The goal of this document is to aid the development of rich and dynamic mobile Web applications. It collects the most relevant engineering practices, promoting those that enable a better user experience and warning against those that are considered harmful.
This section describes the status of this document at the time of its publication. Other documents may supersede this document. A list of current W3C publications and the latest revision of this technical report can be found in the W3C technical reports index at http://www.w3.org/TR/.
a W3C Candidate
Recommendation of Mobile Web Application Best Practices,
expected to become a W3C Recommendation. The
W3C Membership and other interested parties are invited to review
the document and send comments to firstname.lastname@example.org (with
public archive ). Specific instructions to send implementation
feedback on the document is available in the implementation
feedback template . This document was developed by the
Mobile Web Best
Practices Working Group as part of the Mobile Web Initiative
complete list of changes since publication as a Last
Call Working Draft on 6 October 2009
Main changes are:
developer community. The Mobile Web Best Practices
Working Group expects to request that the Director advance this
document to Proposed Recommendation once:
No feature has been identified as at risk at this point. The
Working Group does not plan to request to advance to Proposed
Recommendation prior to
working group decided to consider and
statements that may be regarded as forward-looking when this
document is published (11 February
2010). Examples provided in
the How to do it sections sometimes reference on-going
works on technologies that the group expects to become prevalent in
the development community soon. The group encourages mobile
developers to report on their use of these technologies in mobile
Publication as a
not imply endorsement by the W3C Membership. This is a draft
document and may be updated, replaced or obsoleted by other
documents at any time. It is inappropriate to cite this document as
other than work in progress.
This document was produced by a group operating under the 5 February 2004 W3C Patent Policy . This document is informative only. W3C maintains a public list of any patent disclosures made in connection with the deliverables of the group; that page also includes instructions for disclosing a patent. An individual who has actual knowledge of a patent which the individual believes contains Essential Claim(s) must disclose the information in accordance with section 6 of the W3C Patent Policy .
1.1 Purpose of the Document
1.3.1 Best Practices
1.3.2 Web Application
1.3.3 Mobile Context
1.3.4 Delivery Context
1.4 Relationship to other Best Practices and recommendations
2 Structure of Best Practice Statements
3 Best Practice Statements
3.1 Application Data
3.1.2 Use Appropriate Client-Side Storage Technologies for Local Data
3.1.3 Replicate Local Data
3.2 Security and privacy
3.2.1 Do not Execute Unescaped or Untrusted JSON data
3.3 User Awareness and Control
Inform the User About Automatic Network Access
Sufficient Means to Control Automatic Network Access
Ensure the User is Informed About
Use of Personal and Device Information
3.4 Conservative use of resources
3.4.1 Use Transfer Compression
3.4.2 Minimize Application and Data Size
3.4.3 Avoid Redirects
3.4.4 Optimize Network Requests
3.4.5 Minimize External Resources
3.4.6 Aggregate Static Images into a Single Composite Resource (Sprites)
3.4.7 Include Background Images Inline in CSS Style Sheets
3.4.8 Cache Resources By Fingerprinting Resource References
3.4.9 Cache AJAX Data
3.4.10 Do not Send Cookie Information Unnecessarily
3.4.11 Keep DOM Size Reasonable
3.5 User Experience
3.5.1 Optimize For Application Start-up Time
3.5.2 Minimize Perceived Latency
3.5.3 Design for Multiple Interaction Methods
3.5.4 Preserve Focus on Dynamic Page Updates
3.5.5 Use Fragment IDs to Drive Application View
3.5.6 Make Telephone Numbers "Click-to-Call"
3.5.7 Ensure Paragraph Text Flows
3.5.8 Ensure Consistency Of State Between Devices
3.5.9 Consider Mobile Specific Technologies for Initiating Web Applications
3.5.10 Use Meta Viewport Element To Identify Desired Screen Size
3.6 Handling Variations in the Delivery Context
3.6.1 Prefer Server-Side Detection Where Possible
3.6.2 Use Client-Side Detection When Necessary
3.6.3 Use Device Classification to Simplify Content Adaptation
3.6.5 Offer Users a Choice of Interfaces
3.7 Further Considerations
3.7.1 Consider Use Of Canvas Element or SVG For Dynamic Graphics
Appendix 1: Best
Practice Dependent Device Properties
Appendix 2: References
2.1 MWI References
2.2 Device Independence
2.3 Web, Protocols and Languages
2.4 Other References
Appendix 3: Acknowledgments
The following Best Practices are discussed in this document and listed here for convenience.
