Difference between two Palatinos

[_savage]

This topic was imported from the Typophile platform

I've started to play around with PrinceXML to generate some fliers. The main text font is "Palatino Linotype" but since the text contains some diacritics it seems that PrinceXML automagically falls back to "Palatino". Both are installed on my system.

Palatino Linotype:

PostScript name: PalatinoLinotype-Roman
Full name: Palatino Linotype
Family: Palatino Linotype
Style: Regular
Version: Version 5.00
Unique name: Palatino Linotype Regular April 1998
Manufacturer: LINOTYPE-HELL AG
Trademark: Palatino® is a registered trademark of Heidelberger Druckmaschinen AG and its subsidiaries.


Palatino:

PostScript name: Palatino-Roman
Full name: Palatino
Family: Palatino
Style: Regular
Version: 3.8
Unique name: Palatino; 3.8; 2006-02-23
Copyright: Copyright © 1991-99, 2006 Apple Computer, Inc. Copyright © 1991-92 Type Solutions, Inc. All rights reserved.
Trademark: Palatino is a registered trademark of Linotype AG


So I have two implementations of the same typeface design from two different sources. But both are used in the same document. Does this matter? How much a difference is there between these two?

And more general, is there a tool that helps me compare two types? How do I know which one is more faithful to the original, if that is at all possible? (It seems to me that digital types in some instances are more of an interpretation of older casts than faithful reproductions.)

Thanks!

[George Thomas]

I would not mix the fonts from two sources because the metrics may not match.

As for originality, I would lean toward the Linotype version even though at one point their version likely went through the old phototype unitizing process. Linotype has a history with fonts, Apple is very late to the game. Linotype's type designers were very much concerned with accuracy and attention to detail.

[Nick Shinn]

Yes indeed, interpretation is the name of the game when making digital fonts from metal-era designs such as Palatino. The issue arises because the letter shape on the face of the type changes when it is printed. And the printed shape varies depending on the printer, the paper stock and the condition of the type, among other things. So to which shape should the redesigner/digitizer be faithful?

[Nick Shinn]

“The task of renovating or recreating a design from old impressions is the most difficult of all. The effect of impressing upon damp paper, of worn type, and of the spread of ink, have to be reckoned with; and great skill is needed if, while removing blurred outlines, the subtleties of the original engraving are not to be lost.”
—Stanley Morison, A Tally of Types.

[Queneau]

I have both installed on my system as well, and from what I can see Palatino Linotype has much more typographic features like old style figures and small caps, whereas the other one has none of these but wider language support.

[hrant]

What George said.

hhp

[dberlowgone]

The difference is that one was made by Adobe for Linotype from data unknown, into PostScript, and the other was made by Font Bureau, for Apple, from Linotype data , with Zapf consulting, into TrueType. (I don't think Linotype made either).

[Thomas Phinney]

The one made by Adobe/Linotype would also have been from Linotype data.

[_savage]

Thank you everybody. It seems that I can't use either of the fonts alone :) Using the "Palatino Linotype" implementation falls back to "Palatino" for two characters with rarer diacritics, and using "Palatino" falls back to "Arial" for the ff ligature. Mind you, in these fall-back cases, only that particular missing character is rendered using the other font.

Now I could disable ligatures altogether, but somehow I don't really like that. Isn't there a Palatino that has everything and is a decent implementation?

[Nick Shinn]

Palatino works rather well without ligatures.
It certainly seems to have been designed that way.

[charles_e]

There is a version that we sometimes use, in ttf. It came with some Microsoft software. Actually, we seldom used it, only because designers seldom specify Palatino these days. All ligatures, polytonic Greek, Cyrillic, small caps and os figs, though the combining diacriticals are sparse. What accented character are you missing?

The copyright notice reads:

Copyright 1981-1983, 1989,1993, 1998 Heidelberger Druckmaschinen AG. All rights reserved. The digitally encoded machine readable outline data for producing the Typefaces licensed are the property of Heidelberger Druckmaschinen AG and/or its subsidiaries, represented by Linotype Library GmbH, Dupont Strasse 1, 61352 Bad Homburg Germany. Portions © 1996-1998 Microsoft Corporation. All Rights Reserved.

and the Notice/description reads:

Palatino Linotype is the definitive new version of Hermann ZapfÕs [sic] Palatino, which since its design in 1950 has become one of the world's most widely used typefaces. For this new digital version, Professor Zapf has drawn numerous additional characters to include an extensive range of ligatures, numerals, fractions and support for Cyrillic and both monotonic and polytonic Greek. Special care has been taken to enhance the quality of the letterforms when displayed on the computer screen, ensuring that Palatino Linotype is highly legible whether displayed on the screen or in print. This typeface is ideal for use in extended text settings such as books, periodicals and catalogs.

[hrant]

Oh yeah, I remember that one! It was a big deal when it came out.

hhp

[_savage]

Charles: Thank you for the hint; there seems to be another implementation floating around :) The characters I am missing in "Palatino Linotype" are U+1e41 and U+1e47. These are the only two characters where "Palatino" is used as a fall-back, the rest is set ok in "Palatino Linotype".

hrant: Why was that a big deal?

