Authoring web content to the specifications recommended by the World Wide Web Consortium (W3C) is often referred to as a ‘web standards’ approach.
As the W3C specifications are created independent (although not exclusive) of corporate interests, they provide a standard reference point for web browser developers and content authors. If both developers and authors refer to the same set of specifications, then a webpage should look and function consistently, across all browsers.
A web designer–developer (or commentator) with a passion for things ‘web standard’ is often, sometimes disparagingly, referred to as a standardista.
The term is attributed to Joe Clark and coined in Building Accessible Websites. It is not clear from Clark’s usage whether the term is meant to infer a superficial engagement with the subject, à la fashionista.
For the webpage author, a web standards approach includes:
doctypeto each webpage that declares the specification used
As a philosophy, ‘web standard’ is sometimes extended to include the consideration of usability, content accessibility and website maintenance . For the sake of clarity, these considerations may be better described as ‘best practices’; companion considerations that often differentiate the ‘professional’ from the beginner web designer-developer.
An author must engage with the content to determine the ‘type of information’ they are dealing with…
A standards-based approach could easily be perceived as mechanical, however it is less so when determining how to mark-up content. The elements in the HTML specification are limited, and do not cover all possible forms of written and visual communication.
An author must engage with the content to determine the ‘type of information’ they are dealing with, and then choose the element that best matches the content. For example, content headings have HTML-equivalents in the heading elements
<h1>Heading level one</h1>
<h2>Heading level two</h2>
<h3>Heading level three</h3>
Web specifications are formalised and extended by the World Wide Web Consortium…
The web—like any mass medium—is enabled through a shared infrastructure. Just as television has broadcast standards: PAL, NTSC, SECAM, etc, and radio has AM and FM bands, web-based communication is enabled through a combination of connection and content specifications.
Web specifications are formalised and extended by the World Wide Web Consortium (W3C).
The Web Standards Project identifies the following specifications as web standards:
Source: Sidebar: What standards? (Web Standards Project).
Applied web standards reinterpret specifications to meet the needs of specific content and/or communities. These reinterpretations typically take the form of guidelines and best practice recommendations. For example, the Web Accessibility Initiative Guidelines are a reinterpretation the X(HTML) specification. They provide guidance on how content can be marked-up to improve access by assistive technologies, such as screen reader software.
Initially webpages had a structural bias as the web was used by universities to share research—primarily text-based data. As it became a mainstream communication tool, additional elements were added to the HTML specification to enable greater control over the presentation of content.
Had the web evolved in the same fashion as many other mass-communication mediums, relying solely on a centralised physical infrastructure, all webpages would have been authored using the W3C standards, and programs for browsing the web would have used those same standards to display (render) webpages.
Standards were compromised to meet consumer demands…
Web browsers were (and still are), developed by independent companies, often seeking to gain and maintain market-share. The uptake of the web was so rapid, that standards were compromised to meet consumer demand; for more interactive content, images, animation and other rich-media. Browser developers introduced new, custom (proprietary) elements, not sanctioned by the W3C, to meet those demands. As a result, content authored for one browser could not be guaranteed to work in a competitor’s browser.
At the same time as the competition for browser market-share came to a head (often referred to as the browser-wars), the demand for web design reached a critical mass and visual web-authoring software became commercially viable.
…creating a website no longer required the author to hand-code each page.
With a visual interface, creating a website no longer required the author to hand-code each webpage. Software such as Microsoft FrontPage reused the interface conventions from word-processing applications. Other programs, including Go-live CyberStudio, based the user interaction model on graphic design applications, enabling free-form positioning of both images and text. Behind the scenes, the software would write the code required to recreate the look of the webpage. With code essentially invisible to the author, concepts of standardisation, and semantic markup, were further eroded.
Circa 2001, the web began to undergo a further revolution. The realities of browsers developed without reference to a shared standard, and webpages authored using non-standard or incorrect code also began to be felt.
Authors were faced with creating multiple versions of each webpage or using arcane workarounds to ensure consistent display across different browsers.
The issues were deftly summarised by standards exponent Jeffery Zeldman in his article: To hell with bad browsers. The proposed solution was a return to a standards-based approach.
…equal access and the sharing of information both benefit from a standards-based approach…
A further motivation for a standards-based approach is the adoption of web communication technologies by the public sector.
Unlike private enterprise, government and government-related organisations typically have a mandate to enable equal access to information and services. They are also (at least in the case of the New Zealand government), required to consider issues of interoperability when undertaking implementation projects. A standards-based approach goes some way towards ensuring equal access and enabling information sharing.
… fromDon’t break the web, toDon’t change what developers expect IE to do for current pages that are already deployed.
In 2007 Microsoft released a new version of its web browser: Internet Explorer 7 (IE7).
Depending on the statistics referenced, at the time of the release of IE7, its predecessor IE6 accounted for around 80% of the world market. Put another way, eighty percent of those accessing the web were viewing pages using IE6.
IE6 had essentially remained unchanged since 2001. In the six years since its release, web developers had progressively documented and found ways to accommodate differences between IE6 and the W3C recommendations.
The same web developers had also formed web standards groups and networks that were more vocal about the limitations of IE6. Due to pressure from these groups, the team working on IE7 set about fixing a number of the documented difference — and in the process came close to ‘breaking the web’.
For six years developers had created webpages that needed to work in IE6, and they now expected IE7 to work in exactly the same way: ‘bugs-and-all’. This has lead the IE development team to change its mantra: from
Don’t break the web, to
Don’t change what developers expect IE to do for current pages that are already deployed. .
Making web developer expectations central to the aims of a web browser development team would seem to be a move away from web standards, or at least in the primacy of the W3C recommendations.
Tools that enable people to publish content direct to the web, such as blogs, social networking and photo sharing websites, have created a situation where a content author may have little or no control over markup. These content authors are subject to ‘standards’ imposed by the system developers.
For example, styling a MySpace profile page requires ‘unlearning’ web standards. As an author can only edit content through simple form, changes to the look of a page are achieved by mixing CSS rules with webpage content. Looking under the hood, the code itself is less than web standard, with a proliferation of
<div>s in place of the expected headings, lists, etc.
Perhaps 2007 marks the start of a shift from standards that have been passed down from the W3C, to standards that have been learnt by people using the web via web applications. From web standard ‘best practice’ to web user ‘mass practice’.
a grassroots coalition fighting for standards that ensure simple, affordable access to web technologies for all.