Chart description language?

Hi,

Has anyone considered adding a language for description of charts to 
the W3 family?

Charts have a very broad range of applications: statistics, financial,  
bussines reporting, scientific data, etc..

Such a language could be used to embed charts in web pages, while 
preserving the separation between the actual data and its visual 
representation.

The idea would be to break any chart in three parts:

- The actual data being represented. 

   This data could come from another part of the hosting document,
such a html table or a xform, or even from a external document.

- The chart caracteristics
  
   Such as the char type (bar, lines, pie, 3D, etc.), the scale 
(logarithmic, linear), etc. 

- The presentation of the chart.

    Line colors, font styles for the legend, placement of the legend, 
backgrounds, etc. Ideally this should be done with CSS.

Such a separation has many benefits for the end user. The user can 
copy-and-paste the data to a spreadsheet, choose a different 
representation, etc. 

Combined with scripting and xforms, this allows for very easy 
development of reporting applications. 

One could argue that this could be implemented using SVG + scripting, 
or with some server-side infraestructure. While this is true, such 
implementation would loose large amounts of information, and it is 
exactly the kind of presentation-driven applications that the w3c is 
trying to avoid. 

(In some aspects, it would be like using SVG to represent MathML. It 
can be done, but it's not the same).

And maybe more important, such a standard is easy do define (very easy 
if it is compared to SVG or MathML), and easy to implement. 

Finally, this language would benefit a lot from being developed with 
the W3C process. This would garantee that it "plays well" with others 
standards and with the W3C overall vision, and it would increase the 
likeness of browser vendors actually implementing it. 

I realize that standards are not created with such an ease, but in my 
opinion, there is a strong case for this one.

Has this already been considered? Is anyone interested?

Best,

Angel Faus
afaus@corp.vlex.com

Received on Thursday, 10 October 2002 14:19:45 UTC