Transforming Typeface Design: Moving Beyond Gutenberg and Roman Traditions

adamwhite
adamwhite Posts: 23
in Font Technology
Please read this post at LinkedIn:
https://www.linkedin.com/feed/update/urn:li:activity:7201017969437491200/

and contribute to the topic.

Kind regards,
Tagged:

Comments

  • Thomas Phinney
    Thomas Phinney Posts: 2,851
    edited May 28
    Although I like LinkedIn better than most social media platforms, why would I use the post here to contribute to a discussion there?

    There are a few questionable things in the article. What on earth is “neuro-optical technology for fast reading”? Is that the so-called “bionic reading,” for which we have zero evidence that it actually works (and one study that suggests it does not)?

    “    Advanced rendering engines, such as ACE (an SVG graphics rendering engine), offer unlimited customization in character presentation without losing Unicode information.” Calling Decotype’s ACE “an SVG graphics rendering engine” is at least incredibly misleading. I think it is simply wrong, but even if ACE were to use SVG graphics under the hood somewhere, that is certainly not the point of it, and is almost irrelevant to what it is doing.
  • adamwhite
    adamwhite Posts: 23
    Hi Thomas, thank you for contributing. I didn't say that you need to contribute at LinkedIn, but to read it there. Contributing on LinkedIn would be beneficial for more reach, especially for the people who are not the part of typedrawers community.

    Yes it's the so called bionic reading. I mentioned it as example that is made with a whole new mindset of optical presentation, and not as something as proven or not. In such matter we can say that all classical typefaces (promoted as readable), are not proven except they were the only option offered as reading experience. (Even Gutenberg's pile of calligraphy misused as accessible readable text).

    ACE is not using SVG graphics under the hood somewhere, but it's rendering pure SVG paths compositions. SVG became a web standard so it's nice to mentioned it as that, but in the core of vector graphics it doesn't matter since it's just math calculations presented with shapes. Yes ACE has many more important functions, but understand that people don't know anything about it so I can't present it with any term without missing other descriptions.

    In the years of communicating with people about it, I didn't see that word "Advanced" men anything to a "hard wall".

    I am glad that you are aware of ACE, and this is the one of the main reasons I am starting this subject. It's a hard battle to present such tech to people, unless it's from a "tech giant". Actually it is, but people don't know it's a "giant".
  • Thomas Phinney
    Thomas Phinney Posts: 2,851
    edited May 29
    adamwhite said:
    Yes it's the so called bionic reading. I mentioned it as example that is made with a whole new mindset of optical presentation, and not as something as proven or not.
    It is not a whole new mindset. There have been many, many snake oil concepts in reading and fonts before. The problem is not a lack of ideas, it is a lack of proof. In this case, the ONLY evidence we have says “it does nothing.” The burden of proof lies on the people making the extraordinary claims. Repeating such claims uncritically is a disservice to the public.
    adamwhite said:

    In such matter we can say that all classical typefaces (promoted as readable), are not proven except they were the only option offered as reading experience.
    Nonsense. The question of whether some typefaces are easier to read, and for whom, and under what conditions, has been very extensively tested, for “classical” typefaces and many others. There are entire books on this subject (e.g. Reading Letters: Designing for Readability by Sofie Beier). There is even an entire book on how to do research in this area: https://www.nowpublishers.com/article/Details/HCI-089

    Not all “classical” typefaces are the most readable, of course. But in general, on average, typographers’ beliefs about what makes typefaces readable? They map reasonably well (albeit not perfectly!) to test results.

    There are some interesting and notable exceptions, like the bouma vs parallel letter recognition problem. In this area the typography community’s commonly-held belief was… just wrong. But again, research has shown the way. But that particular issue did not make much difference to actual design of letterforms, which may also have been why it was a challenge to test.

