HZ (Hermann Zapf)-algorithm

An interesting thread from the newsgroup <url:news:comp.text.tex>.

The «hz (=Hermann Zapf)-algorithm» thread

One can try to get the thread via news by getting the start message at <url:news:##iAhnaZ9GA.205@uppubnews03>, download it as gz-compressed unix-mailbox (hz-nntp.gz 5kb) or read the articles that I found interesting here:

1 - 2 - 3 - 4

      From: "Heinz-Dieter Ecker" <hd_ecker@msn.com>
   Subject: hz (=Hermann Zapf)-algorithm
      Date: Sun, 12 Apr 1998 02:27:53 +0200
Message-ID: <##iAhnaZ9GA.205@uppubnews03>

Hello,

does anyone in the (La)TeX world know anything about:

1. the hz program (or "hz algorithm") for typesetting --- formerly developed in cooperation (?) between URW and the famous Hermann Zapf; is it available, do you know any address (WWW or snail mail) to ask for more information?

2. any influences of TeX's way to set type on the general way hz does (and possibly vice versa --- given that Zapf and Knuth seem to know each other well)? With the minimal information I have on hz, it looks as if there existed a certain relationship~...

Thanks for any information and: Happy Easter.

Heinz-Dieter Ecker

-----------
hd_ecker@msn.com
        From: crism@geode.ora.com (Chris Maden)
     Subject: Re: hz (=Hermann Zapf)-algorithm
        Date: 14 Apr 1998 13:48:50 -0400
Organization: O'Reilly & Associates
  Message-ID: <keaf9othtp.fsf@geode.ora.com>
  References: <##iAhnaZ9GA.205@uppubnews03>

In article <##iAhnaZ9GA.205@uppubnews03> "Heinz-Dieter Ecker" <hd_ecker@msn.com> writes:

does anyone in the (La)TeX world know anything about:

1. the hz program (or "hz algorithm") for typesetting --- formerly developed in cooperation (?) between URW and the famous Hermann Zapf; is it available, do you know any address (WWW or snail mail) to ask for more information?

What information are you looking for? There's a synopsis of the algorithm on pp. 189-191 of Robert Bringhurst's The Elements of Typographic Style. He references The Seybold Report on Publishing Systems 22.11 (1993), pp. 3-9, and URW's publication HZ-program: Micro-typography for Advanced Typesetting (Hamburg, 1993).

I don't know how to reach URW; Bringhurst notes that they entered receivership in 1995, and that their type library is now distributed by a corporate entity known as URW++.

2. any influences of TeX's way to set type on the general way hz does (and possibly vice versa --- given that Zapf and Knuth seem to know each other well)? With the minimal information I have on hz, it looks as if there existed a certain relationship~...

Bringhurst notes that HZ applies hyphenation and justification algorithms to an entire paragraph instead of to single lines, and points that HZ is not the first program to do so. I don't know if it uses the same algorithm as TeX, but it's the same concept.

-Chris

-- 
<!NOTATION SGML.Geek PUBLIC "-//Anonymous//NOTATION SGML Geek//EN">
<!ENTITY crism PUBLIC "-//O'Reilly//NONSGML Christopher R. Maden//EN"
"<URL>http://www.oreilly.com/people/staff/crism/ <TEL>+1.617.499.7487
<USMAIL>90 Sherman Street, Cambridge, MA 02140 USA" NDATA SGML.Geek>
        From: Barbara Beeton <bnb@ams.org>
     Subject: Re: hz (=Hermann Zapf)-algorithm
        Date: Tue, 14 Apr 1998 16:43:01 -0400 (EDT)
Organization: Info-Tex<==>comp.text.tex Gateway
  Message-ID: <Pine.GSO.3.96.980414164059.15070Y-100000@sun06.ams.org>

chris maden provides the following information regarding the hz algorithm (by peter karow and hermann zapf):

Bringhurst notes that HZ applies hyphenation and justification algorithms to an entire paragraph instead of to single lines, and points that HZ is not the first program to do so. I don't know if it uses the same algorithm as TeX, but it's the same concept.

i've spoken with peter karow about this.
the breaking-lines-into-paragraphs algorithm -- "certainly it's tex!"
the goal is to achieve the optimal interword spacing before applying adjustments to lettershapes and taking other steps to increase or decrease the actual width taken by the text so that the final interword space becomes uniform. it's the variant lettershapes that are a major contribution from zapf, and i believe the other tactics were invented jointly by zapf and karow.

but so far this system works for text only. peter has said that he has some ideas for re-integrating math, but it's not trivial, and he hasn't had much time to spend on this for several years.

peter attended one of the "wept" conferences in st. malo, and i saw him there when i was attending eurotex. he was kind enough to give me a copy of his recent paper (december 1997) describing "chapter justification", which is a logical extension, on a page/chapter level, to the best-fit paragraph of tex or the hz engine. the concept as described is quite impossible to accomplish without the use of a computer, because of the very large number of parameters and decisions that would otherwise be entirely too expensive to examine otherwise.

the paper cites only four references, and i think this is significant: (quoted from the paper)

1. Listings and explanations of 'all typesetting parameters' could be found in the literature, e.g. Martin Douglas, An Outline of Book Design; published by Blueprint/The Publishers Association; London, 1989, ISBN 0-948905-33-6; page 21 ff.

2. All relevant outline formats are described in detail in the literature, e.g. Peter Karow, Digital Typefaces, published by Springer Verlag; Berlin, Heidelberg, New York; 1994, ISBN 3-540-56509-4; page 139 ff. [...]

3a. Donald E. Knuth, Michael F. Plass, 'Breaking paragraphs into lines', Software--Practice and Experience 11/11), 1119-1184, Nov. 1982

3b. Donald E. Knuth, The TeX Book, Addison-Wesley Publishing Company, Reading, Massachusetts, 1984, reprinted as Vol. A of Computers & Typesetting, 1986

i am hoping that peter will permit publication of this paper in tugboat.

                    -- barbara beeton

        From: Thierry Bouche <Thierry.Bouche@ujf-grenoble.fr>
     Subject: Re: hz (=Hermann Zapf)-algorithm
        Date: 15 Apr 1998 14:00:54 +0200
Organization: Universite Joseph Fourier - Grenoble 1, Grenoble, France
  Message-ID: <skyax7mgzt.fsf@mozart.ujf-grenoble.fr>
  References: <Pine.GSO.3.96.980414164059.15070Y-100000@sun06.ams.org>

Barbara Beeton <bnb@ams.org> writes:

i am hoping that peter will permit publication of this paper in tugboat.

you're not alone!

The Cahiers GUT have republished his URW paper on hz (translated to french) in their no 26. We're extremely interested in any further development.

Another piece of info: if tex had some impact on hz, hz could have some impact on tex, the next feature developped in pdftex will be the use of Multiple masters fonts with a width axis in pdftex to enhance evenness of the color in justified text.

One other very interesting feature of hz was - the ability to mutate fonts (turning any single master font to a virtual MM with at least weight & width axis) - the integration of kernus (and maybe the possibility to kern on the fly not on a pair basis: you don't kern AVA as you would kern TVA).

Does someone know if some of hz will reappear in Adobe's K2? (reportedly, PK was now purchased by adobe)

-- 
Th. B.

Possible future/present implementations?

pdftex

In the last article a future version of pdftex is mentioned as it might use Multiple Master fonts. Do not know anything about chapter justification as mentioned in the third article.

Lout

?

Adobe's K2

K2 is rumored to be a totally new rewrite of Pagemaker. Peter Karow is supposed to be purchased by Adobe, according to the last message.

References

Hermann Zapf, About micro-typography and the hz-program Electronic Publishing, Vol. 6 (3), 283­288 (September 1993)

Article as PDF PDF

Roger D. Hersch, Jacques André, and Heather Brown (eds.), Electronic Publishing, Artistic Imaging and Digital Typography, Proceedings of the EP'98 and RIDT'98 Conferences, St Malo: March 30 - April 3, 1998, Lecture Notes in Computer Science Series, number 1375, Springer-Verlag: Heidelberg, 1998. (ISBN 3-540-64298-6)

Send comments to Aslak Raanes