RE: image sizing

["Grosso, Paul" <pgrosso@ptc.com>]

Okay, so my posting below was too long, so no one
read it, so I didn't get any answers.  (Unfortunately,
there is no way to be terse about graphic sizing which
is why it is rarely done correctly.)

So where can I find out about how DITA is supposed to
do graphic sizing?  As it stands, DITA does not seem
to support what our users need to do in this area.

paul

> -----Original Message-----
> From: Grosso, Paul 
> Sent: Wednesday, 2006 January 11 10:32
> To: dita@lists.oasis-open.org
> Subject: image sizing
> 
> I have several questions about image sizing in DITA.  I
> apologize in advance if we've already discussed these,
> but I find it impossible to use the OASIS document archives
> in any reasonable fashion.
> 
> Let me first point out that there are two areas involved 
> when composing a document containing graphics (aka images):
> (1) the "blank hole" on the page into which the image 
>     will be placed.  This has been called the reproduction
>     area (reproarea) by CALS.  This is what, in XSL FO,
>     is specified by fo:external-graphic's height and width
>     properties.
> (2) the graphic image itself.  This is what, in XSL FO, 
>     is specified by fo:external-graphic's content-height
>     and content-width properties.
> 
> Regarding DITA's image element's height (and width) 
> attributes, the DITA lang spec says:
> 
>   Indicates the maximum height (width) of an image.
> 
> It isn't clear to me whether this is really referring to
> the reproarea dimensions or the graphic image dimensions.
> At first glance, I'd assume these attributes are giving the 
> actual desired image dimensions, but then why say "maximum ..."?
> 
> So just what--in terms of the two areas I describe above--are
> the values of DITA's image element's height (and width) 
> attributes specifying?
> 
> Also, the DITA lang spec says that the scale attribute (on 
> the fig element) is "to allow authors to adjust font sizes 
> on the content of the containing element, primarily for print 
> accomodation [sic]. An image in these contexts is to be scaled 
> only by its own direct scale property.  If not specifically 
> scaled, such an image is unchanged by the scale property of 
> its parent table or fig."
> 
> That seems to make it pretty clear that the scale property 
> of the fig tag shouldn't affect the image.
> 
> Rather, it makes it sound like the image tag should have a 
> scale property, but it doesn't.  Should it?
> 
> paul
>