Introduction

Various user-centered design methods were used during the InstantShareCam design project to understand the user experience and evaluate the design concept.

Produced design proofs (usage scenarios, storyboards, experience videos, crude physical model of camera, website, screen UI mounted on a small monitor) all along the processes goal was to to encapsulate users' experience in a "tangible" and communicative form. Unlike the traditional usability testing processes, these methods are not intended to evaluate the performance issues of the design.  But they were used to understand and evaluate the experiential components of the design project.

The following paragraphs describe some of the key tools used to explore design opportunities, understand users' context and evaluate the design.

Introduction

Observations

Experience Scenarios

Service Simulation I

Service Simulation II

UI Prototypes

Observations

Design questions: What kind of design opportunities exist in media documentation?Who are they?

In the explorative phase, contextual observations on consumer camera users were conducted. I observed tourists at famous sightseeing spots in Berlin documenting their travel with cameras. Each observation (total: three spots) was written in the user observation book and video taped for 30 minutes.

The other contextual observation was at the Orange Carnival festival in Ivrea, Italy. The Carnival is the most important community gathering and tradition for people in Ivrea. The contextual research at the festival lasted half- day each time over two days on two different locations of the town. The second contextual observations were also taped and complemented by short interviews.

The choice of contexts was motivated by the assumption that users have different documentation needs as they are experiencing different social contexts. A major event attendance such as Orange Carnival has a collective significance. The entire local community participates to create the event. On the other hand, the tourism in Berlin is constrained for the traveling group without much involvement of the local community.

Insights:

1. People check the quality of the pictures right after they captured the scene, and such activity generates an immediate gathering around the small LCD. After checking the photos, if they like the picture, they would ask the cameraperson to share it after the trip.

2. Taking a snapshot within a group takes longer to shoot. People take turns to make sure that the cameraperson is included in one version of the group. The cameraperson used several cameras to capture the same shot on several devices. In other words, the captured media was shared by capturing the scene for several times. This behavior is an ad hoc system that users came up to share the pictures on the location.

3. Dodging the camera focus. People have to move around the sightseeing spot carefully to avoid other people's camera shoot. They do not want to be part of strangers' pictures.

4. A higher percentage of video cameras were observed during the Orange carnival.
There was roughly one video camera for every five photo cameras. Whereas, in Berlin, the ratio for the number of videocamera/photocamera seemed much lower. It is difficult to tell precisely if people take photos or videos as the latest models of photo camera takes all video format. A quick interview at Orange Carnival with 3 video camera users gave more insights on how they chose to use videos instead of photos:

-The event is of a high emotional value for the users.
“My daughter is in the parade for the first time!”
-The event is regarded collectively important.
“It is only one a year, and it is our historic tradition!”
-The spectacle is in motion.
“You can only catch the movement of the orange battles in video”

These insights founded the design framework of the project:

-Image sharing system among different camera devices on the location is an unexplored opportunity.
-Users who do not use desktop computers on a regular basis could highly benefit through a system that allow sharing media files from camera to camera.
-The sharing system for video camera is an opportunity to be explored around major collective events.

[top]

Experience Scenarios

Design questions: What would be the experiential drivers of the services? What are the constraints in providing the ideal experience?

During the concept development phase, many scenarios were produced in order to encapsulate the essential experiential requirements of the product.

The contextual observations were largely accountable for shaping the archetypical users in the scenarios. For semi-professional users, the data was collected through my personal work experience as video maker and "hands-on" experience from the in-situ improvisation (Service design Simulation I) in video production. Creation of the archetypical user's profiles for both semi-professional and consumer helped to define the key motivations of the users to use the product. The scenarios define the key task flows and features motivated by these users' motivations.

