Skip to content

Latest commit

 

History

History
1763 lines (929 loc) · 158 KB

2015-ATypI.markdown

File metadata and controls

1763 lines (929 loc) · 158 KB

ATypI 2015

These are live blog notes from the lectures at the ATypI 2015 in Sao Paulo.

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. This is probably FULL of errors. What do you want for free? :) If something here is incorrect it is probably because I mistyped it or misunderstood, and if anyone wants corrections, just should tweet me – @davelab6 - or file an issue. Thanks!

Web Fonts

Original Deck: http://goo.gl/65f2Pq and PDF in this repo

09:30

DK: Web fonts allows us to quantify the impact that type designers have on the world, which is significant and important.

DL: I’m David Lemon, stepping in for Matthew Reichs, my manager, who is unable to attend.

JG: I’m John Gianopollis, and I’d like to show the real world situations with web fonts.

DK: 4 things, Adoption update, technology, and the future.

What is a web font? Here’s a great example of globo.com a local site using Open Sans from Google Fonts.

Adoption: There are a few ways to determine if web fonts are in use. Eg in a browser look at the page source. HTTP Archive looks at the top 1,000 domains and looks at their homepage. The Google Fonts Large Scale Analysis looks at the whole web, including leaf pages, which are often where web fonts are used first on high traffic sites.

We are now at 50% adoption on the Top 1000 pages! Truly amazing

Deep analysis shows even higher: 80% of new web sites use web fonts across the site. We see the 1,000% increase from 2013to 14 and to 15. A massive roll out of web fonts across those sites.

We see 5 Bn pages, 2.5X since this time last year. A huge growth!

There are 75 million domains, +15M in the last year, showing that there is a huge customer base.

Also the rapid shift to mobile - faster than we on our team initially thought. In the last year an 11% increase, and quickly getting to 50% of all web views.

Drilling in by country, we see Kenya at 80% of mobile, but huge counties like UK and Japan are way up.

In script brekadown, latin/latin-ext are the biggest sets; 80% of GWF is latin, 8.9 is latin-ext, then cyrillic. A huge opportunity ahead on non-latin.

Brazil! Looking at counties, Brasil is #2 for our traffic, and that’s fascinating for us. That’s 500M font views per day, we see Sao Paulo is the #1 country within Brasil, 83M font views per day. That is more than every city in the US except NYC! And its fun to relate them to other numbers. There’s 1M pizzas consumed per day in Sao Paulo, more than the total for Italy.

Londrina Solid, by Marcleo, is 576M, and then Montserrat, Lobster and Bitter from Latin America.

The most viewed video on youtube is Gingham Style, now at 2.4Bn. We serve 3Bn font views per day of Open Sans - our top font is PER DAY is bigger than the TOTAL views of top youtube.

John predicted that web fonts would soon grow bigger than flash, and that is now true.

JG:

Its mind boggling how far we’ve come in the last 3-4 years. I wanted to look at the global market. Building content with fonts is changing and to the betterment of the whole indusry. In th e last few months, we knew that the market would shift form flash to html, and in the last 3months in particular, with the Power Saver Plugin, set to On, so you don’t autoplay content; Mozilla is blocking flash for various reasons; Facebook’s CSO is calling Flash EOL.

In the ads space, the largest org for the creative community bpulisherd new guidelines that HTML5 is the preferred format for ads and flash is being deemphasised. After septemreb, major ad networks are no longer accepting flash. Why?

That’s easy. Lots of OPEN technologies out there. Huge Usage and reach. Powerful i18n. Modular. Multi application use.

I think the reach is the biggest thing. So many devices support HTML5 and not flash. Rich media is soemthing users want everywhere. Its a multi app use tech - rich media, ebooks, games - all being done with HTML5 as a global standard. A tremendous amount of browser support globally.

So, a hue upheaval for browsers and tech companies, the open standard, its a perfect storm in the market place. Developers and creatives are looking to take the millions of flash content to convert to html5, and transcoders are converting flash to html5, but its not as successfully as they would hope for, and fonts is a huge topic in the discussions.

The number 2 or 3 point is ‘what about the fonts?’

We see new use cases for fonts, digital ads, email marketing, these are real ways to publish content, using real fonts. Not text as images.

But there are workflow issues to iron out. I go to orgs and talk to creative teams who don’t have matching sets of fonts. They have desktop fonts in 1 team, and web fonts in another team, and so on. that can lead to images of text!

Traditional challenges, of file size, http request minimization, and End User Eduation. Knowing its LICENSED, that you must obey the EULAs, we must continue to educate the market about that.

And new challenges. Digital advertising wants to do cross screen, responsive, dynamic content, SEO’d ads. Web Fonts is misunderstood outside the WWW. You have a 3rd party call for web fonts? that is blocked for a web ad! You lose your brand typography, that’s a problem.

Page view tracking, is considered bad today by ad changes, but counting impressions is the basis for all businesses. What about ad blockers? Apple has made these big now. Anything considered ‘bad’ is blocked. That won’t help. URL domain restrictions will not work for digital advertising.

We must become more flexible. Today, web fonts are tied to domains for security reasons and to support a business model. But in a world where a font can live on any ad server, any device, at any time, a URL restriction will not work. We must address this across the board. We must make our fonts work across the board. , they But there is good news! Last year we couldn’t show you what I will now show you. There are now great tools on there for “in app” web font support- WIX, SquarSpace, Adobe Muse;e digital ads, delta, elite, ResponsiveAds.com, Adcade which are making a great world for designers to alleviate some of the problems. A lot of developers started 10 years ago who are moving into ads, who PREFER hand coding HTML for ads. They try to convert Desktop fonts to web fonts. We know thats not a good thing to do. Tools are helping there, but as an industry, we should focus on how to help users understand fonts are IP and how to respect the licensing.

Who has heard of WIX? They have a web hosting soluition with a DIY UI. They integrate web fonts into that UI. Its not for professional designers, its for the mass market. That is a good thing. Elite, a Valley company ad serving, that has a DIY tool for people to make new html5 ads from scratch. They use free fonts, Google Fonts, custom fonts from a major brand, or go through a virtual type book to find the right font. Here’s a Flite example. The file size total is under 15kb. ResponsiveAds.com in NYC has a DIY tool, here is a video showing it making a RESPONSIVE ad using web fonts. We’ve seen this on the web for pages, but not for digital advertising. This is ALL browser based. Tumult Hype3 is a desktop app that has integrated the Monotype Fonts API. it has subsetting built in. It has all the features like animation that people are looking for.

There are concerns of security. So email marketing uses HTML2 style images and basic CSS. They use safe system fonts. Its good to see at least that use of fonts.

The tide is rising for all! New use cases, new addressable markets, more fonts lead to more content, and more money for everyone in the industry. If you look at all the page views and compared it to the page views of digital advertising, they are 100x what the page views are. The magnitude of the opportunity at hand. The APAC region is thirsty for fonts. Not just being able to use fonts, but to expand the library. Morisawa does amazing work. We want to see the whole region adopt web fonts and move forwards together. We’re not done yet!

David Lemon: Improvements in Performance.

Speed is a key aspect of user experience. Faster = better, reducing file size, smarter downloading, and making more use of browser caches. Web font files are freqently not the largest filetype, but often the most sensitive to delays. Any delay is a bad user experience. So a lot of thought is going into making this as fast as possible. Making fonts smaller is obviously good.

The main area there is improvements in the compression, with Brotli in WOFF2. That is now faster and better than the compression in WOFF and GZIP. Zopfli for prebuilding WOFF1 files (too slow to use dynamically). You can also recognise when TT hints are not needed and remove them - or serve CFF fonts. You can also do subsetting (downloading just a portion that you need.) This is essential for east asian fonts, which can be 5-15 Mb, which is overloading for web pages - yet the market demand is there for the same expressive demand we are accustomed to. Unicode-Range is a possible solution for this.

Here’s a chart from yesterdays working group session on Woff2 vs other solutions.

There are other CSS features for fonts. font-loading helps with more control of loading behavior. Resource Hints allows to front-load the loading of fonts. And HTTP/2 is starting to see adoption which allows parallel downloading, and server controlled push for some resources including fonts.

Browser caching is key. Cross-site font caching is great (at least for Google Fonts, not great for all serving models with monetization models, but people are looking into that.) Also “stale while revalidate” helps.

Subsetting was mentioned earlier. This is a technology that is crucial. TypeKit has a technology for Japanese fonts, I set the font, I paste in some Japanese text, and you see the font loaded. Now some more text, and you see a new subset is loaded.

Opportuniies ahead!

Google, Monotype and Adobe TYpekit stood on the stage last year and each company had said some areas that they saw as future challenges. This year, we saw that we are all on the same page about the challenges. We see the problems as industry wide and all aligned on these issues.

We have great support for web fonts in Latin, but we need better browser support for OpenType layout. Its incomplete and inconsistent, so its a big challenge to do reasonable looking typography. This is esp important for complex scripts. East Asian fonts are working better with subsetting but vertical japanese support is virtually non existent. RTL support is still weak.

We also need more simple flexible ways for licensing fonts that is more flexible and easier to work with. End users want Freedom, Flexiblity and Choice. If we can make is simple and easy then usage will continue to climb.

Email and ads are untapped potential that need more infrastructure, not just licensing. So as the user experience imroves, we’ll see even more use of fonts.

Key takeaways

Its really growing! Rate of adoption in the last 5 years has been amazing, its totally mainstream, with 50%+ usage, and lots of inspiring examples of the new web typography. Our latest microsite for http://acumin.typekit.com/history/ is serving 90 fonts, really fast!

References

The Large Scale Analysis - www.goo.gl/SHegYf the Alexa Top 1M, The httparchive Top 100/1K, and www.stateofwebtype.com are all good sources.

Questions?

Behdad: Complex Text Support in browsers is now better than Adobe CS tools. Any chance to improve that?

DL: Each release improves support, main focus has been on Brahmic scripts.

Si Daniels: A lunch session Q&A will follow.

10:15


Tech Day: Research

“Kevin Larson, Ann Bessemans, Nathalie Dumont, Anand Vijay

Kevin Larson will collaborate on a “state of the union” talk on the legibility and reading research that’s happened over the past year.”

AB: …

Kevin Larson

KL: … Reading app that enhances reading. this mark up enhances reading comprehension by 10% immediately with no training. Dyslexia may be a visual crowding problem. Adding more space between letter, words and lines reduces this and helps dyslexics. This is just a feature to turn on, nots not stigmatizing. Critical for young kids. You can combine the comprehension markup with the spacing.

David Lemon: Curious about spacing. Is there a sweet spot for the amount of space to add. Is what you show tested, or arbitary?

KL: That work has not been done and should be. This is mimicking what was shown to work in the paper (in French and Italian.)

Natalie DuMont

Continuing my PhD at Concordia on iterative design methology. I work with a succession of drafts; I do not start a project with a conclusion in mind. Each series is informed by those before. You explore a variety of ideas. I am often asked “how do you choose?” You set a clear design brief. You have your own design criteria (e.g. I think oblique stress in strokes is more legible that vertical) and the coherence of a concept.

There are different methods; hand drawings, ssXX in Glyphs is easy to do alternatives, there are also Smart Components for parameters and interpolations inside GlyphsApp. There is also most systematic increments of parameters (using Python scripts.)

Here you see 5 neck/spine strokes for an ‘a’ and 4 for the bowl, and using GlyphsApp Smart Components, and scripts, I made a systematised set of iterations, stepping through to make all possible significant combinations, to iterate down into the choice I wanted. The labels show the design space location, plus a final number for manual iterations.

here are the final 4 ‘a’ I developed. I selected the 2 right most.

Anand Vijay at Monotype

Machine Learning is a subfield of AI where programs teach themselves by learning from example. The Natural Language Processing of Siri, Cortana, Google Now, the self driving cars, the recommendation engines of Amazon, the financial projections, the medical imaging… it will all get better. We at Monotype looked if we can use machine learning for fonts? Yes, we did research on 3 topics: Fonts similarity. Fonts pairing, a tricky problem. Fonts attributes, can they become comic, polite, masculine.

Fonts similarity. We wanted to increase the human concept of similarity, e.g. see Arial, and be recommended Helvetica. We collected 200,000 data points from real users, and made a 3 layer neural network. This led to www.FontScout.com which I’ll demo now.

Fonts pairing, a tricky problem. We asked normal people what pairs they liked, but there was no consensus. WE found a chart from Steve Matteson, and applied this to make 11 categories, and paired these according to the rules in the chart. Here’s a mockup for fonts.com where I select a font, and set a complementary font, and then hit a button to look at pairings, and it will generate new pairs. I can star a pair to pin in, and then bit the button to make more pairs. Finally I can drag my pairs into a collections tray at the bottom.

Fonts attributes, can they become comic, polite, masculine? Another demo video. Here’s the 31 emotional attributes, which each has a percentage. i take “comic book” and set it to 50%, then I click again it set to 100%. Now lets set another attribute like Feminine to 100%. Then I remove the comic attribute and see all the very feminine fonts. Now lets say, Happy, and Calm. I can pin a font which keeps it at the top, and i can drag a font into a tray collection at the bottom.

Conclusion

The 3 machines we made work like magic; it captures the idea and has knowledge now so it can work on new data.

We want to use these ideas for typefaces and fonts and continue to explore. The future is great!

Questions

Q for MT: Have you tried applying machine learning to font design? Eg kerning, could you apply appropriate kerning by learning the kerning of other fonts?

A: We have not.

Steve: We have looked at general font design for machine learning, but not kerning. Its a great idea!

Si Daniels: If I disagree with what the brain says the personality of my font is, can I untag that, or does machine says no?

A: If we publish this work then perhaps you could have such feedback. But not today.

Fabio Haag: What was the user data about the fonts were comic or other attributes?

A: We picked 1,200 fonts, and used Amazon Mechanical Turk to ask users questions about the fonts. We used that data and looked if there were consensus for a attribute and a font.

KL: If you take an attribute like Comic, you take mech turk data, then you learn that tag to the shape. How to verify the shapes work? We started with 1,200 fonts tagged, then we ran it on 5,000 fonts and reviewed the results, before scaling it to all 18,000 fonts.

11:00


Tech Day: Tools —----------------

“Type tool gurus will bring us up to date on font development and testing tool advances over the last year, and provide sneak peeks into what’s coming soon.”

Glyphs

Rainer: We started with components, font in the TT spec, and we introduced in Glyphs 2.0 “Corner Components” where you have an open outline that is inserted live into another outline. The main idea was to facilitate serifs. This is an open outline, a direction like any contour, it goes counterclockwise, and you insert it into another node. You see an “A” here with a serif, you can see the curve is bent to make a smooth transition as the A is slanted.

We also did “Smart Components” and this was initially intended for CJK fonts. In Latin, you have the exact same component like an “A” with all the diacritics. But here in Korean you see the same letter used 4 times, but each time it is slightly adjusted. If you do this with components, you can scale them, but this distorts things. So we thought to combine interpolation and components. So you add a couple of interpolation axes to your component, and you can dynamicaly adjust the shape. Its not squished, but interpolated. So you can reuse CJK radical or specially set up components. This works pretty well, and those at Granshan will have seen Toshi’s tibetan font that was made with this.

What can we do about interpolation? Its great if you can do a big font family with 20 masters, but you have to draw them all. Wouldn’t 4 be better? or just 2? But then the problem is that while the instances can be good for many glyphs, some will need more masters. A common case is the ‘e’ where the bar can change. Here is glyphs, the top area is black on white where we draw our 2 masters, and the white on black is a preview showing the instances. We go from very light to a super heavy master. So, you see the crossbar is too light. So we have the ‘bracket trick’

Tim Ahrens contributed valuable code for this. You can have any number of masters on 3 axes, means you now can have more than 1 font family in a single glyphs file. Here is an example, 3 masters, this 2nd master has contrast but the 1st and 3rd do not. 1st is a thin, the 2nd is a high contrast black, and the 3rd is a no contrast black. Now we can export this and see what it looks like: Extrapolate a little bit, you can see display styles. Its easy to switch between masters. You can see the masters visualized; even with bracket layers this is visualized so you can always see what is relevant for interpolation of a glyph. Color coding of contours and components; the colors relate to the order, and also lines to show how the points are compatible. Fixing the orders of paths and components can be tricky, esp with 30 paths in a glyph, but we have a new “fix compatibility” window that allows you to drag and drop the path order. If you have incompatible layers, you get a red stripe on the top of the glyph.

