<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#ffffff">
<pre wrap="">"On a longer
> view it seems to me that dvi is going to disappear as the standard output. "
I wish people wouldn't say things like that. It is worse than unnerving.
PDF is a Black Box solution. You can't analyse its content, absent some grossly expensive commercial software. You can't get usable PostScript from it. since everything you really need is compressed and encrypted, and all the Structure Comments are eliminated. That means that you can't repair the Postscript in a text editor. You can't deal with the annual botched updates---like the one that suppresses all the characters in the Adobe Expert character set---unless you purchase annual upgrades.
There is no sense of archival stability in PDF. (Just try an old PDF containing Expert Character Set glyphs in a new version of Acrobat Reader if you think there is.) It is a nice, fairly compact way of dumping the final, absolutely final, set-in-concrete copy to a commercial printer, <b>IF</b> you can get PDF/X-3 working, which requires yet another purchase of software---the inchoate Ghostscript PDF converter module for PDF/X[1-5] doesn't really work, and GhostScript still produces PDFs that make Acrobat Reader drop the Adobe Expert Character set. I can, in half an hours work per file, hand edit PDF files to get those characters back onto the printout, but I have other uses for those half hours.
The increasing disregard of one of DEK's primary goals for TeX, archival stability, is becoming quite worrying.
Pierre MacKay
</pre>
</body>
</html>