Which font type?

The simple answer:

Any version of Google’s Open Sans that you use is really in OpenType format, dating all the way back to the font’s original 2011 release. You don’t have to worry when Adobe discontinues support for PostScript Type 1 fonts in January 2023: your PDF files will continue to open and render correctly.

As for which version of Open Sans produces a better-quality PDF: there’s no effective difference between the one with TrueType font data, and the other with PostScript Type 1 font data. Text may render slightly differently on low-resolution screens, but text will render identically on high-resolution devices and in printed output.

The more detailed answer:

What you’re seeing in the ‘Fonts’ tab of Acrobat’s ‘File > Document Properties’ dialogue box is something that occasionally happens when you create PDF files from InDesign and other applications.

Even though you deliberately use only OpenType fonts in your working documents, Acrobat identifies the generated PDF file as using PostScript Type 1 and TrueType fonts instead. This can be frustrating – and feels somewhat misleading – but there’s a reason why this happens.

If you notice in both screenshots, the PDF file’s embedded fonts are listed as ‘TrueType (CID), Identity H’ and ‘Type 1 (CID), Identity H’ format. CID is a way of encoding font data to support large and complex character sets, beyond those typically used in German and other Western European languages.

When you generate a PDF file from software like InDesign, an OpenType font may be converted into CID encoding when it is embedded into the PDF. Whether the font is listed as being ‘TrueType (CID)’ or ‘Type 1 (CID)’ in Acrobat depends upon the font data in the OpenType fonts that you’re using. OpenType fonts can use either PostScript Type 1 or TrueType data to describe the shapes of characters; this is why Acrobat’s ‘Fonts’ tab looks the way that it does in your screenshots.

As for which format Google normally uses for fonts like Open Sans: they generally issue OpenType fonts with TrueType font data. These fonts are available directly from Google (https://fonts.google.com/), and from other websites like Font Squirrel (https://www.fontsquirrel.com/).

In my experience, all the versions of Open Sans that I’ve downloaded over the years – going all the way from version 1.1 (2011) through to version 3.0 (2021) – are in OpenType TT format. However, I am aware that versions in OpenType CFF format – that’s OpenType with PostScript Type 1 font data – do indeed exist. If you have a font editor, you can convert font files between the two OpenType formats quite easily.

Which format of Open Sans is better? That’s a moot point: they’ll produce exactly the same output on high-resolution printers and screens. The visual differences that you might observe on lower-resolution screens and with very small text is because of the way that text is rendered in each particular situation. And those differences will depend upon the device, the screen, the operating system, and the application that you’re using to view the text.

Adobe’s own applications – including Acrobat – typically use a text renderer that’s part of their Adobe Graphics Model (AGM) rendering engine. Some applications fall back on the operating system for rendering: a good example is Apple’s own PDF viewer ‘Preview’. Other applications use their own text renderers.

If a font has special rendering instructions known as ‘hinting’ manually built into its data file, the rendering quality of the text you’ll see on low-resolution screens will be marginally better. If no manual hinting instructions are included in the font, you can expect to get slightly poorer text rendering; how much worse will depend upon the typeface’s letterforms.

If your copy of Open Sans is the OpenType version with TrueType font data, then you’ll notice that small text and text on low-resolution screens renders more consistently, but slightly blurrier than the OpenType version with PostScript Type 1 data. This is because TrueType has a feature called ‘automatic hinting’: that means text renders quickly and reliably, without the need for type designers to spend lots of time manually hinting their fonts. Automatic hinting is a nice feature, but be aware that it will produce rendered text that looks slightly bolder and blurrier.

If the intent of the PDF documents you’re creating is for print only, then I wouldn’t worry about which version or format of Open Sans you choose to use. If you’re producing PDFs exclusively for screen reading, then you’ll need to decide whether you prefer the text rendering of the OpenType CFF (Type 1) or OpenType TT version better.

Keep in mind however, that what you see on your screen may not be what your readers see on their screens. Always design with the final intent in mind, and calibrate your own systems to match.

The even more detailed answer:

The general question that you’re asking is a fascinating one. Over the years, I’ve had a number of people ask me a varying questions on the same topic:

‘Is there a difference between the way that PostScript and TrueType fonts render, and does this affect the quality of final output?’

The answer is fairly simple, but the practical results that you observe are not.

First, let’s deal with the question of whether a PostScript or TrueType version of the same font give you better-quality output. In almost every case, the answer is emphatically ‘it doesn’t matter’.

The reason: at the size that we view text, the mathematical differences between the two formats is negligible. Fonts with PostScript data use a form of vector outlines called ‘cubic splines’; fonts with TrueType data use a form of vector outlines called ‘quadratic splines’. TrueType outlines can be converted into PostScript outlines with mathematical exactness. PostScript outlines may mathematically change very slightly when converted to TrueType outlines, but not enough for anyone to notice.

Type designers prefer to use PostScript outlines and cubic splines when they create and edit typefaces with font editor software, because cubic splines are much easier to control than quadratic splines. If you use any vector software like Illustrator, you’re working with cubic splines.

So why would anyone bother using quadratic splines, if they’re more difficult to work with? Quadratic splines are mathematically simpler, and therefore much easier for computers to process. This is a reason why Apple and Microsoft chose to base TrueType font technology on quadratic splines, when they built native support for the standard into the Macintosh and Windows operating systems around 1991.

