Re: [cgmo-webcgm] POLL ISSUE: zero-valued LINE WIDTH, stroke weight

[Lofton Henderson <lofton@rockynet.com>]

My preferences:

1a.
2a.

Comments:

I would actually support a variant of 1a:  put language in the PPF section, 
warning that zero-valued attributes are a bad idea and are not likely to 
produce predictable or expected results.

One reason that I like #2a now (not allow zero) is that it allows us, in a 
future release (e.g., 2.1), to allow zero for stroke-weight and define 2c 
or 2b as the behavior.  Then valid 2.0 behavior and content is a compatible 
subset of 2.1 behavior and content.  Compared to the many more important 
things that we have to define, **and implement**, this is one that we can 
well postpone.

-Lofton.


At 01:18 PM 6/10/2005 -0700, Cruikshank, David W wrote:
>All,
>
>I've read through the thread on this one and will summarize.
>
>There are two questions here and everyone needs to answer them independently.
>
>1)  What should WebCGM say about the treatment of zero width lines (line 
>width specifier of 0 value on the Line Width attribute element in CGM)?
>    a)  Leave it as it is specified in WebCGM?
>    b)  Change the meaning of a zero value to mean "no stroke"?
>Recommendation a)
>
>2)  What should WebCGM say about the allowable values and treatment of the 
>stroke weight style property of WebCGM set through the DOM?
>    a)  Limit value of both NVCC and relative scale to >0?
>    b)  Allow zero values for both NVDC and relative scale and adopt CGM 
> conventions?
>    c)  Allow zero values for both NVDC and relative scale and define the 
> meaning as "no stroke"?
>Recommendation a)
>
>Summaries
>
>1)  The CGM standards allows for parameter values for the line width 
>specifier of the Line Width attribute element to be non-negative.  That 
>means the Line Width can be set to zero in CGM.  In the "informative" 
>guideline for interpreters in Annex D, the CGM standard recommends: "If a 
>zero size is specified in the metafile, it is recommended that 
>interpreters select the minimum available size for affected 
>primitives."  Currently in the WebCGM 2.0 Profile (T20.3), the meaning of 
>a zero value is "Minimum available line width".
>
>2)  Currently the WebCGM 2.0 DOM chapter has a restriction of the 
>allowable values for stroke width.  It limits the relative scale stroke 
>width to >0%, but doesn't address absolute NVDC as far as I can tell.
>
>Personal thoughts
>
> >From a user standpont we have had a requirement for a "line visibility" 
> element in CGM similar to Edge Visibility, but I think that is something 
> that should be addressed at the CGM standard level and not in the profile.
>
>thx...Dave Cruikshank
>
>Technical Fellow - Graphics/Digital Data Interchange
>Boeing Commercial Airplane
>206.544.8876, fax 206.544.9590
>david.w.cruikshank@boeing.com