Department of Informatics, University of Oslo, PO Box 1080 Blindern N-0316 Oslo, Norway
In this paper, we review the public transportation information services in Oslo provided by Trafikanten. Our main concern is on the need for information that is universally accessible, regardless of the users' physical and cognitive abilities. Of particular interest to us are the challenges faced by visually impaired passengers seeking information, and how these can be alleviated by the use of mobile technology.
This article focuses on the importance of accessibility and universal design in public areas and services. All people, regardless of physical or mental abilities, have the right to use these common goods. This belief is exemplified by the recent Norwegian anti-discrimination law (diskriminerings- og tilgjengelighetsloven) (www.lovdata.no, 2009-01-01 ), which explicitly states that the public sector, as well as the part of the private sector that deals with the public, shall work purposefully to promote universal design.
One important service bound by this new law is public transportation. Most people need to get around, but for people with special needs, who might not be able to transport themselves in any other way, public transportation is even more vital. One such group of passengers is the visually impaired. It seems fairly reasonable to assume that visually impaired passengers face a number of challenges, not only when using public transportation, but also when gathering information about it, in the form of time tables or real time updates.
Such information can be made available through a number of different channels, including physical time tables, public service announcements, displays showing real time traffic information, web pages and mobile phone applications. Some of these channels are not fully accessible to the visually impaired, while others can be made accessible through certain design modifications.
In order to facilitate effective use of public transportation, it is also important that traffic information is made available where the user is. This can be achieved by placing accessible information interfaces at strategic locations, such as bus stops. However, the large number of stops, combined with the high cost associated with such installations, can make this an unfeasible approach for the majority of stops. A possible alternative is to allow the users to bring their own interfaces with them, for instance in the form of web capable mobile phones. In addition to being more cost effective, the mobile phone platform offers some interesting possibilities, such as user personalization and position based services.
The purpose of this article is to provide answers to the following questions:
This review is closely tied to our master theses, both of which are works in progress. The rest of the paper is structured in the following way: Section 2 looks into the theoretical foundation for this paper. In Section 3 we present the case, the methodology and the findings. We discuss our findings in section 4, before the conclusions are drawn in Section 5.
In this section we will look into the theoretical foundation for this paper, and at how the different concepts / terms are related and complete each other.
Access to travel information is an important factor for public transport quality (Grotenhuis et al., 2007). Grotenhuis says "travel information includes not only the content of information but also the condition of information (i.e the medium, layout, and ergonomics) and the composition of information". A blind or visually impaired person can use mobile technology to access this kind information if conditions are met by the information service provider.
Next, we look into how human-computer interaction and accessibility are related, and the importance of accessibility in relation to universal design and assistive technology.
There are many ways to define human-computer interaction (HCI). One frequently used definition is: "Human-computer interaction is the study of how people design, implement, and use interactive computer systems and how computers affects individuals, organizations, and society" (Myers et al., 1996)
Computer science as a whole has made remarkable progress in the last 20 years, and so is also the case with the field of Human-Computer Interaction (HCI). HCI has its roots in Interaction Design (ID), which can be defined as "designing interactive product to support the way people communicate and interact in their everyday and working lives" (Sharp et al., 2007). HCI, on the other hand, "is a discipline concerned with the design, evaluation and implementation of interactive computing systems for human use and with the study of major phenomena surrounding them" (Hewett et al., 1996).
According to Sharp, Rogers and Preece (Sharp et al., 2007), "the main difference between Interaction Design and Human-Computer Interaction is one of scope. ID has cast its net much wider, being concerned with the theory, research, and practice of designing user experiences for all manner of technologies, systems, and products."
For the visually impaired, interaction with a computer is based on different preconditions than that of a seeing user. For a blind or visually impaired person the auditory sense is one of the central senses for interaction with the real world (Heuten et al., 2006). Therefore, it is a good idea for interfaces to offer audible output, for instance in the form of text-to-speech synthesis or pre-recorded messages. This functionality can be integrated into the service itself, or the service can be made compatible with external text-to-speech programs (screen readers).
