- From: Chris Lilley <chris@w3.org>
- Date: Thu, 13 Apr 2000 15:37:52 +0200
- To: Dave J Woolley <DJW@bts.co.uk>
- CC: "'www-svg@w3.org'" <www-svg@w3.org>
Dave J Woolley wrote: > > Would it be possible in the next version of the specification > to: > > 1) Use a PDF conversion tool which behaves more like > the html2ps tool used for the HTML4.x and CSS2 > specifications; It *is* the html2ps tool, followed by distilling. > 2) use style sheets to inhibit the rules and links at the > start and end of each web "page"; Right. Problem with html2ps, last I looked, was that it didn't use the standard method for associating style sheets. You had to put the print stylesheet in a .html2psrc file, or something. We could investigate doing that for the next release. > 3) use proper title elements (i.e. ones with some hope > of global uniqueness - HTML titles should make sense > out of context - "Text" does not). Something like "W3C SVG Specification: ,chaptername goes here>" for example? > Particular issues to do with the PDF conversion are: > - the rather unprofessional "local disk" as the top level > in the outline tree; > > - the use of titles (which probably relates to item (3) > above) rather than headings to construct the outline > tree It looks to me as if the headings are used to construct the TOC. > - whilst this may be the best compromise for > commercial quality "HTML", W3C documents are normally > properly structured, with Hn elements used appropriately > and properly nested; How would you tell, in a PDF document, what HTML element was where? > - I very much liked the page number cross-references > generated by html2ps in the hardcopy version. > > I haven't checked whether (2) is possible, in general, or > with html2ps. It is certainly possible in general @media print { .hideme {display:none} } -- Chris
Received on Thursday, 13 April 2000 09:38:03 UTC