In v2.2 we announced in the blog post a new feature. We have corner components, and you can now put in complex paths and even left/right anchors that will be flipped. Also these smart components can BOOLEAN SUBTRACT when combined, which traditional TT components can not.

Finally, web fonts formats. we added this to Glyphs, when you export, you have a web font section, WOFF, WOFF2, EOT, and CFF/TTF options. The autohint will be ttfautohint or the PS Adobe Autohinter based on which you choose. TTFA had an important update to 1.4 and its in Glyphs 2.2 already!

Color Font formats are also new. They are tricky, there are 4 different formats, 3 are now standard (not the Apple one) but we support all 4. You can do classic layer fonts (although no nice preview) then you can do the MS COLR/CPAL tables, then Apple’s color Emoji, and finally with Glyphs 2.2 we support the SVG table, so you see this is a SVG file, and you just add a ‘sag’ named layer, and you can copy and paste a SVG into that layer, and it is exported, and you have everything SVG supports - can see the animation in Firefox! But the CPU required for rendering them is substantial.

Another new thing in 2.2 is taking the Font Family farther. We have a new saving file format .glyphsproject - open a glyphs file, and it shows you all the masters and instances in the file, and you can edit their Custom Parameters. These can control the export process.

Q: This is like Photoshop Actions?

A: Yes

Q: How was the animation made?

A: Plain SVG markup

FontLab 6

Thomas Phinney: This month I became the official President of FontLab, after a gradual transition, and Ted has retired.

Adam Twardoch: I am now Director of Products, and we’ll talk about FontLab 6, an on-going project. I want to show where we are at, and summarize what we have been doing. There are 3 important aspects. First, drawing.

Seriously Better Beziers!

We love curves, and try to make them better. Some designers want to control the form, not just points; balance, tension, Tunni Line and Point.

The “Tunni Lines” a way to control segment tension. In addition to BCPs you can manipulate the tension very easily. You can make adjustments to the BCP handles with the extra point and line that Tunni describes. You can make 2 parallel contours have a parallel tunni line.

We also have Point Sliders. You like your shape but want to adjust, but you have points at extrema. So you can slide a point without adjusting the contour at all, to a place to adjust it, then slide back to an extrema.

Adjust big time in no time!

We have traditional interpolated nudge, per Christian Robertson’s plugin for FontLab 5. The handles are interpolated. We also have Power Nudge. so when you move the points, it finds other notes to move proportionally. You can tag nodes as ‘servants’ to move proportionally.

Power Guides and a Magnet Tool. These tools all attempt to speed up your work. You can attach points to guidelines, and move the guidelines, and the nodes are interpolated. [ Its a mesh transform ] So this allows you to do interpolation with just 1 master.

It now suggest snapping points. If you have standard stems, as well as snapping points across many glyphs, so you can match a structure towards neighboring glyphs. We have font and glyph guidelines, you can tag them, so they are shared across only some glyphs.

Smooth the bumpy Curves!

Some pro type designers are good at drawing smooth contours, but others need help. We took G2 curvature a bit further.

Harmonize, from RMX tools.

Genius Node. This allows G2 smooth to be enforced [ like spiro ]

Curvature

Eraser. Tim Ahrens says that in all other editors, if you remove a move, the curve can be improved, but we have the best algo he has seen

We also have smart corners. So you can make live rounded corners. You can work with the underlying structure, and use

Catch Glitches!

Font Audit is back and improved, thanks to ideas in Tal Lemin’s Glyph Nanny. We polished the UI, and added View Curvature as seen in SpeedPunk and many CAD applications.

Next Gen Drawing!

Power Brush.

For drawing from scratch, you can apply a brush to a skeleton stroke, its live calligraphy, angle, thickness, endings. It won’t give perfect shapes but will give prototypes, you can set the stroke thickness.

We have a Power Pencil, you can add and remove contours with this.

Rapid Tool

I know some font editor makers, they have reservation to the Bezier Pen Tool. Some people don’t get it. Some people are superb with it. So we want a faster way to draw Beziers. So you click once for a corner, two for smooth; you have a single ‘rapid’ handle for off curve points.

We have a traditional Bezier pen too.

No more boundaries!

Fill and Scissors. Self intersecting, overlapping, we see how they work. We have a scissors tool that will automatically restore overlaps, and a fill tool to say what is transparent and which is black. SO you can make a dollar with an S and a | and only have the top and bottom bits of the bar sticking out filled.

We added an ability to extend the nodes.

Shapes!

We call our new components shapes, as there isn’t 1 outline in a glyph and other refer to it. Like subroutines in CFF, you have outlines that all refer to glyphs. so you can edit them anywhere they occur; no need to find the parent. Edit anywhere they occur.

Unlimited Layers!

A new thing for FontLab, not just Mask and Background.

You can navigate, search easily for different properties. You can sort, color flag glyphs, display different properties.

We have live metrics, you can use expressions to refer to other sidebearings, its an expression engine, and we try to be friendly to other workflows and editors, so we can open Glyphs files, UFOs, we use Mutator Math based code rewriten in C++ so its 100x faster, so our interpolation engine can do an unlimited number of axes. So we are friendly to other tools. We don’t tell you move over to only FL6, no, all tools have their strengths, and we want to be part of that ecosystem.

Slanted grids for italics.

New Kerning UI, with previewing of kerning groups; not class based with a key glyphs, like UFO, but new kind of class that means you can’t have conflicts.

New Anchors, that can be positioned with expressions. You can say the anchor is 50% between 2 guidelines.

A few more things!

Yuri: Parameterised Fonts, is interesting. The problem of things like Adobe and another one [ I assume he means Prototype ] is that they all work with some special font. This is PT Serif, opened in FontLab 6.

Now lets show a few glyphs, and sshow the parameters sidebar. This was added to the font simply. You can make the x eighth taller. Or Ascenders higher. descenders lower. Move the bar up. The midline of the caps. You can apply parameters not to some special made font, you can apply them with “power guides” - its not complicated to set them up.

I open a glyph here, “B”, and you see the power line that I drag and the nodes follow. For A-Z it takes me 30 minutes to set up all the guides. I use this to link points, like in TT hinting, but much easier. Here is H, and our ‘Shapes’ are not just whole glyphs but also glyph parts. You can see how easy it is to make smart corners, attaching to any point, you can have all these things combined, smart shapes, power guidelines.

If you work with a glyph, you can easily access similar glyphs.

In CFF you have no components, so when we import a CFF binary, we auto detect linked shapes on opening, so it reconstructs them.

Adam: so now the question is, when will it be available?

Now! www.fontlab.com/fl6 will tell you more about the product, movies, links to preliminary documentation, its available FOR FREE to try out, no built in limited, for Mac. A windows version will come at some point, and we hope you will get this, try it out, give us LOTS of feedback, and we’re here for the next few days, and listening on all the channels.

Behdad: What happened to the name?

Adam: Its FontLab 6 with the roman VI.


OpenType History

[10 mins late]

Vlad: Consumer Electronics starts with standards, even in reverse of typical industry process. I approached ISO and they said, we already have a font standard! Philips licensed PFR from Bitstream, and as a matter of business insurance, they added it as an appendix to another digital technology standard. So PFR was a font standard.

I told them openType is so superior and should be considered, but they said no because “OpenType is Proprietary” - no guarantee that the MS website will maintain the spec there.

I saw that this was a big thing for them. So I approached MS and said, you have OpenType of 6-8 years of investment in development, would you consider releasing all the rights and letting them take over?

Surprisingly the answer was almost totally yes, although I saw some healthy skepticism from some MS folks. A few months went by for lawyers to get an official release, MS went to ISO officially, did all the legal work to release all rights, and in 2006, the document was finalized, and in 2007 the ISO standard was published.

Since then the format is Open Font Format, in the IT / Coding of Audio Visual Objects - Chapter 22; equivalent to OpenType 1.6 with support for context alternatives, Stylistic Sets, and other things; a big change at this point. Not every problem was solved.

A highly controversial item at that time was about the 64k glyph limit. It was an issue then as it is today, and a group looked at how to solve the problem without breaking the backward compatibility. We looked at how to make a higher level fix than OpenType, and that is a new standard, Chapter 28 of the ISO spec, Composite Fonts.

I think this has not got the attention it deserves, its a way to use XML to make a receipt for many fonts to be joined together in a single virtual font. A set of global metrics for the global font, to size and position them appropriately. Languages and unicode ranges can be supported. If used properly, it is a powerful tool.

Meanwhile OpenType kept improving, and color fonts was the next big thing. By 2012/2013, we had a good idea about what could be done for Color Fonts in OpenType. This is the big 2014 MPEG meeting. That was the first time we got such a broad set of companies to join together in font format development.

Here is a MS layer fonts example. Then the SVG example. If you thought Comic Sans was bad, wait for colors fonts to arrive! :)

Greg Hitchcock jokes that he could really see OpenType becoming an ISO Best Seller, but now it has happened! :D

Si Daniels: So that’s a quick over view of the last 18 years!

Behdad: How many downloads?

Vlad: You don’t have to pay to get it, you can see the free link, but you can opt to pay for it, and many people do.

Roger Black; You are saying that there are some failure of nerve, a change or heart, a change of management, that caused abandonment of direction in TrueType and OpenType, and TrueType. I thought apple wanted MS to make it a standard; the GX thing floored me. Its an acid flashback I don’t want to have again. It caused such fall out at our little company that I don’t like to think about it. I think Apple walked away from TrueType at that moment. But you talked about TrueType as a web solution. MS walked away from it like that and we got other stuff.

SI: This is a short version. You need to talk more about Type 1 and TrueType early days to understand that. Everyone I spoke to at MS at the time, thought GX would be what we did, but it was only the legal block with Apple that stopped it.

RB: That was a Sculley thing?

SD: Yes, and then Apple supported OpenType quietly without any agreements in place. I think with the web, EOT stands for Embedded Open Type. It was something that people cared about at the time and the problem related to the competitors moving to open standards. Netscape supported PFR and dropped it because it was proprietary, and there was no interest in standardising EOT as it was build for DRM and other concerns. So EOT stayed as a supported technology from MSIE4 until 8-10 years later.

Chris Lilley: We standardised font downloading in CSS, then Netscape became Mozilla and couldn’t use PFR as they were now open source. We did half the solution; we have a download solution with out an interoperable format. We needed a PC for EOT and a Mac for PFR and it was too much work and people didn’t care. Only when WOFF arrived.

RB: Open standards were great but MSIE had the market by 90% at the time. So any rational designers would use EOT. But no body did that. The EOT offers from foundries were not there. I forgot EOT included the OT part; I thought of it as something else. It wasn’t an open solution as the web community rejected it. We could have had web fonts 10 years earlier if we went for that.

SD: I totally agree.

Johnny G at Monotype: I don’t know if you all know the history Roger mentioned. The work MS did on EOT, that Vlad did on the working group, a lot of the stuff we talked about this morning would not be there. I really mean it, we couldn’t say that we see 50% web font adoption in 2015 without the efforts back then.

Si Daniels at Microsoft: So that’s where we came from. Next is where we are going. There’s an Open mic sessions, we’ll have people talk in order. Everyone gets 5 mins.

Kamal at Monotype: I worked at Monotype for 20 years. The last few years I worked on OpenType table developer to emulate various scripts of the world. By emulate, what do I mean? for many scripts, what you type to enter text is different to what you see at the end. We hide this transformation in OpenType tables, so a font is functioning if you type the right letters and see the right display as a result. The scripts I work on will not be familiar to most of you. So what I would add to OpenType is something I want to show with music notation. What we can do today with OT were a fantasy 20 years ago. As with anything new, initial adoption was fast but what was done with OpenType 15 years ago was much less than we can do now, as we understand it better. We break down characters into classes, base chars and marks. The bases are like the A and the acute accent is the mark. This concept of bases and marks is very common in many scripts. In OpenType we have them, and the main OT interpreter is the MS UniScribe. As assumption was made that marks have no widths and bases have a width. …

Jean-Baptise Levee at Production Type, a foundry in Paris: I don’t pgroam code, or develop, but I design and sell fonts. I wish to ease my work and the work of my customers, the users. The first thing, I heard a lot about variable fonts, parametrics fonts, since the dawn of MM fonts. I have not seen something that is user orietnated. I would like something like a ‘liquid font’, a pattern, not a fixed design, it would support floats for more precise drawings, and that is rendering environment aware to detect it is on this OS/program, so the designer could pick the exact boldness or contrast for that system; this requires a family with widths, weights and contrast would be a design that flows seamlessly in its environment, from a fridge screen to a phone, tablet, laptop. Something a user can adjust to their taste. There are concerns about bad taste ;) But we should trust the user, and overestimate him or her, not the other way around. On some phones you can chose a font for your phone, an illegible floral script, but it isn’t our call to decide what a user should use. The OT rendering engine, that Kamal has already pointed to, but we as designers and users wish for something that looks more like proper randomness, not a small set of variants being rotated. It may not be too useful but it is a request. For kerning tables, there are limits that are frustrating; I would like to be able to kern triplets. That seems like a good option to have.

[ Toshi: Triplet kerning is possible today? ]

[ Thomas Phinney: Yep, well, the format supports triplet kerning ]

Ned at Apple: Often we need shorter glyphs for avoiding clipping, especially in limited screens (watches) then we need to fit text in a bounding box without clipping, especially on multiline text. We have a stylistic set that has a redrawn glyph (here an A that is lower to allow for the ring-acute to work) but this is not dynamic. We detect scenarios where this is likely to be needed. We would like to see something in fonts to support this. JSTF table today has a list of lookups that are progressively applied as needed. The bus might change the width of glyphs, which is an issue for layout engines; e.g. a standard form is a vertical conjunct with marks hanging below, and you might like a horizontal form to fit in a height box that changes the line lengths radically. There are indeed static approaches but we’d like a mechanism that is consistent across fonts.

Behdad at Google: [ lol his 1 slide has 7 sections with 5-8 points each ;) ] So, I think we can integrate GX into opentype. so what you can do with superpolator you can do at run time. And clean up some of the old tables. I have a 2-3 page document. But those are the big things.

Vlad at Monotype: So, you look back at history, you see features can take a couple years to be implemented, and a few more years to become implemented. Some of the previous proposals imply changing the spec in a way that breaks backwards compatibility. I am not saying it can not be done, I think it is inevitable at some point, but please consider the cost of making a change that breaks backwards compatibility. “Cleaning up tables” across all implementations could be a huge cost. Please lets be cautious of this.

Raph Levien at Google: My day job is now the text stack on Android. I care about file size a lot, and speed of rendering; fancy OpenType things are really slow. To address that, a specific technical improvement, is to use WOFF2 style point encoding and eliminating the box. I’d like to see that, which is definitely a break. The smallest representation its the best. I think spiro splines could be best, but maybe not something for 2015. Behdad’s proposal fills in details about variation and interpolation, GX is close, but what if you made a new font and dropped it into a Mac 2, it would work ;) But its just cool, not needed. The speed of layout with state machines in AAT and Graphite is fast. …

Jan Middendorp at MyFonts: I listen to users and foudnries, i know what foundries can do with OT features. I know that type designers and the industry is far ahead of the graphic design industry that makes apps and uses them. We had this choir speaking up about Adobe support of OpenType. We get dozens of complaints a day from users using Laura Worthington’s fonts with all features in MS Word and Apple Keynote. My presentation on Friday will also be about this; I see the industry as a broad thing. I think should bring some knives to the table for normal people, who are not stupid, but do not know as much as we do, to create stuff with what type designers are making. That would really help.

