Garamond Premier Pro issues

[David-Ar...@adobeforums.com]

I've recently started using the Adobe Garamond Premier Pro font. I teach ancient Greek literature and that font contains all the glyphs needed to type classical (polytonic) Greek as shown in the sample files provided by Adobe:
<http://store.adobe.com/type/browser/pdfs/GRPP/GaramondPremrPro.pdf>
<http://store.adobe.com/type/browser/pdfs/GaramondPremierPro.pdf>.

Those .pdf sample files provided by Adobe are displayed and printed correctly under Mac OS X 10.4.7.

However, when creating new documents under Mac OS X (10.4.7) using that font, or switching existing documents to that font, or copying the Adobe samples from the .pdf files to a new document, all the accented Greek capital letters are displayed and printed as non-accented letters in every program I've tried (Word 2004, Nisus Writer Express 2.7, TextEdit, PopCharX 3, the Character Palette) except InDesign CS2, which correctly displays and prints documents that use Garamond Premier Pro.

Switching the font back to Times or Vusillus or Gentium or whatever brings back the diacritics.

So the problem is limited to that particular font, Garamond Premier Pro (in the past I've experienced problems with Minion Pro as well, but currently that font behaves properly).

As far as I can tell, the affected Unicode codepoints are:

1F08 through 1F0F
1F18 through 1F1F
1F28 through 1F2F
1F38 through 1F3F
1F48 through 1F4F
1F58 through 1F5F
1F68 through 1F6F
1F88 through 1F8F
1F98 through 1F9F
1FA8 through 1FAF
1FB8 through 1FBB
1FC8 through 1FCC
1FD8 through 1FDB
1FE8 through 1FEC
1FF8 through 1FFB

Since InDesign deals correctly with the font, it appears to be a problem with the way Mac OS X and its typesetting system handle that particular font. But a colleague with Word 2000 under Windows XP has told me he was experiencing similar problems when using the font.

Is it a problem with the way that particular OpenType font is designed?

Is it an Apple problem? An Abobe problem? An OpenType issue?

It's a beautiful font design and I'd love to use it as my default font, but right now it's just too "buggy", at least when used with Mac OS X 10.4.7.

Thanks in advance for your help.

--
David-Artur Daix
Centre d'Études Anciennes
Département des Sciences de l'Antiquité
École Normale Supérieure
45 rue d'Ulm, 75230 Paris Cedex 05
<http://www.antiquite.ens.fr/index.html>
<http://www.greektranscoder.org>

[Ralph_...@adobeforums.com]

I have examined this font with the Windows version of FontLab Studio 5. The glyphs mentioned by David-Artur Daix, which are all Greek capitals with diacritics, exist in the font, but they are drawn as plain letters without diacritics -- though they do have iota adscript where applicable.

I have also looked at the OpenType instructions, and can see nothing that applies to these glyphs, or to classical Greek at all.

The font lacks the necessary zero width glyphs in the 03xx area for making up classical capitals with diacritics. In any case, these are normally handled by substituting readymade glyphs from the 1Fxx area; and the glyphs here are, or seem to be, defective.

The font behaves in Windows as Mr Daix described for his Mac system. I don't have InDesign. In Windows Uniscribe is insensitive to OpenType instructions for classical Greek, even were these to exist in the font.

I may have missed something here, but I have been designing classical Greek and OpenType fonts for years, and have a reasonable idea of what to look for.

Ralph Hancock

[Neil_...@adobeforums.com]

David-Artur,

A number of applications cannot yet understand the extended charactersets in OpenType Pro fonts. InDesign can. This may well be the reason you are having the problem you describe. For example, QuarkXPress has waited 'til v7 for such recognition.

Neil

[Ralph_...@adobeforums.com]

Can anything *other* than InDesign recognise them?

Ralph Hancock

[Ralph_...@adobeforums.com]

Sorry, I meant anything other than InDesign and Quark Xpress v7. The font certainly doesn't work in Word or any Windows application that I have.

If there is a compatibility problem, wouldn't it be sensible to release the font in a version with proper diacritics on the Greek capitals in the 1Fxx zone, rather than in a version that works in only a few programs? This is, after all, a font available to the general public, only a very small minority of which use either of these programs.

Ralph Hancock

[He...@adobeforums.com]

In should work in most of the Adobed CS2 family.

