The Science of Serif vs Sans Serif Type | Hacker News Reader

The Science of Serif vs Sans Serif Type

alexpoole.info

75 pointsidan16 comments

TL;DR there's no science to back up one being better than the other. Other attributes of a typeface can have an effect for certain mediums (example, large x-height making type readable on screens).

Choose typefaces to match the tone of your content. See also: http://opinionator.blogs.nytimes.com/2012/08/08/hear-all-ye-...

"We should accept that most reasonably designed typefaces in mainstream use will be equally legible, and that it makes much more sense to argue in favour of serif or sans serif typefaces on aesthetic grounds than on the question of legibility."

Exactly. It's an aesthetic choice. Arguing for legibility or readability or any scientific purpose is a secondary argument to simple design sense.

Typographical choices involving both styles of font can work great; it's up to the designer to make the font match the target style for the correct functional and emotional impact. Simple as that.

Scientific arguments over fonts and typography have always seemed to me like analyzing the chemical composition of paint on a canvas to determine its value. Completely missing the point. Is there a certain measurable utility? Of course. But when a design works, and evokes the correct emotions for the presented work in a readable and understandable way, it just works. It's more functional for being beautiful.

Scientific studies of fonts only 'miss the point' insofar as they don't find anything useful... for example, failing to find any real, consistent legibility difference depending on serifs.

If you are not just pleasing yourself whimsically, the point of fonts (like any other thing) is not just to do whatever you want, regardless of whether it is effective or not. Even in the realm of aesthetics. Unless those aesthetics are for my benefit only, I am probably trying to get SOMETHING done.

And unless the way it works is obvious (bearing in mind that many obvious things are actually untrue, and that many true things are not obvious)... it is reasonable to depend on research.

If my art tries to convey sadness but I am so out of touch on how to do that (comic sans) that my peers laugh at me instead, that is likely a failure. If I write a blog post meaning to persuade people to back up their data but rather make people feel content without backups (comic sans), I have failed. If I try to get attention for my product and am ignored because I used comic sans, I have failed.

On the other hand, if comic sans is beautiful to my audience, and if it works for my boss - I had better consider comic sans ;)

The reason we don't follow any recommendations, when we don't, isn't that actually finding things out is point-missing, but rather just that nothing has been found out (yet).

Very observant and lots of truth, thanks for the reply. In my opinion, the art and science of design takes all that into account. It's not simply aesthetic, but also purpose and utility as you say. It's a combination of art and science, but like most things, if you look at it strictly from one side or the other, you fail to grasp the whole and quality will suffer.

TL;DR

  Startup founders, pay attention to the

  1. aesthetics

  2. recognizability

  3. psychology of your logo type.

The article had to do more with the aesthetics than the science behind the types. I wanted to add that Microsoft did a lot of research on this in the 80s and 90s with the development of their Office Suite (more recently here: http://www.microsoft.com/typography/ctfonts/wordrecognition....) . I particularly feel that startup founders need to pay attention to the recognizability of their logos given the word shape and neural network modeling hypotheses given in the Microsoft paper. And it's one reason why I switched my logo from being all caps (like ORACLE) to all lower case (like facebook), although cursive (like airbnb) has its advantages as well.

On the other hand, there's the psychology of serif fonts which was left out in the article (i.e. research to support that documents in serif type is viewed with more importance) and is the reasoning behind law books, legal documents and a lot of résumés written in serif type.

No mention of acutance at all, surprising. There's a lot to be said about the way the human visual system responds to spatial frequencies, above all else.

http://en.wikipedia.org/wiki/Acutance

http://en.wikipedia.org/wiki/Ringing_artifact

The way I see it, serifs may have evolved as a way to emphasise and deemphasise the various spatial frequencies in a given letter. A sharp 90-degree line at the end of a stem may help 'cut off' and locally restrict the frequencies represented by the stem in a way that makes them more easily distinguishable by the visual cortex.

Try flicking between these two images and you might get an idea of what I mean:

Normal: http://i.imgur.com/NPAh3.png

High-passed: http://i.imgur.com/IoHCC.png

This is interesting speculation. But it is premature until you know that serif fonts are more legible - that's not something one should simply take for granted.

From the little I've looked into this claim, it has apparently never been demonstrated despite much trying; if there is any inherent legibility difference between serif and sans-serif fonts then it is small enough to consistently get lost in sample noise. (There may be some recent meta-analysis or something I don't know about, though)

Where serifs are preferred, the real reasons are cultural. Serif letters look like the ones carved into Roman monuments and the ones in all the academic journals. But there is no basic cognitive reason why we might not give that kind of prestige to runes, cuneiform, or Arabic in an alternate history.

I think what has to be explained is not the legibility or discriminability of serif fonts (again since I don't think that has really been demonstrated) but rather the preference for them.