[ 15 minute break ]


[ Panel ]


Challenges of the International Type Market

Indra: I’m a typographer in Germany, I did book design, I have designed type but nothing published (mainly Bitmap fonts) and I’m currently a full time teacher of Typography at HBKsaar, I also do writing and consulting for type makers and sellers (their marketing dept) and I do consulting for companies and enterprises about choosing and licensing type.

I thought I have different views myself - from perspectives of users, academia, larger companies, and type designers - and there are different ideals and goals for selling type.

The interests and problems of the

  • Type Designer
  • Foundry
  • Distributor
  • Graphic Designer / User
  • Enterprise

when it comes to

  • Licensing
  • EULAs, as well as
  • Distribution
  • Pricing
  • Discounts
  • Scripts and character-sets

Type Designer

– IP of their design and its protection – maximum dissemination (?) – maximum royalties, pay

Foundry

– tight EULA that holds up in court – restrictions, licensing extensions (e.g. mod, logo) – licenses that reflect extend of use – licenses for every medium – licensee information – easy handling of purchases – maximum revenue/royalties­

Distributor

– streamlining of EULAs – offering type for widest use-cases as possible – licensee information – right to negotiate certain licenses (OEM) – maximum revenue – share for their part (e.g. distribution, marketing, support)­

Graphic Designer / User

– easy licensing (online) – maximum range of use, extend and duration – understandable terms, prefer. no reading of terms – ease of shopping, accounts, carts – cheap prices­

Enterprises

– worry-free font licenses – covering maximum range and extend of use – right to modify – redistribution – flexible terms, tailored to their needs – one-stop licensing (supplier registration) – clearly quoted pricing­ upfront

Problems beyond that:

– language – local markets, different price levels – quality, extent of product – principle of contracts (who has one with whom?) etc.

Some things we could change

Lawyers

– don’t require all-caps formatting and obscure language

Foundries and resellers

– make it super clear that customers are buying a license, not the fonts – that the purchase comes with terms (whose?) and that customers should read them – list covered points of EULA – make it easy to see online which licenses are available and what they cost­ – present EULA (or list/summary) early in the

  • checkout process or before

– provide info and search for character-set and (OT-) features of fonts – present full character-set, widths, weights, etc … (and many more things I’d have ideas for when it comes to typeface websites)

Or more far-reaching questions

Can we make licensing for different media easier? Maybe with a bucket approach instead of for every case individually? ­ Why can licensing a retail fonts get more expensive than custom font design?

How can we attractively price retail fonts without participating in a destructive race to the bottom?

So the current market has many facets - licensing, EULAs, distribution, pricing, discounts, and scripts.

Licensing is a big topic today. There are different price levels for different regions.

Who are the players?

Type Designer. Wants IP of the design recognized and its restriction. Also may want maximum dissemination of the type. Maximum royalities and income.

Foundry. Sometimes the same as the TD or different entity. A Tight EULA that holds up in court? How tight and complex should it be? A foundry wants it to be really tight so they have means to sue someone. Half my students never read the EULAs, they don’t understand a word in there and they are well educated German students with reasonable English reading skills. Licenses are restricted to some media, so they don’t allow use in logos. You want to make it easy, while maximizing revenues.

Distributor. That works with a lot of foundries, and this gives you different support requests; so they have a ‘house EULA’ and not make special cases which is less interesting to the foundry. They want to offer web fonts for sale, negotiate with OEMs, and they want to maximums their income so want to pay less royalties. They do earn their part, by doing marketing and support.

Graphic Designer / User. They want super easy licensing. Does anyone read the license? I don’t either. They want the maximum range of usage permission, for the maximum time. Even modifying it for a logo, which they may see as their right but it may not be permitted. If you license it from 1 distributor and it allowed and then license another font from another source and its not allowed, its confusing. They like shopping at a huge distributor with a familiar checkout process, support process. And of course cheap prices.

Enterprises. They want a worry free situation. They want a stable price (which can be very high) and that deal can fit their purchase system.

What can we do?

Lawyers. They should not require all cap formatting and obscure language. David Lemon had some thoughts on this:

David Lemon: Its language written by lawyers, for lawyers. it sucks for users. we often go back to adobe legal and try to do something, and are told ‘no’. For me an interesting thing is Matthew Butterick, who is a bar-passed lawyer in California, and he is proud of his plain language EULA that fits on a single page. He has offered his services to consult with other companies to draft EULAs. We on the type team hate it.

Indra: Right. If its 6pt and ALL CAPS, it is so reader unfriendly; some typography can help direct users to key points, like no logo in bold.

Foundries and resellers. Make it super clear customers are buying a license, not the fonts; that the purchase comes with the terms and fonts. make it easy to see online which licenses are available, and what they cost. Present EULAs or list/summaries early in the checkout process (or even before.) Provide info and search for character set and OT features of fonts.

Here are some questions I think of…

Can we make licensing for different media easier?

Why can licensing a retail font can get more expensive than custom designed type? This makes no sense. How can retailing something that exists be more expensive than creating something new. I know a foundry in 90s that has no enterprise deal, only per seat licensing. How long can this last? Or will all custom fonts be yet another me-too Franklin Gothic?

How can we attractively price retail fonts without participating in a destructive race to the bottom? I don’t want to support a cartel, but is a 95% discount sustainable?

I made some notes for this presentation in this Google Doc:

https://docs.google.com/document/d/1KK7bfib0oyZk0Aebf1k1rL8wNhg6pIIp3bCgXIt36Eg/edit

Roger Black: Many foundries do not want to reprice fonts, its ‘market’ pricing, a single price for all customers. In Latin america, there is discussion that USA prices are out of reach, and they don’t want to pirate fonts, so what do they do? Use only Arial? Do you think there can be different prices for different markets?

Indra: I think for custom shops this is the case, but a global online market, how do you determine where that person is?

Roger: Currency!

Indra: But I used to buy fonts in USD because the Euro was strong, but this was in a way cheating.

Yves: Everyone wants the system to their advantage. Being here at ATypI, 10 years ago, the memebership was out of reach for people from some regions, and we did install a system that looks at how much you make and what percentage of an income can that be. So it makes sense that traffic violations should not have a fixed price but be in proportion to income or net worth because a $100 fine could be nothing to someone with a sportscar.

JB Levee: Pricing for digital goods, if a commodity is cheaper in another country, you always have a friend who can get and bring it to you. Graphic Designers are inherently looking to get cheaper fonts. A designer producing a font family in a region with a low living cost can afford to offer the work for prices lower than me in Paris with higher costs.

Indra: Are you worried about this?

JBL: A friend told me, the best business plan is quality. My fonts are not cheap, somewhat expensive, but I don’t see myself selling them for less. Selling a retail font, it used to be an industrial process, with high levels of skills and expertise. It was expensive, retail or custom. Now the price is not low that retail changed but that production as become cheap and custom type for brands is seen so much so that they order new type over and over. See how fast magazine type comimssions are made. This solves all licensing issues, they deal with a single designer who is the foundry and distributor and all legal things are resolved with a direct phonecall. More and more custom fonts are demanded, so everyone will have their own custom font. Its a snowflake effect, everyone wants to be unique so no-one is.

Indra: I think a custom font should be more expensive, since its more valuable, than a retail font. I think this is a big divide.

?: Its an upsell situation, an ‘inside sale’. A 1:1 relationship between customer and type designer is good for selling more things later through the close agreement. and the TDs want to encourage this, so they can upsell the customer.

Joyce Ketterer at Darden Studio: Custom work has a custom fee for the work, and it moves immediately to the retail models for licensing. Its always governeed by the same EULA. This was a solution to a paperwork nightmare. But we realised that it solves the problem you address too. You can offer discounts on licensing, there is a period of exclusivity. We don’t charge for that because we WANT that to happen; we don’t discount if they forego it; it helps us with marketing if its there. There may be more styles or weights in a retail build later, but the apple to apple comparison, yes, its the custom font that should cost more and that MORE should be the cost of the labour, specifically and clearly, and so you can have someone small who buys a custom font and they pay just the labour, but someone who pays just the labour pays more than something that someone else pays. That’s how we see that. I hope that helps.

Henning Krause at Monotype: There are so many arguments already before I said this. One thing is that customers are in fact, in price building models, the customer isn’t paying for work, but for the use they make from the work. The idea of licensing is built on this, not to let customers to pay on the work, which could be conducted in low cost countries, as then the price is lower. Instead licensing is based on value, not labour. There is a crossing point where clients are urged to go a custom route that is similar to a retail font they could have licensed; I think models should be adjusted to allow for that.

Henning: Also at Monotype, I found a company that found a clever way to register

Bill Davis at Monotype: At ATypI in Rome, before I had grey hair, we had a discussion about EULAs, and simplifying them. The problem is of our own making. How do you allow this community to set their own rules? Some allow modification, including it in a logo, others don’t. At Monotype, we make our own fonts, and also distribute other people’s fonts. The goal is to make it easier for customers. We reorder our EULAs to put first things first, and the challenge is to support differnet media. 10-15 years ago, only desktops. Older even, you had printer licensing. You had ATM licensed for 5 desktops. We live with those legacy business models but we keep up with the times, and if a computer allows you to do something, people do it. Every website, fonts.com of myfonts.com or your own website, its all about communicating to the user what you can and can not do. This is a healthy discussion. I’ll be taking notes to see what Monotype can do to help.

Indra: I don’t want to dictate to someone that they must do this or that. But I want people to get type from different places. Companies don’t even have a location anymore; TypeTogether has no central location any more. I work as a print designer and they say “I licensed this font and I will use it for 20 years” and they use Bembo licensed from 1995. Then type designers don’t make any money; novels are printed 1,000,000s of copies and they are typeset by a single seat license. I would love to see way - not like web fonts - that if you have a large print run or a large income, you pay a larger fee. How many freelancers and how many employees do you have?

Jan Middendorp at MyFonts: Its a complex ideal. You want to factor all things into the price? Its impossible. But there is a race to the bottom, and there is a race that already started. The race to the bottom of what a font costs to produce ended when diskettes were abandoned and everything was digital. T Shirts can get cheap, 1 BRL, and they cost per shirt. Maybe in bad conditions. But fonts, as soon as they were downloaded, and the marginal cost became zero, then whatever you want people to pay is always the revenue per copy. And how much time does it cost to make a font weight in 1992? You had to produce it. Today? You use Superpolator, you make good masters, you push a button, and its all automated. So as it gets cheaper to make them, you must find a balance.

Indra: the word of the week! Sliders

Jan: “There are so many fonts! Why not just choose the best”. Price is one thing, but if you have fans, they can raise the attention on a font. Price is a clear quality signal. The race to the bottom now with plans, the fonts are a byproduct of the thing you buy. Or they are free, or you get Google Fonts that cost nothing. So to grow as MyFonts has done, or other things; pricing is a key thing, you compete against zero. Its not an easy thing to do. I want people to get a good sum. But the people who make the most money on MyFonts are the ones who sell cheapeest. If you sell $49/weight for a sans serif, and week 2 will be selling 2 copies, and it goes down from there. What can the alternative be?

Indra: Its all because of the chart system. If the advantage of the chart and marketing and exposure from high sales numbers is not there, so there is less focus on the release and the first sales numbers. In the old days people said it takes time, no sales in the first 2 years is fine, and it was 10 years to peak. But now there is a focus on MyFonts on fresh fonts.

Albert-Jan Pool: Apple never did this with their iPhones.

David Lemon: I want to say that our bestselling fonts are fonts we released more than 20 years ago. That is true at Linotype. One factor in price pressure is that we have a blessing of sorts that there are 10x more skilled type designers, than even 10 years ago. This adds competition to the market. The only solution is to grow the market. I can’t talk about plans in that area, but we at Adobe are definitely hoping to do so - for what that’s worth.

Jan Middendorp: Back to the question of retail fonts being more expensive than custom type! Monotype is better business guys than we are. They make better prices than we do. The solution is not seeing oursevls as being paid for the work alone; that is not what they do, the work was done years ago. We must be more self concious about creating Intellectural Property, that is worth selling by licensing, not just being aid for the work. Obviously there are companies like Montype who make a lot of money for OEM customers, and on the other hand, there is Google giving away fonts for free. When you are a small foundry, you can never compete agaainst zero. You must be like JB LEveee, saying this is the price, and we must be more careful of these things. Competing to zero is an illustion.

Roger Black: There is resentment that the people at the low end; Google is giving away fonts. It was said that it is not wise to price fonts low to make money. I was fascinatined by JB Levee that each custom font is indistringiushable. … Typekit has a buffet pricing. People say to Font Bureau “Give us a trade gothic! Those guys [ at monotype ] want to charge too much for it” but sometimes they want a flavour that is in fashion. The hems go up, they want a font like this, the hems go down, they want a font like that. Or you get CitiBank, that licenses Interstate, and that is the high end; Tobias designed that 18 years ago, and it was a good deal for us to license that, and the work Pentagram did is great. At the high end of the market, we have deals that are fantastic both custom and retail. I don’t think we can push down any piece, its a big giant market.

Indra: I can see corpoations with expensive fonts that are now going for libre fonts just to not worry about it. There is a way to see some things dying down.

Bill Davies: Its like a balloon, if you squeeze too hard, it pops.

Vik Burian: A receipt for against hte race to the bottom is educating the users; conveying the quality of your work. Don’t go below a certain value. It depreciates the type busienss, and being an asset for people to work with; type isn’t an end product. The period of bestsellers is for us 2-3 years until a type really takes off. Perhaps its a diffreence, if you go for display, another story. But for text families, the real stuff, you are looking at another thing. The MyFonts charts are sans and scripts.

Katy Mawhood at Oxford University Press: I work with 3rd party typesetters, and we need broad terms. We look across print, PDF embedding (which can be complex) many foundries price us now per title, and its tricky! Especially legacy titles, we want to publsih a former print product that we distribute online. For instances where we use fonts not for a broad use for a design, but folios, displays, headers or something small, we make a decisoin to pick another font because the licensing terms for broad distribution is very expensive very quickly, and tracking things is hard too. So we want to continue using fonts from former editions, but we are encouraged by pricing stuctures to consolidate our font licensing.

Indra: A lot of font licensing for eBooks is off. If you don’t pay per edition in print but in eBooks it is confusing. I would like to see that simplified. I work as a print designer and I agree with that point.

Jan Middendorp: The charts are important. MyFonts, today, has emphasis on newness. It is the same as pop music; there are new songs in the hit parade. They all sell well, $10,000s a month for each of those families. THe ones undernearth, they do well too. If you are a fan of quality music, experimental music, you don’t grab the hit parade. You know where to find your things. You look in the text section of the newsletter recent issues. If you want hits, you go to the hit list. Bargains, the offers list. But these things do not add up.

Indra: You see things in charts are selling well, and its 1 click from the homepage. If you get on any list, your type will sell well. I see your music comparioson, but prices for music are not as far from fonts. I don’t have any music subscriptions, though.

?: You can buy flatware from Ikea but don’t buy a couch there. You can shop at Target, anywhere, you get what you pay for, and and shut up.

Felipe Sanches: I’d like to highlight something that is underestimate in this panel. I observe that libre licensed fonts distirubted are mostly referred to by the fact ehy are free gratis, but its underestimated to take the opportuntiy to playing in that market, an open market for anyone toplay, anyone from a beginner to a more experienced can play in that market; any libre font is a potential for being hired to work on the glyphs, there is a big market for that. It can be compared to software, its not a communist thing, it has a lot of revenue potential; red hat makes a lot of money from libre software.

Indra: Custom expansion is the traditoinal type amrekt, and if you see libre fonts, targetting new user groups. I hear htis argument all the time, I see it is possible, but I don’t get it that you want to promote open source stuff to get the other clients or income sources.

Bruno Maag: A total falalcy. Something Dave Crossland may verify; its all Latin fonts. Who will order something from that? Open source for fonts is a dead end for font developers. There is no sustainable businss model for that. Have you ever worked with the open source community? The hours and hours you spend with peeople saying that terminal should be different. No. Open Source for fonts is a dead end. Forget it.