In the next subsection, we take a closer look at accessibility as a term.
Accessibility is hard to define, but nevertheless very important when it comes to HCI.
"During recent years, the attention to accessibility for all people has increased and equal opportunities for all people to participate in society are being emphasized" (Iwarsson and Ståhl, 2002)
Kato and Hori (Kato and Hori, 2006) mention that the information may be judged to be "accessible" when it appears to be easily perceivable by the user. But at the same time they point out that its content should not be judged to be "accessed" unless it is cognitively internalized or understood by the user.
Lid (Lid, 2006) argues that today's public transport is not usable for a lot of people with various disabilities. Increasing the accessibility of these services for all is an established political goal. Lid also makes an interesting point regarding the use of passenger surveys, namely that they only say something about the people who use public transportation, and not about those who don't, or the reason why they don't. There is, however, reason to believe that increased accessibility, will lead to more passengers. Additionally, measures that are taken to increase accessibility for all will also improve the quality of the service as a whole, thus benefiting all passengers. Westerheim (Westerheim et al., 2007), however, state that there will still be some environments filled with some obstacles that are in need of special solutions. Westerheim further argues that "Although these special adaptations are not in the spirit of true universal accessibility, in the short term, they will still help with the actual accessibility of the environment." The environment for this paper is reference to public traffic information.
"Many people have tried to coin a definite term for what constitutes accessibility of the environment. One often-used term is universal design." (Westerheim et al., 2007)
Ron Mace made the following definition of Universal Design in 1988: "an approach to design that incorporates products as well as building features which, to the greater extent possible, can be used by everyone" (Osteroff, 2001). This definition has later been refined by others, such as the one referenced by Westerheim, Haugset and Natvig (Westerheim et al., 2007) : "a process intended to promote the development of products or environments that can be used effectively by all without adaptation or stigmatization."
|Principle||Definition||1. Equitable use||Usable and marketable to people with diverse abilities|
|2. Flexibility in use||Accommodates a wide range of individual preference and abilities|
|3. Simple and intuitive use||Easy to understand, regardless of experience, knowledge, language skills or current concentration level|
|4. Perceptible information||Communicates necessary information effectively, regardless of ambient conditions or sensory abilities|
|5. Tolerance for error||Minimizes hazards and adverse consequences of accidental or unintended actions|
|6. Low physical effort||Can be used efficiently and comfortably, with a minimum of fatigue|
|7. Size and space for approach and use||Appropriate size and space for approach, reach, manipulation, and use regardless of body size, posture, or mobility|
The vision of Universal Design is to simplify people's everyday life by developing products, environments and ways of communicating that are useful for as many people as possible (The Center for Universal Design, 2008). The perspective is one of respecting and valuing the diversity in human capabilities, technological environments and contexts of use (Stephanidis and Akoumianakis, 2001). The definitions of universal design alone do not specify approaches to design practice or provide tools for implementing the concept. Therefore the Center for Universal Design in Raleigh, North Carolina, USA, introduced the concept of Universal Design Principles (The_Center_for_Universal_Design, 2008) as shown in Table 1.
"The purpose of these principles is to articulate the concept of universal design in a comprehensive way, and they are intended to be applied to all environments, products and communications" (Iwarsson and Ståhl, 2002).
We will now have a look at the design concept of Universal design and its importance in relation to accessibility. In the next Section we are going to look at how accessibility is mapped to standards, as well as the importance of having standards.
Standards are a crucial aspect of the Web, says Miller (Miller, 2005), and points out how the Web took an existing communication medium, the Internet, and used that medium in a way that made information more accessible. Web standards and guidelines are intended to provide web developers with information about how to create accessible web sites and evaluate the accessibility of existing sites (Shaun et al., 2007).
"All Web content has been created because some person or group of people wanted to accomplish something - whether to provide the weather report for a community or to create a family photo album or to sell cars." (W3C, 2001)
W3C accessibility guidelines, Web Content Accessibility Guidelines (WCAG) , is one of the standard which developers can use as a guide to make its Web content accessible for visually impaired people.
WCAG 2.0 is organized around four design principles of web accessibility, which are aimed towards guaranteeing the ability to access content. The principles are:
For each principle there are guidelines "which define goals and provide a framework to help authors understand the requirements" (Reid and Snow-Weaver, 2008), and for each of the guidelines, a number of testable requirements, called success criteria, are defined. These criteria describe specifically what must be achieved in order to conform to the guideline (W3C, 2008c).
According to professional reviews, qualitative heuristics are important to achieve accessibility (W3C, 2008b). The guidelines recommend usability testing to determine how well people can use the content for its intended purpose.
For a website to be in conformance with WCAG 2.0, the guidelines give a list of conformance requirements and describe what it means to be accessibility supported. By conformance, it is meant that "you need to satisfy the Success Criteria, that is, there is no content which violates the Success Criteria." (W3C, 2008a).
Reid and Snow-Weaver raise a difficult, but at the same time a highly relevant question concerning the user agents and assistive technology used to make the content available to people with disabilities in the form and modality that meets their needs. According to them, the answer will vary depending on environment and language. Since the public information is available via World Wide Web, they propose that the technology may need to be supported by a wide variety of user agents and assistive technology, including older versions, since users may not upgrade their software and hardware for many years.
As we have seen, WCAG 2.0 can be a powerful tool for making web pages accessible to people with disabilities, for instance visually impaired users aided by assistive technology such as screen readers. Assistive technology is the subject for the next subsection.
As mentioned earlier, visual impaired people need assistive technology to access information content which is available on the Internet.
Assistive technology is where technology is used "to either help compensate for a disability or to provide accessibility to information and services, and to in general improve the quality of life of a disabled" (Liffick, 2004).
The term "assistive technology" (AT) will in this paper be limited to describing handheld devices such as mobile phones or PDAs "and/or software that is used to increase, maintain, or improve the functional capabilities of individuals with disabilities" (Liffick, 2004) as well as text-to-speech synthesis.
Liffick (Liffick, 2004) also mentions that "assistive technologies has obvious intersections with many issues that are classic HCI topics: human factors (e.g. range of motion, cognitive abilities, psychology issues); interaction devices (alternate keyboard designs, pointing devices, and selection mechanisms); interaction methods (Morse code, voice recognition and output, scanning techniques, word expansion, word prediction); modes of communication (single or multi switch, audio and voice, alternative languages); web accessibility (mechanisms that interfere with AT devices or techniques, dealing with graphics for the visually impaired, audio captioning); and usability testing (user/device matching).)"
In recent years, AT has also become integrated into ever smaller devices, such as mobile phones. This miniaturization opens exciting possibilities, which we will look into in the next sub-section.
Mobility can be defined as a measurement of how easily one is able to move (Preston and Raje, 2007). Mobility, accessibility, barriers and resistance are related concepts, where one concept often is used to describe one of the others (√˜vstedal, 2009).
According to Dey (Dey, 2001), several researchers have attempted to define context-awareness in computing. Schilt and Theimer were the first to use the term "context-aware" and they, according to Dey, refer to context as location, identities of nearby people and objects, and changes to those objects. Dey also states that definitions that are based on specific examples are hard to use, since the dilemma lies in being able to determine if certain kinds of information not mentioned in the definition should be considered to be contextual or not. Dey's definition, which we will use in the following, is "Context is any information that can be used to characterize the situation of an entity. An entity is a person, place, or object that is considered relevant to the interaction between a user and an application, including the user and applications themselves." Dey mentions further on that if a piece of information can be used to characterize the situation of a participant the situation of a participant in an interaction, then that information is context". After specifying what is meant by context, Dey also provides us with a definition of context awareness.
"A system is context-aware if it uses context to provide relevant Information and/or services to the user, where relevancy depends on the user's task" (Dey, 2001)
Now that we have a better understanding of some of the concepts related to accessibility, we will have a closer look at solutions for conveying traffic information in the next section.
Information about public traffic can be delivered to the users in a wide variety of ways. There are strengths and weaknesses associated with all of them, and some are more accessible than others. Real-time information systems are very common in modern public transportation (PT) and a considerable amount of money is spent on such applications each year (Yeung, 2004). In this respect, the Greater Oslo Area is no exception.
We will now give a brief summary of the information channels used by Trafikanten and Ruter in the Greater Oslo Area, in addition to suggesting a new service for hailing drivers.
Figure 1: Overview of existing technologies for delivering public traffic information
These are placed on all subway stations as well as some of the major bus/tram stations. They are mainly used to inform about irregularities in traffic, not for giving general information. The reason for this is that they are very loud and intrusive, meaning that prolonged usage is unwanted. This may be resolved by using a lot of small speakers instead of a few large ones, as the volume and sound quality would then be more pleasant to the ear. When the PA system is used in combination with electronic information signs, the vast majority of people will be able to receive the needed messages.
More discreet than the noisy PA systems, these are small speakers that are activated by users, either by pushing a button, or by standing in front of it. Because the users will be right next to the speaker, there is no need for high volume. This means that more detailed information, such as time tables or real time traffic information can be made available without being a nuisance to other passengers. This system is currently being tested by Trafikanten at Rikshospitalet tram station. The current implementation does not use text-to-speech synthesis, and therefore requires someone to record all messages in advance. Additionally, the technology required for this system is expensive, which is currently a major hindrance for deploying it at more than a few selected stations.
Traditional time tables are by the far most common information channel used by Trafikanten and Ruter in the Greater Oslo Area. They are cheap, require little maintenance except for when routes are changed, and offer large amounts of information in a small space. Unfortunately, they offer no form of real time information or information about irregularities. For users with visual impairment, they offer no information at all, unless they ask other passengers for help. Using Braille letters to alleviate this problem would probably not be very efficient. Such panels would be much larger than regular time tables, as well as more expensive. Not to mention the fact that not all visually impaired people understand Braille.
Trafikanten has several types of electronic displays placed on selected stations, including the entire subway network. Typically, they show estimated wait times for the upcoming departures. These displays are of no use to blind users, but partially sighted users may benefit from them. Conversations with such users have revealed that they favor the pillar based displays (Fig. 3) over the hanging variety (Fig. 2), as these allow users to stand extremely close to the text.
Figure 2: Real-time screen at bus/tram stop (left) and subway station (www.trafikanten.no)
Figure 3: Multi real-time screen (www.trafikanten.no)
Trafikanten is also contactable through a manned telephone service, 177, which can be used to obtain information related to routes and prices. Giving the users the possibility of talking to another human being instead of a machine means that this service requires little technological insight from the user. There is, however, a certain delay before information is obtained, and this may be detrimental to user experience in some situations. Unlike most other channels of information, there is also a notable cost associated with using this service.
By sending text messages with certain code words to 2050, users will be sent the desired information via SMS. Apart from text messaging being a familiar technology for most users, it is difficult to see any clear advantages of this service over the web alternatives. Additionally, each text message has a fee of NOK 3,-.
Trafikanten offers route planning, time tables, ticket information and real time information on their web pages. These are quite old, and therefore not quite up to today's standards in web development or accessibility. Modern mobile phones are also able to access it, but users of these are probably better off by using the new mobile web pages, or even one of the mobile applications.
For many years, Trafikanten offered an information service through WAP, but this was eventually thought to be a slow and tedious system. Recently, they therefore released a more modern version of the service, making an effort to comply with the accessibility guidelines of web developing. The fact that such established guidelines, for instance WCAG 2.0, exists is one of the strengths of using the mobile web for services that require accessibility. Another benefit for visually impaired users is that mobile text to speech synthesizers often are compatible with the built in web browser, giving the user a familiar interface to work with.
