W3C home > Mailing lists > Public > www-svg@w3.org > December 2001

RE: SVG Coordinate System

From: Thomas E Deweese <thomas.deweese@kodak.com>
Date: Mon, 17 Dec 2001 08:04:22 -0500
Message-ID: <15389.60758.704275.74239@gargle.gargle.HOWL>
To: "Arnold, Curt" <Curt.Arnold@hyprotech.com>
Cc: "'www-svg@w3.org'" <www-svg@w3.org>, "'bac@world.std.com'" <bac@world.std.com>
>>>>> "AC" == Arnold, Curt <Curt.Arnold@hyprotech.com> writes:

AC> That works fine as long as you don't have any text.  Any way you
AC> flip the coordinate system will cause the text to be flipped also
AC> (though you could right it with a nested transform if you wanted
AC> to go the trouble).

    My personal suggestion is that as part of the file format
translation (or generation from the internal format) you subtract all
the Y coordinates from some _fixed_ value.  So if you are doing GIS
stuff use '90', for traditional Cartesian use '0'.  If you are putting
a viewbox on your outermost svg element (as you should) then the
viewer will make sure that the content shows up in the window.

    There is no single coordinate system that can satisfy everyone
(although some choices might make more people happy than others) and
having multiple coordinate systems in one file format is IMHO a recipe
for disaster as it creates 'worlds' of incompatible content within one
format.  It is much better to isolate this sort of incompatibility to
the borders between formats (especially since one of the really great
things about SVG is that it can reference lots of other content).
Received on Monday, 17 December 2001 08:04:35 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 8 March 2017 09:46:51 UTC