JB Levee: I stay vague in general. The whole disussion here is type business. I am new in the busienss, just 10 years. Type business is 2 words. Its easy enough. Many are learning to design type - I teach - but everyone lives in a romantic world where the business doesn’t exist. Its like working in any bsuiness. My advice is to buy books about business. It is boring! But it is useful. Buy and read books about marketing. Its technical, lots of diagrams. It doesn’t make you a lesser artist to make a living out of it. You work with techniques, clients, constraints, prototypes. I feel uncomfortable hearing of basic economics in tehis room. Decrease price, increase volume in larger proportion than you think, so think twice about decreasing the price. Just like the legal side of type, the business is not so hard. Many art schools have business and legal people visiting. Be careful about what you step into.

Indra: I hope at ATypI we can continue a business forum more. I know tools and tech is important, but this side of the field should have a fixed form at the conferences. Thank you for attending and I hope to conitnue the discussion. So lets keep the dialog coming.


We have an exhibition from Reading, Letterform Archive, FAAP, Argetina, and others. Please go upstairs to see the exhibition.

Now, we had an incident today, something was stolen, and for your own security remember not to leave things around, and tomorrow.


Catherine Dixon

Marina: She is a writer, designer, researcher, and lived in Brasil for a while. I think in some ways she is more Brasilian than I am :)

Catherine Dixon: [ In Portuguese ] I am back in my favorite city, and so happy to be here. Thank you to ATypI, Marina, Claudio, Diego, Crys, and Henrique, and all those who dreamed of ATypI happening here. But I’m sorry I can’t speak the whole thing in Portuguese.

I’m sorry that I brought the UK weather with me, its a bit cold in here. There are challenges. I thought of what to say, where to begin. This is a talk with musings and thoughts of mine in recent months. Things that unsettled me. Its still wriggling a little in fact ;)

What do I do? I am a writer, a teacher, and a designer. This is a book cover I designed for Penguin 10 years ago. I was a visiting professor at the Universe of Sao Paulo 2010-2011.

“Mas Tudo Bem”

That is my talk’s title, and talking about challenges can be gloomy. I look for positives in the mix. You can say “the world will end” and brasilians will say “mad todo bem” :) A great expression. But I thought to check for other meanings on Google search, and Nirvana’s Lithium came up “but thats ok” - so while I’m nervous I’m not going to break.

Here are the themes I wish to address:

Workmanship

Articulation

Value

Relevaance

History

Community

Preciousness

I do not have all the answers, just questions mostly.

There are bullet points, but note, “How powerpoint is killing critical thoughts” in Guardian

Workmanship: In 1968, a small masterpiece, by David Pye, about the nature and art of workmanship. Of course gender bias in the name, workpersonship is tricky, but I want to say that in seeking for a term and finding this, I don’t want to give it up.

Workmanship is about execution; production; about HOW things are done. The role of DRAWING in type design, and typographic specification and production. Why not craftsmanship?

Craft is so problematic, loaded, knitting and strange things. Pye said “Craft is a word to start an arguent with.”

There is the workmanship of risk, you don’t know how it will end up when you start; it could go wrong or great at ay movement. always at risk during the process of making.

and there is the workmanship of certainty, that is always found in mass production.

the distinction: is the result predetermined … ?

next

articulation

the difficulty of communicating what workmanship is in typography.

This year penguin, by British publishers, made a series of books, for 80p, in their 80th year (founded in 1935) They are simple, pared down typography. No balance, feel clumsy, heavy handed. Here are the posters; this is not an ellipsis, but 3 full stops.

A simple, effortless typography is not effortless to create. Here is David Pearson’s Popular Penguin Classics. This is also budget work, £2 each, mass produced. They are light, have sensitivity. So you know it when you have got it, but how do you articulate it?

next

Value

Not that I advocate classicism, or to poke fun, ånd I don’t mean monetary value, but our value.

How do you value something you can’t see?

Richness

“Unless workmanship comes to be understood [ the value will be lost ]”

There is little understanding of where things have come from. This relates to the prior discussion about the type business.

Laura Potter medium.com/hand-brain/laura-potter-…. says that the global digital market makes the

Pragmatics

toshi tweeted this this morning, about the tokyo olympics logo, that there is a complete undervaluing of what is happening to design; how long and how things should be done. "The next Tokyo Olympic logo selection will be another spec work. Worse, it only gives you about two months, and requires a new typeface too." https://twitter.com/Tosche_E/status/654241287135227904

A quote about ‘machine creativity’ from a Dublin based software company

Fiona Banner, makes up typefaces, including Avant Garde and COutier. Its in creative review, talking of it matter of factly. “While a little ugly, type purists may baulk at it, it is very readable.”

I was angry that we have explored this territory, with expertise, and we are reduced to fussy twits. It made no reference to contextualizing the work, nothing about that; we are just fussy.

I thought of the antimacassar of design - the dolls put on chairs by grandma to spot them being worn - a ghetto of of OCD purists.

Wider problems with critique…

The value of the expert critic, when everyone is a critic.

So that is some of the things that unsettled me a little bit.

Paradigm Shifts / Things and the times are a changing

I have real arguments with senior art directors in branding, and they talk about this businesses when we get together, and they say ‘the person they appeal to on the streets don’t care about typographic finesse, so why should we? it is the idea that matters’

But I can see campaigns in London where the letters are drawn incorrectly.

There is a shift towards Generalism in state education. Here is a quote from Stuart Baily, “towards a critical facility” in 2006/2007 at Parsons School of Design, when they reworked the curriculum. He has since given up thispotioin.

“A common intuition that the existing model is no longer reasonably accommodating the requirements of today with regard to the ever blurring boundaries”

and the wider ‘dematerialization’ of design. “Findeli argues for an attitude which rejects … the fetisihism of the artifact”

A shift to mediocrity.

“This is not a book” by TomAbba, talking about “the print people who is based on hard earned expertise, and online is consistantly mediocre and becomes acceptable over time.”

So what do I say to these 4 things?

The workship of idea over execution? I am really bored of these lower common denominator arguments. There are many things we can Get Away with, but should we?

Making burgers from horse meat, was a big scandal in the UK recently. People didn’t know so the argument was that they were happy in ignorance. The public rejected this and say no things must be done properly.

David Pye says in 68, “Design is so difficult to learn now simply because the area is in a state of flux”

The dematerialist emphasis on the ‘human context’

Here is twitter, Petr van Blokland tweeted 2 days ago, ‘Where does the thinking of designers come from, that design can be made entirely by thinking in their heads? And then solve world problems?’ https://twitter.com/petrvanblokland/status/653465865845870592

This is a UK shed in the 1950s. We have them for houses with little gardens; store furnative, tools, people like to spend time in them. a little retreat. as a culture they are a big thing. This is a catalog of them. You have the measurements so you can see how they will fit in your garden.

Joseph Roth, The White Cities, reports from Farnce, 1925-1939: “Marseilles is w world where adventure is commonplace and the commonplace is adventureous”.

We should make design exciting again.

back to value. shed catalogs are not expected to be exciting. no big design prizes for that. So there is another problem of invisibility; there are opportunities for working with real human contexts that are missed.

so there are lots of ‘most beautiful dutch X’ or ‘most beautiful german Y’ but I would like to see a book of the ‘most beautiful gas bills’ - things that help people, especially with declining literacy levels especially in the UK.

“Its not difficult to be personally creative whilst culturally inconsequential.” said Norman Potter from his matchbox maxims in “What is a designer?”

I went to a London event about the Murthy Press new initiative by an Indian businessman in NYC to make their canon of scholarship available. My colleague in Chennai studied Bronte and Jane Austen in school, not the literature of her own nation.

The formats were sometimes new, as from an oral tradition. Tiro Typeworks developed scripts (John, Fiona, Titus) and scholars speaking about what this meant, the start of a 99 year project.

This is typography and design at its very best.

There is always mediocrity. Should we limit our ambition? Pye “Why do we accept this as inevitable? We made it so and we can unmake it”

Gibson: “The future will judge us for failing to address the systematic problems that we will be known for being totally aware of.”

What are we doing? Nostalgia.

You might know a UK show called Midsommer Murder. This is my example of a big challenge for typography; the attitude of history.

Here this guy is killed with a printing press! Somehow this printing press, the body has gone into the press, and the pressure was enough to kill him. For this to work, for the body to be there, so the print is on his face, it would be lowered ;) [ big audience laugh about this ]

Rob Green has recently told the story of the ‘lost’ Doves’ Type, where the owners of a private press threw the type off a bridge into the themes, and Rob Green didn’t accept the common sense that it was washed away, he read EVERYTHING before that, he worked out where on the bridge the person would be stood, the tides, when the lowest tide would be, got a diving crew, and found it!

Its not playing fast and loose with history, and also not accepting

I have a pet-hate of nostalgia, especially for letterpress.

Milo CEASA at Projecto Acaia, here in Brasil, that helps poorer kids express themselves.

And L’Automatica, in Barcelona, thinking about the power of a printing press.

Nostoalgia isn’t always a dead end. Rick Poynor quote.

Yep, ok, tradition has value.

Gibson: The fetishisation of craftsmanship for producing status object [ becomes the opposite of what was intended. ]”

Archives.

Archives have value of the history they WILL tell us; they are the resource for examining the inadequate histories we do have. They reconnect us, show us the humility of seeing what has been done before. What they have potential to tell us, is why they are valuable.

I have had some connection to the St Bride printing library, that is in real trouble right now. There is not much news on this, and I want to use this platform to say ‘bear with us.’ Those who gave up on it, hold on! :)

Voices.

Voices, value and lore. The ATYPI list recently had some really great stuff in the old stories; its not what academic researchers are interested in, its the split that Mariane talked about this morning, that lore tells us about the past’s communities.

value vs preciousness.

A barrier to engagement of others.

Some disconnects: font functionality to the design community; there are UI problems, there are EULAs stimulus; termininology; “pro” etc.

I am well known as a typography person, so I often get calls from people with type licensing issues, who have 15 minutes before a presentation to recommend a type and the licensing comes in and

I saw type designer presentation where they said “to stop the silly designers getting it wrong, we did X” and I think this is tricky; those with design experience, and those without. I work with design professionals, but

This is something Laurence Phinney posted from the UK on Facebook, of opentype going wrong. I realize this is a massive physical sign outside a hotel, that has been made. Rather than laugh at this, think about why this has happened, why this happened, and what we can do to stop it happening.

Investment.

Beatrice Wardley and Stanley Morrison educated people why they would need typefaces, and the newsletters and the endless publications sent to printers, the people they wanted to buy from them, so they could know why and what to buy.

They shifted the relationship, in the past. We have nothing close to it; we imply that the education should happen in full time education, but there is no way to get to cover that in a curriculum today.

My own work? I like classification through description; frameworks for engagement with forms, so people have vocabularies for accessing understanding, to read into forms and understand what is going on - not boxes.

So this sense of preciousness is adding to our irrelevance.

It is time to share in order to be heard.

Diversity, ahem. Retro-sexism. In my blurb I had this term, retro-sexism. People I teach with asked me, what on earth is that?

If you read the online chats over the summer and you think you are in the 1970s! I was alive back then but as a kid. Is it now fashionable to be sexist? Someone in a online forum to say “as a woman it is intimidating to be here” and to be interrogated about this, is hard.

So if there’s one good thing to come out of all of that, it is www.alphabettes.org

Sisterhood

I am a bit gender blind, I’m generally chattering without thinking too much about other things. I was thinking about being a member of an org, but lately I saw some things happening in that org that my younger colleague suffered. maybe me being easy going, was enabling. The older women in the org were more sort of tough, and so I thought that the whole org has a part to play.

Are these gender issues - Or ego based? We should be a reflexive community, looking to learn from each other, not just asserting.

How things are done matters, as well as why they are done. Yes the why matters. Typography is not the be all and end all, but its workmanship is not to be underrated. Its not as easy as it looks - e.g. google logo.

Here is Eric Hobsawn in “Fractured times: culture and society in the 20th century.” rom 2014, p.256-7. [ quote missing ]

We need to stop this departure to planet typography. Being our own world.

if we want typographyic improvement, we have to equip a wider circle of people and create spaces for sharing more effectively.

Here is a long quote by Bruno Latour in 2003 (?) on ‘why has critique run out of steam? from matter of fact to matters of concern.”

Conference is not spectacle, most of the time, it is about a community conferring together, celebrating achievements, and beyond that reflecting on who else might become part of the community.

One last word of gringa advice: “The ugliest, most dangerous city you’ll ever love” said the NY Times.

I wish you a great confernce and to go and fall in love with this city!


Bestsellers

Our mission is to greet friends and joke around; we had a studio in the school in Prague, established 130 years ago, one of the oldest in the city. Its in the center, after WW2 it gained University status.

Graphic Design was a key course at that time. Today the Graphic Dept has 5 studios - Graphics, Illustration, New Media, TV Graphics, and Type Design. WE are the final studio.

The school is old, and we have 130th anniversary coming up.

Frazisck Storm was key in moving forward Czech type design. Here are the typefaces from our studio.

Best Sellers. Our students looked at the best sellers and set out to answer the question, “Is it possible to design a bestseller?”

There are all new editors, so many new typefaces, so you need aggressive pricing politically, a screenshot of $89 to $8.90. We have dealt with type experiments at the school for a long time.

Nice quote from Storm on how if he paid attention to best selling, he would never publish his worst selling font.

In 2014, we said to the students: Try to determine from fonts released in the last 203 years, the rents and modern approaches that [ led to best sellers ]”

Students had to organize work groups, collect sales and popularity chart data, process the results, and define what a best seller is. We also worked with the distributors to get private information for the last few years.

We round that most likely successful fonts ahem these attributes: Sans serif, a family of 18 styles, a lot price of $19 per style, and bonuses like alternatives and fun styles. And to start off with a 90% discount.

The students found the winner of all categories is the grotesque. Then brush scripts. Clan sans, Centrale Sans, Nexa, Pluto Sans, and Brandon Text. Wishes Script, Cantoni, Hipster Script, Alana. The 3rd category was the old style - TREND, FRONTAGE, LiebeErika, Sketchetik, and Pady Rene. A lot of slab serifs were designed but they were rarely at the top of the sales chart.

Next step: with the knowledge in mind, design your own typeface with a full character set and high quality diacritics, utilizing local features. e.g., hipster script is far away from slavic folklore. This is the usual work of type design. Image of Fontal has quit unexpectldly.

We needed a good partner to bring the fonts to market. We thought they were a great collaborator and experimental environment. Their team had suggestions no how to improve things. image of email from Florian Hardwig.

So we nmade 12 typefaces, 69 styles, X glyphs, all done in 84 days.

We did 4 types for each of the 3 top categories.

Abnormal. Jan Buble. 4 styles: a sans with reverse contrast, unconventional means of catching the eye. It has monoline thin to a reverse contrast extra bold.

Bali Brush, Andrea Vacovska, 1 style; a bush script with a lot of image glyphs.

Clown, Tereza Smidova, has 8 styles that fit together into 150+ unique type effects.

Joe182, Jan Horcik, is based on the big flat brush sign painter posters seen in NYC in the mid 20th Century.

Moniak Sans, Jan Buchtela, 6 styles. This seems like a trend, there are many new families that do not have italics, or italics come later. This family doesn’t have italics, Thin to ExtraBold.

Monolina, Petra Docekalova, 3 styles, a single stroke with a round nib pen, ait has uppercase, lowercase, swash caps, and alt lowercases.

Rukola, Nikola Giacintova, 1 style, it has small caps and is an expressive swash script.

Sandokan, Matyas Machat, 1 style [ it has a ‘Monster’ vibe ] unusual connectors for lowercase,

Senohraby, Jakub Spurny, 4 styles.

Slurm, Nikola Klimova, 2 styles, regular and bold, each is a handwriting style, condensed, smooth.

Vegan, Vojtech Riha, 12 styles, based on a 60s sans from Czech [ looks like Roboto ]

Zirkel, Ondrej Kahanek, 16 styles, a geometric sans.

Step 3: Presentation. Descriptor the type with 1,000 chars, make a printed pack and specimens. For the summer exhibition, all the fonts were made available on MyFonts.com and this was also the start of our research based on our predictions.

