When did the "Serif vs. Sans" derby started?

Hello, everyone!

While I was writing some stuff, I began wondering when and how this idea that sans-serif type is more legible on screen started.

Do any of you have an idea about it? Cheers!

Comments

  • Thanks, Thomas!

    Well, this sounds obvious enough, my suspicion was also this. But you mention research about this topic; is it something I can still find?

    And what was the inaugural moment for screen serifs? Georgia? I mean, it was almost 30 years ago (1993), so is there probably some earlier successful attempts?

    Cheers!
  • Thanks, James!

    But technical challenges aside, did these early legibility/readability tests were done with high performing serifs? The degree of complexity of a, say, more calligraphic serif compared to a geometric sans is clear, as well as it's difficulty to rasterize; though (and presuming that it could be both done skillfully), how much would a serif underperform? If it would at all, I mean.

    But back on topic: I'm not advocating for one over another; I'm just curious on when and how this idea became popular.

    Great insight, James!
  • Chris LozosChris Lozos Posts: 1,458
    edited May 2017
    The idea became popular prior to the digital era, in the 1930s Miles Tinker's readability testing published in his book "Readability of Print." He concluded that serif type was best for continuous reading. This battle continued for decades with other conclusions like "We read best what we read most".  Screen resolution had a different set of operating factors than print, as Thomas points out.  My guess would be that the push back will continue between various forces with each side having a voice.
  • Nick ShinnNick Shinn Posts: 2,131
    edited May 2017
    Certainly, fonts do have rendering problems at low resolution, but how much distortion is acceptable is open to debate. Microsoft and Apple took different approaches, crisp vs. fuzzy. In as much as Microsoft foregrounded its sub-pixel-implementing fonts (including serifed) in readability theory, e.g. “Now Read This”, it did play to and reinforce the idea that there is something demonstrable about readability, and that sharpness is desirable; however, I concur with Thomas that it was just very very obvious that serifed type didn’t work as well as sans on pre-Retina screens.

    Times was actually not too bad, and with its finely tapered serifs has some finesse when scaled up in size.
  • Mark SimonsonMark Simonson Posts: 1,652
    The original Mac system included a serifed font, New York. It was the default in many of the included apps, such as Note Pad, MacWrite, even the Calculator. It was kind of a low-res bitmapped interpretation of ITC Garamond, Apple's corporate typeface at the time.

    The default was changed to Geneva (a sans serif) in one of the early system updates. I don't know what the reason for the change was, but I don't think anyone complained.

    I remember thinking it was kind of cool and didn't find it hard to read.

    All of this was before the use of outline fonts on-screen.

    It might be interesting to note that the default font in web browsers was (maybe still is) Times. The early web was Times all the time.
  • Maxim ZhukovMaxim Zhukov Posts: 74
    edited May 2017
    scannerlicker said:
    ⟨…⟩ you mention research about this topic; is it something I can still find?
    Chuck Bigelow did quite a bit of research of that subject in early and mid-1980s. Some of his findings and hypotheses were published in The Seybold Report, Scientific American, TUGboat, the RIDT conference proceedings, etc. See, for example:
  • The idea became popular prior to the digital era, in the 1930s Miles Tinker's readability testing published in his book "Readability of Print." He concluded that serif type was best for continuous reading. This battle continued for decades with other conclusions like "We read best what we read most".  Screen resolution had a different set of operating factors than print, as Thomas points out.  My guess would be that the push back will continue between various forces with each side having a voice.
    Well, it might continue. Although, personally, I believe that it's a somewhat stupid question, since there are many factors that makes a design well-performing, as well as ignoring the conditions for which a particular design was trying to respond.

    And "sans vs serif" is just a very superficial question. Which makes it charming.

    Thanks for your input!
  • Certainly, fonts do have rendering problems at low resolution, but how much distortion is acceptable is open to debate. Microsoft and Apple took different approaches, crisp vs. fuzzy. In as much as Microsoft foregrounded its sub-pixel-implementing fonts (including serifed) in readability theory, e.g. “Now Read This”, it did play to and reinforce the idea that there is something demonstrable about readability, and that sharpness is desirable; however, I concur with Thomas that it was just very very obvious that serifed type didn’t work as well as sans on pre-Retina screens.

    Times was actually not too bad, and with its finely tapered serifs has some finesse when scaled up in size.
    The original Mac system included a serifed font, New York. It was the default in many of the included apps, such as Note Pad, MacWrite, even the Calculator. It was kind of a low-res bitmapped interpretation of ITC Garamond, Apple's corporate typeface at the time.

    The default was changed to Geneva (a sans serif) in one of the early system updates. I don't know what the reason for the change was, but I don't think anyone complained.

    I remember thinking it was kind of cool and didn't find it hard to read.

    All of this was before the use of outline fonts on-screen.

    It might be interesting to note that the default font in web browsers was (maybe still is) Times. The early web was Times all the time.
    Nick, I think that Mark summed it up well: although, yes, when within a tiny pixel grid, a sans (due to its simplicity) is easier to produce satisfactory results, there were still examples of good serif typefaces, even in the era of bitmap fonts.