Invited peers among the Interaction Design Institute of Ivrea, students, professors and potential users were involved in reviewing the scenarios. The reiteration of reviews provided feedbacks and critiques to bring adjustments in the scenarios. As the scenarios develop, the media used to convey the experience stories became more communicative. In the final stage, video scenarios were produced in order to convey the purported experience to the potential users outside the peers in the Institute. Theses videos were the inseparable companion for the InstantShareCam service website rendered as part of the design simulation (see below design simulation II).

In short, the scenarios identified users' experience requirements and served to raise critical issues in rendering the designed experience:

- What is the service model for setting up the infrastructure that supports high-speed wireless connections?
- What is the business model? Who provide what and how much does it cost?
- What kind of system architecture would you need?
- Users are more and more demanding in quality of digital imaging and the speed of communication, what is the acceptable speed and quality broadcasted media?
- The leap from 3G to 30G to assure the instant sharing can lead to a safety issue raised by the microwave technology. Are there any other alternatives?
-Who owns the digital media in the service scenarios? How can you be sure that the shared media is not misappropriated? Is it acceptable to give access to surveillance camera contents?
-The Camera interface design. Do you add the functionality as physical or software interface?

All these questions define the design problem areas. In this preliminary research set at Interaction Design Institute of Ivrea, the goal was to shape a clear experience design vision with the intent to foster a consorted multi-disciplinary development effort to solve these problems.

Scenarios summary:
experiencescenarios.pdf (1.6MB)

[top]

Service design simulation I

Design questions: What is the experience of collaborative video production? How does InstantShareCam service would fit in that context?

Some initial assumptions were already forming concerning semi-professional video making based on literature research complemented by my personal experience in professional movie industry. These design assumptions were taken to test by in-situ improvisational technique. 

The simulation of the InstantShareCam service was conducted during the video documentation project of students in Ivrea. Their video documentation set-up presented characteristics where InstantSharecam could likely be used. They were shooting in two remote locations at the same time. The shooting needed to be coordinated in real-time to have a synched video tape.

The simulation consisted of Human InstantShareCam improvisation. Two persons, including myself, improvised to act out the service features of InstantShareCam.

-Camera as a total video production assistant
The concept of InstantShareCam is to service the collaborative workflow of the video production from pre-production to post-production. I wanted to understand how does the collaboration is formed and explore the possibilities to facilitate it. The human InstantShareCam actors were asked to assist the shooting crew from pre-production to post-production by acting as facilitators to evaluate the functionalities that can be created to service the video production.

-Camera as audiovisual communicator between the crews
The key feature of the project is wirelessly connected camera to enable a collaborative and collective video documentation, whether be it in casual situations of tourism or in semi-professional video production (see the users’ motivations in the experience scenarios document).

The initial plan was to set up an Internet system to feed real-time video from one location to another in order to simulate such feature. The lack of Internet connection on site and the scheduling of the documentation crew opted out the former set-up. We video shot the students video shooting without real-time feed from one location to another. We simulated the connected space by a long open phone communication on site. We complemented communication lines by mobile communication while on the go. We focused on observing the information that was exchange through the existing technology, and be a prompt smooth communicator in-between the remote crew.

-Camera facilitating the video editing by real-time logging
For the improvisation, we used simply two sets of paper slates. One set described the likely events of the documentary. The second set is annotation slates that marked good or bad shots. It is far cry from being a real-time logging tool. But my intent was to evaluate if onsite marking would be useful for video editing.

After the session, the actors and the documenting crew were asked to discuss their experience regarding the session and brainstorm the ideas around InstantShareCam.

Findings

-Flexible organization
The documentary making is far more organic then I assumed. The shootings are hard to be planned ahead unlike the scripted movie shooting I have observed in the past.

-The critical factor: speed of interaction between people
In the pre-production phase, the actors used voice communication, to check on the location of the other shooting group and if they are properly setting up the cameras. During the shooting the voice was not used in order to not interfere with the video recordings. SMS for communication during the video shoots was useful but actors and shooting crew felt that the method was not quick enough to coordinate the group.

