ATypI 2013 Day 2

These are live blog notes from the ATypI 2013 conference in Amsterdam Day 2

Usual disclaimer for live blogging: These are informal notes taken by me, Dave Crossland, at the event, and may or may not be similar to what was said by the people who spoke on these topics. Probably if something here is incorrect it is because I mistyped it or misunderstood, and if anyone wants corrections, they should email me ( – or post a comment. Thanks!

In fact I didn’t even read this document once before posting it, so its probably FULL of errors. What do you want for free? bahaha

Ana has a pen

[20 mins of 25 late]

This is Sofia Oslislo’s Quadrato and the ink traps are used as a feature and appear like double strokes or outlines, its a wonderful typeface. Its important to me that students learn to make readable typefaces. You can see the process by the 6th and 7th session.

Paulina Urbanska’a Chiva.

This has these incredible corner point counters. She started with black letter and every session she urned up with something neew and interesting.

Michal Pawlowski’s Yabu is great. He worked night and day on this, an advanced student, and I encourage everyone to publish them. THis is the evolution of his fonts.

The students own the copyrights.

Victoria Grabowska’s Rymex, she’s now an instructor, and its so light it shows the fraktur skeleton in body form. She can do OTL alternates and she made a lot for this. She even did Cyrillic last year. She’s working on a heavy weight too.

Its now breakfast time. I have some copies to give away for people who are teaching type design, or who want to support us publishing the books.

* * *

JB Levée

Amiens extended its course to improve the type design projects that students do. This presentation is a DIY type design course, how to make your own course from scratch.

Put that in context, teaching type today is not he same as teaching it 10 years aog. the situation is different, we had the main schools in EU and that was it: KABK and Reading and back then Ecole Estienne was also active teaching type design.

5 years ago also different! many many courses around the world: cooper type in NYC, a MA in Lausanne CH, Liepzig was there, Argentina, MExico, so we could see a global spread of global type design education.

What was the situation in France?

The same path. Estienne was still active, and ANRT, it died some years ago, and rebooted with Thomas Huot-Marchand (?) as head. With new schools opening their own course, how could amiens differentiate its own course from what is on offer?

You need experiences and know how and thats what the most distinctive thing of any course is. its not dogmas, or ideas. its not research like other places.

its made with a staff of VERY DIFFERENT backgrounds. this is a key aspect.

Sebastien Morlighem, a publishing editor, at Ypsilon.

Patrick Doan is a book graphic designer based here in amsterdam.

Titus Nemeth and Alice Savoie are Reading MATD graduates doing PhDs there. And then me, the only one not doing a PhD 😉

I do retail and custom typefaces. THis is the basic crew.

We have various approaches.

Its good for students to pick form different opinions between teachers. Some focus on research and theory, others on practice, some are well trained in globalised type design.

All of us are Estienne or Reading graduates.

An important thing when teaching type design is working with a small group of students. I can’t imagine teaching a hand to hand skill in a big hall.

its FOUR or FIVE students for 18 months. its a sort of awkward time, but its focused and offbeat and alows them to be well awake to find jobs or pursue their education.

The main thing, I could’ imagine learning type design without learning calligraphy. i don’t think you need to stay close to it, but thats how i was taught and how i am teahing.

calligraphy is not more than a beginning. students are encourage to get engaged with chancery, humantic minisucle, and models afterward to use not as formal reference but to see and feel the quality of strokes, made with a feather or bezier.

you need to feel what a stroke is, expressive. gestural calligprahy helps that.

other tools are used like plain paper with scissors. we focus on aPRECISE shape. you need to work big, simplify, and make choices. paper and scissors don’t allow for imprecision.

this exercise helps to explore parameters and ideas in type deign. weight, width, grey.

letter carving should normally be done in stone or very hard surfaces. the letter carving one of us does here in amine is done in foam, so there is no hard boiled carving. its cutting into a 3D shape. we stuck with it so far but i hope we get to stone later.

