Microsoft to Choose New Default Font to Replace Calibri

2

Comments

  • They seem to be showing up now. I'm not sure why they weren't earlier (other cloud fonts were, just not those particular ones).
  • konrad ritter
    konrad ritter Posts: 204
    Yes to leaving the monoline shtick behind; and yes to making more room for serifs. We have more storage and computation power than ever on our devices, so, what's the excuse for not using more serifs in interfaces?
  • John Savard
    John Savard Posts: 1,135
    edited May 2021
    Yes to leaving the monoline shtick behind; and yes to making more room for serifs. We have more storage and computation power than ever on our devices, so, what's the excuse for not using more serifs in interfaces?

    The problem isn't with storage and computation power, it's with our displays. Even though the Mac has the Retina display, and smartphones also have displays with very small pixels... the default for a PC with a 4K display is to enlarge icons and text only by a small amount. And choosing a 4K resolution will also mean games will run much more slowly.
    So I guess it would be more correct to say that the problem is not really in our displays in the sense of display technology, but rather simply in how the Microsoft Windows operating system lets - or at least encourages - us to make use of the displays we may have.

    I agree it would also be cool for Microsoft to eventually eclipse Segrutiger in their UI widgets, though the multilingual part of that task would be monumental.

    Segoe resembles something by Adrian Frutiger? I mean, I know the story about Arial and Helvetica and Monotype Grotesque, but there I just wasn't paying attention - I never even looked much at Segoe except at low resolutions. After taking a look, though, I guess you could say so - it does resemble the typeface Frutiger quite a bit, not just more than it does, say, Futura or Univers, which could be excused because of the limited design space in sans-serif typefaces.
  • @John Savard Frutiger —> Myriad —> Segoe

    I wish MS would rely Sitka more. One proprietary style for branding, and the rest as the new default.
  • John Savard
    John Savard Posts: 1,135
    edited May 2021
    @John Savard Frutiger —> Myriad —> Segoe

    Ah. And, indeed, Segoe, like Myriad, has a true italic, instead of just slanted characters for an italic.
    As for Sitka: I had never even noticed it. I was aware of Georgia and Cambria. Sitka, having a wider minimum stroke width, is a serif face more suitable to things like interface use.
    There's even a Sitka Small. I tried - using a third-party program, WinAero Tweaker, since Windows doesn't seem to have a setting for this any more - to change the system font to that, but I found the time in the lower right corner no longer displayed properly, as if the colon character was not defined in the system font. Yet there certainly is a colon in Sitka Small.
  • John Hudson
    John Hudson Posts: 3,227
     And, indeed, Segoe, like Myriad, has a true italic, instead of just slanted characters for an italic.
    A true italic was also a characteristic of Frutiger Next, which came between Myriad and Segoe. As I recall, the Linotype case against Segoe was specifically that it infringed Frutiger Next.

  • Yves Michel
    Yves Michel Posts: 183
    I like Skeena very much. Compared to the others.

    But, on my Windows10 with a Microsoft 365 Word version 2104 build 13929.20296 (!FRENCH), not a sign of those new fonts, even if I checked the desired boxes in the parameters.
  • John Butler
    John Butler Posts: 297
    edited May 2021
    I agree Sitka should get wider exposure, and Sitka Small would make an excellent UI font. (I use Arno Pro Caption as my UI font on my Linux machine.) Perhaps it was effectively limited unreasonably by the design brief to promote IE’s Reading View. I get tired of reading every news website in Georgia when Sitka is sitting unused.

    THIS is news to me: according to this document, Office 365 also bundles four weights of Walbaum Text (four as in R I B BI, i.e. counting italics as weights) plus four weights of Walbaum Headline plus EIGHT weights of Walbaum Display. Not sure exactly which Walbaum; I assume it’s the latest Monotype version. I don’t subscribe to Office 365, or I could check it.

