At 09:44 AM 98/12/20 , Randolph J. Herber wrote:

>I repeat and stand by my original statement ``Adobe PostScript level 1 is
>a requirement _from considerations of portability_.''  I work daily with
>scientists scattered around the planet with problems generating, displaying
>and printing Abode Postscript language files and Adobe PDF.  In the interest
>of portability, Adobe PostScript level 1 language files using internal
>supplied fonts or the ``original'' 13 fonts (4 each Times-Roman, Helvetica
>and Courier and Symbol) with proper DSC is the _only_ in-practical-fact
>portable format.

This merely reflects lack of knowledge of how to produce reliable PDF files.
Or lack of access to a system that can produce reliable PDF files.

Like good PS, it is non-trivial, and while we now forget how much education
was required about PS in the beginning, the same is true about PDF.
Just as we had then to struggle with buggy PS clone interpreters so we
now have a bit of a struggle working around bugs in PDF Readers -
including unfortunately the ones from Adobe.  But it can be done.

>The only reliable ``fix'' for Adobe PDF is to convert
>it to an Adobe PostScript language file and, if necessary (which it
>frequently is), forcefully by manual editting dumb it down to language
>level 1.  Nothing else works reliably.

Really?  Before we had PDF we had endless threads on how useless PS is
because so many clone PS interpreters fail...

>PDF fails more frequently than
>anything else---most printers fail if given a PDF directly.

Of course, why would you feed PDF to a printer? (OK, lets ignore PS 3).
And I find printing from HTML browsers completely unreliable, while
printing from Acrobat Reader works with properly prepared PDF.

>As far as I can tell the _only_ reason for PDF is to protect Adobe's fonts for Adobe.

What?  How does it do that?  It is almost as easy to steal fonts from PDF than from PDF.
PDF is an output format that removes the
programmatic aspects of PS (which lead to problems with clone PS interpreters)
the result is smaller and easier to interrpret (does not require a full PS interpreter).

>Ghostview does not handle PDF files as input in any version that I am
>aware of.

While I never us it, I understand recent versions do.

>PDF files are frequently encrypted which causes problems by
>cryptography being illegal to export from several countries as military
>munitions and illegal to import into others without special permissions
>(e.g. France, Russia and the Peoples' Reuplic of China).

What?  The compression schemes are described in detail so you
can undo them.  You can password protect the files, but with a low grade
scheme that any government has the resources to break.  I have never
heard of such complaints.  (Some unethical types on the net
supply code to break PDF encryption - or provide an online service
to do it for you - so you can print and alter PDF files that the author
did not want you to print, alter or plagerize).

>time.  And, there are quite a few printers still in active use which
>are only Adobe PostScript language level 1.  These printers are not
>going to be withdrawn from service just to make it convenient for you,
>Adobe or any one else.

We are talking about PDF here, and while you can certainly generate PDF from
PS level II code, that does not mean you need a PS level II printer to print it - at all!
The Acrobat Reader can print to anything that has a working printer driver.

>I _am quite willing to accept_ a stage of conversion from TeX DVI to
>Adobe PostScript language files.  Until there are converters as
>competent as dvips and dvipsk are for converting DVI to Adobe PostScript
>language files for converting Adobe PDF to Adobe PostScript level 1
>language files, I have no interest in a LaTeX that produces Adobe PDF
>instead of DVI.

This is a separate issue.  It is in fact not clear that TeX -> PDF is now or
will be a viable alternative to TeX -> DVI.  But none of the issues you
or Hans have raised show anything but your lack of knowledge about Acrobat -
not anything to do with why PDF may or may not be a good target language.

>|Maybe tune in to comp.text.pdf?

>        There does not seem to be much there:

Fix your news reader or news server than.  Although, it is definitely not as
swamped with the volume of comp.text.tex  No long flame wars :-)

>|Maybe read http://www.YandY.com/download/pdf_from.pdf

>        If you want this to be read, then put it up as an
>        Adobe PostScript level 1 language file so that it
>        generally can be read.

Absolutely not.  PS is a poor distribution format.  And if you are interested
in Acrobat PDF, the least you can do is install the free Acrobat Reader.

>        Furthermore, to demonstrate the quality of that file:

># xpdf pdf_from.pdf
>xpdf version 0.7a
>Copyright © 1996-1998 Derek B. Noonburg
>Error (0): PDF file is damaged - attempting to reconstruct xref table...
>Error: Top-level pages object is wrong type (null)
>Error: Couldn't read page catalog
># pdf2ps pdf_from.pdf pdf_from.ps
>Error: /invalidaccess in --fileposition--

(1) Transfer the file in binary more, not ASCII!
(2) Get a decent PDF reader.

>|Maybe check out Donald Story's `AcroTeX' web page:
>|http://www.math.uakron.edu/~dpstory/

>        I tried it.  I fail to see how it helps.  I did see that it also
>        has the same problems with respect to Adobe PDF that many WWW sites
>        have with respect to html---you have to have ``bleeding'' edge
>        software in order to use the material.  Which is without
>        consideration of the users' needs, wants, wishes or capacities.

>        For example, general use WWW pages should be written to be usable
>        by Lynx and Mosaic web browsers.  If nothing else, then provide
>        a text only button and a separate set of pages.  If not, then,
>        for example, provide alt= tags for images.

I'll have to ask Donald to rewrite his web site in plain ASCII (or EBCDIC?)

Regards, Berthold.


Y&Y, Inc.  http://www.YandY.com/news.htm  mailto:[log in to unmask]