This is a case where Adobe is well ahead of the pack; Full
implementations of OpenType are the current wave of the future, and
other vendors, such as MS, Corel, etc., are the ones that will have to
catch up. It would be pointless for Adobe to dumb down its products,
particularly when the "STD" or even older versions are still perfectly
compatible with others' applications.

- Herb

[Ralph_...@adobeforums.com]

Is making a product that doesn't work in the majority of existing applications 'dumbing down'? All that needs do be done, for now, is to add diacritics to the Greek capitals in the 1Fxx zone.

I sell a classical Greek add-on for Word for Windows, Antioch, a program that has a fair number of users. And I have had repeated complaints from people who have bought these fonts -- which are, after all, presented as having a full classical Greek set -- and find that they don't work properly.

Most users of classical and liturgical Greek are using quite ordinary programs. They are not professional publishers. InDesign or Quark Xpress would be most inappropriate for them.

At least, information should be prominently displayed on the web site that these fonts are suitable only for people who have one of a small range of applications.

Ralph Hancock

[Neil_...@adobeforums.com]

Ralph,

Is making a product that doesn't work in the majority of existing applications
'dumbing down'?

Actually, it's the other way around! The applications need to catch up. So it is not really something to criticize Adobe for. Speak up to Microsoft, et al for not keeping up. I believe the OpenType standard has been around now for some eight years.

On the other hand, OpenType fonts will function to about the same degree as their PostScript or TrueType equivalents, with the added benefit of being cross-platform. <g>

Neil

[Neil_...@adobeforums.com]

Ralph,

This is, after all, a font available to the general public, only a very
small minority of which use either of these programs.

At least, information should be prominently displayed on the web site

that these fonts are suitable only for people who have one of a small
range of applications.

I certainly would not classify this Garamond as "a font available to the general public" as if the general public would run out and buy it. I'd bet that most folks don't spend a dime on additional fonts. On Adobe's long descriptive page for this gorgeous font <http://store.adobe.com/type/browser/landing/garamond/garamond.html>, it says, in part: Since the creation of the Adobe Originals program in 1989, Adobe Systems has offered progressive new type families and cutting-edge type technology. As a greatly expanded and enhanced OpenType "Pro" family, Garamond Premier Pro includes extensive Latin and pan-European language coverage and a broad palette of weights and optically-sized fonts. Using OpenType's powerful layout features, the process of composing and fine-tuning text is greatly streamlined. Applications which enable these features – such as Adobe InDesign®, Adobe Photoshop®, and Adobe Illustrator® – enable typographic refinements such as ligatures, small capitals, and oldstyle figures, bringing unprecedented control and sophistication to contemporary typography. OpenType fonts, coupled with the advanced typographic control offered by a program such as Adobe InDesign, also let type users take advantage of advanced justification, optical margin alignment, hanging punctuation, and optically sized masters.

Neil

[David-Ar...@adobeforums.com]

Hello Neil.

Some observations: I certainly would not classify this Garamond as "a
font available to the general public" as if the general public would truly
run out and spend $300 for it.

The full package costs USD 200 and is a "gift" to all users who register an application that's part of CS2, which is how I was able to get the font. So while not free, it's neither terribly expensive nor reserved to a happy few.

But that's not really the issue. Before posting a message on this forum I've read all the information available on Adobe's site, including the paragraphs you quote. Nowhere does it say that one _needs_ a CS2 application to simple _type_ accented letters, which accented Greek capitals are.

While I understand why I can't subtly modify the placement of a diacricrital mark or obtain a complex ligature using a common word-processor (however, just play with HoeflerText on Mac OS X and you'll see that one doesn't need OpenType and CS2 to get wonderful typographical features for "free" in every application that supports ATSUI -- but that doesn't sell CS2 copies), I don't see why the font can't be designed in a such a way that typing an accented letter actually gets you an accented letter.

Accents and diacritical marks are not simply toys for designers to play with.

For classical Greek, accented capital letters are absolutely _essential_: it's not a feature that is of use only to designers who could do without accents if the text looked "better" that way. Without their diacritics those letters become spelling errors of the worst kind. The words become literally meaningless if the diacritics disappear.

Minion Pro, another complex OpenType Pro font if I'm not mistaken, is now properly handled by Mac OS X, which is not a particularly archaic operating system and the typesetting system of which is actually very good.

