← Back to context

Comment by zzo38computer

5 days ago

Calibri font has "I" and "l" the same, according to Wikipedia. A better font should avoid characters being too similar (such as "I" and "l" and "1").

Another issue is due to the font size and font metrics, how much space it will take up on the page, to be small enough to avoid wasting paper and ink but also not too small to read.

So, there are multiple issues in choosing the fonts; however, Times New Roman and Calibri are not the only two possible choices.

Maybe the government should make up their own (hopefully public domain) font, which would be suitable for their purposes (and avoiding needing proprietary fonts), and use that instead.

> Maybe the government should make up their own

They have, public sans, courtesy of USWDS, and it does distinguish between l and I with a little hook/spur on lowercase el

https://public-sans.digital.gov/

https://fonts.google.com/specimen/Public+Sans?preview.text=1...

Nothing is more inefficient than the secretary of state thinking about and conducting meetings about the font used in documents. It just doesn't matter in the sense that it "doesn't move the needle".

I expect the leaders of a government deciding on matters that have a real impact on people's live, not on stuff that from a practical point of view is totally irrelevant.

  • > not on stuff that from a practical point of view is totally irrelevant.

    The modern era we live in has far, far too much of this attitude. It's the same force eroding craftsmanship, attention to detail, and human dignity.

    I find it quite reasonable for someone to care about the presentation of official government communications.

    And just so we are clear, I also think Rubio is a horrible person.

  • So, two options.

    a) It's a smoke screen. Do something bombastic and provocative so that the opposition chews on that while something else more "important" passes undetected.

    b) Nah, he's just stupid.

  • In general, yes, but for these leaders... the less sabotaging impact they have, the better.

  • The global impression of the US is worth thinking about. The font is part of that.

    • It's really not. The used font just doesn't move the needle regarding the global impression. 99% of people never ever think or care about the font they use.

      What else should be decided on on the highest level: spacing, padding, allowance of the Oxford comma?

      It is useful that somebody thinks about that stuff, just not the highest level of the government.

      That's like the CEO of Microsoft having meeting about coding conventions, space vs tabs, variable name format etc.

      6 replies →

    • You want to know what the global impression of the US is right now? Here's a translated quote from a newspaper today, from a source in our military:

      > – The US has the most qualified intelligence organizations in the world at its disposal. Both the CIA and the FBI have been politicized under the current regime. I find it difficult to see how we will be able to maintain the trusting cooperation we have had with the US in the past after this.

      The actions of the current administration speaks far louder than any font ever could, and it's tearing down decades of good will and trust.

      1 reply →

    • The Global impression of the US is down the toilet. This only adds to that. I kept being told that I was not American, and America didn't care what the rest of the world thought. Which is it?

    • Speaking as someone who is not from the US I can say that the global impression of the US is not helped by the secretary of state bikeshedding about fonts. There are important issues of foreign affairs that need thought and attention at this time.

      2 replies →

    • It's an interesting thought, given what current global impressions are.

      I'm imagining a scenario in which the President of the United States is doing his usual sort of diplomatic outreach, consisting of waffling incoherently about things he's heard on TV that he doesn't like about their country. At one point he loses his train of thought and starts bragging about how well he's doing in cognitive adequacy tests. The diplomats are waiting until the bit where they get to flatter and bribe him at the end, the bit where he usually reverses his foreign policy, so long as they can get him to understand what they're actually asking from him. One of them speculates whether it's even possible that half the country is actually dumber than this guy.

      A staffer wearing a MAGA baseball cap sidles up to them with some briefing notes. And its just impossible not to notice the notes are typeset in the very same venerable font that was once used as the default for Windows 9x.

      The diplomats are stunned. No sans serif wokeness here. The typeface exudes heritage and gravitas. At last they realize what a very serious adminstration they're dealing with.

    • No one cares about the font US documents are written in. You're not that important.

True though the confusion about that is largely when you're not dealing with words like passwords or hashes. In the context of words it's going to be generally disambiguated by context, I can't think of an example off hand in writing where I and l will that ambiguous. The removal of serifs probably has a higher impact to more people unless I'm missing some common situation where they'd be easy to confuse in context.

  • On the Web I see very frequently foreign names, user handles or URLs where I am confused about whether there is an I or an l, because that Web page has chosen to use a bad sans serif font that does not differentiate these letters.

    Sometimes there is no problem because the words or links containing ambiguous letters can be copied and pasted. Other times there is an annoying problem because either the stupid designer has disabled copying (or like in the output of Google and some other search engines, copying does not copy the visible text, but a link that cannot be used in a different context, outside the browser), or because I want to write on my computer a link or name that I have received on my phone.

    • I disabled fonts on the web browser on my computer, in order to avoid that and other problems. I also disabled the display of non-ASCII characters in URLs (which required adding some codes to make it do that; the built-in settings will only work for the domain name and not the rest of the URL), and changed the font used for URLs, which also helps.

