Are there scientific experiments, measurements on these? Which fonts make you read certain words twice, sometimes letter by letter? Fonts might also differ in how tiring they are for the eyes and brain. As billions of people spend hours reading day by day, even tiny efficiency improvements have a significant value, much bigger value than the one-off cost of designing a most efficient font.

Which font has clearly distinct lI|1, O0, rnm? To make reading faster, less ambiguous, less error/prone: can you design a font with visually more distinct signs: e.g. capital i, lowercase l, 1 (one), | (the "or" sign) should be more distinct; capital o [a circle] and zero [not only narrow, but also tapering, almost like a rhombus]; r before n: rn should differ from m, c before l: cl should differ more from d [the vertical line of d could continue further down; this way would differ from the mirror of b too]. q should have the right curve appendix from its bottom point to make it different from 9 and the mirror image of p.

  • 3
    I would think this would require an in-depth, multi-user, long-term scientific study to answer effectively. Anything short of that would be purely opinion-based. – Scott May 20 '15 at 21:08
  • As it stands, this question is too broad (or unclear) in my opinion. There are several questions here (experiments on readability, the titular question, your requriments). To add to this, many of your criteria are actually not very relevant for legibility. For example, in many contexts the O–0 distinction does not matter, because it is perfectly clear from context, whether you have a letter or a number. – Wrzlprmft May 21 '15 at 5:54
  • 1
    FYI you're asking about two separate things. "Quickest to read" = "readability" where the research is murky regarding typeface characteristics (but much clearer on things like line spacing, measure aka column width, etc). "Least misread" = "legibility" and the research is better quality, used for things like road signs, license plates, etc. Typefaces can be good for legibility and bad for readability (e.g. BLOCK CAPS, monospace), and visa versa (e.g. light low x-height) – user56reinstatemonica8 May 21 '15 at 8:38
  • 1
    In some contexts the O-0 distinction matters. "They wrote me my blood is Type A, but it is Type O." – Janos Ratkai May 26 '15 at 22:18
  • I think the question is interesting, but a bit too broad. I would probably make it more in derection "why font designers don't try to introduce more difference between characters". And this approach obviously could improve readabilty much more than many other font features. – Mikhail V Jul 7 '16 at 21:54

Are there scientific experiments, measurements on these?

Yes! But...

They are usually

  • inconclusive
  • use an incredibly small sample of users
  • are overly narrow in scope
  • tend to lack a lot of context
  • tend to ignore all the other aspects that go in to readability

So, I wouldn't put much weight into it at least on the broad "what is the best typeface" level.

|improve this answer|||||
  • 1
    So basically... "It's a matter of opinion" then. – Scott May 20 '15 at 23:12
  • @Scott in general, yes. I suppose in very specific cases (defined by the narrow research in a lot of these studies) you could argue there's some data, but I don't think I'd call it conclusive data. – DA01 May 20 '15 at 23:45
  • 1
    I agree - the data is pretty inconclusive and seems to show that it's mostly a matter of opinion, personal preferences, and cultural norms. – bemdesign May 21 '15 at 3:58

If you're concerned with readability and distinct letter forms, and not so much with style, then the Dyslexia font (and others like it) might be worth considering.


|improve this answer|||||
  • That doesn't seem easiest for me to read (I don't have dyslexia) – Zach Saucier May 21 '15 at 4:00

Very good question. And a very complex one.

I just make some general statements.

1) Serif letters on printed media are easier to read with small line height spacing. (probably too on a high definition device)

Proportionally sans serif fonts need more line height so the eye can keep track of the line its reading.

The main reason is that the serif helps the eye making an horizontal pattern.

2) A condensed version is a little harder to read than an extended version.

3) A monospaced font is widley used in programming becouse programmers need to see not only the general idea, but the exact information they are reading and writting. This is also true with numbers. The vertical align monospaced fonts gives is very helpfull to spot some errors, for example 10000 vs 100000.

They need to read not only the line on which a character is but sometimes the column onwhere that character is.

An interesting exercise:

I cnduo't bvleiee taht I culod aulaclty uesdtannrd waht I was rdnaieg. Unisg the icndeblire pweor of the hmuan mnid, aocdcrnig to rseecrah at Cmabrigde Uinervtisy, it dseno't mttaer in waht oderr the lterets in a wrod are, the olny irpoamtnt tihng is taht the frsit and lsat ltteer be in the rhgit pclae.

You can not do that in programming or an acounting book.

4) On small blocks of text, the shape and weight of the font helps reading the meaning. For example the word no is easier to read if I do this: NO.

5) In general terms 2 very important aspects are spacing and white space. If you don't have clear spacing and you have uncntrolled white blocks the eye does not make a continuos movement.

This is for the western roman-cyrilic typefaces. For other character sets I have no idea.

|improve this answer|||||
  • 1
    1) is opinion. I've never seen any conclusive data that shows serifs have any bearing on one's ability to read a line of text. 2) Likely true, but context is important. 3) is incorrect. Monospaced fonts are used by programmers to make it easy to align things vertically in the code editor 4) this is often called the bouma and there have been studies that disprove it as much as prove it. 5) absolutely true! – DA01 May 20 '15 at 21:51
  • @DA01 1) I have, but than i have seen the other way around as conclusive too. (hence a inconclusive metastudy :P ) Too many unknown variables to measure it proper. – joojaa May 20 '15 at 22:17
  • @joojaa true, that's probably a better way to say it. – DA01 May 20 '15 at 23:46
  • @DA01 "Monospaced fonts are used by programmers to make it easy to align things vertically in the code editor" Nope, monospaced is still the only chioce in many IDEs (VIM for example) because it would be much harder to implement IDE features like syntax highlighting, selection functionality, and other rendering and interaction functionality. Not much to do with vertical alignment, which is easily implemented with proportional font just by implementing tab stops. – Mikhail V Jul 7 '16 at 21:32
  • @MikhailV you can't have tab stops everywhere in your code. Also, many IDEs allow any font. So, the use of a monospaced font is typically merely a preference by the user...often a preference for wanting columns of characters lining up. I don't see how whether a font is monospaced or not has anything to do with syntax highlighting or selecting text. As for VIM, that could simply be a limitation of the UI VIM is being used within (typically a command line/terminal UI which has traditionally been monospaced due to...I'm not sure...history/habit?) – DA01 Jul 8 '16 at 15:44

Not the answer you're looking for? Browse other questions tagged or ask your own question.