So _if_ that font has been designed in such a way as to be compatible _only_ with CS2 applications not only when used as a designer's tool, but simply as a _font_ used to _type_ text (again diacritics on capitals are not an _option_ in classical Greek: if you can't type them, see them, print them, the font is useless for _writers_), then I agree with Ralph Hancock that a warning should be prominently displayed on Adobe's web site.

To repeat, the problem is not that only CS2 applications "let type users take advantage of advanced justification, optical margin alignment, hanging punctuation, and optically sized masters". That I can understand. It's that one can't _type_ using the font in any other application that is an issue.

Maybe the problem is that Adobe didn't realize that when handling ancient Greek, accents on capital letters are not optional, whereas in most other language they are just niceties? Yet the result is a font that classicists can't use. Which is a shame, because it's beautiful.

[David-Ar...@adobeforums.com]

Hello Neil,

In the interim, I would think that if there is a Type 1 available...

Only the OpenType "Pro" version contains the Greek and Greek Extended sets that I'm referring to in my post and that is at issue here.

It's because that font contains those glyphs that I'm interested in it. Otherwise I couldn't care less and I'd use the STD version. But I don't have a choice.

Garamond Premier Pro is a font advertised as containing the full Greek and Extended Greek sets, and it does, except one can't type nor print those glyphs in any program other than InDesign CS2, which is _not_ a word processor.

Some people still type classical Greek for a living, not just to create beautiful collectors' books. And if a foundry goes to the trouble of creating a font as beautifully done and complete as Garamond Premier Pro, I find it hard to understand why they'd limit its use and usefulness to designers.

I'm not trying to make light of the problem, only trying to offer a brief
opinion and some options.

I realize that, but the only option is simply not to use the font, or only use it in a page-layout program, which is no solution at all when one is a writer, not a designer.

Maybe Adobe only wants to create fonts for designers, not writers. But if that's so, I'd say it's a strange policy.

[David-Ar...@adobeforums.com]

Hello Herb,

It would be pointless for Adobe to dumb down its products, particularly
when the "STD" or even older versions are still perfectly compatible with
others' applications.

To repeat, only the OpenType Pro version contains the Greek and Greek extended glyphs I'm interested in and that are at issue, so the 'STD' version is not an option at all.

And, again, when _typing_ classical Greek, diacritical marks are not niceties for designers to play with, using advanced typographical features offered by the CS2 applications. They are an essential part of the glyphs, which become meaningless when those marks disappear.

As such, the most basic compatibility requires that those glyphs are always displayed and printed with those marks clearly visible.

Which shouldn't preclude InDesign from offering _additional_ typographical features to designers.

_That_ I can readily understand.

Not releasing a beautiful font that nobody can use to _type_ texts.

[Neil_...@adobeforums.com]

David-Artur,

I've read your points, and I can sense the frustration in not being able to use this Garamond as you need to. But my point is that this font was carefully considered and created as an extended whole, making full or near full use of OpenType technology, not a truncated standard font. It is the applications that need to catch up to that and properly embrace the technology.

One anology (maybe not a perfect one) is when Apple moved to OS X when most applications could not yet support it. So folks were pretty much relegated to using OS 9 or imperfect Classic emulation until then. One outstanding case in point is, again, QuarkXPress, a publishing world standard, which took several years before being able to work natively in OS X. Should Apple have waited till XPress got its act together?

Microsoft, and other software developers may consider that OpenType is not important enough to their user-base and their goals to rush into full compatibility with the technology. Most of their customers still use TrueType, dfonts, or PostScript fonts that came with their computers' operating systems and rarely, if ever purchase a single font in their lives. Most designers and typographers avoid these applications if they can.

And of those, a handful of fonts such as Times and Arial get the biggest use. The others simply languish or are terribly misused (is Comic Sans ever correct for a business letter?).

A $200 purchase (or $300 for the Opticals) is just not going to happen for folks who can't distinguish between a virgule and a slash -- or a "typewriter" quote and "typographer's" quote, for that matter.

Most people don't want to have to think about using fonts. They want them as accessible and easy to use as if using a typewriter. They don't want to think about weights other than regular and bold. They don't want to think about swashes, OS figures, or the other niceties of fine typography that just get in the way of what they want to type.