    Plus Tisa, Sabon Next, Quire Sans, one weight of Dante, and two strangely paired weights of Bembo.
  • John Hudson said:
    As I recall, the Linotype case against Segoe was specifically that it infringed Frutiger Next.
    And Adrian Frutiger thought Myriad was too close to Frutiger... There was once a backlash against humanist sans typefaces (just like there was one against neo-grots more recently) for one reason only: too many of them being made. Which is what neuters it for branding purposes (although people could possibly try again in a decade or three).
  • Nick Shinn
    Nick Shinn Posts: 2,216
    edited May 2021
    Even as the relative merits of these “typefaces” are being debated, I would imagine that engineers are hard at work developing the holy grail of One Font To Rule Them All—in other words, a single variable font which may be adjusted by various sliders in layout engines to resemble specific categories of typeface.

    Similar to how CMYK and RGB represent an infinity of colours, rather than colour swatches with names like “Dusty Rose”.

    As I understand it, serifed versions of these new MS faces will be added (along the lines of Plex being a sans/serif superfamily). A logical step then to add such things as x-height variants and horizontal scaling, activated by the apps.

    After all, this was the original strategy of those who developed Word and Quark XPress, with faux italicization, bold, and small cap features, and it does have a certain logic.

    The question then, is which of these new MS faces can be smoothly “slid” between, because if that is possible, there is redundancy that may be eliminated. Or one might say, should there be a slider for how much of a tail lower case /l has, or whether that is an either/or situation?

    The next level of this direction would be a generic “slider” that would, say, adjust the x-height of any font at the app level, without need of making a pair of masters. A job for AI training.


  • Whatever works well as Lucida Grande worked on the Mac will do. :)
    Given the screen example on Twitter, I would choose Skeena without a doubt, although Seaford‘s Italic works better, and not just for the descenders, but because it seems to me  letters like /a and /d have more recognizability at small sizes.
  • Seaford‘s Italic works better, and not just for the descenders
    How so?
  • SiDaniels
    SiDaniels Posts: 277
    Regarding the names. Initially I asked the designers to use codenames that were the names of Alaskan towns. Later I asked them to provide a list of options which went through the name clearance process. Tenorite a mineral name, Seaford a town in New York State, Grandview a scenic railway in Alaska (keeping the train font theme), Skeena a river in northern British Columbia and Bierstadt a mountain that Steve had climbed in his home state of Colorado. 

    John Butler - yes there's some randomness in the set, along with the new fonts we recently added functionality to hide some of these from the menu. So some of the odd-balls won't show up unless they are used in a document. We're still working through issues with this, but will be talking more about it in the future. 

    For those who didn't see the fonts initially or even now, you likely need to restart the apps or maybe reboot the machine. As a work around you could use the cloud apps via www.office.com where in theory they should have been available on day 1. 

    Agree on Sitka, one of the reasons I went with sans-first on this project. If we wanted to switch back to a serif default the choice was hiding in plain sight. 

       
  • Seaford‘s Italic works better, and not just for the descenders
    How so?
    I explained why. :)
  • SiDaniels said:
    Agree on Sitka, one of the reasons I went with sans-first on this project. If we wanted to switch back to a serif default the choice was hiding in plain sight.   
    So great to hear that!

    BTW could different programs have different defaults? For example PowerPoint՝ a sans, but Word՝ Sitka. Or Clippy could some back and say things like: "It looks like you're writing a Masters thesis, not about the bathroom being out of order. Would you like to choose something besides Horsey Sans?"
  • SiDaniels
    SiDaniels Posts: 277
    That's a great question. Lots of people have suggested different defaults, but we see a lot of copy paste between apps and until we have better solutions around that I don't want to encourage ransom note typography. 

    The AI you mention is certainly interesting but some (many) people don't want the bots to read their text. Ultimately through a variety of means we want people to move off of the defaults and select a font that's appropriate for their content. We have a variety of ways to do that today - templates, themes and the "Designer" feature that surfaces options that we've curated - Sitka is heavily promoted through this feature. We have other things to make this easier in the works too. 

    Success here I think is people selecting good fonts, rather than just switching back to Calibri, but even switching back to Calibri is fine with me. 
  • Astute points.
    SiDaniels said:
    some (many) people don't want the bots to read their text.
    Same here, but I think for better or worse (probably the latter) people have become desensitized to that, thanks to hovering digital assistants.