    Some well-known reading research results include these:
    • most of the performance differences between most typefaces are small in testing, but they do exist. More extreme/unusual typefaces can have stronger effects, however! (e.g. throw a blackletter typeface into your test and you will see a stronger effect of variation between typefaces)
    • difference effects are strongest under more difficult/marginal reading conditions
    • typeface performance differences are real and reproducible, and we know a fair bit about what factors differentiate more readable typefaces from less readable ones
    • other aspects of typesetting, besides the choice of typeface, have a large impact on reading
    • we do not “read word shapes” (boumas) but rather, at least primarily, recognize letter shapes in parallel
  • Ray Larabie
    Ray Larabie Posts: 1,425
    edited May 28
    Reading is an amazing process, maybe that's why people often resort to magical thinking and pseudoscientific ideas to explain it.

    @Thomas Phinney has there been any research on how the familiarity of a typeface affects reading? Conducting such a study might be challenging, as a new typeface would require a considerable amount of time to become familiar to readers. Perhaps this is why many new approaches fail to yield results. That familiarity might eclipse any marginal improvements in a novel design.

    @adamwhite Science has proven time and again that LinkedIn sucks.
  • adamwhite
    adamwhite Posts: 23
    @Thomas Phinney, you said:

    "It is not a whole new mindset. There have been many, many snake oil concepts in reading and fonts before."

    When I wrote about it in my post, I didn't speak from the side of point is it a working concept or not. I mentioned it as new mindset in the way it goes away from the "standard" text presentation. It goes beyond the adopted pattern (even it stays in it some part). So I meant in the way it's visually presented.

    "But in general, on average, typographers’ beliefs about what makes typefaces readable? They map reasonably well".

    Now you bring the point which is an argument against you (even I don't try to refute you, I just try to explain and detail on the topic). Typographers' beliefs even being confirmed later is just a belief. They didn't make a research and then produced characters.

    So in regards to this "bionic", did anyone made a research about it and proven it's not working? Again I emphasize, I don't claim that it is working or not. I don't know.

    In my personal experience I don't know is it helping me to read faster, but as I am hyperactive person it helps me to focus on some parts of the text and I see it as a technique to be used in that way. Similar to bluring some parts of the visual content and focusing the clear one. That is the dimension you need to think as well, that we all have different experience with accepting visual content and our minds work differently because of many reasons.

    Capitalization of the first letter in paragraph or a chapter doesn't make you read faster but to say it doesn't help you to visually index the content or focus faster on it would be at least irrensponsible.

    Based on this "are they right or not", we can open so many subjects, like why do we use typefaces as we do for centuries? Why do we insist on uppercase and lowercase letters. Why do we insist on keeping character forms of the medival people and not follow practices of those before. Why do we calculate typeface sizes based on em box even it has no sense. Why do we and why do we...

    That is the reason of my post, to open all subjects and bring many people to the "table" to discuss and open our views.

    As in any tech there are much factors which makes something readable or not as well.

    For an example, in primary school, for me it was quite easy to read handwriten text as we were daily engaged in writing and reading it. Years after when being exposed mostly to print letter forms it became much harder to recognize handwritten characters.

    As well for reading cyrillic which is less and less used in my enviroment.

    Media exposure plays a big role in reading experience.

    One more example is using written emojis like :), :D, :P etc. With the expansion of graphical emojis younger generations hardly recognize these "retro" forms.

    Thank you for sharing resources, I will read it and learn more about. I am not acting as any authority or claiming stances, I try to lead a discussion which will bring attention of many people as possible and bring some solutions.

    @Ray Larabie you said:

    "has there been any research on how the familiarity of a typeface affects reading? Conducting such a study might be challenging, as a new typeface would require a considerable amount of time to become familiar to readers."

    That us the one of the most important points. Getting familiar with certain visual content plays a big role in the speed of recognition. For an example if you see a man from the distance, if you already know him, only few features are enough to recognize him. If you see a new person it doesn't matter how "optimized" he/she is. It doesn't matter until you have certain timeframe of experience with him/her.

    @Ray Larabie I don't claim that LinkedIn is good or bad, or is it the number one platform to speak about this topic. My perception is that opening such subjects will stay gated if only happenes on specialized forums. If you don't have a community of aware users, what's the point of a high quality service. ACE is an excellent example of such. It's the best tech almost no one knows or cares about. It can bring benefit only when people are introduced to it and can recognize benefits.