and then drawing! a lot of drawing, every day, again and again. its not a bicycle, you must practice and entertain as a skill and you will forget it much sooner than you think. if you have 1 week or 18 months, you must practice.

you can not achieve anything without practicing.

i use the type cooker tool, a type design exercise generator. its done early in the course, students work hard and step up their drawing skills fast so we don’t lose time using beziers.

we also focus on using other people’s work as a basis for new designs. the amiens library was giving a gift by a 50s-60s type design active in book design, recontrue de lour, and other type associations. we got his library including some of his owwn type designs.

the interesting thing about revivals is that its a pretext for learning TOOLS AND TECHNIQUES.

so thats what the students do; step into the feet of another person and understand their work, not just reproduce it but understand it.

as i sai dibefore, i think its important to be realistic, type that is not used isn’t good.

i like to make a newspaper masthead exercise, a common task in the professional life, and for students to act as if a client popped in and omeone asked for their logotype to be redone in 3 sways

1. correct it, fix any minor flaws

2. enhance it, retain key factors but change things a little to make it a new style

3. redesign it from the ground up

This helps introduce students to what is originality in type design. students think they need a new form of serif, or find a new way to use ink traps. i steer them away from the idea, to just find interesting and beautiful shapes as well.

This is a exampe from Italia, an Italian newspaper. L’Arena. haha **I** get to pick the mastheads and i pick the most interesting ones.

For teaching teyp design you need a library of books to support seeing shapes and hearing concepts. we have a library with the help of the school, we get to acquire a good set of books old and new, old meaning pre 20thC specimens.

this is useful for saying ‘this idea reminds me of this type from 100 years ago’

we use masterclasses and workshops. students like to hear more opinions, someone who is not from the teaching staff. someone last year, we had albert boton who still draws but isn’t active any more, and older type designer.

Having older people in good to enrich the global teaching.

we have a LOT of visiting people [long list, Reading staff, lo Celso, Kai Bernau, lots and lots]

ANother initiative we like is to organise the life of the type design. exhibitions, lectures and conferences for the last 4 years. we set up 1 or 2 conferences on specific topics, in november normally. we have specialists lecturing on some topic. a designer or a time period.

last year was about Fournier. This year, Albert Boton (?) and you can see these images oft he fascinated crowd.

There isn’t one this year as Sebastian is busy with his PhD but in 2014 we plan something about Didot. We borrowed books from Amiens City Library… and its good for reference.

You may ask, whats the results? of the last 5 years students made a few designs public, one you will hear more about next.

Who are we teaching? future type designers? typographers? art directors?

i don’t know

There is a need for new designs and a consistency. we are here for that need and greeting a new set of students every 18 months.

Sandrine Nugue:

Ganeau: A latine family with three optical sizes.

AT the start i wanted to set out a project for 18 month. ‘to break down open doors’  is a french phrase to state the obvious.

so i wanted to make an expressive running text that is not distribution, so readers focus  on text meaning not the letters.

i read excoffon personal letters and this was inspiring for me. for small and medium text you need legibility, but in the display, the type isn’t so functional but pretty, so the optical size decides the balance of conventional or unconventional shapes.

i made a display, text and caption size for roman and italic.

I have some references: Swift by Gerard Unger, used for many legible typography. we are all recognising his work as it has his signature style. excoffon epxloreated a lot of different styles, many display, some text like antique olive. AO had reversed contrast as he felt we read the top of the letter. he changed traditional proportions of stoke but kept the traditional overall shapes.

i find that balance is important and AO has a daring balance that is still popular today.