Each font had a specimen printed, as each new typeface ideally has a printed specimen. We looked at the issue from the other side, the perspective of design studios who buy fonts, and the type designers who must somewhat respect clients and market demands.

We made a questionnaire for type designers, and

Cornel Windily, lineto.com, says that “the problem may be that all these small foundries feel powerless, the concentraion of the market in 1 company. Also I think that the concept of Google Fonts is not beneficial to type designers.”

So, the final step, create a my fonts account and offer the font for sale.

Her are the first 50 days. Rookie, 532 sales. Clown, 231. Then the rest of the year, the sales ratio was much the same. The total after 1 year, 1,500 total. The discounts were for 30 days to start. This is the effect he discount had a sale for the fist 50 days.

The sales numbers show script fonts were 56%, train and handmade were X%.

So guidelines? Make a script, 1 style, medium price of $40, with bonuses. And start a discount of 50%.

Our results differ from the initial research. Our observations was ttotally different. The task was successful, they tried free market practices within an academic environment.

They continue in individual directions and we are very happy. This project involved many people and we would like to help Jan Middendorp, Dan Reynolds, and Filip Blazek, and all the students who dropped a lot of private life and free time.

And thank you for attending.

http://bestsellers.umprum.cz

Questions?

Q: What software did you use? Plugins?

A: Glyphs and FOntlab, 50/50. Clown was in Glyphs. No specific plugins.

Jan Middendorp: I was the first person to approve this. It was radical! We don’t expect it to be a habit. The berlin team at my fonts looked at this and said they are good designers, that are reaching, and its an interesting research of my fonts, its very short, we are all part timers at my fonts, and so it was interesting to see. We thought Rukola would be a best seller. The sans serifs, the research was good, but you should have tried an 80% discount that would have changed the performance. I don’t like this happening but it is happening. If you are not well known, you must get attention, and use a big discount to do that. We were missing how a new type in a well populated genre was presented at a crazy discount. That’s because you were careful, which is also good. Thank you for the project, we enjoyed it!

Sonia Knecht: Thank you for this, a wonderful project! Great contribution to the business discussions at this event. Can you discuss the topic of names. How did you pick the names> Rukola is an international name.

A: Its original name was “Titi Daisy” and she chose a name already used, and ended up with this. The name is a big deal for any designer, and they did considered this in their business plan work. It is so difficult, and some students did pick unusual names.

SK: Did you recommend to change the names?

A: It depends. they own the projects, they decided their own way. We try to help them of course but the name was their decision.


Arabic Type

Christian Sarkis and Lara Captan

CS: WE collaborate and are both from Lebanon. We weill share with you some of our conversations to make the wider discourse stronger.

LC: ARabic region is in the middle east, north africa, to north indian subcontinent. As the script spread from east to west africa, there was a lot of invention. Here is the 8th century Kufi, square and geometric, and you have diverse integrations of the same shape. Here is Morocco script, playful and rounded.

The system remains today, and Arabs have mastered page layout, headine scripts and text scripts. Further west was the Nastaliq script, in Persia and in Pakistan.

CS: Much is said about the script and arabic expression. But this is not our focus; we look at the gems in the scripts for innovative typographic conventions. you have this, handmade, and this, loq quality print.

In latin, there was a long process of evolution. In arbic, there were disjointed jumps, and lot lost much of the core characterized. What i try to do is star with current arbaic type, and bring back calligraphic features. What lara does is start with the written script and move forwards to digital type.

Challenges! Firs the cultural context. There is a clear direction for westernization. Everything seen in the west is validated, It has an impact on letterforms too. There are trends, like the Frankenstein trend, chopping up latin letters to make arabic ones.

The last trend, most prominent, with strength, is what people are dubbing Kufi. This was never something I understood, what they referred to, but his. There are no clear references to the existing Kufi. I polled on Facebook for designers and typographers, what do you mean by Kufi for contemporary types? And I got an inner answer, its based on the the square Kufi some way.

LC: I tried to verify this. Here is a 17th century mosque room detail, using bricks, glazed and non glazed, with square forms; you have a square grid, you have 7 lines and you make something that isn’t totally readable but is a nice pattern. My xperiment was to reduce the line width, to round the corners, and to compare this to typefaces recently published.

KS: We think experiments should also happen on the other styles, the cursive styles. They may be text or display but they are all meant to be read. We narrowed it down to a few parameters; that we can make digital type with. All these scripts are made to be read, not patterns.

Parametrs: cascading, slant, proptions, contrast, elongation, and alternation. lets walk through these and look at the challenges of each.

FYI, arabic has attached and detached letters in 1 word.

LC: There is slant in the baseline that we would like to see reintroduced. letters slip into each other. Also the link from the top, cascading. This is something that exists from the birth of the script. Everything has thing, and we can avoid it in type, sure, but we ought to have a choice for that.

Bala is a text type I am working on for that, and I constrain myself with the characteristics of naskh but to do some things differently. I do this with Decotype’s ACE.

To do Bala I played with the angle. I lookout at 5’, 0’ and 1’. Once there is angle, I must make sure the horizontal connections are correct at that angle, and also at the different levels.

Cascade letters can continue, I use non words to demonstrate this. I make my initial letters as stable as possible, since the middle letters always change. the top letters can connect, too. I felt this was becoming out of control. So i made the exit point of all letters that can be final the same.

Cascading makes white space underneath, but with ACE you can place letters under each other. I wondered if cascading is really working for text, and I designed options for users to use full cascade to more horizontal versions. Instead of using full horitzonal, I wanted to balance tradition and modernity. See here, I use option 2 and 1; I dropped option 3.

KS: at type and media i wanted to use a diwani style, and bring back the levels of that. I looked at their grouping and quickly this failed. the words looked deformed, went way above stem height. it was meant for display, with a bold baseline. far form calligraphy; i wanted a modern type, not a traditional one. After a while, looking at the script, I had a epiphany, and the calligraphic slant of baseline is a human thing, for human hands doing writing, so i dropped it. then after a regular flat baseline was made, i then made a slanted baseline that was typographic.

It was still the first type that i know that works like this. The stacking, came after. I built in my mind the idea of translating calligraphic features and I tried to do a Naskh with stacking. This is similar to the conclusions that Lara got to; and I saw that some stacking was too wide. Here’s 2 options.

For me, these are stylistic optimizations for display usage. This is where I differ from lara.

second, the script grammar. what is this? The rules that are functional or visual, for arabic script, to avoid confusion. we have a letterform known as ‘tooth’ which often are repeated. some types make these the same height and width. in hand writing, they always have different forms.

So in my first naskh exploration, i made this variation. The seem, looks like 3 teeth next to each other, so we needed these forms to avoid confusion.

It took me time to find what common ligatures were needed. Here are 4 typefaces, from calligraphic, to super modern compressed hairline, using the same ligatures.

LK: I also use script grammar. I want to now the grammar. in future type, if i have the full nkolwedge of the grammaar, i can use it it my design process. here you see some teeth, from 1 to 5. You can have 100 teeth and this algorithm will work for all.

When designing arbiac, with no references about where to begin, i decided to start my own analysis of the script, with help from ecotype, grouping into 3 sets. the red and black letters here show the variations; connections can be across, up or down in direction.

parameter 3 is the proportions. you have a rule in calligraphy for the dot of the pen giving a unit system.

KS: as lara said, we start from the script basis, but for me i want to make modern type, not going back to old classics. Here is an experiment to make counters much larger, even to break the stokes. Also to make the height of the counter larger. We do not know the effect of this on the script.

I did a height system, compressed to narrow to regular to extended, and there is no experiments of this that i knew. with a lot of personal research, on this, I realized that the counteroffers of arabic are working so differently to latin; in latin you have a circular form, but arabic counter forms have a trianglular basis, which must not be lost.

parameter 4: contrast. also a dear parameter to me, as i started experimenting with this at type and media, and this is like the gerri noordzji. he told me, in arabic, you have thick strokes going into each other! Wow! that was a revelation for him. in latin you avoid that to avoid clogging.

In arabic, you have pen angles at 115, 75, and 12. You have a lot of rotation too.

LK: In calligraphy, you take the pen, the lines are always being rotated, all the time. if you are not a skilled calligrapher, you can pick up, rotate and put down the pen to try this. The flat areas are translation of the pen edge. the dashed line here is drawn in parts. There is also pen shift, where the contrast is at the smallest point.

Here is a comparison of DecoType Naskh with the stages of the development of Bala. Krystian told me ‘arabic has triangles’ and this was a great idea that set a direction for the design, I took it more high contrast and then brought it back a little.

You must observe the script thoroughly to make it right.

I found at a point in the process that I would make it less curvy, having more flare, and these details are where we think people do not experiment enough. the sky it the limit to explore where the pen changes direction.

KS: I experiment with contrast, and when i worked on 2 typefaces, if you may, its like sans/serif in latin. you adjust the angle from 60’ to 30’ and this changes the letters radically. We learn latin contrast conventions, but in arabic, where to place it, and how fast it fluctuates, is not as well known. Also, the relation of the inner and outer strokes. The high contrast ones have the pen as influential, but in lower contrast i am more free from the pen. another experiment, there were 4 widths and 10 weights for each. What happens to letters when you make them thicker? in arabic, the whole latter grows taller as the stroke becomes thicker; unlike in latin. I tried, when i go darker, i bring back more pen like contrast, compared to the light styles.

Stacking them up, you can see the movement of the pen becoming more pronounced.

parameter 5: elongation and text.

I was working on the Louvre Abu Dhabi, and the arabic is always shorter text when Enlish and French on the signage. Elongation was key for this! Its a basic opentype feature, and i wondered why it wasn’t happening more and more. you select some letters an you can expand them like this. I do not allow any extension, but most are supported. Here is an example of the madda feature.

parameter 6. altenrates and swashes.

LK: The arabic script allows for variety and alternation. the typographer has the joy of picking! I made alternates in Bala. The meem can be open and cascading, closed and cascading, or closed and flat. With ACE this can be for all unicode.

KS: The last 2 parameters are important but less essential than the first 4. it is a shame to lose those things. so a word on technology, we have so many possibilities with ACE. five years ago, there are more software. we need a to of coders to make tools for arabic. This is an example, made by Erik Vank Blokland, to make at tool for kerning and mark positioning, privately.

LK: I would love to see ACE and openType working, or ACE into OpenType. I would love to see the shapes you want to make and answer the things raised in the panel yesterday. I really like the minimal amount of glyphs; not more than 500 glyphs! and it offers all that you saw today.

KS: I am fine with OpenType, but we need more experiments to push it to the limits. All my work is OpenType.

LK: It will be nice to see a conversation with all type designers to begin making a base for arabic, to trampoline on arabic, to make it better, more diverse, and i would love to see that happen.

KS: I think the effort to experiment is more important than the results.

LK: I want to think the Creative Industryies Fund NL, to make the Bala family, and Mirjam Somers, my consultant for this project.

KS: I have a special announcement. The last 5 years I worked with Peter Bilak on arabic types, and a few hours ago we launched the website for Typoteque Arabic. This foundry is intended to make new type for the arabic script, to start a platform for arabic designers.


Kevin Larsen

[ 10 mins late ]

1 black and white pair is 1 cycle.

Here is a 250 dpi monitor graphed by low to high contrast on the y axis and spatial frequency from low to high on the x axis.

Here is 300 DPI. We care about the gaps between 1 and 2 pixels, its important for legibility of CJK, but applies to Latin as well.

Here is lowercase /i, a single line and dot. At 96 DPI, we can all agree this is an i. But maybe not the Candara i. It looks like all other i at that DPI. At higher and higher rez, we see more of the curves. At 192 DPI you see a 2 pixel stem width, and at 300 dpi, a 3 pixel stem width. From 400 to 500 spin we see the shading along the edge to make a subtle curve in a more refined way.

Questions?

Pedro Amado: Mobile displays? You try to render 34 cycles per degree (?) but couldn’t a lighter weight help?

A: Would you be happy in your latin fonts if you were told you could only use 1 weight, ever? :) Hinting is a way to push lines around to make things work with what we have. The question is, was that beyond human perception? And the answers no.

Okay lets make this perfectly dark as we can. Now your eyes adept o the black.

Ideally we wait 5 mins, but we will jumped ahead on this. If we waited then this circle would appear very white to you. It might seem little grey now.

I will not change the try circle at a ll, but make a white circle around it. I just add white, but your perception changes.

Now, I put an even whiter circle and the colors are not changing at all. I just add another circle. So soothing you saw as white, you now see as grey.

And now I put ANOTHER circle.

And another one! You guys are experts, you should recognize white right! A

And another. That is six rings, now seven rings. So my point si that contrast is not an absolute in the world, its all relative. We look at he blackest thing in the encironment and the most white. It is variale=on your eyesight and the environment.

So, another question, this is the chart I showed earlier, of low to high spatial frequencies and contrast. The lines do go all the way to the top, your perception ill NOT go all the way to the top. We will see contrast at some spatial frequency than others.

The front of the room says the lines are highest more to the right, and the back of the room says more to the left. We see things 3 to 4 cycle per degree, and as we go higher contrast we need to change.

Here is an “E” in the image. …

If you have a high quality monitor and a screen without reflections, that will give you better contrast. the phone screen in sunlight, the white on the screen needs to be as white as he sunlight, and your screen can not do that today. and its age related, as we get older, less light gets through our eyes. At 20, full, at 40 about 50% and at 60 years old about 40% of the light gets through the eyes.

So high spatial frequecny and low contrast can go beyond our ability to perceive it.

Here is a fun one, this is the same image, with einstein in high sharp lines, and the blurry areas are low spatial frequency so we don’t pay attention to them, but smaller, we see only those that make out the image of marlin monroe.

Here is letters in the same way, F is seen when the image is big and D is seen when small and the E in the middle.

Take aways!

You all were right and jobs was wrong :) We need CONTINUOUS options from 0-50 cycle per degree for perfect resolution.

Contrast is an important as resolution, and for designers here, if you want something to be high contrast to see it, we want larger sizes and more contrast so people can see it - a high quality computer in a dark room will give you high contrast but a mobile screen outside will be lower contrast.

High contrast is critical to good legibility.

Thomas Phinney: I keep on seeing advice in places that black no white is bad, and not as legible. Are there situations or subgroups like dyslexics where max contrast is bad advice, or is it a urban myth?

A: Cataracts have distortions and white on black can help you there. Its no lowering contrast. You get discomfort problems if the light is more than 3x the brightness of the environments; here in this dark room your laptop set to max will be 20x and that can give you discomfort. You always want high contrast as i define it.

TP: So designers should use high contrast and the user shouldd turn down their screen if needed?

A: Yes.

Q: You showed us a slide with various /i renderings. The form of the pixel is square. The light beam is round.

A: On an LCD screen, a white pixel is 3 rectangular stripes of RGB, but in CRT they were round. Trixels are other shapes; in VGA screens in the 80s they were also not square. The last 20 years, they have mostly been square.


Ann Bessemans

Here I will define reading comfort, how to measure it, and arthymic types seem better for some people.

Last year at ATYPI I presented my research into rhythm. it is considered and repetitive, a game of black and white. I look at latin, the vertical strokes like in the word minimum.

Vertical letter strokes are alternated with the white shapes inside and outside the letters, making a strip pattern. For a white paragraph of text, I made a curved line to how the frequency of rhythm. And also a font with the top and bottom removed to leave only the stems.

Reading comfort is not well defined. Its an ambiguous term, to which all different type people will give different readers, with a common sense core, that reading should be comfortable.

Unger says it is making reading pleasant, and never a tiring activity. Squinting can play an important role in fatigue. Blink rate is thus linked to visual fatigue, disocvered by Luckiesh in the late 1930s. He put all causes of strain to 1 thing, though; not just pupil dilation but also mental activity. Eye trackers can not measure reading discomfort. Frowning can also reveal discomfort, and a web cam can record this.

EMG of the under eye cheek muscle and eye brow, is accurate, but in contradiction with comfort.

MS Reading Tech group is seeking a method to measure comfort without effecting reading performance, nothing found that really works yet.