Inform the User About
Automatic Network Access Provide Sufficient Means to Control
Automatic Network Access Ensure the User is Informed About Use of
Personal and Device Information
This document sets out a series of recommendations designed to facilitate development and delivery of Web applications on mobile devices. The recommendations are offered to creators, maintainers and operators of mobile Web sites.
Readers of this document are expected to be familiar with the creation of Web applications, and to have a general familiarity with the technologies involved, but are not expected to have a background in mobile technologies or previous experience with Mobile Web Best Practices (BP1) [MWBP] .
The document is not targeted solely at developers; others, such as interaction and graphic designers, site administrators, and tool developers are encouraged to read it.
These recommendations expand on the recommendations of BP1. Where the focus of BP1 is primarily the extension of Web browsing to mobile devices, this document considers the development of Web applications on mobile devices.
The approach in writing this document has been to collate and present the most relevant engineering practices prevalent in the development community today and identify those that: a) facilitate the exploitation of device capabilities to enable a better user experience; or b) are considered harmful and can have non-obvious detrimental effects on the overall quality of an application.
The goal of this document is not to invent or endorse future technologies. However, there are a number of cases where explicitly omitting a Best Practice that referred to an emerging technology on the grounds that it was too recent to have received wide adoption would have unnecessarily excluded a valuable recommendation. As such, some Best Practices have been included on the grounds that the Working Group believes that they will soon become fully qualified Best Practices (e.g. in prevalent use within the development community).
In building a Web application, it is not necessary to implement all Best Practices. Instead, each Best Practice should be considered as a possible measure that might be implemented towards the goal of providing as rich and dynamic an experience as possible on a mobile Web browser.
While the focus of this document is the Best Practices that apply to applications running in a Web browser, in many cases these recommendations are equally applicable to other kinds of Web run-time, such as the W3C work on Web Widgets [ WIDGETS ] and also in a number of vendor-specific initiatives.
In a world where the line between mobile and non-mobile is necessarily blurred and a document that restricts its focus solely to best practices that are uniquely mobile would most likely be very short. With this in mind, the focus of this document is to address those aspects of Web application development for which there are additional, non-trivial concerns associated with the mobile context. This applies equally both to the limitations of the mobile context (e.g. small screen, intermittent connectivity), and also the additional scope and features that should be considered when developing for the mobile context (e.g. device context / location, presence of personal data on the device, etc).
This document builds on some of the concepts described by the Ubiquitous Web Applications Working Group (UWA) and the Device Independence Principles [DIP] . It also discusses device and delivery channel characteristics, which the UWA has named "Delivery Context" [DCODI] .
Additionally, some Best Practices are relevant only if the device exposes certain capabilities (for example, access to device information such as location). In these cases the requirements are summarized as a separate sub-heading.
Saying that applications should be sensitive to the Delivery Context implies that some level of device knowledge and content adaptation is required. For Best Practices specifically related to this area, see 3.6 Handling Variations in Delivery Context .
These recommendations are complementary to the recommendations of Mobile Web Best Practices 1.0 (BP1) though their focus is somewhat orthogonal. Whereas BP1 focused on delivering a good experience on a broad range of devices, this document's focus is on making use of advanced device capabilities to deliver the best possible experience on those devices that can support it. For this reason, while readers of this document are likely to benefit from reading BP1 it is not a pre-requisite.
Also, the terms "AJAX" and XMLHttpRequest (XHR) are used to refer to any asynchronous browser request.
Most applications have the need to store data of various forms, both intrinsic content (e.g. the emails of an email application, the calendar events of a calendar application) and supplementary personalization settings (e.g. preferred theme, default view, etc).
These Best Practices relate to the appropriate technologies and techniques to use for managing a Web application's data.
Cookies are a common and effective means to store small amounts of state on the client. They are appropriate for simple personalization data and are commonly used to store a token representing user identity in order to enable automatic sign-in.
Information stored in cookies, however, is sent to the server for every request and so using them for excessive amounts of data can negatively impact performance, particularly on a mobile network.