In short, the content of the communication concerns where the other group is (geo-location) and what they are shooting currently (video recordings). Actors and shooting crew felt that video feeds from the other location would have been a desirable feature that could facilitate their coordination. This was brought up as a result of the incident we have experienced during the video shoot. There was a blackout at one of the shooting locations, and the documentary making had to be terminated. During the incident, one of the InstantShareCam actor put on his nigh-vision to keep on documenting. The crew felt that if they could view through the night-vision the other location, it would have caused less panic and would have informed “in a glance” what was going on.

-Camera operation
The participants thought that the digital slate feature was not a useful feature (rendered here as a paper slate). The circumstances of the production changes as time passes. The crew has to improvise as they go, and they felt that they could not really interrupt the flow of video taping by inserting paper slate. If the insertion of video marking were integrated in the camera operation, such as attached as an easy button on the camera, they thought that they may use. It was more of a reasoned motivation to suggest such feature as we felt that during the post-production, the visual markings can help to quickly scan the video taped to edit it.

[top]

Service design simulation II

Design questions : How do users get in touch with the services? Is the purported services echo the users’ desire?

A fictitious marketing website of InstantShareCam was used to test the acceptance of such service among potential users. The website was designed as if the service existed. It explains how the product works and how much the service cost and how it can be used (Video scenarios and testimonials). The website hinted how the business is possible by creating a fictitious partnership between the known brand names. The Sony Playstation Portable was used for the concept vision.

The review of the website happened naturally by way of the pseudo-viral marketing. The news on the existence of the service was spread once a major blog site has reviewed the website. The cascading references by other bloggers created a small buzz and attracted visitors to the website.

“Viral marketing and viral advertising refer to marketing techniques that seek to exploit pre-existing social networks to produce exponential increases in brand awareness, through viral processes similar to the spread of an epidemic. It is word-of-mouth delivered and enhanced online; it harnesses the network effect of the Internet and can be very useful in reaching a large number of people rapidly.” Wikipedia

The marketing technique was put to use in this project to reach the opinion leaders in the TV broadcast community and peers in interaction design community. In May 2005, the site recorded 32783 hits. The former number is calculated as [total hits 50698] - [hits from the network 13209]-[hits from Italy 4706]. The actual visits to the website were 980. The high interests came especially from the professional market ranging from major broadcasting companies to semi-professionals.

Web stats May 2005.

The website has a disclaimer to let visitors know that the service is a research concept. Regardless, three emails from users who missed the disclaimer wrote to inquire about the service pricing and availability of the beta testing program in their region. The emails came from an individual, a non-profit organization and a broadcasting company. These reactions to a fictitious marketing website do not assess the performance of the provided services. However, if the promise of service experience echoes to users’ desire to experience it, I believe that the service has defined the right user experience framework.

In the ideal testing scenario where resources and times are plenty, an in-depth focus group testing should be conducted in order to decipher the details on the reactions to InstantShareCam concept.

Side Note
The recording of hits from US military, and subsequently from the US political parties is not such a quaint visitor. It shows the subversive nature of InstantShareCam where the line between a democratic surveillance system and participatory entertainment system is thin. We live in the world where such ambivalent system is culturally acceptable and desirable. The allure of easy sharing and prospect of new video making technique seem to transcend the notion of surveillance in favor of new creativity, at least for the semi-professionals.

[top]

UI prototype

Scenarios reached the stage of refinement where the context and experience are plausible enough to the point of attracting positive reactions publicly (Simulation II). The next step was to start looking closely the design problem of what to render as software interface and what to leave as hardware interface. Investigation on the product design started as the rough foam core models to evaluate the size and weight. Some interface ideas for hardware was drawn as sketches. In parallel, the interface ideas for software started as flowchart to categorize the tasks seen the in the scenarios. To test some of the interface ideas for software , I started as a test through low-fidelity user interface schematics.

Design question:  How low is the resolution of the small LCD screen?