[hrant]

If you'd like to have those two characters added to Palatino Linotype, it's not difficult (if you have the right software) but the EULA probably prohibits any modification. I know people in Linotype and I can try my luck with getting permission if you like. If interested: hpapazian at gmail dot com

The big deal: just that it was supposed to have been the definitive version. And maybe it is. But as Charles hints I guess Palatino is no longer in high demand.

hhp

[charles_e]

Unicode 1E41 is a (precomposed) lowercase m with a dotaccent above. U+1E47 is a (precomposed) lowercase n with a dot accent below. Neither is in this Palatino font in precomposed form -- nor, likely, any Palatino you can find. You're in romanized Indic script country.

However, you're in partial luck. The Palatino I'm referring to does have U+0323 (dotbelowcomb) as a combining accent, and it seems to have mark positioning. The font does not have U+0307, the combining dot accent. It does have U+02D9, the spacing modifier dot accent. Of course, there is no mark positioning with spacing modifiers, it isn't their purpose. Still, if the occurrence is infrequent, you could kern it back & rekern to properly space the following letter. The Unicode in the file will be a trifle compromised. Or, you could use a proper combining dot accent (U+0307) from, say, Charis or Gentium -- funny thing about those periods, they look a lot alike. Just get the size & spacing close in the application program.

The license for the font I'm talking about forbids modification, even for personal use.

Modification
You are not allowed to edit or modify this font, even for your own use. Please contact Linotype Library GmbH if you require a customized version of this font.

Again, any dot accent from any Palatino would look the same, if you can find one with the combining accents to preserve syntactic meaning. Sadly, most fonts don't have any.

On the other hand, an Adobe font does allow modification, if purchased before 2011, even if for one of the Monotype conglomerate companies' fonts (of which Linotype is now one). Otherwise, no time constraint.

Tell your program to default to *that* Palatino font for the one character.

Ain't EULAs fun? Isn't the new Monotype great? If you see one of Monotype exec's crossing the street, remember what the accelerator and brake pedals are for. Careful, don't get confused...

[hrant]

Isn't the new Monotype great?

It isn't just Monotype, the no-mod clause has sadly become the norm, with very few exceptions (such as Adobe and Monokrom). Now, if we could figure out who started this trend, I for one might have trouble hitting the right pedal at the right time...

hhp

[Té Rowan]

Looks to me that TeX Gyre Pagella does have U+1E47 though not U+1E41. As far as I can see, the applicable licence (the GUST Font License and (by extension) the LaTeX Project Public License v1.3c) does not limit commercial exploitation.

[Nick Shinn]

Shinntype allows mods.

[hrant]

Ah yes, thanks for the reminder! And even more thanks for allowing.

hhp

[charles_e]

Lots of the smaller foundries allow modification. Christóbal Henestrosa is another (though you're suppose to tell him), some of the fonts marketed by Village, etc.

Sometimes it is automatic (in the EULA), sometimes you have to ask. The ones that turn you down are more often, but not always, involved in the advertising side of type use.

I've researched this some, though I have not gathered hard data. What fits the questions posed & answers I've gotten is the "no modifications" clauses began as an effort to limit/stop piracy. The small foundry (publisher) tells the lawyer what they are trying to stop, he/she words it a certain way. When that is all the publisher is after, they will usually allow modification, as long as it is well done and does not affect their product negatively.

However, some publishers look at modifications as an extra revenue opportunity. This would include FontFont and the Monotype conglomeration. Even so, they are quite different companies. FontFont at least, is quite generous with both their backup and embedding policies. The new Monotype is not; I find them similar to some of the more aggressive pharmaceutical companies. You know the line about drug prices -- due to "research costs" -- esp. researching what gets you higher sales with your print & TV advertisements...

There is a funny story about FontFont, maybe even true. They had screwed up the encoding on a font, and a somewhat miffed customer finally agreed to pay to have it straightened out. Only at that point did FontFont asked "well, what's wrong?" That's asking a bit much of your customer, what? (Believe this was posted on Typophile many years ago.) But "modifications" aside, their EULA is still one of the better ones for many type users.

[hrant]

Charles, if you ever do compile some hard data on this that would be great.

hhp

[charles_e]

If you mean hard data on why people made certain decisions, I'll never do it. An academic paper, that one.

If you mean hard data on what the individual licenses are, I use the compilation already made up by FontFont, to be found here

http://www.fontshop.com/licenses/

Not everyone covered, to be sure, but a lot of information on the site.

[hrant]

I meant about which EULAs [don't] contain the no-mod clause. That FontShop list is probably the best single place to start compiling such data.

hhp

[charles_e]

Again, this won't be me.

Whether or not a typeface has a no-mod clause can be important in my business, but making a list of all fonts that have such is not. There are only a small set of fonts I care about. Put another way, I'm not interested in identifying fonts I can repackage & sell. If people that design the interiors of books for scholarly book publishers aren't interested in a typeface, it's of no value to me, regardless of the EULA.

[hrant]

I think what could be useful about preparing and publishing a list of which foundries [dis]allow modification is that it would lead to more business for the lenient ones, putting pressure on the strict ones. This would lead to a future with more good options.

hhp