    We can always say that most of the stuff related to typography is not necessary. In many areas as well. But as we want a variety of possibilites and to act as a modern progressive society, then it's shamefull to stay on some patterns which are chosen by few biased people.





  • John Hudson
    John Hudson Posts: 3,137
    Aside, re. ACE and SVG: my understanding is that this was specifically the mechanism developed to render ACE layout on the Web, essentially a ‘font’ consisting of a folder of SVG graphics of radicals from which to compose text. Part of the flexibility of the ACE rendering model is that it is not tied to a particular font format, or to fonts as commonly conceived at all.
  • Thomas Phinney
    Thomas Phinney Posts: 2,851
    “ Typographers' beliefs even being confirmed later is just a belief. They didn't make a research and then produced characters.”

    Not true. There have been many studies that worked exactly that way! Some done by psychologists, some by type designers. I literally cited entire books, which cite many such studies.

    “    So in regards to this "bionic", did anyone made a research about it and proven it's not working? Again I emphasize, I don't claim that it is working or not. I don't know.”

    Not only has there been that research (as I already said!); there has even been a second study, one that also finds no statistically meaningful effect. 
    https://pubmed.ncbi.nlm.nih.gov/38723450/
    The first study is here:     https://blog.readwise.io/bionic-reading-results/

    (Note that the effect is actually negative, in both studies. But small and not statistically significant.)
  • adamwhite
    adamwhite Posts: 23
    John Hudson said:
    Aside, re. ACE and SVG: my understanding is that this was specifically the mechanism developed to render ACE layout on the Web, essentially a ‘font’ consisting of a folder of SVG graphics of radicals from which to compose text. Part of the flexibility of the ACE rendering model is that it is not tied to a particular font format, or to fonts as commonly conceived at all.
    Yes I meant when rendering on the web, since Tasmeem plugin is deprecated. In any whey the final output can be whatever, it's the least problem.
  • adamwhite
    adamwhite Posts: 23
    Thomas Phinney said:
    “ Typographers' beliefs even being confirmed later is just a belief. They didn't make a research and then produced characters.”

    Not true. There have been many studies that worked exactly that way! Some done by psychologists, some by type designers. I literally cited entire books, which cite many such studies.

    “So in regards to this "bionic", did anyone made a research about it and proven it's not working? Again I emphasize, I don't claim that it is working or not. I don't know.”

    Not only has there been that research (as I already said!); there has even been a second study, one that also finds no statistically meaningful effect. 
    https://pubmed.ncbi.nlm.nih.gov/38723450/
    The first study is here: https://blog.readwise.io/bionic-reading-results/

    (Note that the effect is actually negative, in both studies. But small and not statistically significant.)
    I just hardly noticed quote link below as an option. So much to say about UX at this site.

    I meant in early stage of typeface design developement. Not about later research. Typeface design historically is not based on research but on the timeframe practice.

    I could say as well for this "bionic" unless they make claims which are falsely presented as correct and not and pure belief or opinion.

    The first link you sent couldn't be open. It has some error, but I will read these other you sent. Thank you.
  • Thomas Phinney
    Thomas Phinney Posts: 2,851
    edited May 28
    Sorry that link does not work for you! Maybe it is geographic limitations? Snell’s article is also posted in many other places. Here are some of them (or just search on “Snell bionic”):
    “I meant in early stage of typeface design developement. Not about later research. Typeface design historically is not based on research but on the timeframe practice.”

    Typeface design is often influenced by current understandings of legibility, and this in turn is influenced by research.