Walter Tracey in Letterrs of Credit says reading comfort and visual comfort can be easily connected, but not the same.

Understanding visual comfort can come first, though. There is strong evidence that there is the rhythm of a typeface and in past year’s presenation i showed that slanting letters will reduce the verticality and making reading more comfortable.

Here is Meta Serif from 2007. The 21st century model uses equal letter widths, but this was not true before 1700. Then, the letters were very different. This means the rhythm is more irregular than that of type made popular today. Type with vertical contrast also became popular. Here is a Fleishman.

To the end of the period, the 1800 - 1900 period, the letters became more condensed; the vertical contrast meant that reading comfort was impacted.

after 1900, reacting to the complaints of the 19th C, letters got a slant, but the equal letter widths was maintained.

In visual perception studies, you view some patterns,

spatial frequecy, width of stripe and spacing, size pattern, and contrast. visual discomfort is from unpleasant artifacts from grid or patterned lines that are illusions. this can happen with the patterns of marks found in text for reading.

Young readers benefit from a font that is not set too small.

Here is a para where the letters are printed twice, just a few mm apart on the x axis.

WHat bad effects can be? Illusions, eye strain, headaches, and even epilepsic seizures. Risk groups can be those who suffer from migraine, eye strain, visual stressed, and beginner and poor readers.

Moving the head towards the page can indicate discomfort, since it will reduce the lines seen in your core vision. More line height can also help.

I like quizzing readers, a questionnaire about events that indicate visual discomfort. all the questions were answered on a 3 point, never sometimes always.

Take aways.

I made a definition: Reading Comfort is the physiologically related and subjective information …

2nd, …

3rd, I would like to emphasize that rhythm should be more considered. people with good vision are fine with a more regular rhythm type design. but several groups at risk of visual discomfort will benefit form type with more arthymic design.

I hope to relate legibility, comfort, spatial freeuency usage, and rhythm (stripe patterns.)

I hope this will help type designers to improve their text types.

To be continued! Thank you.

Pedro Amado: Legibility has so many parameters. can a mask effect, dividing the information in to small chunks, can be related?

A: Yes, what Wilkins did is to study some typefaces. he looked at a condensed type and saw that as the type become more condensed, people move their head towards the pager. as kevin say, the one at the back see more lines than the ones at the front. so that makes reading more comfortable.


Homages

Petra Weiss: Bernado, a brasilia typographer, passed; he was a member of the Just In Type team. ????, the Argentine type designer, of Belen, who worked with Dario, also passed.

Erich Alb (typographer, publisher, Switzerland): There are 1,000s of images of Frutiger’s work. He passed this year, best known for Univers, and his book “signs and symbols, their design and meaning.” Please don’t photograph this deck. Form and counterform, he said that counters make the beauty in a typeface. This Esset was cut with sissors.

After a hard days work on concentrated type design, he drew freeform art, again form and counterform. Here is a wood cut art piece. Here is this ‘life cycle’ triptych. Birth, motherhood and tenderness. He did sculpture work, it was his teenage dream. After visiting Lankuzi (?) and seeing his perfect forms, the dream was gone. Then on, he worked on type design only ;)

Here is type history cut into wood plates, as a final piece at the school of art in zurich where he graduated summa cum lord (?)

Here is the school work he did in Zutich in 1950. A sans serif that looked lie Univers already. In paris 3 years later he started the big project. Here is the lumitype in the 1950s, looking like a control center of a space ship.

Since a teenager he collected signs and symbols, and his book was translated worldwide. I bought a copy for $29, and now on ABook it was $10,400.

OCR-A was not allowed in europe. After 6 years of designing, solving problems, he came up with OCR-B. I said, when a designer strays too far from the base form, you get frustration and illegibility. The figures are higher than cap height, this was a smart problem solution by Adrian. Here is B and 8 overlaid; you see the difference. You see here in my passport that they are higher. OCR-B is used for all passports; its a world standard for bank drafts too.

For the IBM typewriter, adrian designed the univers type. here is the musee paris logo. Here is there arrow for Charles de Gaulle airport in Paris. Based on the square. Here is the Frutiger typeface, made for the CdG signage in 1976.

here is a 200 meter long wall with symbols by him. Here is paris Metro signage. Here is his Univers Devanagari beta. He was helped with Mahendra Patel. Swiss Post used his Frutiger type. Adrian also designed the type for the post busses, in 3 languages.

JB Levee: Jean ? studied calligraphy. 73-85 Jean worked in paris, and from 95 he taught calligraphy outside france, and spencerian style and this won him a lot of fame. he made many workshops and some great t shirts. … he was a maverick, with 2 periods. first the type, and then the calligraphy. he made designs that were very different to what he is famous for. fascinated with the designs of the 70s, he loved optical art. he published a few books on optical patterns.

Gerry: Richard SOuthall, a schoolmate of M Carter, and was from a elevated part of society. No reason for him to go to typography, he was destined for some job with the state we would not hear much of. He took over from M Carter a printer job when photo type was becoming a thing. This is a photo of him at the Automatic Type Design conference in Nancy, his last public appearance. His bio is a who’s who and what’s what of what was most interesting in type design when he was working. In the 80s, drum scanners would be the person who wrote the software that drove them. He was involved with lumitype, aiding the transition to the new domain of computer software based type. He had an uneasyy job description as he explained for other communities how they worked. His book “printers type in the 20th century” is a standard for any type curriculum, and its not too expensive yet ;) He recently wrote a chapter for the History of the Monotype Corporation. He made a paper about the Colorado typeface, made for a US telephone directory. Long before that, encoding letters were discussed a lot. Mandel designd curves, they made pixels, image setter dots really, dots of ink on a specific press, and then made systems to describe those forms. designing what the rendering engine will produce and work back to a digital encoding has a long tradition. they interrogated typeface in a way typeface designers normally do intuitively. here are graphs of the counters changing as the sizes are increasing. Peter Karow did similar things, and they were important and relevant to our practice today. Sadly his work seems t appear in boring publisations, that are not accessible. Not in public libraries, not well indexed on the web. “Designing a new type, the designer is concered first with that appearance of character images, and only secondarily with the shape of the object that gives rise to them.” This quote is hard to find. Richard tried to formalise the system of writing about these things. So his texts are essential to discussion about type as a discipline of knowledge.

Sumner Stone: Herman Zapf needs no introduction. He passed recently. He inclufenced a huge number of people. Including me. You can see what I wrote on my blog about him and my experience with him. I want to say more today. Thanks for Julian Waters for these images, a very big fan of his, who is good about collecting images and taking them when he visited.

I can’t talk about him without his wife Gudrun, who is alive at 98 years old today. She did type design too and Herman would say he married the competition; I know of no other couple where both are notable type designers. perhaps only recently; they were formidable, the two of them. I love this image of them together. Diatom Classic is one of her types. There are a few type designs like his, ornamented type that they designed, one Saffir, and Samarg (?) which bounced ideas off each other. She is an accomplished binder and we put on an even in SF called Zapfest, and here’s the logotype I designed for this event. For both of them.

The way I met him was by a film, now online, and at the start there’s a blackboard which he draws on with chalk. I saw this with my teacher, and i went to kansas city for 2 years to study because of it. He was visiting there, spending time with us idivudually, showing us his special technique and tricks. he did this one morning, for some product. he wished to preserve and enhance the quality of the hand in his lettering and type design. this is 1/4 of the size of the finished poster that was 24 x 36. he picked up an edge pen, it took him 20 minutes, and then he took it downstairs to hallmark, blew it up to this size, and silk screened it. I snatched one and still have it.

he did various work for hallmark, including designing type for them seen in a book by rick kusik (?) “what type design needs today”. here is examples of the work he did for hallmark.

This is his signature! ;) [ an A to Z in swash caps that create a circle ]

Rosenberger was the punch cutter at Stempel he worked with, and here is some of the work that he cut.

He did MANY typefaces, and something missing from my blog is that he said over and over, that we should not just copy the lpast, we should make type for today. that was a theme for what he said all the time, in books and lectures. he sawys this and you look at his work, his work is classical. and yet, you coan not know that when you see his work that he did it.

he had a personality coming through and it was always an incremental modification of classical forms. one of the typefaces that i think when it first appeared was astounding to the type community was optima. this was a step outside the kind of things that herman did outside that time. in the 60s and 70s, people would ask me what i thought of it! he was no t very forthcoming about where it came from, early on. when i worked within in the early 70s, i have the feeling that it was a favorite of mine, and one of his. he used it a lot when he did typography. his book of alphabets is set in it. these are sketches of it.

he went to italy after WW2, and that is where he saw the letters, in florence. in florence, there is an article by paul stiff out these images. Now we see it. Herman drew these letter sketches while he was looking at this, on italian bank notes that were worthless. Here are images of them.

this is a medal he designed with optima like letters, knocked out with a pencil. after his initial fame from palatino and optima and also mealier which is not survived as well but was a hit at the time. after that, he branched out after type design and calligraphy. teaching at RIT was a big ting, he got a special position and influenced many people. fro m around the world, who went to those classes, including julian waters and gerry keller. (?)

He wrote a manual for rooting, a beautiful thing. he worked with URW and ade his first 19th century sans serif.

He did a number of projects with URW, including a way to deal with hyphenation and justification to make optimal results. he saw himself as a frustrated engineer. if he was in a richer family, he would have become a person involved in science and engineering, and he thought digital age was great and popped right into tit.

in the semi retirement period, he did some art pieces, using graphite technique, with wax and scratching.

Here is fontographer, looking at his own designs again, for PS digitization. he worked with another collaborating, akira at linotype, he was in his 80s and going each day tow work. that maybe kept him alive longer, and i concluded that i am not retiring. Palatino sans!

To leave you with a thought: “Certainly the art of writing is the most miraculous of all things man has devised.”


Jan Middendorp

This talk was submitted in spring, when type designers were pissed off with pricing, discounts, and generally that people don’t like things done by monotype. Don’t get me started ;)

So I proposed to talk about the A BC of type, art business and culture.

The letters a white white and the background is grey ;)

I have 3 hats, author editor and translator of serious books about type design and graphic design

I am a professional type writer

And i worked as the chief consultant and editor of the my fonts newsletters, and also the last 3 years as the principle gatekeeper for the bestsellers list

ANd I organized a poster competition in Dubai

I am a user, a critic, and a salesman of type

After submitting my talk, I changed my mind. I started in May, Peter Bilak presented FontStand, welcomed by many independent foundries, that would never work with other distirubutrso before.

Peter started his talk by talking of talking about the business of type.

This guy, bankrolled gutenburg, then sued him, took his stuff and his apprentice.

Type is a tricky balance of type, business and soap opera.

Fontshatna it a lbit early to discuses but there was the Q&A with Kupfers and the bestsellers, and we look forward to stephen coles keynote tomorr.w

so the business is reduced here. i decided to make it more personal and i will ask questions that have been bugging me about type culture, and weave in my work with my fonts. not because i am presenting my fonts - just presenting personally today - but i use my position to do something good for type culture.

What is the state of the art for typography and type?

On one level, professional levtterform design, including lettering, is wonderful. Our raft, if i may use such a word, is never before better technically informed, young designers have os many techniques to design type, to learn both tech and craft, to see them as complimenting each other and a very wide knowledge of all that came before, no more modernist laws of what is allowed and what is not, and all feeds in to all today.

Technically we have new cultural connections, a global view of type production, we can make a single font with many writing systems, working cross platforms, and more apps automate many stpe.s

For this who are curious there is a toolkit that has no precedent.

AS type guess better and better, the design of letterforms has no use, there is more and more bullshit.

“Never been better”? “Too much rubsbish”??

The Shallows by N Carr, say that the net makes us more superficial. Googleitius and wickets, replacing books and memory by quick searches. Young people count their likes on bechance to measure their success.

Not he web if something doesn’t suck then it is awesome.

The teaching! The type teachers should be more like music; you should already play music well, hearing well, even pitch, and understand music and history before you can study. You do not have to show these skills to be a teacher of type.

It surprises me how much you want to know about the history to be a professor of an art university.

I have a degree in something else though ;)

As a sconsultant for myself,ts i see how the stream of rubbish is made, and the attitude of people who offer this stuff. The people who are humble often have talent, and the arrogant ones are often hopeless.

When I began with my fonts 8 years ago, I was suppressed that there was no quality thinking at all. It was idealism and commercial thinking. They started in 2000 to offer a lot of fonts very fast. They also thought about being an open level playing field for type designers, and that is due to the fact that my fonts was bitstream; no one had much understanding of type design, they were engineers and organizers. They wanted to make shopping for fonts easier. In the 90s you needed to know the names, the designers, to buy from, and the bitstream guys wanted a festive font, and not aknow anything to find something nice. They wanted equal opportunities to make and publish type, to just go for it. Myffonts did do something positive there, to help young designers find an audience immediately with their own foundry. They were the amazon.com of type, and not be neutral. No one thinkks that amazon is bad if they have mediocre books, but my fonts was seen as needing to have gatekeeping responsbiliities.

The open gate policy was a good thing? “A fantastic game changer”

I joined in 2007 and saw how more and more people came to my fonts. Nick Sherman made a statement that the stupidty was gone, and the clueless hobbyists kept coming, so i amnd more in the team pushed for quality control. it took years to turn things around, and in 2012 I started a foundry review process. Tyrol, in California, and Dan Reynold and Florian Hardwig, part time. We accepted only stuff that is good and useful, and to give feedback to those who need improvement.

3 years since we stared .we have 1100 applications. 1 per day. we look at each font submitted, and we approved 25%.

Werejected 65% and the remaining 200 foundries are still working or agave up. Here is an example of the to do list.

We need to tell more foundries that their fonts could be removed.

We need to tell them how to do it.

We see the beginners have he same mistakes and we write the same thing citing the same resources. There are not enough good books for beginners - no english language book that does this. I mentioned to spanish people the Como Crear Tipos, and essays and tools by KLTF and Twardoch and ILoveTypography all in once place. We propose to published a series of guides to foundries.

Who should provide guidance to beginner designers? Not a bad idea for my fonts. This should be released over this weekend.

I worked with an illustrator, Beatrice Davis in Berlin, and I will published a series of guides for foundries. One on naming, originality, and so that will be a series to develop over the next year.

Some type designers are negative on monotype, I am not an outsider or an employee. But I can say that there are good things of being in a larger company. yFonts was limited when we were at bitstream. Now we can sponsor more events, the foundry guides, and hire the best people we think of to help us out. We help typographic culture, not on monotype’s corporate conditions but our own.

So, the culture of type. This is a lecture that could be an hour. I have tried to do good. There is a downside too, coordinating the review team, I have no time for writing articles, and its the most boring work I ever did.

I want to make more spiritually rewarding work. I a not a black and white thinking person, and Type culture is broad for me; sjhapring text, origin content, spreading the word, politics, economcis, and it can influence those things.

To foster change can help these things, and type culture is something important.

I think we should remember graphic design was invented by artists to foster change, and some stop being artist and become propaganda designers.

This is something I made, TRUE and FALSE rotational ambigram.

This is a russian artist, a publicist for the revolution, and later a copywriter.

The 20th c most interesting type design, Dwiggins, and this is the title cover of The Crew of the Ship “Earth” - similar to the bucky fuller thing - and this book reminds us that we sail on the ship and be a team, and the human race is not so good at this.

I think we should mobilize all design disciplines to politicians and changing concentrated power from short term concerns. Type is a small artifact in the world where we have total control of all details. I like to have the skills that I have to make ideas speak more clearly.

The one time I was able to do this, when I work fro FontShop Benelux, before Fontshop was acquired, Rudy was leading fontshpo in a more punk way, and with the “Druk” magazine i had a way to do this. Left is an issue about Cuba, with an illustration by Erik van Blokland, and I have tendencies for subversiveness.

Mentality is more important than prestige. I would refuse to work for an oil company, and I would prefer to work for a radical publisher than whatever.

Repeated was founded this summer, when my sister who lives in cornwall, joined a community of radical book publishers.

My friend Lorp drove down from Bristol, and said the same thing.