Also, in the mobile context, cookie support cannot be relied upon since it may be disabled either in the device configuration or by the mobile network. For this reason, applications should endeavor to remain functional even if cookies are unavailable. See BP1 [COOKIES] Do not rely on cookies being available for more cookie related caveats.
If supported by the device, client-side storage APIs provide a mechanism to store more extensive amounts of data than would be appropriate with cookies. At the time of writing, work is still undergoing to unify these APIs, see the W3C WebApps working group [ WEBAPPS ] and Device APIs working group [ DAP ] for more information. Additionally, some examples of existing technologies that support client-side storage APIs are BONDI [ BONDI ], HTML5 [ HTML5 ], and Opera Widgets [ OPERA ].
Making use of client-side storage in Web applications is a powerful technique that brings Web applications into parity with native applications in terms of start-up time and responsiveness. Two key advantages are worth noting explicitly:
[ Client-Side Storage ] Requires: Local Storage API.
If a client-side storage API is being used the data in it is not visible to the user's other devices. Whilst this is appropriate for some forms of data (e.g. preferences and state relevant only to a given device), it is often necessary to send this data back to a server in order to provide a consistent view across devices (e.g. between mobile and desktop instances of an application) and make it possible to recover data if the device is lost or damaged. See 3.5.8 Ensure Consistency Of State Between Devices for further discussion on these topics.
As a rule of thumb, data that needs to be shared with other devices or recovered in the case of a lost or damaged device, should be replicated back to the server as soon as possible.
The technologies that provide client-side storage APIs provide facilities to detect the current network connectivity. For example, HTML5 provides a property on the navigator object ( navigator.onLine ) to indicate whether the client is currently online, and dispatches two events on the Window object to indicate a change of network state ( online and offline ).
However, these APIs should be used with caution. Even if the browser is reporting an online state, on an intermittent network this is no guarantee that a subsequent connection will succeed. The most effective approach is to fail gracefully in the event of a connection failure, store unsaved data in a queue of uncommitted changes, and set a timer to try again later.
[ Client-Side Storage ] Requires: Local Storage API.
Use trusted information, and protect all personally identifiable information.
18.104.22.168 What it means
22.214.171.124 How to do it
Where possible, instead of parsing JSON data by executing it with the eval() function, use a JSON parser (for example: http://www.json.org/json_parse.js ).
If this is impractical ensure that the data contains no user-generated content (e.g. the server is responsible for the content of all fields in the datafeed) or that any user-generated content is correctly escaped.
Allow the user to control application behaviour that might not otherwise be apparent, such as access to the network and access to device data. For example:
Note that where possible it is preferable to rely on the browser's native functionality to notify the user of these activities, however the Best Practices below provide further advice on appropriate application behaviour in situations where the native functionality of the browser may not be sufficient.
Ensure that the user is informed if the application needs to access personal or device information. The user should be informed of the types of information that will be used by the application and whether / how that data will be exchanged with the server.
These notices should be provided when the user first accesses the Web application, or on first access to user information. It should provide the user with enough information to judge whether or not they want to allow the application access to their data.
In many cases use of APIs that provide access to personal or device information causes a native confirmation dialog to be presented to the user. In this case the application should not force the user to confirm again at the application level, but should make clear in the UI that displayed data has been accessed from the device.
If the user declines a prompt to allow application access to personal or device information, the application must recover gracefully. For example, if a request to a device API fails, do not automatically retry if this will lead to the user being presented with repeated native confirmation dialog boxes.
[ DEVICE DATA ] Requires: Device Data APIs.
If an application requires user identity it is usual to prompt for user credentials (username and password) and provide the option to sign-in automatically on next usage session. This is especially important on a mobile device where data input is more difficult than on a desktop.
Note that if automatic sign-in is enabled, a sign-out link should also be provided.
User credentials can be stored in a cookie or in local storage. However, it is important not to store unencrypted password information since this is insecure. Typically, a securely hashed token which, if necessary, can be revoked on the server, is stored locally in order to enable automatic sign-in.
The most effective way to ensure that applications run smoothly and with low latency is to minimize use of device memory, processor power, and network bandwidth which are more limited on mobile devices than on the desktop.
Compress content for efficient delivery.