Vendome was made in the 50s and that influenced me. vendome was the idea of modernising the garamond. its an alive design, full of vibration, and it was a hit for 20 years, not for running text. typographers didn’t really know how to use it.

i wanted to look at the latins. in the 19th C that painters and lithographs were used. they gave freedom to the letter forms. how with these shapes i can design a running text?

its a challenge to make a lot of choices. in exercises i made these shapes, cutting black paper with scissors is like a sculpture. my tutors said my work was like gerard angers! i had to find my personal approach. we have massive shapes. we have in mind a latin serif.

i made some early digitiziations. i like to make bold letters thirst to defiane the shapes. i started my optical size, but it wasn’t the best way, as i had to change things 3 times in an ‘adhesion’ – tarts the test word we draw.

then i continue on the caption and display size. in may we went to University of Reading. Gerry told me a sage advice, he said too many ideas,  5 ideas, retain just 1 and use the other 4 for the next 4 fonts.

So i made the shapes quieter. i make connections stiffer, …, changed curves, homogenised counters. its faster work and you can see the result in text size.

you see the italic and it was the trickiest part of the family. you put the shape part and you have to make it quieter. i didn’t know what to expect, so i tried a lot of things. i chose the smiler width for roman and italic, and an same typographic colour or grey.

now the text size is stabilised i moved to caption size. i questioned the utilisation of the caption. it will be more for notes and discontinuous reading. so i wondered about the utility of serifs.

so i tried a lineal thing. i thought about a flared stem too. this gave me problems. with a laser print at home or school, the serifs are rounded. but at the end i had the opportunity to use offset printing and there it is sharp.

here is the final caption size. I”m quickly trying to make a combination italic.

then the display size. its something like a gift or something to design. i don’t need to mind legibility so much.

i saw vendome, make it quickly, and thought something more different; on the left you see the roman serif and italic serif less, and the opposite on the right. but so many combinations was a mistake, a hice exploration. then i looked again at a flared stem and one of the first clssications fby thibaudeau was based on serifs. you see the 3 different terminals in the family.

Hoefler said, “one size never fits all.” I am a big fan of that.

Of course the sans structure is across all 3 sizes. i used tim ahrens’ 9 steps definition for optical sizes.

caption needs

1. larger width and more spacing for more white

2. we need a larger x height

3. we need lower extenders

4. we need more stroke weight

5. we need larger counters – this can totally remake a K or J

6. caps need less classical portions, more uniform.

7. they need lower contrast

8. they need skeleton changes to reduce complexity

so we need a complete family that i think bout the size and the utility for reading. to do this at amiens is nice, we have time and the support of teachers.

it was perfect.

finally, my idea at the start, the solution was to make a personal typeface.


* * *

Reading Arabic

Arabic is a connected script, some glyphs look the same in the 4 positions (start, middle, end, on its own) and others change a lot. … There are dots on vowels. ….

Old manuscripts have a rich tradition, elaborate formations, lovely work over many centuries, and if you look at this character and the formations varying on what comes after. THis is context aware forms that change and react to what comes before and after.

This is Thuluth and Naskh. The small text is Naskh. Its a calligraphic style from the ARbaic word for ‘to copy’ and its used for text settings mostly. Long texts for reading are in Nashk style in calligraphy and typography.

You can find nice calligraphy, but most books in stores is like this. Very static. the Books written by hand are lovely and what we read in schools growing up is SHOCKING.

This is the book typography i grew up with. you feel the loss between the two.

Thanks to Thomas Milo and family, you can do typography in the rich tradition of arabic writing. This is the range of variation possible in reading arabic text.

So, when reading Naskh there are 3 main typographic integreprations of the style. Each bring in an almost numerical approach.

A arabic typesetting book says the 4 positions, you design them and you are done. That was a compromise for metal typesetting. this is argued too simple, but its what i call the tradition. linotype simplified this further, to 2 forms, so middle and end were 1 form and medial and singleton were 1 form.

you have traditional and simplified. then you have dynamic. you don’t have this sat next to that, you have awareness of what is before and after. its not 4 forms per letter, its looking at the letter string and understanding how it is put together.

we can discuss the names but there are 3 kinds of ararbic type, traditional, simple, and dynamic.

You have a few examples of this. Lotus from the LInotype library is a popular simple style. this is Decotype Naskh, a dynamic one.