So repeater wanted me as their typesetter. I did their first 2 books, women and working britain, and also a book, about 220 pages in a single sentence, no paragraphs and no full stops in the whole book. They said it would easy, just 1 paragraph ;)

It became a challenge for the reader too. I ahem MA in the theory of literature. This book didn’t have paragraphs or chapters. I read it, I realized I could take the most extreme words of each page, and place them where the chapter heads normally go in the header. I made samples, and the author said, This is what I wanted! I twas a radical idea, and so when the author approved it, he reviewed all my choices. That was a great project, very interesting for me.

This is Vito, by Type Jockeys. The primary typeface is chaparral.

So for me type is part of a lrrger thing. You can enjoy that and admire it, and it is a tool that customers need to use without obvious mistake. The smugness of people who make mistakes. Here is a photo by lord of a typeface not working out of the box at a hotel signage.

This is an email from Thierry Blancpen, from a post office, using 4 typefaces in 1 email.

“Those stupid graphic designers” say type designers. (a) the type is too complex or (b) the software doesn’t work with the fonts. Some fonts are not made beyond 256 characters, I saw this in my airplane, the quote marks have the HTML entity in the txt! on the airplane!

So, why don’t type designers design the new Google logo?

Petr said “Design is making it difficult for people to do it wrong.”

Henrique says what it is to design type to every day people. One of the parts where we fail is one wikipedia You can detest it, but you can improve it. Why do us not contribute to the type design pages. They have a project typography, none of us participate. This is stupid.

What is business to type design ? “Type IS business, isn’t it?” J B Levee was right the other day that business basics are essential to the craft.

We have a guide to pricing type. This is very emotional and it is that as much as an economic factor. That low prices degrade their owkr. But the economic reality, compassion comes with the trade, and you numbest find an audience and convince to buy yours not the others. Higher prices doesn’t mea higher income. Lower prices reach new vocational font buyers. Font prices are about choosing your niche in the market. You can pick clients with a high price, or sell to the masses with low prices until they say instead of $30 a font? Crazy? Or, wow a whole family, yes, that is nice. So I don’t tell you how cheap to go.

Are there ever too many typefaces? Frutiger said is there enough wine? yt Yes, I think there is. We see fonts that look like others, and we can not refuse them all, we are just a shop, not a chop.

So we need to be reasonable, and the bottom is already here. The font s that cost nothing get better and better. We should explain that it takes special skills to make type that is bulletproof.

And things go wrong in some companies. [ The video of Adobe MAX last week showing Project Faces “I often have a problem choosing a font, and I put some placeholder get in u=illustraotrr, and then i look at fonts on my system and some projects you want something custom. we take existing text, outline it and drag around bezier points. But first that is time consuming, and changes to 1 letter, its hard to apply that to the other letters. So lets jump to the iPad, an app we prototype. instead of an outline, we have a skeleton or the center line, and here I have the P, we see the skeleton line. We derive two points from either side for the thickness. we can change the skeleton without the curves. We have a skeleton font, and with sliders, we can create any thickness we want, adjust the width, for condensed or extended, we can select a set of fonts with the crossbar, and this is cool, its moonlight, and important in typography is the contrast, the thick and think parts. if we adjust the contrast, we can see a more classic style font. If i jump to the A you can see that working. We can slant for an italic version. this is missing serifs, so we have them we need to slide them out. We can control the curvature, and I will back out here and talk about effects. We also want to apply interesting effects, to create a thin center line, we can make it wider and adjust the offset, to make cool styles. We can also deform the path, like this jitter, will create a stroke that will look hand drawn. So, back to my first problem, I type in my title of the poster, I increase the width, the weight, the curvature, and then I can apply a stencil effect to it. when i ram ready to export, you can export it as a vector or as a full font with all the changes. I exported this, I called it Max Font, and you see this actual font file from the iPad in illustrator. ]

So, I am glad you enjoyed that torture. So adobe will replace type designers with an app, that is today’s news. Thank you very much.


History of Adobe Type Team, Miguel Sousa

I work with Adobe in the Typekit team, I write tools, development and sometimes I do type design.

My Brasilian friends, call me Miguéu, its not correct ;) Its my first time in Brasil... last year I had 2 tickets for the World Cup but this little guy came into our lives so I was not here. Yesterday I walked aorund the city and I found it so nice to be in a foreign country with everything in my mother tounge. Last night I got this dessert, a full size Calzone with banana and chocolate - a candy store on a plate!

So, down to business. There are 2 people in the room on the front row who know more than I do, as they lived type at Adobe. I refer to Sumner Stone, the former director of typography, THomas Phinney, the former product manager for Adobe Typography, and David Lemon, teh current senior manager for type development.

Adbe has more than 30 years of experience in software development, and in the recent times succeeded to enter digital services. This company is a font technology and design powerhouse since the beginnin. Lets take a journey form the start of DTP to today's rise of open source fonts.

Chapter 1: The beginnings. Adobe was founded by Charles "Chuck" Gescke, and John Warnokc, who met at Xerox PARC in the mid70s. While there, Warnock had the idae for a page description language. They tried to convince Xerox to use it in their products, and after 2years failing at that started Adobe in 1982. They started with an idea for high end software and hardware for printing, using John's software for connecting them. Here's images of other such integrated systems of the time.

A PDL can describe a page, images with X and Y dimensions, a text block with Times at 12pt, and so on.

As Adobe was working on its PostSCript system, they were approved by Steve Jobs. He was developing the Mac and a laser printer for offices. He tells them that he has the hardware and needs the software solution; they adobe guys know it would fit the bill. At first they declined, they wanted to sell the full system, and the printer was a low res 300 dpi unit. But later they agreed, and ni Jan 1985, the laser writer is announced. The mac and laser writer were part of apple's vision that was different to IBM's that dominanted. But apple would see the huge revolution happen: DTP.

Mac came with a word processor and drawing app, MacWrite and MacPaint. But to do more complex layout on the same page you were out of luck - unless you know how to write PS code by hand, and send that to the laser printer. that would work, but designers could not do that.

SO when Aldus PageMaker in 1985, that changed things radically. a GUI simmilar to what we use today, and supported the PS language. People could creat their own documents and prnit them.

The Mac compuater, the Laser writer, the pagemakrer app, and the ps langauage, were the things that sparked the desktop publishing reolvution. And one more element that was crucial to the sucess of DTP, and that is scalable fonts.

The tehings on this page were all made by companes with A in the first name, Applle, Adobe, and Aldus.

So, a WYSIYWG toolchatin was not so bad. It is a big deal that office ocmpauters looked like this, office printers were lineprinters like this, and toosl for copmositng pages were compass and rules. You gridded a layout by hand, specifying type and its sizes; this was sent to a typesetter who made mocks to cut and paste into the layouts. So DTP was really cool.

AS I mentioend, PS was what got adobe started; PS was responsible for the type development efforts. PS was developed and others tried to solve the problem. A device indepdnetn graphics system, a single languega for computers and printers, handling text and graphics, and resolution indepentet, from a 300 spi lasrer to a 1,200 offset printer.

PS was special for its scalable fonts, not just bitmaps. Before PS all fonts were bitmaps of a fixed size - here are the mac system fonts. PS chagnd this, and these fonts lived in teh memoey ro fhte printer. the could scale to any size and be output at the max size of the device. Here is a smaller and a larger S. Adobe launched with 13 fonts, which included the basic collection of designers - helvetica sans, times serif, courier mono, and a symbols font. They licensed them from a 3rd party and courier and symbol were designed in house. As a new comer to a industry that traditioanlly copied and cloned each other, they wanted to send the message they cared about type, by licienging the fonts from the rightful owner. They etablsihed a taskforce in the comapny that led to the type deveopment team.

That team is still alive today, proving that type is a timeless need and that adobe cars about it.

To help set up PS as a standard, more digitla fonts were needed. the fast way was to license existing designs. they wanted to use typefaces owned by other type companies. they licensed from ITC new collections to convert to PS, AGFA, Compugraphic, MOnotype, and others followed.

Sumner Stone was hired in 1984 to engage with 3rd party foundries to create the collection and make new designs; he starts the first adobe type originaly, Stone SAns, Serif and Informal. The first PS superfamily. Apple's 2nd printer, LasterWriter+, had more linotype licensed designs, and also ITC. That was 22 new font files, 35 total. They were all developed at ADobe by the type team. Realising teh commercial pontential, they release the Adobe Type lIbrary in 1986, with more than 100 fonts, all Linotype and iTC designs except one. Sonata, by ? Huggins at Adobe.

WIth this release, type was Adobe's first shrink wrapped prodout, a year ahead of Illustraotr 88. Sumner hired Robert Slimbach with whom he worked at Autologic, and like Stone, he wanted to work on original designs, and starts a Garamond revivial first, and then Utopia. The first ADobe Originals collection.

In 1988, Carol Twombly joined the adobe type team. her first typefaces were trajan, polipihious, and released in 1989. Around teh same time other team members concerted wood type to the new medium, and th 1989 collection had these fonts. Reviving the art of fine typography, to surpass teh old, was starting to be real. Here is Jim Wasco, Fred Brady, Robert and Carol.

As DTP took off, so did the originals program. The new digitla types stretched the boundry of type sophistacation and scope. they wanted top quality type and adobe was the perfect home for this. the adobe originals collection grew, and in 1990 the 2nd wood type was released, by barbara lind, joy raddick, under fred brady and carols' supervisoin.

Another revivla, adobe caslon, by carol twombly. Minion by Robert Slimbacj, that is now earned a place on the list of workhourse classics. To supplement the inhouse type designers, Adobe commisioned outside designers, and Tekton was the first. Then the Wild Type collectoin in 1993, a failed attempt to broaden the library style.

Here's a samepl of them, Exponto, Mess by Michal Harvey, and Jimbo by Jim Parkinson.

More recent families, are Garamond Premiere, the 2nd garamond. Tis is more authentic, with all the optical sizes cut by Granjon. Arbo, also by Slimbach. Hypatia Sans by Thomas Phinney. Adobe Text by Robert. Trajan Sans, by Robert. Also noteworth is ADobe Clean, the new ADobe corp brand type that are exclusive to Adobe.

A few lesser known fonts are handwriting fonts, for Acrobat's signing feature.

Next, Ground Breaking Technology. This ability came with a big challenge. THis was what fonts were like on screen; bitmaps were poorly displayed. You could design and compose the page but what you see and what you got were different. To solve this, Adobe made Display PostSCript, so fonts were 'printed' to the screen so the fidelity was closer. As Adobae licensed PS to apple, they approached apple to licensed display postscript. But they declined; Jobs was no longer at Apple and the current management did not want to do it. They were unhappy with the licenseing fees they paid to Adobe for PS already.

MS had become a big player in this peroid with MSDOS. Apple and MS were so eager to avboid adobe, they started developing an alternative to PS and the Adobe secret format. The news broke at Seybold in 1989, when MS annoucned to cross license their PS, Apple made TrueType and MS made TruePage. Warnock was furious, and he wasn't afraid to show it. [ great image of Warnock as a jedi ]

He saw these companies teaming up to destroy his company; 65% revenue was PS licensing, 10% fonts, and 25% applications. So clearly the future of the company was at stake. he didnt't take teh rumours that precedd the conference lightly, and he announced a new type rendering ror screns, ATM, and the public release of the Type 1 speification. ATM was released a full year before the first TT fonts in MacOS 7. This meant type on screen was not heavily jagged.

PUblishing the type1 spec, adobe was no lnger a gatekeeper for fonts that rendered well on screen. There were 2 font formats, type 1 with hinting that was closed, and type 3 that was open but without hinting. In 1992, adobe introduced Multiple Masters. This meant you could have all the weights, widths and styles. You could have ATM and Myriad was the first fmily to take advantage of this; this allowed for optical sizing too. Small sizes have surtdy features while large sizes have refined features. Here is Arno Caption, with less contras, and Arno Display, meant for larger sizes.

The MM technology didn't lsat long as a consumer product. The tech survived as a type development tool, thugh. In 1996, ADobe and MS parterned on OpenType to end the font wars. This had many advantages over Type1. The fonts are cross platform, the small caps and ligatures are included in a single file instead of separate fonts. And the format enables correct shaping of complex scripts like Arbaic and Indic.

Which brings us to Non Latin Type. Minion Cyrillic was the first, and it was Robert's idea. He saw that it would be the most useful thing he could do as a type designer at an intnertioanl corparion. There was very little information on this, and so we commisioned a report from Thomas Milo at DecoType. We focused o nthe main east asian languages, CJK. PS was again the main driver of this effort as Adobe negotatied with the printer manufacteroers. ADobe sought licensing with Morisawa for conversion of its font collection to PS.

Kuzoko Mincho, released in 1997. Minion in 2000 was released in OpenType, with the Cyrillic glyphs along with the Latin, and this was also the first to include Greek. Lithos also got Greek support in 2000. The latin, design is based on inscpirational greek. Carol had drawn much of the greek in 1989 alongside the Latin. Myriad greek by Carol and cyrillic by Fred. WArnock, released in 2000, was teh first with LGC develpoped together.

Adobe ARabic and Hebrew in 2006 were our first middle eastern designs; at the time we outsourced these from Tiro Typeworks. Tim Holloway did Arabic and John Hudson the hebrew. Thai and Develanagi also by Tiro, Thai by Tim and Devanagari with Fiona. The deva was released in 2012 when it wass upported in adobe apps. Kazuraki was a new vertical japanese font with a brush style, and this twas technolgiclaly challenging, pushing hte limits of vertical typesetting.

In 2009 the AFDKO was released with arabic support and we could do complex script development in house. Myriad arabic and Myriad hebrew was done in house, design by Robert Slimbach and the production was done by myself and Paul Hunt. Other complex scripts were made ffor other indic scripts. Adobe Gurmkhi in 2012, Vaibhav andPUal, and also Adobe Gujuari by David Breina, and Adobe Tamil, by Fernando Mello. Neelakash did Adobe Bengali, released last year. There are more indian typefaces in the pipeline.

Then, open source effort. being a tradioital software copmanu, adobe didnt catch on to open source until recently. August 2012, we released Source Sans, our first open source family. It was adopted by many companies for branding and online. Its broad lanague, greka nd cylic, and high quality, set it apart from other open source prpojects. Later in 2012, we release Source Code Pro; the reaction was even bigger. We were blown away by how often it was downloaded. Source Sans and Code were designed by Paul Hunt. But they are incomplete without a Serif design. Frank G made this in house. This has the same weight range, but only supports Latin. An Italic style is in the works.

Source Han Sans, is a pan CJK family, a passive project that is not possible without a partnership with Google and 3 asian foundries. The family is made up of many weights and each font has more than 65k glyphs. It support CCJK in a signle file, the first pan CJK family ever, and the glyph count is at the limit of the font format. Here is a glyph chart of all 65k glyphs.

WE also open sourced type tools and tehcnologies. In May 2013 we annouced the Adobe Glash rasteriser was contributed to freetype, that program that converts glyphs outlines to the screen. it was an adobe crown jewel, it was a big deal to annoucne this for adobe and the many comanies that use freetype. in september 2014, we released the afdko on github. we git a lit of support and sometimes contributions from 3rd party developers which is great. If you want to get the toosl and fonts, go to github.

What's netx?

Te font industry changed a lot in the last 30 years, adobe led the way and continues to be a front runner with font design and tech. the copmanyfocus n revenue has changed a lot, and type has remained a core copmonent. the acuiqison of typekit in 2011 brought in the expertise for serving web fonts to the web and mobiles devices, and typekit is the backbone of a smilar sevice that delivers fonts to your destop.

I donte xpect much change, tehre will be more commercial and open source fonts, more type tools, and more type technology from adobe. we know that the only way to stay in business and remain relevant. type is our passoin.

Roger Black: Source Sans Han, can you email it?

Miguel Sousa: It was several megabutes. Mac OS allows to select a file and preview it, and it crashed the Finder.

Raph Levien: Source Hans Sans is 15Mb per weight, and 7 weights.

MS: It covers everything.