Most fonts are purchased (and Thomas Phinney will correct me on this point if needed) by type-savvy designers, typographers, publishers and corporate entities for specific purposes -- but that is just is just a small overall portion of the software market. It is amazing that there are as many fonts as there are. Consider that the overall market small enough so, except for some of the well-respected type designers, most cannot support themselves on profits or royalties type design.

To be honest though, I'm not sure how broadly educators, historians, technical writers, scientists et al have embraced OpenType at this time. And there's the question of what font technology is most prevalent among those who compose in Greek or Eastern European languages who use, say, Microsoft Word?

I'm not trying to be argumentative; only tryng offer some thoughts of what, from my point of view, is happening here.

Neil

[David-Ar...@adobeforums.com]

Hello Neil,

To be honest though, I'm not sure how broadly educators, historians, technical
writers, scientists et al have embraced OpenType at this time. And there's
the question of what font technology is most prevalent among those who

routinely compose in Greek or Eastern European languages who use, say,
Microsoft Word?

I suppose they mostly use the fonts that come with their computers, or, if those do not work properly, free alternatives they can get on the internet. And they certainly don't care about the font technology used by the foundries. Most probably haven't switched to Unicode and still use older encodings, at least Classicists.

However, even though OpenType is a modern technology that allows for many complex and advanced typographical features, I still don't understand why one would create font using it that doesn't pass a _basic_ linguistic compatibility test with one of the languages it's supposed to support.

Why make the diacritical marks dependent on the proper use of an OpenType feature so advanced no one but Adobe knows how to use it properly? Those marks are an integral part of the glyphs and, pardon the pun, "are too important to be left to the typographers" ;-)

Is there any valid technical reason why Adobe made that choice? Or is it just to make sure only Adobe's applications can make use of those fonts?

I'd really like to know whether Quark 7 can display this font properly. I doubt it.

It's the only font that works that way, BTW: Minion Pro which also includes the Greek and Greek Extended sets does not treat diacritical marks this way. There are modern, professional OpenType fonts out there that work properly in every applications as far as accented Greek capitals are concerned. So why make GPP such a headache?

I'm not trying to be argumentative; only tryng offer some thoughts of

what, from my point of view, is happening here. And unfortunately, your
particular needs, at least for the time being, may be caught in the middle
of this.

I've submitted a bug report to Apple. Maybe they'll improve OpenType support in the OS so that it can deal with fonts like Garamonf Premier Pro. The same way they did in the past to make sure Minion Pro works properly. Under mac OS X, it's probably not an application issue: even Word 2004 relies on ATSUI for much of its work.

What I dislike about the whole affair is the distinct feeling that there are no valid reasons, except business ones, to limit the compatibility -- and by that I mean _basic_ linguistic compatibility -- of Garamond Premier Pro to CS2 applications. Frankly, as a Mac user, I'm really starting to see the advantages of Apple Advanced Typography (AAT) features over OpenType: at least with AAT the font _works_ everywhere and _if_ you use it in a AAT-savvy program, them you get all the typographical goodies for free; with OpenType one is simply forced to use Adobe's programs just for basic compatibility.

To repeat, those diacritical marks are _not_ a typographical feature and should _not_ be treated as such. At least not exclusively. There should be a "fall-back" strategy included in the font design for compatibility -- not application compatibility, linguistic compatibility.

Just imagine a font that can be used to type French, but in which all the accents, "cédilles", "ś" and "ć" would only appear in InDesign CS2: in every other application, it would be displayed and printed as simple ASCII. Would such a font be well designed? Sure typographers could play some magic with the accents' placement, but everyone else would be quite upset.

Someone has to stand up for Homer and Plato ;-)

[Neil_...@adobeforums.com]

David-Artur,

I wish I could respond to your points better than I have. If Thomas Phinney happens by, I'm sure he could offer better insight.

Neil

[Don_Mc...@adobeforums.com]

Why make the diacritical marks dependent on the proper use of an OpenType
feature so advanced no one but Adobe knows how to use it properly? Those
marks are an integral part of the glyphs and, pardon the pun, "are too
important to be left to the typographers" ;-)

There are scads of cases in the InDesign forum where a scholar has supplied a file with some kind of accents or unusual character mapping that results in the file being unusable. The file worked fine in Word, which apparently works on a garbage in, garbage out principle.