HTTP 1.1 compression, which uses the gzip and DEFLATE algorithms, is widely supported. Web servers should be configured to serve appropriately compressed responses. Where supported, alternative compression formats (such as [ EXI ]) may also provide benefit.
Note however, that the cost (in time and battery usage) of decompressing data should be balanced against the gains in transport efficiency. When configuring HTTP 1.1 compression note that:
This section elaborates on the Best Practices of BP1 ( MINIMIZE ). Smaller applications will download and execute more quickly and more reliably than larger ones on constrained devices.
"Minification" / "optimization" may take a number of forms from simple removal of white space and comments, to the global substitution of tokens (variables, method names, selector names) with shorter alternatives. In general, minification based upon a lexical / grammatical understanding of that source is less fragile and is preferred to simple regular-expression based tools.
The redirection of requests (using HTTP 3xx status or HTML meta refresh) is typically used to exchange information between servers (e.g. account authentication). The delay incurred by redirects is much higher over mobile networks and so the number of redirects should be kept to a minimum.
Try not to use redirects. If more than two redirects are required consider using an interstitial page to communicate to the user that the application is still working.
Establishing the necessary connections in order to complete an HTTP request can take significantly longer on a mobile network than on a fixed network. Even though bandwidth is typically more restricted on a mobile network it is still preferable to make fewer, larger requests.
Consider the following possibilities when designing an application:
126.96.36.199 What it means
A Web application typically requires a number of resources (style sheets, scripts, image, etc) each of which requires an HTTP request. As above, HTTP requests are particularly expensive on a mobile network and so fewer, larger requests should be favored over a larger number of smaller requests.
188.8.131.52 How to do it
As far as makes sense after taking into account 3.5.2 Minimize Perceived Latency combine all style sheets into a single resource and all scripts into a single resource. If multiple scripts and style sheets are required as part of the authoring process, then try to arrange that they are merged before the page is served.
184.108.40.206 What it means
Web applications often depend on a number of static images to provide icons, buttons, etc. If served as a separate image each one incurs an additional HTTP request which is detrimental to performance.
220.127.116.11 How to do it
Combine images into a single image for transfer (spriting). To optimize efficiency combine images of similar sizes and color palettes. Combine images that do not change often. If one of the component images changes, the entire combination image will need to be refreshed. To render individual components of a resource use CSS positioning and clipping.
[ CSS ] Requires: CSS2 Clipping and Positioning Support
18.104.22.168 What it means
Visual effects (e.g. background images and gradients) are often used to improve the look and feel of an application. These can be included in CSS as base64 encoded strings in order to avoid an additional HTTP request
Note that base64 encoding adds around 10% to the image size after gzip compression and this additional cost should be weighed against the benefits of fewer requests.
22.214.171.124 How to do it
Background images can be encoded using the data URI scheme:
[ CSS ] Requires: RCF2397 data uri support.
126.96.36.199. What it means
Dynamic resources that change occasionally (e.g. a user's avatar) can still be cached by identifying them with a URI that includes a hash of the resource content. Using this technique means that the browser does not need to check the resource headers in order to validate its cache, instead, any change in the resource will lead naturally to a corresponding change in the resource reference.
Where the actual resource at joeblogs-[xxx] could be either generated in some offline process or served dynamically.
188.8.131.52 How to do it
Expiresheader to a date in the far future.
184.108.40.206 What it means
If possible, data designed to be accessed by AJAX requests from the client should be cached in the same way as primary content.
220.127.116.11 How to do it
The standard caching techniques (
Cache-Control header), as well as resource
fingerprinting (see 3.4.8 )
can be used on AJAX data as readily as primary content pages.
18.104.22.168 What it means
Static resources do not need cookie information and so performance can be improved by serving them from a path or sub-domain for which the application's cookies are out of scope.
22.214.171.124 How to do it
Use a different domain, sub-domain, or path name for static resources to the main application, and restrict the valid path of cookies such that they will not be exchanged when they are not needed.
Application data served from /myapp will receive cookie information.
Static data served from /static will not receive unneeded cookie information.
The in-memory size of the Document Object Model (DOM) may be limited on mobile devices. Large / complex pages may exceed this limit and cause unpredictable errors.
Limit the amount of information in the DOM at a single time using pagination or other appropriate techniques.