When I was finishing my MA diseration at Reading, i went into the simplification script reform. i concluded we can do various designs, but to recommend what to do, we ned legibility studies of arabic. until we have science to back up arguments, we just have opinions of designers.

so i wanted to see what is changing; to measure complexity and vocalisation. the 3 kinds go from from 2 to 4 to many. the effect of vocalisation is a linguistic question.

with normal reading you don’t have the vowels.

Hypothesis: The complexity of word formation decreases readingg speed, and this effect is expected to eb more profound in younger readers. the presence of vocalisation marks will reduce regressions in eye saccades.


I wanted to reduce variables in the designs of different families, so i made this family with varying levels of complexity but evenness in other aspects of type.

Its based on this specimen from Istanbul from XXXX AD.

I started in the middle with Afandem Traditional. I looked at Lotus and other traditional designs. I was inspired by the manuscript for the forms.

At the top is Afandem and under is Lotus. You see the similarities. This is intended. There are some differences, you ave these ‘blind’ characters with blobs not counters. I added an eye to them. I kept some traditional details.

Now these 2 fonts are in paragraph setting. You can see these 2 types are overall the same style. That’s the point.

This goes back to the manuscript. I kept the same look and feel and spirit of the manuscript for a consistent design.

You have open counters, think to thick transitions….

So then I made Afandem Simplified.

You can clearly see the simplification of the forms by comparing them here. You see the fluidity of motion is disappearing. There are some changes that need to be done so characters can fit together.

Arabic fonts must FAKE the continuity of stroke movements. If you can’t do this, the type doesn’t work.

So we had to change some shapes so letters could connect.

This design is MEANT to be very simplified, and you can compare this to another simplified design and see they are similar because of this simplification. the Afandem bottoms are not so flat as Yakute (?) but they try to keep the same spirit of word formation.

Its not easy to make a sophisticated typeface like this. The reference was the DecoType Naskh. I then typeset my test text, converted the Afandem Traditional to Outlines and then made it into a piece of lettering. So no layout coding complexity.

So here are the 3 designs. you can see i changed the spacing. the definition of a complex style is that the spacing varies. I tried to minimise this, so if some part is close another part is far. This is a 4 letter word.

So, who has a background reading in eye movement? if you have questions, ask me later.

we have a perceptual span that is asymmetric, forward scanning, we can see 7-8 letters at a time clearly, see 3-4 behind and 14-15 letters ahead.

in a fixation, the recognition happens in the first 50-60ms to pick up on that. we are doing other processes in our brain to see after that.

[DC interpretation: this was tested by fucking with people and fast computers, they have cameras on peoples eyes cordinated with the screen, and as people read their eye saccade fixates on a word or word part, and the computer then disappears that word after so many milliseconds, and we did this faster and faster until people noticed]

Are we reading in parallel as we fixate, or is that mechanics of fixations? thats what psycholingustists research. There is one theory i like, the E-Z Reader model.

fixation position is influenced by what you are reading. what happens in the average 250ms of a fixation? you have 50-60ms of ‘visual analysis’ which happens in parallel.

then you have the level 1 and level 2. l1 is, do you know this word? l2 is lexical access, you know what the word means.

then I stage, you integrate that word into the meaning of the other words.

so when you saccate you tell the eye to move, a start stage where it can be cancelled. then it moves and the previous process runs. all in 250ms.

legibility isn’t just visual, its about other factors includigon cognition. the process of reading must be understood before knowing how legibility works.

Slattery & Rayner in 2009 (these guys are some of my favourite researchers) talk abut how easy it is to extract the fixated word from the letterforms to start processing the word. you can read a spanish word if you don’t know spanish, but its perfectly legible because you can decode the letters.

this shows comprehension is not a method to test legibility.