    • Yeah I understand it's an issue other places but I don't think it's actually a significant issue in government documents and forms written in English which is the usecase here. The choice doesn't have to satisfy all requirements it just needs to be a good choice for government writing.

Come to think of it, I vs l vs 1 vs | is one advantage of serif fonts.

  • The crossbars on the capital "i" are not serifs.

    But sans-serif fonts are certainly the prime offenders of rendering a lower-case L in place of the capital "i".

    • The crossbar of a t is not a serif, but those of the capital I are definitely serifs.

      Only on computer screens it is possible to confuse serifs with crossbars, because of the very low resolution, which forces the increase of the width of a serif to 1 pixel, possibly making it as wide as a crossbar.

      To convince yourself that capital I has serifs and not crossbars, just look at high-resolution photos of some Roman imperial inscriptions, like that on Trajan's column, which are the gold standard for the design of the capital letters in serif fonts.

      Most letters of the Latin script are made of 3 elements, thick lines, thin lines and serifs. The width ratio between the thick lines and the thin lines is called the contrast of the font.

      Serif fonts normally have a higher contrast and sans serif fonts not only have no serifs, but they also have no contrast or only a low contrast.

      Serifs are even thinner than the thin lines (which include some of the crossbars), except in sans serif fonts (which have no serifs) and slab serifs fonts (where the serifs are as thick as the thin lines).

      Both the sans serif and the slab serif fonts are fonts typical for the 19th century after the Napoleonian wars, when they were used mainly for advertising, where they attracted attention due to their anomalous serifs and they also allowed a lower cost by using cheap paper and printing machines, which would not have rendered well the standard serif fonts.

      In several programmer fonts, where most characters are sans serif, a few characters are made slab serif, i.e. with serifs that are as thick as a crossbar, with the purpose of distinguishing them clearly from similar characters. Thus capital I is made with thick serifs looking like crossbars, even if that is not the standard capital I shape. The reason is less to distinguish it from l, which should have a low hook even in sans-serif typefaces, but to distinguish it better from vertical bar, which is important in programming languages.

      Moreover, because such programmer fonts are fixed-pitch, a few narrow characters have slab serifs that do not exist in variable-pitch fonts, in order to avoid excessive areas of white space between letters. Such slab serifs added for blackening are put at the top of the small i, j and l letters, not only on capital I (but on the small letters the slab serifs are unilateral, not bilateral, like on capital I). Such extra slab serifs on the narrow characters are inherited from the type-writing machines, where they had the purpose to diminish the pressure of the hammer hitting the paper, to avoid making holes in the paper.

You are right, but if legibility had been the reason for change, Times New Roman is a rather poor choice, even if better than Calibri.

Among Microsoft typefaces, Georgia would have been much better than Times New Roman, especially when read on displays, but even when printed.

There are of course even better choices, but Georgia is a familiar typeface for most people, it is similar enough to Times New Roman and the older versions of Georgia are free to use by anybody.

Georgia is not as condensed as Times New Roman, but here Times New Roman is the anomaly, as it is more condensed than a normal font, for the purpose of fitting within narrow newspaper columns.

From Windows 3.0 to Windows 98, I have used Times New Roman as my main text font in documents, because Windows did not include anything better, but immediately after the introduction of the superior Georgia I replaced Times New Roman with it for some years, until eventually I stopped relying on the bundled typefaces and I have bought some typefaces that I liked more, for use in all my documents. (Windows 3.0 did not have yet TTF fonts, with which the licensed Times New Roman was introduced later, but it already had a metrically equivalent Times font).

Yep. Any font that neglects to put crossbars on the capital "i" should be eliminated from consideration for any practical application.

See this policy of return to Times New Roman really works. People are debating particular letters after (both) rulings have been made instead of the fact that president protects pedophiles.

  • Only rich ones. Lowbrow pedophiles who hang out in pizza parlors are a whole different thing.

> Calibri font has "I" and "l" the same, according to Wikipedia. A better font should avoid characters being too similar (such as "I" and "l" and "1").

Only when used in a context where they can be confused. This is a situation where HN is going to give bad advice. Programmers care deeply about that stuff (i.e. "100l" is a long-valued integer literal in C and not the number 1001). Most people tend not to, and there is a long tradition of fonts being a little ambiguous in that space.

But yes, don't use Calibri in your editor.

  • > Most people tend not to

    Except the whole rationale for going to Calibri in the first place was that it was supposedly more accessible due to being easier to OCR.

  • > Most people tend not to

    Yeah because normal people never have to deal with alphanumeric strings...

    • > Yeah because normal people never have to deal with alphanumeric strings...

      Natural language tends to have a high degree of disambiguating redundancy and is used to communicate between humans, who are good at making use of that. Programming languages have somewhat less of disambiguating redundancy (or in extreme cases almost none), and, most critically, are used to communicate with compilers and interpreters that have zero capacity to make use of it even when it is present.

      This makes "letter looks like a digit that would rarely be used in a place where both make sense" a lot more of a problem for a font used with a programming language than a font used for a natural language.

