirely on the context of use, but subjectively they seem to be rather disappointing, with less colour range or sharpness than a typical 5MP compact camera, and more reminiscent of a 3MP phone camera. The audio quality is fine for speech but is not of sufficient quality for listening to music or engaging with a video clip.
Interaction Media Experience: The two interaction media that Glass supports are video and audio. The visual display was clear and sharp and was usable outdoors and indoors. The Glass outputs audio either through the skull bone (bone conduction) or through an earpiece that plugs into right hand arm. In either case the audio is clear in a normal domestic or working environment. For use in traffic, the volume would need to be raised, perhaps to unsafe levels.
Glass also has speech recognition, which seemed able to recognise places names and other data as well as programmed instructions. I worked well indoors, but not at all outside in traffic.
Networking Experience: Networking is fundamental to the Glass experience. It has WiFi and Bluetooth capability built in and a USB port for use as an umbilical for transferring data.
In order to experience and use the full range of services, it is expected that the devices couples to a phone using the Bluetooth connection and then access the Internet through the phone. This interaction is configured using a dedicated app (MyGlass) on an android device or an iPhone. When attempting this on an iPhone, MyGlass confirmed the bluetooth pairing and access to the Internet, but no services requiring Internet access worked. As an alternative, the Glass can be configured to work with WiFi either through the phone or a computer. The WiFi access credentials are entered into the website (and are therefore given to Google) and a QR code is generated which is read by Glass to enable access. This worked both through the computer and the phone.
Services such as Navigation did not however function when theGlass was connected to the internet through an iPhone configured to run a personal hotspot. Photos taken with Glass were transferred seamlessly to the google photo service in the cloud and were available through the web, so this type of service worked seamlessly and without the need for explicit manual synchronisation.
As the whole Glass experience is predicated on functional network and Internet access, which could not be achieved through an iPhone, the experience of using a Glass transparently for the full range of mobile services was not possible. This rendered the experience of engaging with Glass as a service provider platform impaired to the extent that it could become irrelevant.
Service Interaction Experience: Interaction with the services/apps supported by glass is achieved by:
- Tapping and swiping the arm
- Speaking items from a menu presented on the display (active items are presented in white, inactive menu items in grey)
- Nodding, where available
- Winking, where available
From the users perspective these interactions all worked well in ideal conditions. In all cases though, in day to day use there interactions were less than optimal for the following reasons:
- When using the glass by tapping the Glass arm whilst busy doing other tasks (the act of moving a hand from whatever task it has been busy with to reaching for the Glass arm and making the correct gesture in the correct location on the arm) is a skill that will take time and practice to perfect. Navigating to the picture taking service and then taking a picture with Glass arm taps whilst cycling, for example, is difficult and I believe unrealistic. Enacting tapping and swiping gestures whilst in a conversation is not “normal”. The conversation halts because the participants are reacting to the apparent loss of attention on the part of the glass wearer and the non-Glass-wearing participants in the conversation become distracted and/or concerned trying to understand what service has been interacted with and what media has just been captured, if any.
- Speaking commands do not work in noisy situations and need to be loud enough to be intrusive in quite situations. There are many situations where spoken commands would not be appropriate (meetings, art performances, church etc).
- The nodding gesture, as well as being a strange gesture to witness, seems to require extra processing causing Glass to lose power significantly more quickly than if this feature is disabled.
- Winking as an interaction is a very strange gesture to witness. Winking is generally an involuntary and necessary act which is required to keep our eyes lubricated. This can result in unintended gestures having unwanted consequences.
Service Functionality Experience: Glass has a very limited platform for communicating information to the user, based on a small screen and audible comments. In situations where the information supplements the task in hand (viewing an artefact in a museum, looking at a cityscape) Glass could be valuable. In situations where users are engaged in other tasks and are interrupted by Glass, attention can be lost from the task in hand, or the notification from Glass can be missed and lost. At best this can be irritating, at worst dangerous, for example while driving or operating machinery. An example could be using the recipe app whilst using a blender or food mixer, and then receiving a completely unconnected twitter message or email notification. The driving context is contentious because some have likened the Glass to a phone based or even dedicated SatNav system. The difference is that when used as a SatNav, notifications from other apps can’t happen, as in the case of a dedicated SatNav device, or can be disabled, as in the case of a SatNav app being used on a smart phone. Glass is not mature in this sense at this time.
Glass is perhaps best seen as an information portal device for apps running elsewhere. For that to be successful, the distribution of information between devices and synchronisation and connectivity between devices needs to mature. Glass also requires a user to become immersed in the Google eco-system to gain maximum benefit from services such as calendar and contacts. For users with an existing workflow based in IOS the experience is frustrating because some functionality is available but not all. For people using Windows or Blackberry mobile platforms, it is necessary to move data into the Google services or to duplicate data across the platforms.
Operating System Experience: Operating system interactions are concerned with getting a system into the state where it can be used to achieve usage goals. For some users this configuration and customisation is fun … for most it is not.
As with any operating system, there are issues concerned with navigating to the point in the options available to address the specify aspect to be configured. In Glass this involves swipes backwards and forward and up and down. As there is no single user manual, but rather a set of videos and manual pages online, firstly finding functions and then invoking them can be quite time consuming and tedious.
Conclusion: As a means of displaying contextual information in the visual medium, Glass can work very well. It is not uncomfortable to wear and the display is well placed to be visible without interrupting the field of view. Services such as Navigation and SkyMap and FieldTrip are good examples of this type of service functionality. This does not mean that information appearing on the display does not interrupt attention. Services that require interactions and that generate many notifications are much less successful on Glass. The interaction methods are just too unnatural and awkward, and socially unacceptable and intrusive when engaged in tasks or conversations. For this reason, contextual information services would need a degree of intelligence, centred principally on image and audio recognition to reduce interactions with Glass to a minimum. The push notification aspect needs to be refined considerably because the notifications are not synchronised with task workload, so may cause attention interruptions at critical moments.
Vocational activities where access to contextual information such as maintenance and repair, medical triage and diagnosis or field exploration could benefit from bespoke services on Glass.
The media capabilities of Glass are generally poor; a compact camera or smartphone is much better understood in social situations and in most cases will give considerably better results.
In conclusion, in all the use cases that I have been able to imagine, there are products that do a better job, and none that Glass does better. Even in the most compelling case, contextual information display, too little information is presented through the audio or video at too low quality that I would imagine quickly reaching for another device.
Please feel free to add comments or examples of good practice use of Glass below.