User interface schematics were designed in Adobe Photoshop by using a small LCD screen. Each schematic comp was transferred on digital videotape for viewing on the camcorders LCD. The process informed about the visual needs (font/icon size, colors. The test results lead to leave out textual interface to a more iconic representation for the readability of information. This was not surprising as many of the camera interfaces relied on iconic representation of the interface in the past. However, with the increase of the resolution in the display system in the near future, the framework for on-screen visual design is likely change.

Design questions: Is it confusing to have multiple video streams on small screen?

Seeing a static Photoshop UI schematic is quite different when the UI schematics start including real-time videos. The videos are showing constantly different movement and it could make the interface difficult to read.
To test this hypothesis, a crude prototype with real-time video feeds was created by using Macromedia Flash and FlashCom Server over the local network at the institute. The prototype used four computers, two webcams, two regular video cameras connected by Firewire cables to the computers. One computer was linked to 6-inch monitor to simulate the camera interface.

The test was conducted among the students and reviewers at Interaction Design Institute of Ivrea (4 semi-professional users). The small monitor that acted as the camera interface was a passive test material to check on the perceived affordance of the interface. The prototype shows how it feels like when one of your friends in the Buddy group signals you while all of them broadcasting their respective camera view to each other. The signal is indicated by blinking on-screen Buddy icon.

Before the test, users were shown the experience scenario storyboard of “Guerilla Video Shooters”. They were told that the small monitor is the prototype version of what “Myriel” (the character name in the scenarios) see on his camera. For the ease of testing, the three videos on the quad-view were pre-recorded except the fourth quadrant, which is the tester’s webcam attached to the small monitor. I asked users with the small monitor in hand to keep the focus on one subject. While they are focusing on to the subject to film, I pressed the buddy icon on my interface screen on computer to see if they would see the signaling.  Afterward, I interviewed the tester how they feel about the interface.

Despite the crudeness of the prototype, the simulation provided more feedbacks on design. 

- Users noted about the speed of the broadcasting. We observed often the slow frame rates of the streamed pre-recorded videos. The pre-recorded videos intentionally filmed a subject with lots of movement so that any slowness in frame rate is perceived. The reaction to such slowness was “it is funny” and “so weird”.

Such reactions confirmed again the importance of efficient video distribution design discovered as the problem area through scenario storyboards evaluation (Cf. Section Experience scenarios). For the next interactive prototype, the broadcast speed needs to be carefully considered in order to not distract users from the primary tasks.

- Some thought that quad-view Multiviewfinder was meant to guide them to complement what other people are recording with new video streams. They found that it may be an interesting interface to use it as a reference interface to frame better their video shoot.

The Multiviewfinder[fig. 18] was initially designed for displaying information of the peers shooting videos and not for recording the event they are watching. It is under “Producer” mode as the screen allows producers to supervise the camerapersons on the fields and direct them. However, it seems that such interface can be included as the interface for shooting video. This reminded that semi-professional users are not as strict in the role distribution in the collaborative video production. Someone who leads other camerapersons can be at the same time cameraperson. Thye are more flexible and has interchangeable role. The feedback was taken into account to change the application flowchart. The Multiviewfinder, which was in Producer module, has moved into the Camera module to support multiple-stream video recordings to fit the hobbyist style of filming videos.

- In reaction to blinking icon, users touched the icon though it is not interactive (on the small screen). 
Users are capable of discovering the interface despite of the busy interface screen with multiple video streams. Users have moved the camera rendering the action distractive to their main goal of filming the scene.

The proposed UI schematics seem to be acceptable for semi-professional users. They already have a good level of visual literacy to not be distracted by the moving video images. The testing environment and material are too crude to draw any conclusion on the general performance of the UI schematics. For the next phase, the prototype needs to be more robust to take the test in the outdoor context where user will be in a more distractive environment.

[top]

Conclusion

fig.18. MultiCam ViewFinder interface prototype in Macromedia Flash.