InDesign, and other Adobe applications, are by necessity quality out programs, and as a result some consistent level of quality in is required. OpenType, with its potentially full (few fonts are full) Unicode support, provides that quality in.

The trick is, as Neil mentions, waiting for the rest of the world to get up to that same level of quality.

[David-Ar...@adobeforums.com]

Hello Don.

There are scads of cases in the InDesign forum where a scholar has supplied
a file with some kind of accents or unusual character mapping that results
in the file being unusable. The file worked fine in Word, which apparently
works on a garbage in, garbage out principle.

I'm sorry, but NO -- sorry for shouting, but I'm getting riled up here --, in this case, the problem has _nothing_ to do with an unusual character mapping or strange accents. And _nothing_ to do with Word per se. _Every_ application out there can't deal with Garamond Premier Pro _except_ InDesign CS2. Period. Well-written apps, badly written apps, legacy apps, modern apps: not one of them works. Mac OS X, which is not known for its lack of typographical support/features, doesn't support that font properly. Only Adobe does.

And I'm not some eccentric scholar trying to use that font with a custom and obsolete encoding scheme. I've actually created a program which converts fonts using older encodings to Unicode <http://www.greektranscode.org>. I know all about Word's shortcomings. I've dealt with dozens of fonts created for many different encodings over twenty years. I've worked around myriads of compatibility issues, all of which I understand quite well. Reading, writing, composing, translating, "converting" ancient Greek is my job every day of the year. I know that subject inside out.

The issue I'm raising here has _nothing_ to do with older fonts/encodings/programs. It has everything to do with Unicode-compliance, precisely. I'm sorry, but as it stands, Garamond Premier Pro is _not_ fully Unicode-compliant in its treatment of diacritics on accented Greek capital letters.

Let's get technical. The relevant chapter can be found here:

<http://www.unicode.org/versions/Unicode4.0.0/ch07.pdf>

A polytonic Greek character can either be represented by a base character to which one adds combining diacritical marks, or by a precombined character which already includes in its design the diacritics.

Garamond Premier Pro seems to be mixing both techniques, in a incoherent and incompatible way: small letters are precombined, but capital letters are _not_ and the combining diacritical marks are treated in such a way as to appear _only_ in InDesign CS2. Use another app and it's as though they do not exist in the font.

Why use such a complicated design, incoherent in its treatment of small and capital letters, when Minion Pro for instance, another Adobe OpenType "Pro" font works quite well, offers the same features and is _compatible_ and _standard-compliant_?

I don't buy the "quality out" line, sorry. Word is not the issue, nor Microsoft. It's not a platform problem, nor an OS problem. One can get "quality out" using Minion Pro. And I still don't know why GPP has to be _incompatible_ to be a good font, when Minion Pro does not. It's Adobe vs. Adobe and it does _not_ make sense at all.

As demonstrated in my first message, the problem with this font affects a limited range of characters as defined by the Unicode consortium. The design of those glyphs is not left to interpretation: they are an integral part of the classical Greek language and it's certainly not up to Adobe to decide what is and what is not part of the glyph. And if Adobe decides to use combining diacritical marks, why use them in an incoherent manner? Why use them this way _only_ for Greek capital letters, as the accented capital letters in the Latin Extended block seem to be fine. Is it because those diacritical marks are put on the left of the Greek capital letter, not above it, as is the case for Latin capitals or all small letters, Greek and Latin alike? Now _that_ could be an explanation, with possible typographical reasons, except Minion Pro does fine without using the same troublesome technique as GPP. Is Minion Pro "garbage", according to your choice of terms? I don't think so.

I'm not a dogmatic fellow. i'm not here to trash Microsoft, Apple or Adobe. In fact I love Garamond Premier Pro. It's a beautifully designed font and the Greek glyphs are simply gorgeous. I _want_ to be able to use it as my _default_ Greek font. What's so upsetting to me is that I can't because it's not compatible with _any_ word processor out there.

So I'd like to know _why_ it's been made that way. What's the _reason_ behind that weird and maddening design choice.

Please stop giving the line: "Adobe is just pushing the envelope". Here it simply does _not_ make sense. It's a _bug_, not a feature.