It seems to me that many font connoisseurs and font-makers attach a lot of prestige to immersion in the traditional craft, and find less-traditional geometric constructions to be ugly (i.e.: as a cultural judgement, like finding Middle Eastern music dissonant). It happens that this geometric tendency is easier to find in sans-serif fonts.

Maybe this is also part of why sans-serif fonts have become so important to advertising, where there has been a premium on the new and advanced for at least a century. (And this would also add another reason for preference of serif fonts: sans-serif is associated with mass production and consumption, and the default fonts on computers; thus more subject to snobbery).

I did some work here.

There's a lot of problems with legibility research. It's very, very hard to test for "legibility" or "readability" -- as, not surprisingly, the typeface is only a small part of what's going on when the human brain turns symbols into thoughts then plays them back later.

To the extent that we could get valid data, people did best when they were familiar with the typeface in question. And back in the age of newspapers, the closer to the local newspaper layout and typography, the better they scored with news-like information.

melloclello's comment hints at this part: serifs are high-frequency data. So there's more "information" in a serif face than a sans-serif face. In fact, simple filtering of a serif face makes something that looks very much like a sans-serif face. But whether this extra information is helpful or distracting to readability or legibility depends on what the reader is used to. (But even that didn't matter much in our tests!)

I've gone through Design for Hackers by David Kadavy a few times already, and Part V is fantastic to gather some context and understanding of fonts and typography.

PS: the whole book is IMHO a great read for people who want to know more about the fundamentals of design. At the very least it will make non-designers able to communicate with designers better.

Maybe it is confirmation bias, but I can't imagine reading a book or even a long magazine article in sans-serif.

In my experience, it's every bit as bad as you imagine:

When I was an undergrad, for whatever reason Univers was the default typeface used for computer document production (much of which was done using the Scribe markup language), and as a result, tons of lengthy documents were printed in Univers, including a lot of stuff that was book-length. [Why it was Univers I'm not sure. I don't think they had very many fonts though.]

It was absolutely dreadful: reading page after page of text set in Univers was downright exhausting.

I think Univers is not such a bad typeface in the abstract, but because of the way it was used, I came to loathe it (and right now looking at some Univers examples on the web, I still feel a twinge of disgust...!).

After I had been there a few years, other document production systems (such as TeX), and newer printing technology (like the Laserwriter I with PostScript!) came into use, and the number of supported fonts increased. Many people shifted to serif fonts for their documents, which was a huge relief...

Oh man, my first university used Univers (and not just Univers, the condensed version) for all their signage and handouts as well.

I almost feel like with a font like Univers the similarities between letters could saturate out your pattern recognition quite easily (imagine looking at these same stems and bowls all day long), and that perhaps serifs (or at very least, differing lettershapes) add a much-needed bit of stochastic resonance to the whole process. Maybe the uniformity of modernist typefaces isn't all it cracked up to be. </idle speculation>

An exaggerated example of what I'm talking about: http://i.imgur.com/Ypwtm.png

(I wish I had a Discrete Fourier Transform plugin for Photoshop so I could try it with these two examples)

Although I've heard many people say that sans-serif fonts are only for low DPI environments I can't stand serif fonts anywhere no matter the size or display, not even on paper or my Kindle.

I don't know what is the cause, I don't have any vision problems apart from myopia. When I need to read more than a few sentences in serif fonts (like in OS X's Dictionary app) I usually give up and copy-paste the text to somewhere I can view it in a sans-serif font. I've also configured Firefox to show everything in Ubuntu Regular[1].

It started as a preference but maybe by doing that I'm getting increasingly lazier and losing the ability to comprehend different symbols.

[1] http://font.ubuntu.com/

Care to tell us how old your are? It might just be what you're culturally accustomed to. Just like blackletter or half-uncials look weird to almost anyone nowadays.

For me it's basically the other way around, I find almost any sans-serif (and most slab serifs) to be esthetically unpleasing. It's like reading a book set in highway/emergency signage. Don't even get me started about Helvetica... (And I'm not even that old - 34. Maybe I never should've read that Tschichold book)

Oh, sorry didn't mean to leave that out. I'm 22 years old, I recently started using glasses for myopia (it didn't affect my preference regarding text) and I don't have astigmatism.

I agree with you on the aesthetics of serifs, I prefer it when I'm admiring how beautiful a piece of paper looks but when I'm reading I find them illegible. Same goes with cursive, although I can read it just fine I hate every minute of it. I also don't have a nice handwriting (don't know if these tell you something).

It is funny how aesthetic choices differ. I unintentionally associate serif fonts with infographics, ads etc. that I find overdesigned and not my preferred ways to get information efficiently.

Don't let the art schools hear about this! LOL. Not that they're big on empirical stuff anyway. (Except as an occasional brief aside to "prove" why their arbitrary choice is "better" than anyone else's arbitrary choice.)

Reply on news.ycombinator.com