    For example, Carter's Georgia is another example of a typeface that ruled the web and low-resolution screens for decades. Still, when you mentioned "sharpness" as something "desirable", I read it as "clear identification of form", so we're on the same page, here. :)

    Thank you both for your input! Always good to learn from you guys!
  • scannerlicker said:
    ⟨…⟩ you mention research about this topic; is it something I can still find?
    Chuck Bigelow did quite a bit of research of that subject in early and mid-1980s. Some of his findings and hypotheses were published in The Seybold Report, Scientific American, TUGboat, the RIDT conference proceedings, etc. See, for example:
    This is gold, thank you Maxim!
  • Hello, everyone!

    While I was writing some stuff, I began wondering when and how this idea that sans-serif type is more legible on screen started.

    Do any of you have an idea about it? Cheers!
    Hi. I never noticed there was a derby :smiley: 
    I have been working with screens and typefaces since beginning of 80s in professional typesetting equpments and later on for the web. As a graphic designer I tried to use the most suitable fonts for the quest. In the beginning of web was only "Times" but when neaded sans-serif I did a graphic pic for that.
    There are sans-serifs that works bad on the screen and there are serifs that works perfect, and viceversa... And have always been. Seen and heard people mentioned this subject, but I really don't understand.
  • SiDanielsSiDaniels Posts: 277
    Some interesting notes on TNR being the web default... http://www.typophile.com/node/63435
  • SiDanielsSiDaniels Posts: 277
    In the early days of computing monospaced bitmap fonts were mostly sans (with the exception of cap I, lc i and j) because it was hard to cram serifs into an 8 x 8 pixel grid. I also think sans became the most popular style when we moved to proportional fonts as they could be a little narrower - getting a little more text in a dialog. 
  • Mark Simonson wrote:
    The original Mac system included a serifed font, New York. It was the default in many of the included apps, such as Note Pad, MacWrite, even the Calculator.
    Actually, Apple never seemed to make up their mind on this front. They switched back and forth between New York and Geneva several times.

    André
  • Mark SimonsonMark Simonson Posts: 1,652
    edited July 2017
    Hmm. Not sure about that. I bought my first Mac a month after they were launched in 1984, so I've used every version of MacOS. I could be wrong, but my recollection was that the switch only happened once, from New York to Geneva, as part of one of the early system updates, maybe 2.0.
  • 0.97 Used New York
    1.1 Used Geneva
    2.0 Was back to New York

    André
  • Ray LarabieRay Larabie Posts: 1,376
    The Commodore Amiga had a system font that was part serif, part sans.

    Amiga Regular at MyFonts
  • One aspect of this discussion, especially in digital screen context, that I find interesting and largely unexplored is the notion that stylistic aspects of the sans are of importance in its supposed better legibility, or at very least in its perceived dominance in this domain. When compared to serifs typefaces, sans can be ascribed a more utilitarian, clear and rational tone (in most types, anyway; certainly grotesques, most geometrics, and surely a good amount of humanists as well).
    It could be argued that computers, and screens as their typographic user interfaces, needed to have a typographic voice that echoes the functional approach of the machine itself, and how we humans expect it to look like. Or to put it a tad poignantly, a device that can in a split second make more calculations than a human in a lifetime simply has no room for the grace and refinement of an antiqua.
  • Nick ShinnNick Shinn Posts: 2,131
    edited July 2017
    Working as an art director in the early 1990s, I would try out different typefaces for new projects, and was struck by how old-fashioned serifed styles made things look—and retro was not in fashion except for scripts.

    This was particularly notable for body text, where the mantra had always been that serifs were de rigueur for legibility. So that idea was abandoned (or mothballed, at least).

    There were exceptions, Scala, for instance.
    And slightly earlier, before digital really took over the media, Licko’s Matrix was ubiquitous.



  • Mike DugganMike Duggan Posts: 237
    beautiful experiments. bitmap fonts ROCK

  • Mark SimonsonMark Simonson Posts: 1,652
    Interesting.

    After getting my first Mac in 1984, I signed up for the Apple developer program in order to get the "Font Editor" program that was part of the developer tools. Also included was an early looseleaf copy of Inside Macintosh. In the section about fonts, I recall that there were earlier, more generic names for the "city" fonts in screen captures illustrating font menus. "Chicago" was called "System" and so on. Unfortunately, I seem to have lost or tossed that copy of Inside Macintosh long ago, so I can't verify it.
  • André G. IsaakAndré G. Isaak Posts: 626
    edited August 2017
    @Mark:

    Here's a link to a .pdf which might be the one you had.

    https://archive.org/details/bitsavers_applemacInionalEdition1985_53888388

  • Mark SimonsonMark Simonson Posts: 1,652
    The one I had was earlier than that. That one is from 1985. The one I had was a much less polished draft version from mid 1984.
  • The one I had was earlier than that. That one is from 1985. The one I had was a much less polished draft version from mid 1984.
    Perhaps this copy from 1984 for two binders (instead of the earlier 3 binders):
    https://archive.org/details/bitsavers_applemacIn84_27699101
    (Here’s the The Font Manager Programmer’s Guide section.)

    and Volume 2:
    https://archive.org/details/bitsavers_applemacIn84_23664048
  • Mark SimonsonMark Simonson Posts: 1,652
    edited August 2017
    Thanks! Those are really interesting, but apparently not the copy I had. They used to mail packets with chapter updates every few weeks. Seems like it was in one of those.

    Anyway, this is getting way off-topic.
Sign In or Register to comment.