Ben Mitchell's typo blog charting the excitement, activities and challenges of my 12 months' studying the MA in Typeface Design at Reading University.

Opinions are all my own.

Last week Nance Cunningham joined us in the Department to talk about the Burmese script. Nance works as a lexicographer, working on English-Burmese dictionaries, so is familiar with the difficulties of Burmese fonts. She stayed a couple of nights, which meant she could give me quite a comprehensive understanding of the Burmese letters and how they should work together.

Burmese, like other Brahmic scripts such as Thai, is an abugida or alphabetic syllabary, which means each syllable begins with one of 33 ‘initial’ consonants. Each consonant has an inherent vowel sound, which may be modified by adding diacritic vowel marks above, below or around the initial consonant. In addition to the vowel marks, an initial consonant can also be marked with one (or more) of four ‘medial’ consonant signs, which are written below, beside or wrapping around the initial consonant. As if that isn’t complicated enough, a Burmese font needs to compose, decompose, position and kern all these marks in a consistent order.

A virama can be added to any syllable to modify the vowel sound or create a dead syllable (similar to the mai thanthakhat or karan in Thai).

The ‘a’ vowel sign needs special handling, as it takes the same form as some of the other letters. So when these letters take the ‘a’ vowel, it is a taller variant form to avoid confusion. A further level of complexity arises when this tall version of the ‘a’ vowel needs to use the virama sign, as they both use the ascender space. In this case, the tall ‘a’ sign is replaced with a double-headed form. This will be handled in the font by straightforward OpenType glyph substititions.

Consonants may also appear stacked one below the other, in a kind of ligature, and are accessed by pressing a special kind of shift key (u+1039) which sends the next keyed character below the baseline under the previous glyph. Fortunately this only occurs at the join between two syllables, so vowels are not involved and do not require repositioning.

As well as instructing me how the writing system works, Nance brought along a bunch of Burmese books and magazines, some dating back to the 1950s. These have become a key point of reference in the research for my own typeface, as they show different styles of Burmese letterforms, but at the same time illustrate what needs to be consistent to maintain readability.

From travelling in Burma several times, I had noticed around me that Burmese letters often took horizontal stress patterns, with the heavy parts of the letters along the baseline and x-height. I had also noticed a few cases where this had been reversed, to give a more familiar look for those of us accustomed to the Latin script. However, when I remarked about this to Nance, she contextualised my observation: in fact I had been mostly looking at vernacular signage, intended to draw attention and therefore on the whole using display faces. A quick look at her books and magazines showed me that the only way prose is written in Burmese is with monolinear forms, using the ‘spaghetti-like’ strokes of Helvetica.

For my dictionary type project, I’m going to need a family of Latin and Burmese fonts, so I was keen to explore the conventional ways Burmese typography indicates hierarchy: do they use bold and italic in the same way as Latin? Funnily enough, one striking thing we found was that headings and emphasised passages have been chiefly set in smaller type sizes, with bolder forms and increased tracking between the letters. As for italics, we did find them, but very scarcely, and never used in passages of roman text for emphasis. Instead, italics seemed to be used only in their own right where a different typographic style was required. In places where italic might be chosen in Latin typography, it turns out condensed is a valid choice.

Nance’s books showed that Burmese has adopted speech marks and parentheses (though the speech marks are often not designed to fit harmoniously with Burmese — perhaps there’s room to explore French-style guillemets? — and parentheses are generally oversized to accommodate the large extender zones). A full-stop (period) is not used in Burmese, as there is a traditional ‘stick’ section sign. However, the ellipsis is occasionally used, but the dots take the form of little circles to fit in nicely with Burmese shapes.

Finally I wanted to explore how possible it might be to move away from monolinear strokes and introduce some chirographic styles into my type family. I looked at Burmese cartoons and advertisements, and found a few examples of pen-formed strokes with angled contrast. So far I haven’t reached any conclusions about whether this might be worth pursuing for my dictionary face, so more research will be needed.

With thanks to Nance Cunningham.

Photos of Burmese typography over on Flickr.

Posted at 12:58pm and tagged with: Burmese, type, font, book, typeface, letterforms, script, OpenType, Brahmic, one column,.

My last few weeks have moved really quickly, as I made my decision to come to the MA programme only at the start of August. Since then, I’ve had to finish off some freelance design jobs, sort out my enrolment and accommodation in Reading and get some vfb font files in good enough shape to send to the publishers (unfortunately no sign of a release date yet, as I’m going to need to concentrate on the MA for the next 12 months).

  To get us started for the MA, Course Director Gerry Leonidas sent us our reading list and a couple of exercises to work on over the summer. Whilst sometimes quite dense and technical, I have found the books largely interesting. My favourite so far is Twyman’s British Library Guide to Printing as I didn’t know much about the different kinds of printing presses. I’m currently getting to grips with Kinross’s Modern Typography, which is not quite so friendly in style (rather dry and detailed).

  

   Our first practical task was to get familiar with FontLab. Gerry sent us an image of some (hand) lettering, and asked us to convert it into font vector outlines, then set the word ‘Condensed’ in InDesign. Accidentally, I used the tiny thumbnail version (200px wide) of the image, rather than the full-size image, which explains the round corners at the stroke terminals. This was an exercise in curve-wrangling, and although Gerry suggested we auto-trace the image in FontLab, I chose to manually draw the nodes and position the handles intentionally, rather than end up with an untidy mess of extra nodes and bumpy curves.

   I also chose to normalise the /o/ and /s/, which seemed unbalanced in the original lettering and difficult to space in a working font, where the letters have to work no matter what combination they appear in (unlike in single-word lettering).

   My spacing (letterfitting) is still rather uneven, as I’m not very experienced with this style, or italic forms, and would probably want to normalise all the letterforms a lot further to obtain a harmonious font. I did put in the key dimensions and generous alignment zones (the hand-drawn nature letters are different sizes), in case I ever need to build this into something more.

   I wondered if I slightly rushed this (it took about an hour), but then figured this was intended as an introductory exercise, not a full-on practical project.

   Our second assignment was to sketch some selected letters from the fonts FF Tisa, Plantin, Electra and Adobe Jenson. I also sketched some letters from other fonts, thinking it might be useful to see how different forms are constructed. I was particularly drawn to find out what makes a text face ‘warm’ so here there’s a Fleischmann /a/ and some italics from Quixote. (I also looked at Tiempos Text and Espinosa Nova.) It was a fun assignment; my biggest difficulty was getting the overall proportions to match the originals. My /k/s for example invariably had the knee way too high to start with.

Posted at 12:00am and tagged with: Assignment, FontLab, Letterforms, MATD,.