[Neil_...@adobeforums.com]

David-Artur,

I've just put in a request for some additional assistance on this issue.

Neil

[Thomas_...@adobeforums.com]

I just wanted to say that the type team has been reading this discussion, and we're talking about it internally. I hope to have a response within the next day or two, although it may take longer (no promises).

Regards,

T

[Victor...@adobeforums.com]

Thanks, T. As someone who uses polytonic Greek in his work, I would be very interested in seeing this fixed (and a reasonable upgrade path for those interested!).

Thanks too to David-Artur and Ralph. Your description of the problem is spot on.

(Incidentally, in Mellel you can produce cap Alpha with smooth and rough breathings, as well as combined with an acute accent, but no other combinations and none for none of the other vowels.)

By the way there was a typo in the URL David-Artur's conversion site. It should be: <http://www.greektranscoder.org/>

(Anybody know what happened to Sean Redmond's online converter? I found it extremely useful, but it's been down for ages.)

Best,

Victor

[David-Ar...@adobeforums.com]

Hello Thomas,

I just wanted to say that the type team has been reading this discussion,
and we're talking about it internally. I hope to have a response within
the next day or two, although it may take longer (no promises).

Thank you and your team for taking the time to consider the issue. I'm looking forward to hearing from you.

I realize that, by having to explain my point repeatedly in order to make it clear that a boiler plate defense of OpenType "Pro" fonts and Adobe's applications, or a boiler plate attack against Microsoft Word, were not satisfactory answers, I may have, conversely, sounded a bit harsh in my description of what I consider a "bug" in the font's design.

As I've pointed out in my last post, I believe I know at least some the typographical reasons why Garamond Premier Pro uses some advanced OpenType features to deal with diacritical marks on Greek capital letters. Because those marks are located on the left of the capital letters, not above them, they present kerning issues that Latin capital letters do not.

However, as I've also pointed out, Minion Pro deals with the issue differently and in a way that is currently compatible with Mac OS X 10.4 (I can't speak for Windows users).

Moreover, I've been testing several documents in InDesign CS 2 and I've encountered several problems with Garamonf Premier Pro's kerning information for those diacritical marks. I've created a small web page to demonstrate the issue, so that you "see" the problem:

<http://dadaix.free.fr/garamond/>

The first example deals specifically with accented capital letters. The second one shows an issue that affects small letters.

So, to sum-up, while there are obviously some interesting typographical challenges to deal with when working with diacritical marks on Greek glyphs, especially capital letters, but not only, I'm really not sure the approach used in Garamond Premier Pro is warranted -- given that it makes the font incompatible with any application other than InDesign CS2 -- nor particularly effective -- given that it does not always provides good results.

I realize that creating a font as complex as Garamond Premier Pro is a huge and complicated task, and I want to reiterate here that I think that font is beautiful and beautifully done. But I also believe there are serious problems with the treatment of diacritical marks in the Greek Extended set.

[David-Ar...@adobeforums.com]

Hello Victor,

(Incidentally, in Mellel you can produce cap Alpha with smooth and rough
breathings, as well as combined with an acute accent, but no other combinations
and none for none of the other vowels.)

I haven't tried Mellel: thank you for looking into this issue. It's interesting that Mellel can "see" at least a couple of diacritical marks, even if, in fact, it only adds to the mystery.

By the way there was a typo in the URL David-Artur's conversion site.
It should be: <http://www.greektranscoder.org/>

Again, thank you for your help.

[Neil_...@adobeforums.com]

David-Artur,

Please understand that as similar comments about OpenType compatibility would properly elicit the earlier responses in this topic and elsewhere in the Forums, clearly you have demonstrated a different scenario.

I look forward to hearing the Adobe type team analysis and response.

Your patience and detail of the issues you've experienced are certainly helpful to everyone here.

Neil

[Ramón_G_Castañeda@adobeforums.com]

David-Artur,

I may have, conversely, sounded a bit harsh in my description of what
I consider a "bug" in the font's design.

Even though ancient Greek is Greek to me, I could immediately see your point and share your frustration. We were at the same stage with Spanish, French and Russian 30 years ago.

You have performed a good public service with your tenacious discussion of the issue. Good for you!

[Thomas_...@adobeforums.com]

Thank you, David-Artur, for bringing this to our attention. We care very deeply about both the quality of our typefaces, and meeting our customers' needs.

