Tag Archives: non-Latin fonts

The Sinhala Script

Before joining Dalton Maag as a font developer, I had the opportunity to explore the Sinhala script for a whole year during my MA in Typeface Design at the University of Reading, UK. I spent many hours collecting and studying material about this beautiful and fascinating script. The font that I designed following my research is this year’s Granshan Grand Prize winner, and I will share some bits about my research in this post.

Sinhala, also known as Sinhalese, is a South-Indic script and one of the two official languages of Sri Lanka, the other being Tamil. According to the most recent government census, Sinhala is in use as a first language by more than 80% of the total population. Including minority ethnic groups that use Sinhala as second language, the script is used by around 17 million people, only considering those within Sri Lankan boundaries.


Situation map of Sri Lanka.

Script Characteristics

Sinhala is derived from ancient Brahmic and as with all modern Indic scripts it has a complex structure. Even though it is written from left to right and it does not have joining characters, Sinhala has little in common with the Latin script. Beyond the clear differences regarding the letter shapes, Sinhala does not have differentiation between upper and lower cases, and most importantly, it uses consonants as the basic unit for word construction. All consonants have an inherent short vowel “a”, so in order to represent different sounds it is necessary to add vowel marks (called Pilla) that can be used before, after, above or below the base-consonant.


The basic scheme of Sinhala’s typographic anatomy.

Origins: Stone Inscriptions and Palm Leaf Writing

The Sinhala script has an extensive written heritage which exists in the form of stone inscriptions and palm leaf manuscripts, mainly devoted to religious subjects. In palm leaf writing, the letterforms were incised in prepared leaves with a pointed stylus and the grooves inked with charcoal powder so that the writing became legible. The writing material has influenced the script appearance, since it was necessary to make the shapes more circular in order to avoid damaging the leaves. The best models of Sinhalese manuscripts were used by European colonialists in the creation of the first Sinhala moveable printing types.


A palm leaf book. This is written with large sized letters from a skilled hand, to be read at night, during recitation of the text, from 1778. (Wellcome Library, UK)

Sri Lanka was disputed and ruled by three different European colonial powers for 443 years, a period of intense transformation of Sinhalese letterforms. The Dutch were the first to print with Sinhalese type on the island, in 1736, and decided to keep the features of native Sinhala script, following the writing practice on palm leaves. The type was cut to be monolinear in a rather geometric fashion and typeset with no separation between words in the early documents. In opposition to the Dutch approach, a new style of Sinhalese letterforms emerged in the second half of the 19th century and it was notable for its high contrast appearance. This style was introduced during the British ruling period and gradually replaced the existing monolinear model as the preferred style. Until the present day, the high contrast style has prevailed as a very popular model for text typesetting in printed newspapers, books and magazines in Sri Lanka.

Rafael Saraiva


Left: The title page of Singaleesch Gebeede-Boek from 1737, using the Dutch monolinear design (Google books). Right: The high contrast typeface used in a governmental publication from 1957. (Personal archive)

Eleni’s Greek Workshop

Dalton Maag Font Developer, Eleni Beveratou, presented a Greek type design workshop recently at our studio. Eleni is a native Greek speaker and reader, so I decided to attend her workshop to better understand the nature of the Greek alphabet and to help break some of the bad habits many font developers (like myself) have acquired through a lack of understanding. We started out with some basic handwriting exercises where we learnt about the proportions of characters, the correct way to draw them (where to start and finish the stroke) and some of their alternate forms. It was interesting to hear that whilst the alternate forms appear to be very different to each other, they are regarded as equally valid, and equally legible, shapes for use in both handwriting and type design.


We then learnt the correct pronunciation of the characters (‘Vita’, not ‘Beta’’ for example), and from this we all had a go at transliterating our names. This was surprisingly simple, especially after learning how to build sounds from combinations of letters.

It was clear that we were here to glean a basic understanding of the Greek Language as much as how to draw the correct shapes, so only after raising our hands and having our handwriting marked by our teacher, did we start sketching Greek letters to compliment some of our Latin designs. This came quite easily to everyone after having such an insightful introduction to the alphabet.

IMG_20130921_110133DSC02560 IMG_20130921_101949

The first thing was to identify the key character of our Latin typeface and how to apply that correctly to the more cursive nature of Greek. As a group we all sat around with our tracing paper and pencils, and once we had an idea of the direction of our Greek typeface, we got busy with the beziers.

