XHTML in Mobile Integration

The word what of WAP 2.0 is going to be XHTML Mobile Profile. This can be a big step towards one Internet for fixed and wireless worlds, because the same site could be browsed both from the PC along with a mobile phone. To have an ordinary consumer, the change from WML to XHTML MP isn’t that dramatic, but, with respect to the browser, the “mobile Internet” may look a lot more like the Internet that individuals know in the fixed world.

A familiar feel and look is likely to increase user satisfaction when utilizing WAP services the very first time. At the time of penning this paper, there aren’t many WAP 2.0 devices and services available on the market. Also, there aren’t any guidelines for developing usable mobile services with XHTML MP. It might be beneficial, however, if there have been usability guidelines for WAP 2.0 already only at that early phase. Usability guidelines or strategies for WAP 1.x happen to be presented in a number of publications.

Nokia Research Center conducted a usability test of 20 test sessions for XHTML MP applications. The aim was to discover the user interface solutions that users prefer and discover easy to use along with the solutions that users find hard to use and therefore should be avoided. We wanted to discover whether you will find cultural differences between your preferences of various nationalities around the user interface solutions.

In line with the results, along with the results of countless earlier WAP 1.x usability test sessions, we compiled a summary of usability guidelines for WAP 2.0 service development. Within this paper, we focus on the differences found between WML and XTHML MP usage, and between different user types. At the conclusion, we list some differences between designing applications for fixed and cellular devices.

Procedure

The tests were through with Nokia 65xx cell phones, running a prototype version from the Nokia XHTML browser. The applications were utilised via a GPRS connection as WAP services, and also the implementation was pure XHTML Mobile Profile. The amount of subjects was 20: 12 in Helsinki, Finland and 8 in Boston, USA. The topics in Finland were from various Countries in europe or from Japan. The topics varied from active people that use the current mobile Internet to ones which had never used it.

Each interim management knew, however, a minimum of the principle of either WAP in Europe/USA or Japanese mobile Internet (i-mode, J-sky or EZ-web). All subjects were built with a mobile phone in daily use plus they knew how you can type having a mobile phone keypad; they’d either been writing SMS messages or inserted names towards the phonebook from the phone. The consumer interfaces utilized in the test were specified and developed due to this test.

There have been two services within the test: a news service for information retrieval as well as an auction service with interactive forms. Because we desired to make general guidelines, we required to compare the usability of various kinds of interface solutions. We selected basic interface solutions utilized in WAP, the web, or Windows today and built the help in three various ways. The three user interfaces contained exactly the same data, however the way the information was presented, navigation, and also the usage of the sun and rain varied in each interface style. Different users tested the applications and interface styles in various orders.

To prevent associations that will make one style much better than other, we named the styles based on fruits: Banana: Long pages, flat hierarchy; Tables; Selection lists; Images Orange: Short pages, deep hierarchy; Wizards; Option for text input or list selection; Images Apple: Accesskey shortcuts; Keyword search prioritized; Textual input; No images When making the user interface styles, we targeted the Banana style for novice users, Orange for intermediates, and Apple for experts. In Banana and Apple, we placed a couple of logical entities of information on one page, whereas in Orange we even split one logical entity onto several pages to reduce scrolling. Examples of these different interface styles follow.

Main Results

It would appear that the variations in the users’ preferences on different interface solutions weren’t between the subjects from different countries, but between your experienced mobile application users and novice users. There have been differences between countries within the experience level, however: all Japanese subjects were very experienced, whereas in Boston, the topics were not that acquainted with the mobile Internet.

The knowledgeable users emphasized finding information and making selections quickly. They liked to utilize a wholesale biscuits search when performing the duties, and requested direct links towards the main pages from the service. Text input was less painful for experienced users, plus they could better bear in mind the selections they’d made in forms.

Experienced users also easier understood the dwelling of the application, making it easier to allow them to know where these were. They were also more patient in awaiting the page to download. Because they services were getting used for the first time, many expert users liked the easy user interface styles and felt the Apple style, created for them, was too complicated. An extended study period could be needed to discover whether frequent use would boost the success from the Apple style.

Novice mobile Online users had to learn how to browse and employ the pages with no mouse. They didn’t always scroll down enough to determine the whole page. Novice users preferred a tree hierarchy to some search functionality in navigation, because seeing their email list of sections helped these phones understand the structure from the application and also to increase the feel of control.

You that were unfamiliar with text input using the phone keypad didn’t like to input text, but preferred selection lists. These were uncertain whether or not they used the machine as they were likely to, and when a webpage was downloading too slowly, they desired to cancel the download. These users preferred the Banana and Orange styles, however the ones who used the search function within the Apple style liked it.

Some Differences between XHTML Mobile Profile and WML

The brand new user interface elements supplied by XHTML MP include push buttons, check boxes, radio buttons, multi-line text fields, animated image format, and tools for sophisticated page layout. XHTML MP doesn’t provide decks of cards, timers or any other events, softkey assignments, input filtering, or multi-page forms. Exactly what does this mean for that user interface design?

Page Length

In WAP 2.0, information is downloaded as single pages, less decks. It has an effect on the perfect page length: oftentimes, it is better to possess a few long pages as opposed to the same information split on several short pages. Within our tests, subjects felt the download duration of a short page was more than that of a long page (time perception associated with amount of information expected).

The right page length depends upon the information onto it: target pages containing information that the user was trying to find may be even 20 screenfuls long, whereas the navigation pages normally shouldn’t contain a lot more than 10 one-line links.

Forms

It seemed that users began to lose control when the form was more than two screenfuls or maybe the form was split onto several pages. A multi-page form perform, if the form was short, familiar, there was no need in a situation to return and change the values. Users did take advantage of viewing all of the selections on a single page. When the users didn’t remember the selections they provided in a peaceful lab environment, they’re even less prone to do so inside a disturbance-prone mobile environment. Users also appeared to think that the input they provide is immediately saved, with no need to send the information to server. If there are many ways to proceed in the form page, the input might not get saved whether or not the user designed to do so.

Push Buttons

Users appeared to understand well the command buttons to complete the final actions for that tasks. It had been better when the labels in buttons weren’t technical terms; terms like Update or Submit are not as easy to understand than Save Changes or OK. Push buttons shouldn’t be used for cases when more information is required before the operation can be accomplished. In those cases, use hyperlinks rather than push buttons.

Text and Number Input

The conventional XHTML MP doesn’t provide a method to define whether a port field is numeric or textual. It is then hard for that end users to go in values, given that they have to do the mode vary from text to numbers, or the other way around. We recommend avoid requiring a mixture of numbers and letters in a single field, but putting the unit in the correct input mode would want a change in the XHTML MP language. Nokia devices will understand a cascading style sheet (CSS) property “-wap-input-format”, which defines the input mode and prevents input errors already around the terminal side.

Conclusion

Even though move from WML to XHTML Mobile Profile helps the applying developers in multi-platform development, there still remains possible to make the applying usage a hardship on the end user. Insufficient decks means the website structure should be different than in WML, and also the developers must use extensions to be able to have smart input fields. The developers must learn how to use push buttons and hyperlinks inside a consistent way. WAP 2.0 isn’t the only change the mobile Internet will face soon. At the same time, quicker networks (like GPRS), more WAP capable devices, and displays and input products are coming into use. They are likely to make WAP more usable also for that end users.