We're still discussing this issue internally. I made a proposal for how to change our code that would achieve the same effect in InDesign, but would have the default forms for "dumb" applications be the accented forms, instead.

Even if we were to change this in future new fonts, it's another question as to whether we should change this in Garamond Premier Pro, due to compatibility issues with the existing fonts. Plus, even if we change this in Garamond Premier Pro, it would be an open question as to when. Given the other things on our schedule, it would be some months before we addressed this.

Regards,

T

[Ralph_...@adobeforums.com]

Thomas Phinney wrote:

... it's another question as to whether we should change this in Garamond

Premier Pro, due to compatibility issues with the existing fonts.

Does this mean that repairing the spacing fault that appears with this font in InDesign, where the diacritics on capitals are hung off the left margin of the letter and mess up the word spacing, would cause trouble because if the new font were used with an existing document, it would cause lines of text to turn over?

Even if so, I think it would be advisable to make the change. I've been looking at the samples on David-Artur Daix's page <http://dadaix.free.fr/garamond/>, and the fault produces an ugly and unprofessional result, especially with the wider diacritic groups. You don't really want to leave it like this, do you?

Ralph Hancock

[David-Ar...@adobeforums.com]

Hello Thomas,

Thank you, David-Artur, for bringing this to our attention. We care very
deeply about both the quality of our typefaces, and meeting our customers'
needs.

Thank you for taking the time to look into this issue.

We're still discussing this issue internally. I made a proposal for how
to change our code that would achieve the same effect in InDesign, but
would have the default forms for "dumb" applications be the accented forms,
instead.

That would be very nice. I suspect that is what's happening with Minion Pro, since, when using that font in InDesign, it displays the same kerning issues (I hope that is the right term) as Garamond Premier Pro, and so treats diacritical marks in a comparable way, making them "hang" over the blank space before the capital letter, yet remains compatible with other applications.

Even if we were to change this in future new fonts, it's another question

as to whether we should change this in Garamond Premier Pro, due to compatibility

issues with the existing fonts. Plus, even if we change this in Garamond

Premier Pro, it would be an open question as to when. Given the other
things on our schedule, it would be some months before we addressed this.

I realize that fixing such a problem can't be a priority. However, in the meantime, may I suggest that someone from your team contact someone at Apple so that in the future Mac OS X can deal better with the OpenType features you're using in Garamond Premier Pro (and maybe other fonts developed using them)? Especially if that font is not updated for "compatibility" reasons (in which case it would remain _incompatible_ with any application other than InDesign).

Last, even when working exclusively in InDesign, Garamond Premier Pro displays some "buggy" behavior when trying to _type_ accented Greek capital letters, which should incite you to correct the font. In fact, when trying to produce those capital letters by _typing_ them, the only glyphs that appear correctly are:

1F1E
1F1F
1F4E
1F4F
1F58
1F5A
1F5C
1F5E

_All_ the other accented capital letters appear unaccented. In _InDesign_. Not some "dumb" application, as you put it.

Switch the font and the diacritical marks reappear.

The only way to obtain those letter is to insert them using the glyphs palette, since typing them does not work.

Last, when importing texts, InDesign produces the correct glyphs _most_ of the time, but not always. And when it does not, one must replace the defective glyphs with the correct ones using the glyphs palette, since one can't type the correct ones directly in the application.

So, all in all, I'd say the font is really problematic and does not work as expected, not even in InDesign. Not if one has to actually _produce_ some text, not just import pre-composed texts -- and even then it's not a smooth ride.

I'm going to update <http://dadaix.free.fr/garamond/> with a couple more pictures.

Best regards.

[David-Ar...@adobeforums.com]

Hello Ralph,

Even if so, I think it would be advisable to make the change. I've been
looking at the samples on David-Artur Daix's page <http://dadaix.free.fr/garamond/>,
and the fault produces an ugly and unprofessional result, especially with
the wider diacritic groups. You don't really want to leave it like this,
do you?

Thank you for looking into this issue.

I've updated the web page with a detailed explanation of the problem as I see it and more pictures to better demonstrate the issue.

In my opinion, currently, when one relies on the Greek Extended set, Garamond Premier Pro can only be used on an existing and "finished" document in InDesign.