Given the additional complexities of interacting with an application on a mobile device, special consideration should be given to the overall user experience. User experience is influenced by a number of factors, including: latency, interaction method, and data consistency.
User experience is strongly influenced by the initial start-up time of an application.
Offline Web application technologies like HTML5 AppCache [ HTML5-OFFLINE ] bring Web applications into parity with native applications in terms of their start-up time and their ability to be used even where network coverage is intermittent. The following steps should be considered to minimize the start time of a Web application.
Consider the following techniques to help minimize application start time:
Lowering perceived latency is an important factor in improving the overall usability of a Web application.
A number of techniques can be used to lower perceived latency:
Interaction methods vary across devices. Three main interaction methods should be considered when designing the UI:
The optimum configuration of UI elements varies depending on the interaction method used by the device. Ideally, the UI should be adapted based on a knowledge of the interaction methods supported by the target device. If this is not possible, then the UI should be designed to provide a good experience for each of these different interaction methods.
Particularly where navigation of content requires multiple links (i.e. back/forward in a carousel) the following factors should be considered:
focus method can be used to move the
focus to the part of a page that has changed. However, if
unexpected, this can confuse or irritate the user, especially if
returning to the previous focus is not easy.
focus method only if it is
essential to the use of the application, and does not inhibit user
126.96.36.199 What it means
Web applications can switch views without a full page reload by showing and hiding sections of content. However, this means that the browser <back> button doesn't work by default, and it is not possible to link directly to specific views within an application. Usability is enhanced by enabling both of these features:
188.8.131.52 How to do it
Assign a URI with a distinguishing fragment identifier (e.g.
to interrogate the browser location in order to determine which
view to display.
For further discussion on this topic see: http://ajaxpatterns.org/Unique_URLsNote that showing and hiding content in this way can have adverse affects on accessibility if not carefully handled. See [ ARIA ] for more information on writing accessible rich Web applications.
Standardized URI schemes have been defined for some common device functions, e.g. making phone calls and managing address books. These URI schemes, if supported, can enable users to easily use these functions from Web applications.
The most broadly supported scheme is tel: as described in [ RFC3966 ]. Code such as the following can be used to enable "Click-to-Call":
Note that [PHONE-NUMBER] should always be entered using the full international prefix (e.g. +1-201-555-0111) to ensure that it works outside of its home country.
184.108.40.206 What it means
On small screens it is important that paragraph text flows both so that it doesn't require horizontal scrolling and so that it will re-flow if the view orientation is changed. See BP1 [MEASURES] for more details.
220.127.116.11 How to do it
Do not use absolute or pixel measures. Use percentage and RELATIVE measures for containers so that text can re-flow automatically.
18.104.22.168 What it means
This recommendation builds on the recommendation in BP1 ( 5.5.1 Thematic Consistency ) and expands it to consider the application preferences , personalization data , and state that form part of the overall experience on a mobile Web application.
User credentials valid on one device should be valid on other devices. User preferences captured on one device should be accessible on other devices. Data updated on one device should be viewable consistently on other devices.
An important example of this is offering a consistent experience where data entered on a desktop is available on a mobile and vice versa.
22.214.171.124 How to do it
For any application data that is not exclusively relevant to the current device, favor storing it on the server so it can be shared by other devices. See 3.1 Application Data for more details.
Network-initiated content delivery ("push") methods allow notifications and updates to be sent to the user even when they are outside of the application context.
Push method support may be disclosed through a User Agent Profile document if published by the device vendor, or through a device classification repository.
If supported by the user agent, options for Push methods include:
Certain classes of browser attempt to display desktop pages on a small screen by automatically zooming the display. This can be problematic for applications that have already been optimized for a small screen. The viewport meta tag tells the device at what scale to render the page.
A typical viewport setting looks like this:
<meta name="viewport" content="width=device-width, initial-scale=1.0"/> ,
and is inserted into the <head> element of the HTML document. This setting informs the browser to always render the page at 100% (e.g. no browser based scaling) and is appropriate for pages specifically designed for the target screen-size.
Variations in the Delivery Context (such as different device capabilities) is a prominent feature of the mobile Web. Web applications should adapt to known or discoverable properties of the Delivery Context by adjusting the content, navigation or page flow, with a view to offering a good user experience on as broad a range of devices as possible.