Meanwhile, Adobe chose to promote its PostScript Type 1 font format over TrueType, and integrated it into their creative software like Illustrator, InDesign, and Photoshop. These applications also supported TrueType, but most of the professional typefaces that graphic designers licensed in the 1980s and 1990s were released in the PostScript Type 1 format.

The choices that Adobe, Apple, and Microsoft made over which font format to prefer had long-term consequences. Graphic designers generally used expensive PostScript fonts to produce high-quality work. But during the 1990s, a large supply of inexpensive TrueType fonts of varying quality appeared for sale everywhere. If you were around at the time, you may remember those ‘1000 fonts for only $29.99’ compact discs.

This flood of dubious-quality TrueType fonts gave the format a bad reputation: you never really knew what you were working with. Meanwhile graphic designers have preferred used PostScript fonts – assuming that they’d be more reliable – right through to today.

In 1996, the era now known as the ‘Font Wars’ was ended, when Adobe and Microsoft worked together to produce the OpenType font standard. OpenType fonts can have either PostScript font data with cubic splines, or TrueType font data with quadratic splines. OpenType’s native support for the two older standards made it possible for typefoundries to easily convert their existing fonts into the new format. And that’s where things still stand today: OpenType is the default standard, and it’s constantly revised to add new features as needed.

What happens if a type designer has a typeface in their font editor, and builds both a PostScript and TrueType version of the same design? Nothing much: you simply get two versions of the same typeface, just with slightly different font data. The quality of any printed output will be practically identical, even if you use very small text in your artwork. But if you view PDFs of the same artwork using the two fonts on-screen, you may see a slight visual difference. The reason: an obscure font feature called ‘hinting’.

The reason hinting exists is that digital fonts are mathematically described as vector artwork, and screens and printers are bitmap-based devices. That means the text you see on-screen has to be converted from vector to bitmap format before you can view it. This conversion process is called ‘raster image processing’, or ‘ripping’ for short. That’s where the word ‘rip’ – as in ‘to rip audio or video’ – originates.

How text is converted from vector outlines to a bitmap pixel grid is a complex and sophisticated process. Many techniques have been developed over the years, and each offers particular advantages and disadvantages. As mentioned above, the way you see text rendered on-screen is very much dependent upon your device, screen, operating system, and application. Change any one of these factors, and you may see the same text typeset in the same font render slightly differently.

Fonts with TrueType data and quadratic curves are simpler to process, and have the advantage of a feature called ‘automatic hinting’. That means that the type designer doesn’t have to do anything special to get consistently-rendered text. The quality of TrueType on-screen rendering is both reliable and good. But a side-effect is that automatically-hinted TrueType text renders slightly blurrier than manually-hinted PostScript text.

Fonts with PostScript data and cubic curves are easier to work with in font editing software. But when rendered on-screen, support for automatic hinting is limited or non-existent. That means a type designer has to spend extra time in their font editor, adding special instructions that define how their font should render at small sizes and on low-resolution screens. These manual hinting instructions are difficult and time-consuming to do, which explains why professional-quality fonts cost so much.

A good type designer will spend the necessary time adding manual hinting instructions to their fonts. Other designers don’t bother, producing fonts that print well, but look worse on-screen.

It’s fairly easy to identify whether a font has manual hinting: small text will render more reliably on-screen, and it’ll appear slightly sharper and lighter in weight. Fonts with no hinting – and TrueType fonts in general – look slightly blurrier and darker in weight.

Which is more preferable? That’ll depend upon your design, and the fonts you choose to use. Plus to a large extent: your own personal taste.

But I wouldn’t lose too much sleep over it, unless your text is rendering in a way that makes it illegible. Then it’s probably best to reconsider your design and your choice of typefaces. Open Sans shouldn’t cause too many problems: it’s deliberately designed to be as legible as possible at all sizes, on all devices, and in all conditions. That works in your favour, and benefits your readers.

I hope that answers your question. If not, please let me know.

Best regards
Andrew

–30–

Google fonts are all OpenType. Internally, an OpenType font may use TrueType (quadratic) or PostScript (Bezier) outlines, but in either case it is an OpenType font. A given font file may not implement most features of the OpenType format, especially if it was simply converted from its original format to OpenType with a utility such as FontLab's Transtype, but it remains an OpenType font.

The Adobe announcement about ending support for Type 1 fonts applies only to actual Adobe Type 1 font files, not to OpenType fonts. This is a case where the unfortunate ambiguity of the terminology makes a simple situation quite confusing for most people.

You don't say in what way the appearance differs between the two versions of the document, but you have given a good example of one of the frailties of relying on cloud-based font services. If a font version changes, a document that was set using an earlier version may have unexpected text reflow or other issues. For this reason, it is good practice to download a copy of the exact font used in a project and keep it available. Font managers such as Extensis Suitcase Fusion can store multiple versions of a font and will use the correct version for a given InDesign, Illustrator or Photoshop document so that issues don't arise.

Open Sans is appropriately named. It is open source software and its repository is public here on GitHub. It is not "stored by Google in its repository." Apart from the fact that GitHub itself is owned by Microsoft, open source projects are not "owned" by anyone other than the people actively contributing to them. You can reach the font developers by reporting what you are having difficulty with on the "Issues" page of the repository.