[rfc-i] The problem is SVG-Tiny

Phillip Hallam-Baker phill at hallambaker.com
Tue Jan 21 07:49:43 PST 2020


Looking at the responses, it appears that several things are being
conflated:

1) The restriction prohibiting color in the diagrams.
2) The restrictions on the use of fonts.
3) The restriction to a subset of SVG-Tiny.

It is the last that requires a massive amount of time and effort to support.

I have read the posts on the list and the only technical responses I can
find are from Brian complaining that the conversion is non trivial. I am
not a slow coder by any stretch of the imagination, I have developed the
most comprehensive converter so far and it is clearly a non trivial amount
of work to complete it.

The SVG-Tiny profile was being discussed back in 2002 when the mobile web
was a very different place. The mobile web was crippleware. The operating
assumption being that the mobile web would be a subset of the actual Web.

That assumption was completely demolished in 2007 with the launch of the
iPhone. SVG-Tiny was published in 2008 and has never received any
appreciable level of support. The only product that supports SVG-Tiny
native is Adobe Illustrator and that certainly isn't going to change.

The only substantial use of profiles of SVG is in SVG fonts which represent
a very different problem. There are really good performance reasons for not
wanting your font description language to have multiple layers of inherited
attributes, multiple coordinate systems, etc. SVG-Fonts is not intended to
be a drawing spec.

In 2014, when the decision was taken to use 'a profile' of SVG, it was
stated that the decision could be re-opened later. Well that is exactly
what I am doing now.

I have code that shows this problem is seriously complex. If people want to
challenge that assertion, let them demonstrate it with facts, not vague
references to proof of their position that they haven't produced or
endorsing the expert opinion of an unnamed person.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.rfc-editor.org/pipermail/rfc-interest/attachments/20200121/6c765e88/attachment.html>


More information about the rfc-interest mailing list