LaTeX makes things more legible by following certain rules, and there is a strong emphasis among LaTeX users on obeying typographical conventions to make things more legible.

What happens when we break those? Can you give examples that demonstrate why we have certain typographic conventions by providing counter examples? i.e. if we run the following code we can see that the text is still legible, but much harder to read. I think this would be a fun & easy way to learn why we have them instead of just saying "Do this, don't do this"

For example, I use a font called ProFontWin. I use it for programming as you never mix letters up with one another, but it is terrible to read body text in as the eye doesn't flow over it. Similarly using a monospaced font makes each letter easier to see, but the text harder to read. These are both typeface examples: What are some typesetting examples? Preferably with horrible, terrible, no good, very bad, counter examples.

The most obvious one I can think of is spacing: If you changed the spacing between words it would be harder to follow the text (Example: Justify anything in MS Word)

The story behind the question: I recently had to jump through a bunch of hoops due to changing regulations at my university, and for a few moments contemplated turning in one of the required documents in the most illegible typeface I had installed (Blackadder ITC, Old English Text MT or something), so that it was technically handed in, but really annoying to read. I decided annoying the department secretaries (aka the real power in the department) would be a Bad Idea, but this question came out of it.

  • 1
    If someone wants to set this as a communitywiki question I'm fine with that; I wasn't sure if it should be one or not.
    – Canageek
    Nov 24, 2011 at 1:35
  • 11
    Take any standard LaTeX document and load the wordlike package and see what happens. Nov 24, 2011 at 2:22
  • Huh, didn't you have that as an answer?
    – Canageek
    Nov 24, 2011 at 4:51
  • Yes, it was an answer, but I do not have time now to add explanations, so I decided to delete it and simply make a comment. Nov 24, 2011 at 4:58
  • 3
    See page 31 of Guidelines for Typography in NBCS: How not to format a document.
    – lockstep
    Nov 24, 2011 at 17:54

4 Answers 4


Readability is not simply "being able to read each word", but also being able to understand the relationships between the words, between the paragraphs and so on.

Titles are put into evidence because that's their role: stick out as much as necessary for the reader to spot them and know that something has changed. Paragraphs are indented for the same reason and lists are given a different treatment because they are different from normal paragraphs. The interline leading is not too wide so that the eye can catch easily the next start.

Here's an example where some of the "rules" above are broken: one can probably read that text, but not as fluently as if it were well presented.

enter image description here


Avoid line breaks within abbreviations, and use a thinner space than between normal words. LaTeX provides \thinspace for this, and a shortcut \,.

This example shows how it looks

  • without the space
  • normal spaces, where line breaks can occur within abbreviations
  • a non-breakable space, so no line break but the abbreviation looks too wide
  • a thin space, non-breakable

Please ignore the wide interword spacing because of justification in a narrow column. In such cases, generally better use ragged text instead of full justification, here it's just to keep the example short, the focus is on the abbreviations i. e. and e. g.

spacing abbreviations example

Dummy text, i.e. some blind text just for a test, e.g. in a minipage environment.
Dummy text, i. e. some blind text just for a test, e. g. in a minipage environment.
Dummy text, i.~e. some blind text just for a test, e.~g. in a minipage environment.
Dummy text, i.\,e. some blind text just for a test, e.\,g. in a minipage environment.
  • 5
    The thin space in i.e. is a "germanism". :)
    – egreg
    Nov 24, 2011 at 11:49
  • 1
    Ok, so it's at least for German typography, even if we have to write in English language here. And the German user base is big ;-)
    – Stefan Kottwitz
    Nov 24, 2011 at 12:16

There is a package called manuscript, which is a small gem. Try this to as a demonstration of bad type, wide paragraphs and double lined texts.


Part of the results:

enter image description here

Submit all drafts of your thesis using this style to inject typographical awareness onto your thesis supervisor!

  • 5
    40% of my mark for 9 credits is solely at his discretion, and he has been commenting on how professional my LaTeX documents look. There are many other people I'd rather punish with bad typesetting. Also: Dear lord, that would take lots of ink.
    – Canageek
    Nov 24, 2011 at 20:17

This is a slightly lengthy comment.

Nothing happens if we all do it. Same applies to Comic Sans font or running naked in the city with an obligation to salute people in the most interesting ways or any other widely accepted dogma-ish concept out there.

The special fact about typography in general is that it heavily relies on experience and user feedback (Incredibly and unfortunately, this also applies to Comic Sans, that's why it resists to vanish.). So they are not created out of affinity or acquaintance to some lost religious cult but real trial and error, and boy, mostly error. I don't know too much about typography but the difference between older! versions of windows and macs show you how it makes a difference if you work on it for hours (which is only a very limited aspect of bad typography. By the way, this comes from someone who dislikes Macs, I digress).

Lastly, I give the floor to the master for the discussion on taking it for granted :P

  • Thus why I am trying to ask for demonstrations of what makes things harder to read; I'm training to be a scientist in real life, so I want to see the experimental evidence for certain things being harder to read. We've found some things make it easier to read: English switched scripts a few hundred years ago (Shortly before 1100 AD as I recall, but I could be a bit off), German moved from Fraktur to (mostly) Latin script, and so on.
    – Canageek
    Nov 24, 2011 at 4:54
  • @percusse Yes, a lot of typographical "rules" are based on meeting the readers' expectations -- but this actually the point: in doing so, the design of the text doesn't distract the reader from reading and understanding. This is actually also the point of having an orthography. But most of these typographical rules are not arbitrary but based on centuries' observations of what works best. You can also see this from the fact that some of the most crucial rules (e.g. line length) are observed internationally in nearly all writing traditions.
    – Florian
    Nov 24, 2011 at 12:00
  • @Canageek OT-ish: In Germany there was a lot of discussion on the readability of Blackletter (Fraktur is just one of its styles) vs Antiqua and experiments were conducted that resulted either way. But in the end the decision was made on aesthetical and mainly political grounds. Writing styles in the pre-printing times are a much more complex story still, but there was no such thing as a simple switch at a given time.
    – Florian
    Nov 24, 2011 at 12:12

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .