The three submissions for internet vector graphics now on the docket of the World Wide Web Consortium (W3C, www.w3c.org) reflect competing power blocs within the graphics industry as they maneuver to establish hegemony on the post-XML web. Vector graphics offer a number of important advantages to web designers. Unlike traditional web graphic formats, JPEG and GIF, vector-drawn images are not fully rendered images drawn pixel by pixel. Rather, they are sets of algorithms which instruct the client system how to redraw the image. As such, a vector image can be transmitted in a much smaller file than its bitmapped equivalent, and the end result is much more efficiently scalable to differently sized windows and monitors. The first vector graphics spec, titled Schematic Graphics on the World Wide Web, was based on SGML and presented by the UK’s Orwellianly-titled Council for the Central Laboratory of the Research Councils (CCLRC). W3C staff member Chris Lilley praised the submission for its evolution from drawing tools and its attention to ISO standards for 2D graphics. The W3C added that if enough interest was shown in a 2D vector graphic markup language, a Working Group would be formed. That interest was not slow in coming. The next salvo was fired by an IBM-led consortium that included Adobe, Netscape and Sun. Their proposal, Precision Graphics Markup Langauge (PGML), built on Adobe’s achievements in PostScript and Portable Document Format (PDF). The PDF developer community immediately pounced on the PGML submission, describing it as Acrobat Reader-free PDF and the next big thing on the web. Unlike Schematic Graphics, PGML is expressed in XML and supports cascading style sheets (CSS). However, in spite of the fact that it pioneered the field, Adobe no longer has the vector graphics market to itself. Microsoft led the most recent consortium to submit a spec, and admits that its Vector Markup Language (VML) pretty much describes the way vector graphics are handled in Microsoft Office. This is the first time some of the things in Office have been documented, said Thomas Reardon, who was the Microsoft representative in the group that signed the submission. That proprietary bias didn’t prevent Autodesk, Macromedia, HP and Visio from appending their names to the document. VML is also expressed in XML and supports CSS, but unlike PGML it adds HTML and Document Object Model (DOM) support to its arsenal. W3C’s Chris Lilley commented on the differences between VML and PGML in some detail, saying that PGML may have more expressibility where VML is more editable. Microsoft’s Reardon responds: VML includes macro-ization concepts and can tie different graphical objects together, which might make it more editable. How that detracts from ‘expressiveness’ – by which I think Chris means the richness of what it may render – I’m not sure. All parties are publicly upbeat about the possibility of uniting the competing specs into a single, unified and open standard. Adobe’s vice-president of engineering, Tom Malloy, told CNet: There’s a whole lot of commonality between [PGML and VML], and we expect to get a standard out of the process that’s greater than the sum of the parts. But if, as Reardon’s comments suggest, the different submissions reflect differences between the proprietary implementations on which companies have staked their businesses, the divisions may be more political than technical. The W3C says its next step is to set up a Working Group with all three submissions on the agenda. Whether the formal process of the Consortium can contain the ancient antagonism of the players remains to be seen.