Where possible, use the evidence available on the server to determine the properties of the Delivery Context, and adapt the responses to the client before transfer, thus improving the user experience and avoiding transfer of unnecessary or incompatible data.
In its most basic form, the minimum evidence from the requesting device is the HTTP request header fields. Typically, the following header fields provide evidence of device capabilities:
Where it is not possible to determine certain properties of the Delivery Context from the server, this information may be available at the client. Once obtained at the client, the information can be used directly to adapt the presentation, or it can used to request alternative, adapted content from the server.
There are a few client-side solutions available to the developer:
CSS Media Types: CSS Media Types allow different stylesheets to be associated with different media types (e.g. print, screen, mobile) and are traditionally used to repurpose content for these destinations. Since the list of recognized media types is limited, however, and devices are notoriously idiosyncratic in their interpretation of types, it is in general not a helpful technology in this context. See Media Types [ CSSMT ] for more details.
CSS Media Queries: Media queries are an extension to the "media-types" paradigm that allow developers to apply specific style rules based on the device display characteristics (e.g. screen width, orientation, or resolution). At the time of writing this specification is not fully supported, but can provide a useful way to modify the page layout (for example to re-flow sections of text) in a more maintainable, declarative way than is possible with script. See Media Queries [ CSSMQ ] for more details.
If a large number of devices are being targeted, or if the application is sensitive to the permutations of a large number of configuration properties, the number of application variants required might quickly become unmanageable.
To combat this, classify target devices and build a single application variant for each class. This allows you to exploit device capabilities with a manageable code base.
Identify the target devices for the application and assign these to device classes of varying capability. Focus on application variants that work in each class rather than building device-specific exceptions for every variation in device configuration.
Device classes should be defined on an application by application basis, so that the variants can be tailored accordingly.
Example 1: Possible definition of application classes based on rendering and device API capabilities:
Class 3: Advanced device APIs, for example: access to location API, device PIM data, or application cache.
Example 2: Possible definition of application classes based on supported user-interaction modes:
Class 1: Pointer Based.
Class 2: Touch Based.
Scripted and XHR based applications are not supported on all browsers. If broadest reach is a primary concern then consider providing a variant of the application that uses synchronous FORM posts in place of XHR requests. This Best Practice is related (albeit with a differing focus) to BP 1 [ OBJECTS_OR_SCRIPT] .
Not only is device characteristic detection imperfect, it cannot always account for the differing use cases of an application. Therefore automatic detection is not sufficient as the only mechanism for determining which version is appropriate.
Where multiple versions of an application exist (e.g. to support the various device classifications) always offer the user the opportunity to change the selection.
Always attempt to default to the most appropriate UI on first use.
Always remember the user's preference for future visits in a cookie or local data store.
Canvas and SVG provide alternative options for incorporating graphics in a Web application. Support for these technologies varies across devices so in many cases the choice of which technology to use will depend on the target devices for a given application.
SVG is well-suited for graphics that must be scalable and whose components need to be modified (e.g. panning and zooming a map) whereas canvas is better suited where a static bitmap is sufficient (e.g. drawing a scatter-chart or creating visual effects).
If speed is important, canvas may be more effective. However since Canvas generates a flat bitmap it is not inherently accessible and so should not be used as the sole means of conveying information.
See http://dev.w3.org/html5/spec/Overview.html#the-canvas-element for information on how to use the Canvas element.
See http://www.w3.org/Graphics/SVG/ for information on how to use SVG.
The following device properties included in the DDR Core Vocabulary ( DDR-VOCAB ) are of particular value in supporting best practices recommended in this document. They should be available in any DDR supporting the W3C's DDR Core Vocabulary:
Mobile Web Best Practices 1.0, Jo Rabin, Editor, W3C Recommendation, 29 July 2008 (see http://www.w3.org/TR/mobile-bp/ )
The editors would like to thank members of the BPWG for contributions of various kinds. The editors would also like to thank contributors to the public list, and contributors of Last Call comments whose comments have been taken into account in the creation of this document.
The editors acknowledge significant written contributions from:
* Daniel Appelquist, Vodafone
* Jo Rabin, mTLD Mobile Top Level Domain (dotMobi)
* Phil Archer, W3C
* Jeff Sonst