So, the effect of complexity type style on reading speed?


are you familiar with crowding? we saw this at atypic 2 years ago.

the retina has a little space. when you put a cmpex visual shape there, you might not recognise everything. things are so close you can’t see them separately. so glyphs with many strokes take more time.

so, strokes that are too close are less legible. complex shapes take longer to decode.

i go from linguistics to arabic and back 🙂

you can’t imagine reading arabic if you can’t. we have many joining glyphs, sometimes vowels and sometimes not. we usually read without them unless it drastically changes the meaning. to kill or to be killed. or you can easily make swear words without the vowels. imaging reading english with the vowels removed.

so we have a grammar for vowels that helps. when we make words in arabic, we use 2 morphologies. derivational and inflectional. we have a root, 3-4 consonants, k t b, then you have a patterns and apply it on top of the consonants. when you put three sounds … this means the action ‘to do’ and when you put that on k t b it becomes katana means ‘he writes’ and [DC: this was too fast to note, arabic grammar is cool! 😉 ]

so you need to have an idea about what you are reading to guess what it means.

there haven’t been many studies of reading arabic. roman and pavard in 1987 was an early one. they said when you put the vowels on, reading will go slower. i agree with the hypothesis.

342ms is average fixation compared to latin 250, but often a phrase will be found in a fixation as its a dense script.

most studies of arabic were done in israel as they have research universities there. they have people who learned arabic first and then hebrew as a second language. and people read hebrew faster. both are RTL, both often don’t have vowels. the difference is that its not a joined script.

arabic’s connecting nature make it slower.

the fixations in latin are left of cneter. why? maybe direction of attension, its left to goto right hemisphere, but in arbaic its more variable.

there it depends on the root. the fixation follows where the root of the word is. there was a study in french script too thats similar findings.

when you read in general, left hemisphere does the work. but after 500ms both are used. in arbic its needed as a word shape ‘m l k’ can be 9 other words as there are all possible and if you keep reading, you’ll understand which word it was. so the right hemisphere is used to keep all 9 possibilities around in parallel.

in arabic the RH can’t tell the difference is there is just 1 dot.

so we did a beirut univeirsyt study with 72 students, we used the 3 types, the independent variable to measure was style and voclaisiatoin and age. the dependent variables were reading time, etc etc.

we did the sme text, 3x2x2

i have a video but no time


reading time.

we have averages, but we use inference stats to validate its not just chance. for reading time totals, the only main effect was that of age. so good we measured this.

fixation duration. there was an effect. simplified faster than traditional than dynamic.

same thing for vowels. when you have them, it takes longer to read. 4.7% difference.


the more complex, the less legible. vowels give clarity but take time. for kids its useful.

when you are older you read faster 😉

the extra cost to word decoding

the benefits of …

the question of authenticity? do we use the elegant style that looks better, or do we see the beauty of a simplified style that can be read quickly, or the beauty and enjoyment of text. there is a need for all styles, and there is now technology to do all.

this is not about one style for all.

there is a crisis of literacy, we need to design more types! 🙂


* * *

Freatures of glyphs, whole glyphs, and words are important to reading.

reading a paragraph

I think reading all caps book text is annoying only on convention. but at distance when you compare 2 styles at same point size and blur, uppercase performs better. but uppercase takes up more space. so if you make the uppercase equal the x height, then the legibility is equalised in tests. and uppercase is more condensed.

The big thing in legibility history is the question, is sans more legible than serif? helvetic versus times… they are equally legible. serifs play no role in reading, they say.

that tells us nothing. there are so many variables in the 2 types. but if you just test the presence of serifs in a design, with same metrics.

mary dyson said about the rapid presentation of text on screen, at text sizes serifs were better and large sizes there were no differences.

then compare univers 689, baskerville 168 (?) and gill sans medium.

the hypothesis is that in a h and u the aperture closes, but in a ‘ij’ it sets the stems apart.

So i wanted to look into this. i took a sans and added slab serifs. i compared them for distance reading.


* * *

Type and Emotion

I teach in the UK at Camberwell and run my studio.