      8 replies →

    • No, because normal people can read "l00l" as a number just fine and don't actually care if the underlying encoding is different. AI won't care either. It's just us on-the-spectrum nerds with our archaic deterministic devices and brains trained on them that get wound up about it. Designing a font for normal readers is just fine.

      4 replies →

  • Yes, exactly this. Judging a document font based on how well it functions as a programming font is weird.

  • "Only when used in a context where they can be confused."

    So what are you supposed to when you're typing along and suddenly you find yourself in such a context? Switch the font of that one occurrence? That document? Your whole publishing effort?

    Capital "i"s without crossbars aren't capital "i"s. They're lower-case Ls. Any font that doesn't recognize this should be rejected.

    • > Capital "i"s without crossbars aren't capital "i"s. They're lower-case Ls. Any font that doesn't recognize this should be rejected.

      You have asserted this at least thrice in the past thirty minutes. What makes you feel so strongly about it? "Rejected" for what purpose? Do you understand that you've just trashed Helvetica, to take a famous example?

      2 replies →

No. I don’t want the gov wasting money making a fucking font.

There’s a few dozen off the shelf fonts that would work for 99.99% of people.

For those who it doesn’t work, deal with it. It’s a font. Or fallback to system font.

  • You know the fonts on our roads are standardized? And a lot of other official documents?

    Designing a font that will be public domain forever costs next to nothing. It's a one-time cost that pays dividends into the future and that will probably outlive us.

    The government would create something standard and accessible, and anyone could use it. No encumbered licensing.

    I think companies refreshing design systems is a waste of money, but the government doing it is actually incredibly prudent.

    • I don't think you understand how gov spends money lol.

      What you think is "next to nothing" will 99% turn into $300 million dollars and 10 years later about $4 billion will have been spent.

      And 100% there are people waiting to milk the gov doing this. Maybe you are one of them? In that case...

      1 reply →

  • Neither Calibri nor Times New Roman are free to use, although they are free in certain contexts for Windows users. The US Government is paying plenty for them.

A font was the en_US version of fount. A fount was a particular example of a typeface. A typeface is something like TNR or Calibri. They all seem to have been munged into a single set of synonyms except for fount which has been dropped (so why do we still have colour and all that stuff)?

A print, then typewriter, then computer typeface emulates a written script but also takes on a life of its own. Handwriting in english is mostly gibberish these days because hardly anyone uses a pen anymore! However, it is mostly "cursive" and cursive is not the same as serif and sans.

English prides itself on not having diacritics, or accents or whatever that thing where you merge a A and E is called, unless they are borrowed: in which case all bets are off; or there is an r in the month and the moon is in Venus.

So you want a font and it needs to look lovely. If your O and 0 are not differentiated then you have failed. 2:Z?, l:L:1? Good.

I use a german style slash across the number seven when I write the number, even though my number one is nothing like a german one, which looks more like a lambda. I also slash a lone capital Zed. I slash a zero: 0 and dot an O when writing code on paper. Basically, when I write with a pen you are in absolutely no doubt what character I have written, unless the DTs kick in 8)

  • I thought I was the only one that still crossed a seven and slashed a zero. I don’t dot an ‘O’ however.

    • I cross my sevens, slash my zeros, and use a hook on lowercase T to avoid confusion with plus signs. I think I developed the hook-T habit in college math classes.

      1 reply →

    • That's good, because the "O" should never be dotted. You use slash OR dot for zero, unless you vaguely remember them both as useful for disambiguating but forgot that both marks are for zero and vary by typeface. Mostly dotted zero was just during the dot matrix era. I wouldn't mind being shown counter examples.

  • > English prides itself on not having diacritics, or accents or whatever that thing where you merge a A and E is called, unless they are borrowed

    Its called the letter “ash” and its borrowed from... (Old) English. Though its functionally reverted to being a ligature, which is what is was before it was a letter.

    (Also, English has &, which was a letter even more recently—its current name being taken from the way it was recited as part of the alphabet [“and, per se, and”], including the effect of slurring with speed—and which also originated as a ligature.)

  • The use of the "font" spelling variant rather than "fount" is any case a clearer indication of etymology. After all, a "fount" of types refers not to its role as a fountain of printing (fons fontis L -> fontaine OF -> fountain) but the pouring out, melting and casting of lead (fundo fundere fudu fusum [fused!] L -> fondre / fonte F).

  • The linked A+E thing is called a ligature:

    https://en.wikipedia.org/wiki/Ligature_(writing)

    Same root as "ligament" and "ligand."

    • It's a ligature in modern English, but it's a proper letter in Anglo-Saxon.

      Ligatures or contextual letter variants (such as s being written with a different symbol when it's at the end of a word) are a sin to encode as characters. They should be part of the presentation layer, not the content layer! And don't even get me started on OCR which thinks such things are good to "preserve".

  • There's no pride in not having diacritics, it's a sign of an insufficient script. It's the reason why English writing gives no hint of pronunciation.