Document size proposal - take 2

[Graham Klyne <GK@xxxxxxxxxxxxxx>]

Following my previous failed attempt, I shall attempt here to float an
alternative approach to the document size measurement issue.

It appears that there are two cases to be considered:

(a) documents and rendering systems that have no concept of physical
dimension:  for these, the usual measure of size is pixels.

(b) documents and rendering systems that do have some concept of physical
dimensions:  these are typically characterized by the physical dimension
and resolution.

I think there is also a third case:

(c) documents that have no inherrent concept of either physical size OR
pixel size.  I am thinking here of data like pure text (with no indication
of font size), and (maybe) fractal image format.  I do not know if there is
a rendering device equivalent to this (a laser light show, maybe?).

If we allow both cases (a) and (b), and avoid introduction of "notional"
dimensions and/or resolutions, then we also need to consider how to match
data corresponding to one case with a rendering system corresponding to the
other.  This proposal attempts to do this by indicating how and when the
different forms of measurement should be used.

The basis of this proposal is to have separate measurements for both pixel
and physical dimensions.  I have reservations about this approach, but
think that it should be considered.

---begin proposal---
2.1a Image Pixel Size

   - Media Feature tag name(s): 

     pix-x
     pix-y
    
   - ASN.1 identifier associated with this feature tag:

     ***New assignments by IANA***

   - Summary of the media features indicated by this feature tag:

     These features indicate the display size of the recipient for
     display or print, measured in pixels; they indicate horizontal
     (pix-x) and vertical (pix-y) dimensions.

   - Values appropriate for use with this feature tag:

     Integer, >0

   - The feature tag is intended primarily for use in the following
     applications, protocols, services, or negotiation mechanisms:

     Display applications where different media choices will
     be made depending on the pixel size of the recipient device.
     These features would typically be used to describe devices
     whose capability is dominated by the pixel dimension (e.g.
     common video display systems or hand-held device displays),
     and data intended to be displayed on such devices.

     For example, a web application for use on a 240x480 display
     might use different HTML pages than one intended for use on a
     1024x768 display.

     If the entity described has known rendered image physical
     dimensions, then those should also be described using size-x
     and size-y features.

     Data that is suitable for display on both low- and high- resolution
     devices should also be described in terms of physical size and
     resolution (size-x, size-y, dpi) feature requirements.


2.1b Document Image Physical Size

   - Media Feature tag name(s): 

     size-x
     size-y
    
   - ASN.1 identifier associated with this feature tag:

     ***New assignments by IANA***

   - Summary of the media features indicated by this feature tag:

     These features indicate the size of a displayed, printed or otherwise
     rendered document image;  they indicate horizontal (size-x) and
     vertical (size-y) dimensions.

     The unit of measure is inches (to be consistent with the measure of
     resolution defined by the feature tag 'dpi').

   - Values appropriate for use with this feature tag:

     Rational, >0

   - The feature tag is intended primarily for use in the following
     applications, protocols, services, or negotiation mechanisms:

     Print and display applications where different media choices will
     be made depending on the size of the recipient device, where the entity
     being described has some known physical dimensions.

     These features would typically be used to describe high resolution
     devices whose capabilities are dominated by physical size rather than
     the number of distinctly realizable image points (e.g. laser printers),
     and data originating from or intended to be rendered by such devices.

     When describing capabilities in terms of physical dimensions, some
     indication of resolution (dpi) would also be generally specified.

     Data suitable for display on both low- and high- resolution
     devices should be described using both pixel (pix-x, pix-y) and
     physical size (size-x, size-y) feature requirements.
---end proposal---

NOTE:  at the core of this proposal lies the idea that smaller,
low-resolution devices (say <=100dpi, <=1024 pixels) tend to be
pixel-limited and pixel-described, and larger, high-resolution devices (say
>=200dpi, >=1600 pixels) tend to be limited more by physical size.  In the
former case, higher resolution does not generally mean better quality;  in
the latter case, higher resolution generally does imply better quality.

The "crossover" case is handled by indicating that data from or for either
type of device is described using both sets of feature tags.

I note that case (c) described above would be covered by either note
specifying any size (pixel or physical), or by indicating a minimum pixel
size needed to present the essential content of the image/document.

#g

------------
Graham Klyne
(GK@xxxxxxx)