When you go to the London olympics, there’s all this branding stuff and when you get there they just did the signage on a whitebaord. 10 years ago i would say ‘ITS WRONGGGGG WRONGGGGGGGGGG’ but now i think its full of life and I’m drawn to that sort of visual communication.

This is a textbook. its on a grid. i like monotype grotesque. Its a bit wrong.

The caps are too bold.

The space is weird.

The number are all really weird.

Its sort of great.

I use it a lot. I want a letterpress feel.

As a student I used William Sandberg’s style, I liked to make things that look old.

This is ok. its what we used to do.

A few years ago I found a type specimen, an Ampersand that is like an hour glass with a comma on the bottom right. i don’t know much about its history, i will not tell you about that.

the research and the journey i go on is not academic. it exists as I’m doing work in the stuiod. i would love to spend 6 months finding out where this is from. but i run a design studio. so my research is curiosity driven.

it looks like an egg timer. or a dog. i have no problem using it in my work.

it gives off a curiousouty and says an emotion.

again, there is more graphic design where ei used this. other things in this layout, the time is, 3.30pm, and the arms of the T make the clock face of that time.

strangeness makes my work now. modern typeface as SO FANTASTIC.

So in many ways the soul goes a little bit. they feel a bit cold.

this feels more, warmer. gives off a warm emotion.

this is a print i did recently, of a concrete poem from the 60s, ‘This is a square poem’. Each sentence is a whole sentence on its own. Its laid out in a square paragraph block. its badly printed. blurry.

i was asked to redesign it. i came to the conclusion i couldn’t make it better. so i reset the poem over itself, to make a new-old design.

This is more typography from my studio.

£ensure – a naff book about making money from hobbies.

these ideas get into my head and get remixed.

this is another example. a letterpress card with the letterpress dept at camberwell.

* * *

So the web allows interactivity, and typography leaves something to be desired. it emulates print.

i want it to be better. but we stub our toe, i don’t see new things.

i see typography receipes:

50-80 chars per line

1 modular scale

1 typographic grid

2 awesome font families

combine all these into 1 page, add leading to taste.

so could you reduce typography to a formula?

no. i tried once. failed.

these are JUST guidelines. they make the same boring designs over and over again.

your friend wants to make an interesting typographic work. what do you tell them?

‘it depends’

or, whats a good line spacing for my document?

it depends

or, whats the best text font?

it depends

we can go on and on about that. humanist, versus modern, always:

it depends.

what does it depend on?

we give the sme answer over and over. in other areas of life this wouldn’t be cool. how much is that car?

it depends!


typographic depends on the RELATINOSHIPS between all these things. we don’t WRITE or DESRIBE that. its like a cake.

imagine desiring the ingredients for cake and their amounts. even the steps to make it.  but not what it should taste like. so cook book receipes always work best in the authors kitchen.

the best cooks can improvise based on a target taste with any ingredients.

this means context. relationships. suitability. expressoin. good quality ingredients.

this sounds familiar.

what document are you margin type for? what relationships between type size. suitability, quality of materials, etc you know what that means for type.

its hard to describe relationships in a broad way that is specific.

the interaction design community has been looking at this. the airbnb UX is good, they have the sme problems, layout, visual information design, and a new thing in typography:


You lead people through pages for creating an experience. they don’t have a general theory for good designs every time. like engineering or material science which are more concrete.

software is a bit more liker hat. in the 1990s, a few people asked what good software developers were like, who made bug free programs?

it wasn’t they were so smart. they were seeing patterns and reusing solutions for multiple problems as they come up. how to re use them and relate them to toehr problems.

they have abstract relationships to convey.

christopher alexander created ‘pattern language’ for architecture. this focuses on a language to describe relaiontships, and a process to make thoughtful descisoins to solve a problem.

software and UX communities picked this up and ran with it. these are from erik gamma, ‘designing patterns’ – bridge, composite, decorator, adapter, singleton, factory, proxy….. are all code patterns.

ux patterns: extras on demand. centre stage. escape hatch. visual framework. global navigation. clear entry points. illustrated choices. titled sections. progress indicator. overview plus detail.