DSC02544 DSC02546 DSC02557Taking the sketches into FontLab, we all went through numerous iterations of our characters, with Eleni giving us each lots of pointers on how to achieve the correct proportion and texture for our font with lots of print-outs along the way.

The final stage was a group crit, where we all laid out our designs on the floor and took a step back to view the fruits of our labour. There was lots to learn from everyone’s designs, some ideas to remember, and others to remember to forget.

Overall, I’m confident that we all have a better understanding of the Greek alphabet, the language and some of the typical mistakes to look out for.

 Matt Burvill

DSC02591 DSC02570 DSC02586 DSC02588

The Square


No matter what the form of writing, calligraphy or type, from the micro to the macro view, the square is the spirit behind Hanzi.

History of the Square

Writing within the square, by Qing calligrapher Yao Meng Qi

Writing within the square, by Qing calligrapher Yao Meng Qi

As I wrote in a previous post, the pictographic origin of Hanzi meant that it evolved into an ideographic script. But at that point it still held a very human, lively and vibrant aesthetic. There were always people attempting to regulate Hanzi, however it was not until Hanzi matured, during the rise of Kaishu in the Tang Dynasty (‘Kai’(楷)means “regulation” and “system”) that the square came into existence. It is rumoured that the Tang Dynasty’s calligraphy grandmaster, Oh-Yang Sheun (Ou Yangxun), invented the square grid to form a loose regulation of Hanzi’s structure, consistency and readability. His legendary calligraphic works have been used constantly as models for practice, for at least a thousand years, and continue to be used in the present day.

Philosophy Behind the Square

Calligraphy writing based on the ‘米’ square

Calligraphy writing based on the ‘米’ square

I recently read a book about the aesthetic of Chinese calligraphy, and although I’m not sure of the authenticity behind this, it is still interesting to share the theory with you.

Traditional Chinese culture doesn’t believe we have to change nature, or indeed that we could, because we are part of nature. We realised that there are always limitations, as humans living on the earth, and that there are always regulations among human beings as well. It doesn’t mean the Chinese create boundaries for themselves, but rather, while limitations are actually there, it is always good to realise them and then aim at breaking them. The theory is then that the square was introduced for philosophical reasons, to place a limitation on the shape of character, rather than purely aesthetic ones.

The Square in Hot Metal Type

Metal type in Ri Xing Type Foundry, Taiwan

Metal type in Ri Xing Type Foundry, Taiwan

There was wooden type in China before the invention of hot metal type, however it was made using a loose system and didn’t produce a good standard of printing.

It wasn’t until American missionary William Gamble, who was appointed by the American Presbyterian Church and sent to Shanghai and Hong Kong in 1858, that things moved on. He adapted the proportional Latin movable type system to the Chinese ideographic character set, and unified the character cutting method by introducing the electrotype type-making method to the Far East. The nature of the technology in use reinforced the square boundary that had previously been introduced to the script. The ‘square’ spirit has since then been transferred to all the Hanzi typefaces.

Metal typesetting of Korean Hangul, picture: Hangeul Typography School

Metal typesetting of Korean Hangul, picture: Hangeul Typography School

 Hangul are used to be drawn within the square, picture: Hangeul Typography School

Hangul are also drawn within the square, picture: Hangeul Typography School

The legendary Shuetai font family, Kana, Kanji being set in square. picture: @miucaffe on Flickr

The legendary Shuetai font family, Kana, Kanji being set in square. picture: @miucaffe on Flickr

Japanese Kana and Korean Hangul’s ideographic nature also made them fit into the ‘square’ system, since the three ideographic scripts are together called CJK scripts, as invented by Dr Ken Lunde in his classic book ‘CJKV Processing’.

CJK Typefaces in the Square

Shuetai font family’s Kana was being drawn within the square

Shuetai font family’s Kana drawn within the square

As CJK typeface designers, we are all very familiar with the ‘CJK unified ideograph’ character set, which takes up half the contents of any Unicode codebook. The CJK typefaces went through the phototypesetting age and, in today’s digital age, the solid form is not a technical constraint anymore. The square is thus not as prominent as in the hot metal age, however, CJK typefaces are still being draw in 1em square box (1000x1000upm in font design software), and with an inner square used to define letter spacing among characters. We have noticed that there is a trend that new Korean fonts are also now being drawn in a rather condensed rectangle bounding box, so perhaps new technology is now dictating limits to other scripts in the way that it has to Hanzi in the past.

Julius Hui


Creating Nokia Pure Bengali

The Indian Nokia project started with Devanagari and Tamil. After those writing systems, we made our way towards West Bengal to design Nokia Pure Bengali. The project was a steep learning curve yet rewarding.

The research scope for Indian writing systems is wide, whilst the digital source of references is very restricted. Tackling the Bengali writing system was very challenging for both production and design. Luckily, we had the chance to have Fiona Ross* consulting for us, as well as Jo De Baerdemaeker; without them, the project would not have been what it is now.

Designing a typeface is a team effort. Digital typography creates a link between linguistics, design and technology. This tri-party collaboration becomes even more tangible when it comes to designing a non-Latin script involving complex syntax rules.

The first difficulty we had to face was the glyph list, also called encoding. This list answers the question: “what do we have to draw?”. The encoding covers both linguistic and production needs. The linguistic glyphs are the ones used to write down sounds transcribing the language or any languages (transliteration should not be forgotten especially in the syllabary systems). The production glyphs, and often alternates, are created to help the proper behaviour of the final product, the font. As the designer and engineer develop the project, they understand the technical needs and the tricks to create.

Bengali shares resemblances with Devanagari, but its feel is very different. Even in the case of a neutral sans serif, one should not forget the pulp and essence of the designated writing system. Bengali is a very vibrant and dynamic script; you can notice its liveliness in many out-strokes, for example in Da (see below).


Most of the conjuncts** in Bengali are shaped vertically. This becomes a real issue in digital typography. Designers and engineers are working within a bounding box, and the available space below the Latin baseline is rather small; vertical space tends to be lacking. Therefore, when developing a Bengali typeface, the designer is greatly encouraged to restrict the depth of the characters and conjuncts as much as possible. To add more design complexity, the conjuncts are often visually crowded, so this problem has been addressed using many optical modifications and corrections (see below).


Bengali conjuncts are built upon base glyphs; most of them are visually self-explanatory. However, a handful of them are not and need special care and research in order to make sure of their accuracy, like in KTa (see below).


After many months of development, the design and production eventually came to an end. You can appreciate the result compared to its Latin companion (see below).


Amélie Bonet

* Fiona G. E. Ross is the author of The Printed Bengali Character and its evolution, first published in 1999 by Curzon Press. Reprint by Shishu Sahitya Samsad in 2009.

** A ligature is a glyph representing the combination of a consonant and a vowel. A conjunct is the compound of two consonants together to shape one single glyph (see below).


Chinese Font Design

blogtypeNei Ho/Ni Hao! which is “greetings” in both Cantonese and Mandarin. I’m Julius Hui, a type designer for Dalton Maag, and I’m both a born native of, and currently based in, Hong Kong. I’m very glad to be able to share with you my knowledge of Chinese (Hanzi) type design – this is probably the very first time that Chinese type design has been properly exposed to the world.

It has long been a myth, not only believed in the Western world, but also by Chinese people, that the Chinese are not willing to share their typographic knowledge with a third party, or only to disclose it to a very limited extent. There is also a lack of willingness to systemise the knowledge available. This is mainly because of the Chinese culture of the craftsman, which is not limited to type design, but add to this the incredibly huge character set, and the number of Chinese type designers is very limited as a result. In fact there are probably no more than ten type designers who are capable of developing a Chinese character set entirely on their own. To sum up, it has always been one of the toughest careers of all in design.

But I have been very lucky. Five years of design studies at The Hong Kong Polytechnic University provided me with plenty of time and space to get myself focused on type and typography research. Upon graduating, I decided to take a Chinese type apprenticeship under a type master, learning to draw Chinese characters stroke by stroke, as well as studying massive weight-balancing and structure refinements, the three main processes of Chinese type design.

3 years of apprenticeship was tough, but extremely meaningful. I set out on my own afterwards, until I met Bruno last summer. I then joined up with my lovely colleagues, who were coping with the fast-growing client needs for Chinese type design. Having learnt a comprehensive set of skills, and combined these with my academic background, they have asked me to be an evangelist for good Chinese type design, demystifying that ‘wild west of type design’ and all the challenges ahead – stay tuned with our new blog as I’ll be writing more!

Dor Jei/Xie Xie!

Julius Hui

julius blog image