Dr. Peter Karow Award for Thomas Milo

blokland's picture

’s-Hertogenbosch, the Netherlands, 26 October 2009

The second Dr. Peter Karow Award for Font Technology & Digital Typography has been awarded to Thomas Milo for the development of the ACE layout engine (the heart of the Tasmeem plugin for InDesign ME) for Arabic text setting.

The Dr. Peter Karow Award for Font Technology & Digital Typography is presented once per five years to a person who makes an exceptional and innovative contribution to the development of digital type and typography related technology. The first Dr. Peter Karow Award was presented to Dr. Peter Karow himself at the third DTL FontMaster Conference at Castle Maurick in 2003. So, in this case it took actually an extra year before the jury (Dr. Peter Karow, Dr. Jürgen Willrodt, Peter Rosenfeld and Frank E. Blokland [chairman]) came to an unanimous decision for the second award.

Thomas Milo and his company DecoType developed with ACE, which is an acronym for ‘Arabic Calligraphic Engine’, new advanced technology for Arabic text setting, which needs a far more sophisticated approach than for instance the Latin script, based on a thorough analysis of the Arabic script. Not only served Milo’s typographic research as the fundament for the ACE technology, clearly it also formed a basis for the development of the OpenType format, although this is a less known and acknowledged fact.

Thomas Milo’s importance for the development of digital type and typography is evident and in line with the position of Dr. Peter Karow in the field. As one consultant of the award jury stated: ‘Dr. Karow made type digital in a way we know today (description of shapes as outlines, rasterization, hinting, greyscaling, plus page-layout improvements). Thomas Milo added the “smartness” needed for scripts that ask for a more sophisticated behavior than Latin’.

On the 18th of November Dr. Peter Karow shall personally present the award to Thomas Milo at the Type[&]Design 2009 conference in The Hague.

The jury of the Dr. Peter Karow Award congratulates Thomas Milo with his impressive achievements!

‘Arabic script is the result of a millenium and a half of intense and competent design development. Like for Latin script, computer technology should facilitate such a heritage. The first line presents DecoType philosophy in a nutshell. In black, the result. In colors, the letters. In outlines, the structure. The typeface is DecoType's Nastaliq, the language is Persian. The second line, for comparison, shows the same text in OpenType. The typeface is Adobe Arabic.’

speter's picture

Congratulations on a well-deserved recognition.

Miss Tiffany's picture

Congratulations, Tom.

Frank, I also published part of this in the NEWS feed as well, I hope that is ok.

blokland's picture

Tiffany, thanks for publishing the info on Thomas Milo and the Dr. Karow Award on the NEWS feed.
Some links on Thomas Milo and his work:
About Thomas Milo:
— Thomas Milo on Arabic Script, War in Lebanon, and More, an interview with citizenreporter.org
— About designing typefaces for ACE in Tasmeem (PDF):
— Titus Nemeth: Tasmeem 4 Typefaces 2009
— Titus Nemeth: Tasmeem · Typeface Design in Arabic
— Titus Nemeth: A Primer for Arabic Typeface Design for the DecoType Arabic Calligraphic Engine in WinSoft Tasmeem
— Titus Nemeth: The Current State of Arabic Newspaper Type
and Typography

Additional videos demonstrating ACE by way of Tasmeem:
— Abdallah sez ... Aphorisms set in Ruqah and Naskh
— Using Tasmeem’s Text Shaper to expand text and fit it on the page
Using Tasmeem’s Text Shaper to expand a line
— Using Tasmeem’s Word Shaping

behnam's picture


Certainly Tom Milo's work on ACE is the first ground breaking technology in service of Arabic script. The appreciation will only grow in time.

Vladimir Tamari's picture

Congratulations Tom - you have certainly earned this award! Having seen ACE in action (I have the honor to be one of the font designers for Tasmeem) I feel ACE can yet be applied to other scripts - Latin? Chinese-type ideograms...who knows? It is a really smart-font technology that knows how to build up fonts using a few basic shapes, Lego-style, but with a seamless and sophisticated printed or displayed output that varies according to context.

Hello benham this time we rejoice in a happy occasion!

John Hudson's picture

Vladimir: I feel ACE can yet be applied to other scripts - Latin? Chinese-type ideograms...

When humans invent a hammer, they start to look around for things to hit.

Since almost all writing is composed of strokes with at least some stroke shapes repeated, any script can potentially be broken down into the kind of pieces that ACE employs. [ACE can handle different levels of decomposition: some fonts might only decompose to the archigrapheme level, i.e. to undifferentiated letter shapes, while others might decompose to smaller stroke units.] The key question to ask would be whether other scripts benefit from such treatment in the same way that Arabic does and, because we're not talking only or even primarily about fonts here but also layout engines, can a general case be made for handling text in this way for other scripts? For instance, I can imagine something like ACE applied to a cursive Latin type such as Bickham Script Pro, and can also see how such a design would benefit from the kind of ACE-based variation that Tasmeem exposes so nicely. The OpenType version of Bickham Script Pro is less static than most Latin fonts, but it still is quite limited in the number of ways in which in can display a particular string of characters: the contextual lookups give the appearance of a lot of variation, but in fact they constrain the variation. Something like the Tasmeem word shaper and dissimilation functions, exposing all the possible ways in which a word could be written with the glyphs in the font, would be a clear benefit and enable more variation. On the other hand, the vast majority of Latin text neither requires such handling nor clearly benefits from it, simply because the norm of the formal book hand in this writing system is disconnected and aims for consistency rather than variation.