this made UX not an amorphous design.

so design patterns, what are they? they have 4 things: problem context, a solution of relationships, examples, and a name for the pattern.

e.g., architecture entryway. CA said you want queues to show you are in one place then in another. e.g. a gravel pathway feels different to pathway, sounds different. plants give a smell sense. revolving door takes attention but is smooth to pass.

so you can take examples, case studies, and unite them. a pattern needs a name.

my research found many teams in software and interaction use these names in everyday conversation. this makes a design temrinpolgy, reduced confusion, and disagreements in teams.

tahts a pattern. a pattern language is a set of them. you take a problem, fit it to patterns, when you find a fit, you apply it.

imagine a cook book. a cook book with no receipes. just flavours and relationships between flavours. e.g. you want popcorn – salt, butter, truffle oil. you may see wine goes well with that. then you add cheese. dairy, starch. so want something sweet not salty as pop corn is salty. exists and is like this.

what is a pattern algnauge for tyopgrpahy?

newspaper -> news deign, economic text setting, typographic identity, variegated type family. reading typeface, headline typeface, newspaper proportions, malleable letterforms, economic widths, large apparent size, and duplexed widths.

bigelow and holmes did some research related to this.

we could make a classficiaton system that uses the patterns in the document or type.

i think with digital media becoming default, its like introduction of guternburg bible. huge change. digital is infinite free canvas, an interactive medium to think about. we must think about so many new things going on.

right now, we treat digital type as fancy paper. i want to go past that, whats different about digital? to do that, advance the digital medium, what are our hidden assumptions? make them explicit. A pattern language helps to find assumptions and make them known.

a newspaper type influenced by many factors. economic; print small, low leading. lot of cols, so proportions must be a certain way. the proportions of news type are driven by a business model, not aesthetes. that is a print business model. this means hidden assumptions.

a pattern language makes clear why are we doing what we are doing. so we can reevaluate things.

thats my pitch. i think a a pattern language would be great for typography. what do you htink> email me, or @robmck

homework, please eat dinner and think about the relationship of flavours and talk about that. don’t tweet some typography is cool, talk about the relationships that make it cool.

bon apetite!

Q: how to encourage creativity after a pattern algnague is set up?

A: matthew carter has a good name for it. they help you define a parameter for your design you HAVE to meet. then you see what parameters are free form. MC made a diagram for olympian. you need stems that are HERE. then you are free to design the serifs. you’re requirements are locked and denied, so you can focus on what you can build and go wild with.

Q: I find i do better design when i can feel a constraint. having no constraint makes designing hard. i try to translate what you say about teaching. i teach relationships, thinking about the whole. where i am blocked, is the phrase ‘pattern language’.

A: what i describe is, sit down withe experienced people, create a book with the patterns. like software and ux and architecture have done. ‘designing patterns’ for coders did this. you have a name for the pattern, and i work at microsoft, and people say ‘make a decorator for the output of that factory’. its a terminology, a design language that is specific. thats why CA said it is a pattern LAGNUGEA. its a domiain specific terminology, a lgnague. UX uses this too, and people learn concepts faster with this teaching. solve this problem with these patterns, its fast. they refer to each other. it makes you walkt the process. they help you think through things well.

Q: so its like decoding intuition?

A: its encoding expertise so you can decode that as intuition.


This entry was posted in Uncategorized. Bookmark the permalink. Post a comment or leave a trackback: Trackback URL.

Post a Comment

Your email is never published nor shared. Required fields are marked *

To create code blocks or other preformatted text, indent by four spaces:

    This will be displayed in a monospaced font. The first four 
    spaces will be stripped off, but all other whitespace
    will be preserved.
    Markdown is turned off in code blocks:
     [This is not a link](

To create not a block, but an inline code span, use backticks:

Here is some inline `code`.

For more help see

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>


  • What we do

    Understanding Fonts is a type design training business. If you'd like an event in your college or city, let Dave know: