RE: Future of DSSSL: What about PDF?

Subject: RE: Future of DSSSL: What about PDF?
From: Avi Kivity <Avi@xxxxxxxxxxxxx>
Date: Fri, 5 Mar 1999 12:03:02 +0200
On Friday, March 05, 1999 03:18, Didier PH Martin
[SMTP:martind@xxxxxxxxxxxxx] wrote:
> Hi Avi,
> 
> I am trying to understand if your answer is: Yes it is useful or no
> it is
> not and here is why.
> 
> About what you said:
> 
> a) A jade derived engine can have attached "back-ends", as you know,
> 
> actually we there is output in: MIF, RTF, Tex, HTML, XML and SGML.
> However,
> to obtain a PDF document you need a Tex to PDF converter. Here is
> the
> process:
> 
> document--->Dsssl engine (jade based)---->Tex------>TexPdf------
> >pdf
> 
> So the question now is: is this process more useful?
> 
> document----->dsssl engine (jade based)----->pdf
> 
> Or said differently, if an environment provide direct PDF production
> from a
> DSSSL script, is this useful or not?

Useful, but not because of the directness, but because of the increased
power.
This is the process I would like:

document ---> dsssl engine <---> formatter ---> pdf/ps/win32 gdi/X

Simply cutting Tex and TexPdf out would just reduce processing time, but
upgrading the computer is more cost-effective. Having a tightly-coupled
formatter improves the quality of output, not processing efficiency. Take a
look at general-indirect-sosofo and page-model - I don't believe they can be
implemented via tex (they certainly can't be implemented via rtf)

The formatter is independent of the output format, but interacts with the
front end. It makes layout decisions that are not made by current back-ends.
Because it is tightly coupled with the front end, it can make available
information to the front end (via general-indirect). Since it is
dsssl-oriented, the more complex layout features of dsssl would be supported
more directly and correctly.

So, my answer is:
- 'yet another jade back-end': not useful, merely convenient
- a full featured formatter with pdf output capabilities: very useful; a
major step forward 

> 
> Regards
> Didier PH Martin
> mailto:martind@xxxxxxxxxxxxx
> http://www.netfolder.com
> 
> -----Original Message-----
> From: owner-dssslist@xxxxxxxxxxxxxxxx
> [mailto:owner-dssslist@xxxxxxxxxxxxxxxx]On Behalf Of Avi Kivity
> Sent: Thursday, March 04, 1999 4:22 PM
> To: 'dssslist@xxxxxxxxxxxxxxxx'
> Subject: RE: Future of DSSSL: What about PDF?
> 
> 
> On Thursday, March 04, 1999 22:58, Didier PH Martin
> [SMTP:martind@xxxxxxxxxxxxx] wrote:
> > Do you think that a PDF backend for a DSSSL engine is something
> > that
> > can be
> > useful? To be more precise, having PDF output from a DSSSL script
> > processing.
> 
> A tightly coupled 'mid-end', or formatter, is highly desirable, to
> support
> non-simple page-sequences and generated flow objects.
> Various back ends (pdf, postscript, X, win32 gdi) should be easy to
> attach
> to the formatter since they implement essentially the same model.
> 
> Once you get to the page description level the format no longer
> affects
> output quality (correct me if I am wrong), so the format should not
> matter.
> There are near-lossless converters for the formats specified above.
> 

---
"The only words which have meaning are the last ones spoken"


 DSSSList info and archive:  http://www.mulberrytech.com/dsssl/dssslist


Current Thread