  • John Savard
    John Savard Posts: 1,135
    Even as the relative merits of these “typefaces” are being debated, I would imagine that engineers are hard at work developing the holy grail of One Font To Rule Them All—in other words, a single variable font which may be adjusted by various sliders in layout engines to resemble specific categories of typeface.

    Similar to how CMYK and RGB represent an infinity of colours, rather than colour swatches with names like “Dusty Rose”.

    As I understand it, serifed versions of these new MS faces will be added (along the lines of Plex being a sans/serif superfamily). A logical step then to add such things as x-height variants and horizontal scaling, activated by the apps.

    This reminds me of one of the Metamagical Themas columns by Douglas Hofstadter (author of Gödel, Escher, Bach: an Eternal Golden Braid) wherein, by using different versions of the letter A in various typefaces, he criticized what he saw as an overoptimistic comment on font parametrization by Donald Knuth.

    But while one couldn't invent, say, Calypso by twiddling dials on a parametrized font not designed explilcity to include that kind of typeface as a possibility, more pedestrian things like x-height and serif type or lack of same certainly could be parametrized.

    So one could design a parametrized font that had, say, Helvetica, Univers, Frutiger, News Gothic and Venus at its sans-serif end, and Jenson, Bembo, Baskerville, Times Roman, Clarendon and Century Expanded at its serif end, and parametrize other trivial things like x-height (so you could get faces that looked like Ionic No. 5 or Corona)... and claim that by twiddling the knobs on it, while you couldn't get every face, you could at least approximate anything that was "any good" for general reading.

    Akzidenz Grotesk, however, is not 80% Helvetica mixed with 20% News Gothic. And you couldn't even approximate Gill Sans or Futura on this. Of course, if one threw in Futura, by mixing Futura with Jenson, turning the serif dial to Futura and the letter geometry dial to Jenson, could you approximate Kabel?

  • Nick Shinn
    Nick Shinn Posts: 2,216
    I expect that ultimately typographers (“users”) will be able to parametricize between any two static fonts, with AI to smooth out the artefacts.

    Then the trick will be for type designers to come up with stuff that resists AI’s blandishments, as well as novel conceptual axes.
  • Fingers crossed it will cause the infillers to wake up.
  • John Savard
    John Savard Posts: 1,135
    To me, the obvious sneaky use of this technology would be: parametrize the difference between one old typeface the design of which is no longer protected, and that of the typeface whose design you wish to pirate without getting sued... and then extrapolate some distance beyond the latter, so you have an "original" typeface at the other end.
  • I'm not a lawyer (thank the gods) but AFAIK if it can be demonstrated that you used an algorithm to modify protected outlines, you're in hot water. In fact this once happened many years ago when Adobe (?) reverse-engineered the method somebody was using to modestly modify fonts en masse and put them on a CD to sell. If it was Adobe, David Lemon (and maybe Thomas Phinney) might remember more details.
  • Mark Simonson
    Mark Simonson Posts: 1,739
    That was the Southern Software (SSI) case. They had scaled the outlines to 98% or something and renamed the fonts, thinking that would make it legal. I remember seeing the collections sold at computer stores in the nineties.
  • Regarding Sitka, I recently noticed MS added intermediate weights (semibold) in all optical sizes to Windows 11.
  • Seaford is certainly the most flavorful and humanist of the bunch, which is a good thing. There's just a bit too much «elbow» in that flavor for my taste, though (same as in Syntax). What's up with that /three/?
  • John Hudson
    John Hudson Posts: 3,227
    Regarding Sitka, I recently noticed MS added intermediate weights (semibold) in all optical sizes to Windows 11.
    Sitka was rebuild as a variable font, so yes, has a new semibold named instance as well as interpolable weight and optical size axes. A complicated project about which I will write something or give a conference presentation some time.
  • So, did a winner ever get decided, or is that debate still on-going?
  • John Hudson
    John Hudson Posts: 3,227
    The MS Office team has made a selection, but I don’t think there will be any public announcement until the version of Office apps implementing the new default is released.