The brilliance of ACE is that it proceeds from the Arabic writing system, taking into account the features of that system as a system and the subsystems of specific styles. It seems to me that the real lesson of ACE is not whether this particular technology can be applied to other writing systems, but that the same process of devising a technology from analysis of those systems might bear similarly excellent fruit.

OpenType tries to be all things to all writing systems, and ends up being only some things to most writing systems. ACE seeks to be all things to the Arabic writing system, and largely nails it (I'm not yet happy with the spacing results, although the idea is sound and doubtless the implementation can be refined). It's a great Arabic hammer, and perhaps one could use it to hit things other than Arabic. But wouldn't it be more in the spirit of ACE to invent hammers optimised for those other writing systems, based on analysis and modelling of their characteristics?

Vladimir Tamari's picture

Hello John,
>>...When humans invent a hammer, they start to look around for things to hit...>>
How true- that is how technology develops. I am fairly new to digital fonts and can only appreciate in retrospect how revolutionary was Tom's idea. In my time (the early sixties) I patented a handful of common letter endings to abridge the number of Arabic letter-press types, but that is like a stone to Tom's high-tech hammer.

It is true that ACE has been developed for Arabic, and initially for one or two particular styles of Arabic. This reminds me to express my admiration for Tom's scholarship dedicated to the study of the historical development of Arabic calligraphy, particularly the Ottoman Naskh style - he deservs another prize just for that.

I must mention another great advantage of ACE - the same font consisting of relatively few glyphs (archigraphemes) can be used for Persian, Urdu, Pashtu and a host of other similar scripts. Ace shifts the necessary marks constructing the shapes as it goes. It succeeds brilliantly, but like any computer software it has to be tweaked and adjusted to deal with unexpected issues, and indeed the question of spacing is one that is being dealt with. Adjusting the parameters of ACE to suit the expectations of a particular font is an interesting way to further enhance it. There are other possibilities but the Decotype team have enough on their plate for now.

The reason I mentioned Latin scripts is that the whole family of languages that use its basic letter-shapes and marks (Swedish, Vietnamese, Turkish etc. stc. ) can benefit from an ACE-like system in the same way that Arabic-related languages have. The font designer only creates the minimum number of necessary shapes, and the software places the attached marks when and where they are needed.

Yes, variation is a key advantage ACE gives the font designer, blurring the difference between calligraphy and font design. Another way it can be applied is to scripts with swashes and other 'artistic' devices. For example if a fluid script like the so-called kusa brush writing of Japanese (in vertical lines) can be applied in an intelligent font the result will be as exciting as Tasmeem is for Arabic.

behnam's picture

The concept of ACE is beyond that. This is the first time that technology -any technology- has looked at Arabic script in its natural behavior.
Arabic script is not more 'complex' than Roman script. It just behaves differently. If Roman text is an assembly of characters, Arabic script is an assembly of words. ACE is the first computing technology that understands that. The immediate outcome of-course, is the work of Tom Milo in tasmeem. But this is not only a calligraphic tool, although Tom may have initially thought it was. ACE essentially is about computing how Arabic script behaves. We can't have a true 'Arabic typography' without being able to draw characters in their natural behaviors. But beyond typography, any matter of computing dealing with Arabic script can not be resolved unless the natural behavior of the script is understood. An assembly of words not characters.
Think of hand writing recognition in touch screen devices for example. It is impossible to do unless the device view the script the way ACE views it, word by word. So the concept, in my view is far reaching, beyond its original intent.

Vladimir Tamari's picture

Exactly, benham. The success of ACE in dealing with Arabic (and of course that means also Persian, Urdu et al) is that it is built on Tom's thorough research into the Arabic scripts concerned. What you say about 'any technology' not having dealt with Arabic word shapes before ACE is true- but only if the hundreds of lead letter types with which early Arabic printers hand-set such scripts as thuluth is not considered a technology. Arabic is fortunate in coming late to printing, because the calligraphic tradition overlapped the advent of computer technology.

Latin scripts had to go through the few hundred years since Gutenberg when the handwriting of letters and documents developed separately from printing, which became set in the regimented way the technology dictated. Tom is well aware that the Arabic fonts developed in the last sixty years (including my own) in imitation of Latin aesthetics, are a departure from the true genius of connected Arabic writing. I have written elsewhere about the possible advantage Arabic has where each of its words has a unique outline, enhancing legibility.

