W3C home > Mailing lists > Public > www-xsl-fo@w3.org > June 2001

RE: line thickness

From: Ian Tindale <ian_tindale@yahoo.co.uk>
Date: Thu, 28 Jun 2001 07:14:40 +0100
To: <www-xsl-fo@w3.org>
Message-ID: <NFBBJAKBAMNBCEAHLICEIELICDAA.ian_tindale@yahoo.co.uk>
> -----Original Message-----
> From: www-xsl-fo-request@w3.org [mailto:www-xsl-fo-request@w3.org]On
> Behalf Of G. Ken Holman
> Sent: 26 June 2001 23:41
> To: www-xsl-fo@w3.org
> Subject: Re: line thickness
> At 01/06/26 21:08 +0300, Ilkka Hartikainen wrote:
> >I'm trying to make thin vertical lines (between columns and to page
> >boundaries) with xsl:fo using thin, black block-containers, but my
> >PDF-output (made with FOP) seems to render all blocks thinner than 0.5mm
> >in 0.5mm. Is this a feature? How can I make my blocks thinner than this?
> >0.5mm lines are far from hairlines.. please, if anyone has
> ideas, let me know!
> What length specification are you using?  Have you tried "1px"
> for a single
> device dot thickness?

Actually, apropos and all that, may I refer you all to a thread currently
developing over on www-svg@w3.org where I'm putting my opinion forward that
'device' space units should not be used, or at least, should not be your
first choice. I suggest that real-world values be used, and let the
implementation and device and OS sort out how pixels map to the result. If
people specify pixels in media that could end up being printed as easily as
it could be found on screen or television, then that exposes to the danger
that some devices will take it literally and not give you what you thought
you wanted.

(Re: Question on spec, if you're going to hunt for the messages)

I remember a decade ago, a certain version of the Harlequin RIP, on an Ultre
imagesetter engine, that decided that all QuarkXpress 'hairlines' should
equal one device pixel (as it does on an Apple Laserwriter of the day). The
result was that people's hairlines were disappearing, but actually on closer
inspection, there was a thin streak of 1200dpi single pixel line only just
showing up (and no chance of making it through repro!). The next version of
the RIP fixed that behaviour, but it just goes to show.

Cheers. Ian Tindale
Received on Thursday, 28 June 2001 02:12:49 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:58:25 UTC