    I can even think of a case where a particular feature, in a particular (very popular and widely-used) typeface, is there because the type designer misunderstood a particular term/phrase in legibility research.
  • Craig Eliason
    Craig Eliason Posts: 1,433
    adamwhite said:
    The first link you sent couldn't be open. It has some error, but I will read these other you sent. Thank you.
    If you mean the nowpublishers link, it's because the URL picks up the close-parentheses at the end in Tom's post. Clicking the link then adjusting the url to omit the  ) worked for me. 
  • Thomas Phinney
    Thomas Phinney Posts: 2,851
    Ah, quite right. Perhaps some helpful admin can fix that in the original post.
  • adamwhite
    adamwhite Posts: 23
    @Thomas Phinney Yes @Craig Eliason was right it's because of the parentheses at the end.  Why admin, can't you just edit your comment and delete both parentheses?

    Thank you for all resources. As I mentioned stem width in my post. Can you share any resources about decission practices for a starting size of a stem width in typeface production throughout history?

    Frank E. Blokland writes many details at https://www.lettermodel.org/ about origins of character proportions and possible grid systems, but I didn't encouter any sources for stem width specially. 
  • adamwhite
    adamwhite Posts: 23
    adamwhite said:
    @Thomas Phinney Yes @Craig Eliason was right it's because of the parentheses at the end.  Why admin, can't you just edit your comment and delete both parentheses?

    Thank you for all resources. As I mentioned stem width in my post. Can you share any resources about decission practices for a starting size of a stem width in typeface production throughout history?

    Frank E. Blokland writes many details at https://www.lettermodel.org/ about origins of character proportions and possible grid systems, but I didn't encouter any sources for stem width specially. 
    I mean he has this article: https://www.lettermodel.org/stem_widths.html and a LeMo method, but I mean some other sources like books, articles, and researches.
  • Thomas Phinney
    Thomas Phinney Posts: 2,851
    (Ah, I can indeed just edit the post! But for most of the history of this website, they disallowed edits on posts that were more than 4 hours old. Once one has gotten used to the limitation, one stops trying. Perhaps this is new with the recent software update and redesign.)

    Blokland’s website is an introduction to his PhD dissertation, and it links to the dissertation online at https://openaccess.leidenuniv.nl/handle/1887/43556

    You can buy the printed book version online from websites such as this one: https://www.lulu.com/de/shop/frank-e-blokland/on-the-origin-of-patterning-in-movable-latin-type-bw-trade-edition/paperback/product-1erzj7j6.html?page=1&pageSize=4

  • John Savard
    John Savard Posts: 1,121
    It may well be true that designing typefaces with a mindset keyed to Roman characters and mediaeval penmanship limits their applicability to new media in the digital age. But I was amused when I read that statement none the less: because I remembered how one of the most notable typeface designs which attempted to break out of the traditional straitjacket, Hermann Zapf's Optima, ended up unsuitable for laser printers (at least at low resolutions near 300 dpi) never mind screens, because instead of serifs, it used a gentle tapering of the strokes that was hard to reproduce with discrete pixels.
    When Marshall McLuhan wrote "The Medium is the Message", he had something else in mind also; that even if old content fits a new medium without problem, its meaning is transformed by the fact that the new medium gets used differently.

    First, let's look at what my Optima example suggests. Letters aren't written on the screen of a computer with a quill pen. So this definitely does give great freedom in the shapes of letterforms. But this is not new. Think of Calypso or Bifur. Obviously phototypesetting gave a near-absolute freedom in designing the shapes of letterforms - although phototypesetting would normally be followed by lithography, and while it didn't have the same level of problems with ink traps that letterpress did, I'm sure it still imposed restrictions.
    And just look at a lot of 19th century specimen books, which often contained very ornate display typefaces.
    If the freedom to design radically different letterforms has already been around for a long time, might it not be reasonable to conclude that digital media, of themselves, will not change the habits and preferences of the human beings reading text on the other side of the screen?
    Perhaps the biggest change is that the screen glows of itself, instead of the white background of text simply supplying a matte reflection like paper... which makes eyestrain a greater concern in the digital age.
    With respect to technologies with greater destructive potential than the art of typefounding, "we must because we can" has often been a very bad idea.