John Berry: Thanks for making adobe's development seem planned and coherent from start to end. MM didn't last because service bureuas had toruble with it. It was a de facto resistance. Also Adobe itself shot itself in the foot, despite the fonts teams efforts, the app develoeprs were not co-operating and no-one took the approach that you needed a high end tool, it should be a no rbainer.

MS: Right. Illustrator had a little panel to specify new instnces in the app, InDesign still supports the fonts if they are installed, but not much more than that.

John Berry: peple thought you either wanted expert control or simplificy for the masses.

MS: ALso MM fonts werre hard to market. Its easier to say here's a family with 40 files, or 1 single file, the same money. Selling that was hard. And technology problems that you mention.

Bruno Maag: Is han certified for mainland china?

MS: Its not certified yet but is in process.

BM: If its 15Mb, how to serve for the web.

MS: We already do this.

David Lemon: As I mentioed on wednesday, we use subsetting and on the fly augmentionat, so its totally managbale.

BM: What about any old corporate who wants to not use Typekit?

MS: They have to. Well, its open source, so they can roll their own solution, but we have a working solution today.

DL: You must use subsetting and streaming.

MS: We analyse the page and only deliver what is needed.

DL: Monotype and Google offer subsetting solutions.

BM: Who wants to partner with Monotype?

MS: Where are you going with this?

BM: A corp doesnt want to go through awnyone, what can they do?

Stuart: There are multiple versions with differnet file sizes and subsets, so if you want just korean, you get just korean; there are maybe too many flavors. There is 1 giant OTC which works great as a mac desktop font that has everything. A korean font with WOFF is 2Mb. Both Source Sans Han and Noto versoins.

MS: Answered? Good.

Henrique: What about Project Faces?

MS: I can't comment as I know as much as you do. We were as suprised as everyone else. It was a separate team, with their own ideas. We have mixed feelings. What they showed doesn't mean it will be released, it is a preview of what Adobe is exploring in house. We are now in contact with them and trying to understand their vision.

Sumner Stone: Its amazing to see as employee 25 from 1985, if you wrote a memo you wrote a PostSCript program to do that, and Warnock insisted on this, he taught you himself how to do it.


Stephen Coles: State of the Mart

[ PDF with original speaking notes available in this repo ]

Catherine’s other keynote was great! A high bar for me :)

Sao Paulo is a great city, so great to be feeling so welcome here, and great to see people here who design fonts for sale now or will soon? about 15%. Who has you own website where you sell these fonts? About half of those.

I don’t make fonts, I’m in awe of those who do, and I’m not much of a graphic designer. I use them when I can. I teach at Type Camp sometimes and do some day lectures, and I’m in awe of full time teachers too like Catherine and Indra.

I’m almost a journalist; I try to report on news in our industry and add my own comment, but I’m also not full time there.

What is left? I’m a consultant. What is that anywhere? Not a more generic term available.

Here’s a stock image photo of a yuppie, boring shirt and tie consultant ;)

I’m more like a Font Broker, like a wall street trader buying and selling fonts.

I connect designers with fonts, font sellers with font users. That really is me, type designers in 1 ear and type users in another year.

www.Typographica.org is one place we publish essays on the scene. In 2008 I published an article about options for “Taking your font to market: Foundry, Reseller or go Solo”.

Tonight is an partial update of that - but check the full piece.

I also work on www.fontsinuse.com and this has articles by Florian Hardwig, Indra, and others like Mark Simonson, and others who contribute to the site. I learned a lot about how type is chosen and used on FontInUse.com

And I now run www.typedrawers.com where people chat about type in progress, politics, business and tools. I took it over last year in 2014, and we have a team of 7 moderators.

This along with a few years on the creative team at fontshop, 2004-2011, shaped my role as a font broker.

This is all about the western world; other regions and scripts need study but that is not my expertise; Toshi has promised he’ll do a talk on the Asian font market.

Toshi: I didn’t!! I remember you asked me

SC: And this is all biased towards indie families too. Monotype, Adobe, and Google have done big things, including sponsoring our conference, but theirs is not the perspective I take today.

What is an “Independent” foundry?

One thing could be that is is designer owner company.

Another could be that it isn’t dependent on a single external company for their livelihood, like a single big sales platform.

So there are 4 things: making type today, selling type today, and making and selling type in the future.

First definitions. You type designers will have to bear with me while I go over some definitions, as we should speak from the same language.

Buying and selling fonts means selling retail licenses. not work for hire.

A font maker is someone who draws letters, does font production, or holds font monopoly rights.

People don’t shop for fonts physically, although here’s a photo of someone in east berlin with physical carved letters in a shopping trolley ;)

Font Buyers, and users, are graphic designers, ad agencies, web devs, device makers, startups, megacorps, church newsletters, and individuals.

Foundries! We know where that term originates, maybe its antiquated for metal shops, but i like having something for us that is specific; this means a company that makes fonts. It could be 2 or 3 people, a single person, or a dozen employees; and someimtes its a larger foundry with dozens of employees. Eg DaMa, Hoefler, FontFont, Monotype, URW. The people who are behind the company of the bigger foundries can be invisible, and for me the best make a point of showing off the designers behind their brand.

Retailers: A reseller, disturber, shop. Like my fonts, font.com, fontshop, font spring.

Small retailers: A boutique shop, like Village, Type Trust, and Font Stand.

And speciality retailers, like FontDeck and Web Type, who offer small collections for a specific purpose like web fonts.

Type Foundries Today!

We know what things are like but not really specifically until recently when Ruxandra Duru (who works designing book covers at Penguin Randomhouse) who attended ATYPI last year in Barcelona.

Her thesis project was a study of indy type foundries. She examined the western players, to see what it is like to make and sell fonts in 2011. What occupies their time, how often they collaborate, and how they distribute their fonts.

It was PDF that got some attention, and when I found it, I asked her to make it into a website. After a lot of hard work, with her and her web designer and a co editor, we published the report.

Its a 2013 update, a census and update on the indy foundry scene. Sadly she can’t be here today with her day job. the best I can do is play a little video of her now.

RD: I wish I could be there! I know its in good hands with Stephen Coles. I thank all those who contributed especially those of you here today. As of today its just a start and will grow in the future. Thanks again!

SC: I wish she was here too :)

Demo!

http://census.typographica.org

Currently its not responsive for phones, and it seems to think this projector’s resolution is so small that its a mobile :(

Census criteria: there are 3 sections in the report, how we got to this point, who the players were for the first years, in a great and short summary. Then a timeline with what all the key moments were, with new companies, new software launches to speed font production, and then the census part.

How to get a comprehensive set of how is in the font scene? There are over 1,000 foundries but some are no longer active and some may be controversial to call independent.

First, they retail typefaces. This includes graphic design studios like A2, and excludes people who do not charge a pay-per-copy retail license fee. This excludes larger foundries, not those owned by a large conglometerate or a business with other revenue; and active within the last 5 years and with a website showing their types for sale.

She checked Type Navigator, by Jan Middendorp, the most recent and thorough showcase of foundries in print. Typebase, was revised in 2012 and invited updates.

She also looked at a twitter list collection by Ivo G at FontShop.

Also the Indie Fonts series in the early 2000s, they were specimen books with a good documentation of the explosion of fonts in that period.

Those were her main sources for the 300 foundries in the census.

So, in 1980s, Elsner and Flake, Emigre, and others were started, but the total was 2x by 1992, a huge growth period. TT introduced, Ikarus on Mac was completed, and FontLab and Fontographer. A few foundries made grunge fonts but they are no longer active today so they are excluded.

In 2000s, font editors matured, and the number of sales platforms increased. Being a retailer became easier for small indies.

In the 2010s, there were even more.

The low number for 2013 is an outlier; the sources didn’t have the most up to date info.

Here is Linotype in 1919, the factory gives a sense of what it looked like then: a huge industrial operation with 1,000s of employees.

Here is DJR with his dog on a laptop in the sun; this is 1 of the dozen staff at Font Bureau, who are spread all over.

This is Victoria Rushton and others at the FB Office.

Here is Just Another Foundry with a husband-wife pair, working at home.

95% have less than 5 people.

5 foundries have more than 15 people.

Font production is not limited to specific parts of the world.

Africa, Middle East and Asia are excluded.

Most places with many are traditional places for print and design.

Zooming into south america, in 2013, we see the spread across the continent.

Henrique Nardi will help me with the Brasilian map

HN: I promised! ;)

And what we need for the next update is where the locations are across South America. Just 10 years ago it was much more barren.

We love interactive sliders, so we hope that in future you can combine the date and map over the years.

Font editors have got a lot better. Font production is faster, Fontlab, Glyphs, RoboFont, and others.

The production process is faster, the barrier is lower. This leads to a lot of poor fonts, but the number of professionals grows fast too, and many have advanced degreess in type design.

In 1998 there were NO type design focused courses; by 2000 there was Reading and Type Media. In 2010, the USA got one at Cooper Union, and just 2 years later, there are 11 places offering degrees in type design worldwide.

San Francisco, Mexico City, France, FADU and one here in Sao Paulo. There is a great Wikipedia page on this.

The volume of the production of fonts is staggering today.

More fonts!

What does this mean? More competition for makers, more choice for users. Florian Hardwig: “Since there are so many, its harder to pick a good one.”

Usage trends also spread faster now. Helvetica released in late 50s but didn’t get huge until 20 years later; the association with 1960s modernism, didn’t stop it from being popular all the time since then.

Today it takes weeks for a font to become popular. I can’t verify the timeline, but a few months after Apple introduced iOS 7, web designers followed suit, so they added a long shadow, especially to type , and it was a visible trend on Dribble. These logos are all within 3 months.

[ SC captions the Google Fonts logo as Luky Vj ]

You know when it is a trend where there is a generator for it, and a backlash.

Users determine what kinds of fonts get made; there are obiovus trends in font usage but also in the design of type. Commissions by users are common and this effects retail market trends.

Type designers reposed to what is popular. The fast response to trends, something on a bestseller is replaced within months or weeks, as we saw this week.

Chromatic fonts are now possible and all these geometric sans with color and layers are all within a few years.

Trendy fonts do sell! but temporarily. they flame out. I see why type designers go down this path, its practical, but I am more interested in type that sets trends.

Lets consider the perspective of the type designer. If you finish a font, what are your choices?

Submit it to a foundry. Usually an exclusive distribution deal, the royalties can vary. Most foundries pay a percentage of the final price, and so the fonts can be resold in many channels. Sometimes that can be a percentage of a suggested retail price, so that the cut for the designer is always the same. Thhe benefits are no business effort to bring the type to market, and technical help with the production. You should know about business and the contract, of course, but uyo can spend more time on designing type.

You can also go to a reseller like my fonts, and have your own foundry brand; you set the price, and each reseller has different customer bases and different marketing efforts; ask about how much effort they will do for you.

Some offer fonts through many resellers, like Mark Simonsson, and others use more select resellers. This requires more works; your own brand, license and policies, and completing the fonts to a level of quality.

Most retailers handle sales, but the main advantage is reaching a wide audience. The main drawback is the competition with the other vendors and being buried.

You can also do it all yourself. This brings you 100% of sales, of course, but exclusivity has its benefits as Hoefler and Co, or LineTo can attest. It gives your brand more value. But it can be a hard path to follow, to bring traffic to your site. It may look like a lemonade stand, on the roadside, waiting for people.

You don’t know how much you could have been making if you went elsewhere. But you can make a proper shop, make a fully independent living. You need to be biz savvy, spending less time on drawing type, and cover the overheads. But you maintain a relationship with each customer - a huge advantage and disadvantage!

Is this a pro or a con? I think mostly pro. I’ll explain why in a moment.

There are other models. WE are in a state of flux. There are new platforms, there will be more in the works, and among these I see things happening that can improve the state of the market.

Retailer curation. As the svolume explodes, this is more important. Be selective on quality, the drawing craft, and also a sense of taste, originality and a vision of the future. Curation sets trends, not the other way roaring. FontShop set the baseline for this, the FontFont slogan was ‘by designers for designers’ and the brand was meant to extend this to a retailer.

As the collection grew, they had to balance a big inventory with latest releases against not burying the other products in the inventory. I saw this first hand.

Resellers like Fontshop do a good job overall, but the bar is not high enough for others. Most of the most interesting foundries are not totally indeptent, Swiss type, Process Type, Lucas Fonts, TypoNine, and Typotehque. This sia slow trickle, but these are bellwether foundries.

FontStand, allows trying fonts for free, its great for users, Monotype had SkyFonts, and Adobe has TypeKit Desktop Sync. FontStand is set apart with its selective list of foundries, which is attractive to makers who know they will be in good company; they say they will grow slowly and deliveratebly.

At my fonts, we heard about how they have a review board and reject more than they accept, but their MO was to let the market decide, and until last week the my fonts lists were based purely on sales. Today they list staff picks, and there is a new hompage as of last week. But the big discounts linger on the homepage.

When you enter a Gap store, but the latest stuff on full price is up front to see before you go to the sales to the back. You don’t have to put price cuts up front, but MyFonts have a clientele to cater to. But there is a need for retailers in other markets to do something else.

I think the font market of the future will be defined by a direct relationship between maker and user. I met users who say “it is a my fonts font” or “it is a they work for you font” but people see humans make fonts more and more.

The Creative Characters newsletter is focused on individuals, and type designers need to know their users too. Who are they, what do they need, what problems do they have with their fonts?

Font designers tell me they love closer relationships with customers, and unless you sell directly, most of that is lost.

Florian Hardwig said, as above, many fonts makes choice harder. I don’t have all the answers, but I know that buyers and sellers both need a better market that respects both their needs.

The farmer’s market is like this. They grew a lot in popularity in California and Berlin where I live. Living close to agri biz means there is more concern for Grow Local food movement. We can get our staples, Artisan, Organic, and other buzz words. Instead of a supermarket that buys all the goods and resells them, we buy direct from makers.

The people at the stand often work on the farms, and the market committees are a combiantion of retailer and maker, with direct connection of users to makers. We know the quality because they literally stand behind the product. We give them direct feedback, we have mutual respect.

Can this work in the font market? Sometimes the answer is not in the market; typographica and fontinuse are editorial.

I am working with Marina, Rob Meek, Nick Sherman, Floraian Hardwig, Chris Lewis, and David Jonathan Ross who advises us. They are some of the best people in the type world. I waterski behind them.

I can’t say more about this top secret project, but what I can say is, a MyFonts Killer! ;)

No! Its not a my fonts killer, that is a joke. We expect to co-exist with My Fonts and help the ecosystem. Its a guide to the type market and the best places to find them; foundry websites, or a retailer, whatever the foundry prefers.

Expect an email from our team soon.

These are solutions for retailers and new retail concepts under development. I see hope. I want to see type designers to retain independence, do good work, and respect themselves.

I love this craft and I love these people, I want to see them continue to do what they love.

Thanks!

Jan Middendorp: In HTML you have a thing for italic, and slash italic. So, slash my fonts, I think that what I think is hardest for those making choices on line is that there is too much. Even news and politics, its too much. What we all need is less. I agree what you say. I also talk to my fonts about starting something that offers less, a curated collection, but it wouldn’t have much credibility if its part of the monolith. so you have great people, Florian is a guy I work with in Berlin, and I wish you all the best with this effort. [ big applause ]

SC: If you are a budding type designer, please ask questions. Or lets chat at the party.

JB LEveee: A corerction, type education. Lots of type designers also have an educational background, and I want to stand against the idea that before Reading and KABK there was nothing for type education. Me and many others were trained in the type course of the last 30 years, and there are many good places for learning type around the world. It was not a sudden thing at that time, and there are older courses.

SC: Great! Lets make a better map. I wanted to speak of post-graduate programs, but I agree totally.

Jose Scaglione: Nice presentation! I wonder if the graphic deign community needs us to curate the fonts that we offer them. Why they do they need us?

SC: I think there is always more. Education is top of the list for those making or selling fonts. People want to be informed and people want it. Type selection is a huge topic and I want to encourage more people to write on this.

?: If the curation and education is done on the same page, that is good.

SC: Yep

[ applause ]