ACE-like software can help revive the tradition of Latin calligraphy in the natural way it was written combining disciplined letter shapes with spontaneous variations according to the place of the letter in word or in the sentence and page.

blokland's picture

Due to traveling abroad, the Dutch minister of Education, Culture and Science apologized for not being able to attend the Dr. Peter Karow Award presentation to Thomas Milo at the Steigenberger Kurhaus Hotel in The Hague coming Wednesday, but he will be represented by his Director-General. This very much underlines the acknowledgement of Thomas' achievements for digital type and technology.
Some photo’s of the preparations for the conference session at the Royal Academy:


© Jorn Henkes

sofyaani's picture

I just want to say the first script that wrote its below "technology designed for Arabic" it's not Arabic actually , it's Iranian (Persian) Script that called Nastaliq!!!!!!!!!!!!
It's not Arabic at all!

Thomas Milo's picture

Dear Sofyaani,

If the Arab armies in the year 732 had won the Battle of Poitiers (http://en.wikipedia.org/wiki/Battle_of_Tours) the example could have been in a native Dutch Islamic style - but technically the script would still be valid as a sample of Arabic :-)

Thomas Milo

PS - Because the Arab armies did win other battles, the sample text happens to be Persian. It means "in personal experience". It is in the nastaliq style (which is widely used in the Arab world in a form simplified by Ottoman calligraphers-designers and called ruqah script), because that style is a typical example of Arabic script that defies all outside, industrial attempts at simplification. The second line has the same text, ergo, it is also in Persian, but in a typeface that represents the tendency to strip Arabic of any characteristic feature that requires other technology than what is available for Latin.

sofyaani's picture

how a ridiculous reason

sofyaani's picture


behnam's picture

Tom Milo developed an Arabic Naskh before this Persian Nasta'liq. Both of them belong to Arabic script. I don't see your point sofyaani. Do you have any or just spamming around?

blokland's picture

On Wednesday the 18th of November 2009 at the end of the very successful Type[&]Design 2009 Conference session at the Steigenberger Kurhaus Hotel in The Hague, the second Dr. Peter Karow Award for Font Technology & Digital Typography was presented to Thomas Milo by Dr. Peter Karow himself. The ceremony followed on an inspiring talk by Thomas on the history of his company DecoType and the development of the ACE technology.
Prior and during his talk Thomas underlined that without the assistance and support of his colleagues at DecoType, who are his wife Mirjam Somers and his brother-in-law Peter Somers, his smart font technology would not have become as sophisticated as it is today.
Mirjam Somers was educated at St Joost Art Academy in Breda, the Netherlands, where she was a talented pupil of the famous calligrapher Chris Brand. After two years she continued at the Gerrit Rietveld Academy in Amsterdam where she graduated Architectural Design in 1974. Having worked as a carpenter and free-lance typographer she encountered Arabic script in 1982, when together with Thomas Milo she designed and produced the Arabic Letter Box, a wooden prototype of a computer program and an educational tool at once. It taught them that Arabic requires an unconventional approach.
The inspiration was actually provided by a Chinese character cut by Mirjam during her St Joost days: it gave the clue for the penstroke-based approach. With Tom she reduced the traditional Arabic Ruqah script to 70 penstrokes and defined a mechanism to generate all and any letter combination correctly: the smart font algorithm. Mirjam played a major role in designing the Naskh typeface (1993–), and created the Emiri (2004, a revival) and Nastaliq (2008–, an original analysis) typefaces.
Peter Somers received his degree in æronautics at the Delft University of Technology in 1971, and joined Fokker Aircraft’s R&D department of advanced materials and processes in 1973. Later he joined the Optics Research Group of the Delft University of Technology in 1998 as senior researcher for the development of non-contact methods for the inspection of aerospace structures. There he also continued the development of real time speckle interferometry.
The combination of the font box and the out-of-the-box font, which Mirjam Somers developed together with Tomas Milo, inspired Peter Somers to write revolutionary smart font program code. He created a software based equivalent: a logical system for context sensitive selection of character variants for Arabic script, incorporating the concept that the Arabic characters in calligraphic composition can be decomposed into a limited series of basic shapes.
Peter designed an appropriate data structure for these shapes and a logical system to connect them. Available methods such as digital image acquisition and image processing, very common today, but advanced and hard to get by at that time, were used to implement these ideas in software. His efforts led to the first ‘intelligent font’ and resulted in the DecoType Arabic Calligraphic Engine (ACE) which powers intelligent fonts that provide high quality dynamically shaped calligraphic results.
The photo shoot below was taken during the ceremony, and show Mirjam Somers and Peter Somers besides Dr. Peter Karow and Thomas Milo.








blokland's picture

Tuesday 17 September 2013 Nicole Minoza posted on Adobe’s Tyblography the news that Dr. Donald E. Knuth will be presented with the third Dr. Peter Karow Award for Font Technology & Digital Typography at the ATypI Amsterdam 2013 conference.

Dr. Knuth will receive the award in the presence of the previous two winners: Dr. Peter Karow (in 2003) and Thomas Milo (in 2009; see also this Flicker set).


Syndicate content Syndicate content