    On Star Trek: The Next Generation, a computer screen in the background depicted the language of one alien race, consisting of characters which were animated spheres with colored patterns on them.
    Our technology now allows us to prepare documents for the use of such beings, should the need arise.
    But we are not such beings, whose psychology is such as to make that a more effective means of written communication for us than plain old letters.
    Our needs and wants are such that Parkinson's Law still works - we use new technologies to fulfill our needs and wants until their limits are reached and we need to deal with the consequences. Some new technologies do provoke new desires: computers let us play video games instead of just board games, for example. But the process doesn't need to be forced, especially for something as basic as reading - which works best if texts are in such a form as to be easy and quick to read.

    This response, of course, only presents one side of the question. I admit that new technology in media probably does open exciting new possibilities in typeface design. But what these possibilities are - what innovations which would be genuinely useful, instead of being different for the sake of being different, and hence being rejected - will be visible to people with rare levels of insight. Those without such inspiration are better off not wasting their time, I fear.
  • adamwhite
    adamwhite Posts: 23
    @Thomas Phinney Thank you.

    @John Savard Those are interesting examples. So many factors to think about.
  • Vasil Stanev
    Vasil Stanev Posts: 775
    edited June 21
    I have carefully read through the thread (although the books so kindly cited will have to wait for a more appropriate time) and have been interested to a certain extent about the topic. It seems to me that this ties with a bigger trend in modern society - the trend to optimize everything, even the things that (perhaps) don't need fixing. I was first made aware of this trend when I got spammed with a video on YouTube (I think it was), titled "Ten ways you have been drinking water wrong".
    I happen to be fond of reading and have read many books. At no point did I think that reading should be faster to save me some time - devices and apps already save me more than enough, at least in my case. I think rather the opposite should be true - since it gets easier and easier to use AI to work with text, summarize it and so on, reading a large text should retreat back to a laid back activity - one gets a paper book from  a brick and mortar bookshop or library and allows themselves to spend several pleasant hours with an object that doesn't feature pop-up ads, with no phone ringing, just a break from the insane rhythm of work life. If I would need to speedread through a technical text, instead of changing the font, I would convert it to audio and listen to it at 2x speed. It has been maybe 2-3 years since I started "speed listening" through YouTube videos, and some of them were audiobooks.
    I remember that my grandmother used to be able to read a book of about 200 pages in 1-2 hours, while I was able to do it in about 40-50 pages per hour. But this doesn't mean I was able to note down important topics and remember everything. I recall that I used to e.g. read the word "tea" in the book and my mind would spend 1-2 minutes free associating words until I realize that my eyes were going over the text while I thought about Asian tea trade and remembered nothing. Is this going to get better with another font? What if by speeding up we actually get into other problems? This free association might be a sign that the book is boring - it is the job of the author to make it engaging, to make you want to read from cover to cover in one sitting.
    Saying that, it is commendable that the author of the topic wanted to do a good deed for many people *applause*, but I think it's unnecessary.
  • John Savard
    John Savard Posts: 1,121
    edited June 22
    Vasil Stanev said:
    If I would need to speedread through a technical text, instead of changing the font, I would convert it to audio and listen to it at 2x speed. It has been maybe 2-3 years since I started "speed listening" through YouTube videos, and some of them were audiobooks.

    I don't know about you, but when I read a technical text, I want to be able to look back, and perhaps reread what I saw in a previous paragraph. That is much easier to do with the printed page than with any kind of audiobook. Also, technical texts often have diagrams and tables.
  • Vasil Stanev
    Vasil Stanev Posts: 775
    @John Savard
    True, true.
  • Mosh
    Mosh Posts: 1
    Thomas Phinney said:
    • we do not “read word shapes” (boumas) but rather, at least primarily, recognize letter shapes in parallel
    Hi, Thomas,

    do you have any link or reference to this? I tend to mention the bouma theory to my students when they start learning type composition, but it would serve me good to read about negations to the theory.
  • Nick Shinn
    Nick Shinn Posts: 2,188
    edited July 14
    Is this anti-Latin thing woke or neo-liberal globalization? 
  • Thomas Phinney
    Thomas Phinney Posts: 2,851
    Mosh said:
    Thomas Phinney said:
    • we do not “read word shapes” (boumas) but rather, at least primarily, recognize letter shapes in parallel
    Hi, Thomas,

    do you have any link or reference to this? I tend to mention the bouma theory to my students when they start learning type composition, but it would serve me good to read about negations to the theory.

    https://learn.microsoft.com/en-us/typography/develop/word-recognition

    The bouma theory is basically bullshit. Parallel letterwise recognition is the primary factor in reading words. There is a vast amount of evidence that says so.

