freenode/#clim - IRC Chatlog
Search
19:58:07
jackdaniel
I'm wondering what the specification of the line-style-effective-thickness should say. namely how should we treat :normal?
19:58:22
jackdaniel
specification says: ":normal---thicknesses and lengths are given in a relative measure in terms of the usual or "normal" line thickness"
19:59:02
jackdaniel
what is the usual (or normal) line thickness? later it says, that it should be comfortably visible thin line (because CLIM aims to be density independent)
20:00:31
jackdaniel
currently on CLX and other pixel-based backends it is simply "one pixel", while on postscript it is 2/3 of a point, so these lines have clearly different thickness on some displays (depends on a density)
20:01:58
jackdaniel
since I'm working on a specification I thought it is a good time to settle the "normal" value, which would be approximately based on the british technical drawings standard, which mentions, that a thin line should be one of 0.18 | 0.25 | 0.35 | 0.5 | 0.7 | 1.0 | 1.4 | 2.0 millimeters
20:03:45
jackdaniel
https://cadsetterout.com/drawing-standards/line-weights/ (the standard reference)
20:07:35
jackdaniel
(and to make it more pixel-friendly, hence the word "loosely based", it would 1dp that is 1px on a 160dpi screen)
20:10:33
Inline
with higher resolution or pixel density it can get thinner and thinner on pixel based backends
20:42:53
jackdaniel
but as I've said, I took it only as a reference of a "usual" thin line, we are free to pick whatever we want
20:44:30
jackdaniel
it is up to the backend to round, having something what rounds nicely is good though