The other branch of mobile services consists of independent applications for various platforms. At the moment, such applications exist for java capable phones, iPhone, Android and Windows Mobile. Of particular interest is the possibility to use GPS to pinpoint the user's location, thus using context-awareness to minimize the amount of input needed to select stations and routes. However, accessibility might be an issue for such applications, and this is something that must be addressed. As of version 1.5, the iPhone application supports VoiceOver (Apple, 2010), which is a gesture-based screen reader for iPhone. Initial testing done by Trafikanten suggests that this combination works well.
Figure 4: Real-time iPhone application (www.trafikanten.no)
While there are currently many ways to receive information, visually impaired passengers have little to no possibility of informing drivers of their intent. This can be a problem when multiple buses stop at the same stop at the same time, or when a bus requires waiting passengers to reach out an arm to signal it to stop. Ben√Ętre et al. (Ban√Ętre et al., 2004) describe Ubibus as an application designed to help blind or visually impaired people to take public transport. Their focus is on the tasks of signaling to the bus driver to stop, and knowing when the correct bus arrives at the stop. To achieve this, they propose using PDAs or mobile phones with WLAN or Bluetooth capabilities.
As we have seen, Trafikanten provides a multitude of channels for information about public transportation. This is a good starting point, as it increases each user's probability of finding a solution that is suited to their needs. For instance, solutions that are practical for visually impaired users may not be usable at all for a deaf person, and so on.
The original information system for public transportation is plain time tables printed on paper. These are still available at almost every station, and quite often they also provide the only available information at a station. Partially sighted people may be able to use them, but for blind passengers they are inaccessible. In theory, time tables printed in Braille could solve this, but the sheer size and complexity of such tables would be impractical.
LED displays are a more modern and flexible variation of the paper time tables, the obvious advantage being the possibility of giving dynamic information. Examples of such information are real time traffic information and important messages. These displays offer nothing to blind users, but partially sighted passengers may benefit from the increased font size compared to paper time tables. This does, however, require the displays to be at eye-level, making them possible to read at a close distance. Because of costs, and unlike paper based time tables, these displays are not available everywhere, nor are there any plans to make them available everywhere.
For visually impaired users, PA systems are similar in usability to what they are for passengers without disabilities. Unfortunately, this usability is rather limited, for a number of reasons. Firstly, it is a push based service, meaning that the recipients cannot choose what kind of information they want. Secondly, PA systems can be noisy, and quite annoying to the general public, so they are mostly used for high priority messages only. Lastly, such systems are only available at some stations, further reducing usability.
Systems that utilize audible information on demand are an improvement over the regular PA systems. They are less intrusive towards other passengers, and it is therefore acceptable that they provide more detailed information. Trafikanten's solution currently relies on pre-recorded audio, but text-to-speech synthesis will be implemented in future versions. The main problem with such systems is the cost associated with installing them. In a cost-benefit analysis, it is simply not feasible to place such systems at stations without a large user base.
The call center and SMS service have the benefit of being based on familiar technologies, which helps users feel more comfortable while using them. The call center in particular can be very appealing to those who struggle with modern technology, as it allows users to talk with an actual person. For the visually impaired, this is a very relevant service, not only because sight is not required to operate it, but also because it's available everywhere, as long as one carries a phone. The downside of these services is that they are by far the most expensive to use, particularly from a mobile phone. Additionally, they are considerably slower than some of the other channels.
The services based on web pages are probably the most versatile of all, but given the need for a quick response, it is vital that the interfaces are simple and effective to operate. This has recently been improved upon, but doing a straight forward search may still take a minute or two, during which several decisions has to be made. The solution to this, which many users are unaware of, is to bookmark search pages for easy access in the future. This is a particularly important trick for visually impaired users, as the full search process is more cumbersome with a screen reader.
Mobile applications are very similar to the mobile web pages, but the interface can be made much more streamlined, with features such as auto complete. One of the most interesting possibilities, which made the iPhone application an instant success, is the introduction of location-awareness in the form of GPS/triangulation. This eliminates the need for typing in long station names with tiny keyboards, as the user's position limits the relevant stations to only a few. Choosing between these is then as simple as selecting one of them. For visually impaired users, it may be problematic to get such applications to work with their screen readers, as compatibility on some mobile platforms can be quite limited. If compatibility with existing screen readers is a problem, an alternative would be to include audio input in the programs themselves.