    There is a single very vociferous advocate of word-shape and “boumas” in the type community. They claim that immersive reading is different than anything being tested, and that under the right conditions it will be proven to be so. But they have no evidence for this. My understanding is that pretty much every test that has been done that would distinguish word shape from parallel letterwise recognition as a theory of reading, which has shown a difference, has favored parallel letterwise recognition as the primary factor.

    Note that some of what makes word shapes more different is ALSO making the individual letters more distinctive. Lowercase letters contribute to individually distinctive letters as well as to distinctive word shapes.

    Addendum: one of the many problems with word shape is that there are an awful lot of similar word shapes. e.g.
    Lowercase
    Immersive
  • John Savard
    John Savard Posts: 1,121
    edited July 23
    Nick Shinn said:
    Is this anti-Latin thing woke or neo-liberal globalization? 

    I think I have to be careful in replying, as we don't want to get into politics here.
    The LinkedIn post contains this sentence:

    Traditional typeface creation is still deeply rooted in Roman characters and medieval print craftsmanship, a legacy that limits its potential across a wide range of media applications.

    So the issue isn't primarily the potentially political one of limitations in typeface design as currently practiced in supporting non-Latin scripts, such as the Arabic script or scripts using Chinese characters.
    The design of types for the Latin script - and Cyrillic and Greek - is indeed heavily rooted in calligraphy and punchcutting.
    Sure, we have display faces like Bifur. But when it comes to graceful artistic typefaces for setting books... nothing seems to beat a Jenson. So we have the Doves type, Centaur, and so on as the most praised typefaces, and considered the most beautiful.
    If people like it and it works, why fix what ain't broke?
    These days, ink on paper isn't the only thing to which literacy is applied. People have already done a lot of work on making redesigned typefaces that work better on computer screens, and such work will continue.
    But I think I've already stated my opinion on this. It's true that the computer screen is a new medium. But it's also true that typography has many years of experience behind it, and so while the adaptations needed to cope with low-resolution displays were obvious, attempting to optimize type design anew for the modern age from a fresh start will require designers who are geniuses, not mere iconoclasts.
    The issue isn't wokeness, it's the impatience of youth.
    In fact, I think I can make it more explcit.
    (I claim no originality for what follows. I am insufficiently well-versed in the classics of literature to be able to name who it is, among the authors of ancient Greece or Rome, or perhaps of the Enlightenment, that I merely parrot below... but I know enough to suspect that is the case.)
    Let us begin with the premise: The computer screen, and other visual multimedia, consitute a new environment for print, one which is significantly different ink on paper. Therefore, we should be seeing original and novel typefaces suited to the new medium.
    Then, if that is not the case, three possible explanations suggest themselves:
    • No one has had the wit to devise such typefaces yet.
    • Even when new typefaces are devised that are of genuine merit, unless they directly meet someone's immediate need, it's easy for them to be overlooked, and hard for them to get noticed.
    • The typographic community is hidebound in tradition and prejudice, always turning to the old and not open to the new.
    Now, what is the difference between these three explanations? The first two are... discouraging. They indicate that it may take a long time for there to be new typefaces for the new media, and don't immediately suggest a plan of action to change that.
    But the third? Ah, that one gives something to do, and even someone to blame! So it's human nature to opt for the third explanation.
    And it's not as if there's no evidence for the third explanation; the typographic community is very fond of the old typefaces it has that have stood the test of time.


Categories