As we have seen, phone based services are currently the only real alternative to paper time tables at a large number of stations, and this will not change in the foreseeable future. One of Trafikanten's strategies is therefore to improve these services, so that dynamic information can be accessed effectively from anywhere. With the release of new mobile web pages, as well as applications for most major mobile platforms, passengers are already reaping the benefits of this strategy.
Upgrading the stations and stops themselves will probably be a much more challenging task, as the demands for infrastructure will be greater. This is, naturally, a result of having to provide the users with physical interfaces, instead of allowing them to use their own in the form of computers or mobile phones. These interfaces are expensive, and there is a lengthy beauracratic process involved before they can be installed. As a result, Trafikanten evaluates the costs and gains associated before a decision is made. Typically, the LED signs are placed at stops with more than 2-300 passengers per day. For the moment, the requirements for placing push buttons with audible information are much stricter. At the time of writing, the only such system in service is located at the Rikshospitalet tram station. This makes sense, as the vicinity to a large hospital means that the percentage of visually impaired passengers is probably higher than elsewhere. For the same reason, there are plans to install the same system at the station closest to the offices of the National Federation for the Blind.
Due to the costs involved with installing interfaces at stops and stations, it does not seem very likely that they will be available at every little stop in the foreseeable future. Therefore, we believe that there will always be a need for portable devices, such as mobile phones, to provide users with information where there are no other channels.
While there are many ways for the visually impaired to access information, it can be much more challenging to get a bus to stop, or to board the correct bus if there are more than one available. To solve this, one would have to device a way for the passenger to signal the driver to wait. In other countries in the past, this has been done in a variety of stigmatizing ways, such as holding up a sign with the desired bus number. Today, however, it should be possible to find a solution based on wireless communication.
Trafikanten provides a wide variety of channels through which information can be accessed. This helps ensure that passengers, the visually impaired included, should have at least one service that is suitable for them.
The systems that consist of a physical installation at the station can be quite powerful, but they fall short of universal accessibility due to the fact that they only cover a percentage of stations. This may change in the future, due to cheaper and more available technology, but there are currently no such plans.
Services using mobile phones, however, suffer from no such limitations. The vast amount of available information, as well as the possibility to access it from anywhere, is the main advantage. With a compatible screen reader, visually impaired users can also benefit from these technologies, particularly after usability testing has been performed.
While it is possible to find out when a particular bus is leaving, it may be very difficult to identify it for visually impaired passengers. At the busier stations, this is a serious challenge, which should be addressed as soon as possible
As we have learned, visually impaired users of public transportation are faced with a series of imposing challenges. In light of these challenges, we would like to do an extensive accessibility evaluation of the services offered by Trafikanten to users of the public transportation system in the Greater Oslo Area. The details of this evaluation are not set in stone, but there are several things we would like to investigate, if time permits:
Heuristic evaluation is a powerful method to discover usability problems in a system. A set of heuristics, or principles, is defined, and a small group of experts perform an evaluation of the system against these heuristics. In this case, the heuristics will be WCAG 2.0, and the experts will be students with an interest in HCI. This will help decide whether or not the new web pages are WCAG 2.0 compliant, and to which extent it is compliant (A, AA, AAA)
To be able to get an even greater insight in the matter at hand, it is vital that we spend time with the actual users of the services. If all goes well, we will be able to perform usability testing of one or more of these services with visually impaired users. Which exact services that will be evaluated depends partly on the wishes of Trafikanten and the National Federation for the Blind, but likely candidates are the mobile web pages, the iPhone application and the system for audible information on demand.
We would like to thank B. Flyen and M. Bentzen from Trafikanten for their valuable insights, and our supervisor, J. Herstad, for his ongoing support.