It can't be used to create nor edit a new document, not even in InDesign itself, since it can't be used to _type_ Greek Extended glyphs in a reliable way.

And even when dealing with an imported document, it remains problematic, since some glyphs lose their diacritical marks in the process and have to be reinserted using the glyphs palette.

So I agree with you: I think the font should be fixed. It really does not work properly.

BTW Minion Pro displays the same "kerning" issues I've described, though it seems to use different OpenType features, since it remains compatible with other applications.

I'm surprised that no one has pointed out those problems to Adobe before.

Best wishes.

[David-Ar...@adobeforums.com]

A quick follow-up.

I've installed an evaluation copy of XPress 7.0 to see how that application handles the font, since it's supposed to offer full OpenType support.

Short answer: it does not support Garamond Premier Pro. When importing a document into the application, either already set in GPP or set in another font and then switched to GPP in XPress, all the diacritical marks seem to be lost (of course switching to another font beings them back).

More interesting, when looking at the content of the font using the XPress' own Glyphs palette, one discovers that all those accented Greek capital letters, while listed in InDesign's Glyphs palette with a different GID number for each glyph, but under the same Unicode reference as the one for the unaccented glyph that is actually displayed by default, have the same GID number in XPress as in InDesign, but an incorrect Unicode reference.

An example: capital epsilon letters with smooth breathing, GID 992 unaccented, GID 993 with diacritics, both shown as Unicode 1F18 in InDesign's Glyphs palette, appear as GID 992 and 993 in XPress' Glyphs palette, which is normal; but then, in XPress, only the unaccented glyph (GID 992) is listed as Unicode 1F18, while the properly accented glyph (GID 993) is listed as Unicode E2B8, which is of course incorrect and actually does not correspond to any known character.

There is really something strange going on in this font. The OpenType features at work seem to have XPress 7 confused. Only InDesign seems able to make sense out of them.

One other thing: when inserting the accented glyphs using XPress' Glyphs palette, one actually gets a combining diacritical mark and a capital letter, each glyph being a separate entity, instead of combined into one unique glyph. One can delete just the letter, or just the diacritical mark.

So, in InDesign, the fact that those glyphs are treated as "precombined" characters when one selects or deletes them is probably another clever trick on the application's part.

Which tends to prove the point that precombined glyphs and combining glyphs are "mixed" in Garamond Premier Pro in a way that only InDesign understands.

That is all utterly confusing.

In conclusion, I would like to strongly urge the Adobe team to clean up that font. Compatibility with already set documents should not be enough of a reason to prevent that beautiful from being updated and, dare I say, fixed.

Best wishes.

[Neil_...@adobeforums.com]

David-Artur,

Your expertise and observation is truly appreciated here. Hopefully some resolution will be forthcoming.

As you can see the word «Hermai» is too close to the preceding «hoi» because
the diacritical mark is drawn over the blank space between the two words,
using the default (metrical) kerning information stored within the font.
The good news is that at least here, in InDesign CS2, one can see the
rough breathing ;-)

I'm a graphic designer by trade, and do not know Greek. However, from strictly a visual standpoint (correctly or not) the words you cite appear to have word spacing that is too open when corrected as you suggest. Could it be that Adobe was using optical spacing as the criterion when designing the font, and not correct Greek usage?

Neil

[Ramón_G_Castañeda@adobeforums.com]

In the Quark side, I would not expect the regular demo version of QPxp to work at all for Greek. Maybe the Quark Passport version?

[Ralph_...@adobeforums.com]

Neil Keller wrote:

I'm a graphic designer by trade, and do not know Greek. However, from
strictly a visual standpoint (correctly or not) the words you cite appear
to have word spacing that is too open when corrected as you suggest. Could
it be that Adobe was using optical spacing as the criterion when designing
the font, and not correct Greek usage?

Classical Greek, with its plethora of marks, does indeed require rather open word spacing, wider than would be considered necessary in a roman font. Consider, for example, the often found sequence (in visual order)

delta apostrophe space lenis capital-Epsilon

(Capital Epsilon with lenis has the code 1F18.) Since a curly apostrophe is either almost or exactly the same shape as lenis, there needs to be a substantial gap between them to make it clear what is intended. The problem becomes more severe when the last letter is capital Epsilon with lenis and acute (1F1C); in this font the diacritics now completely occupy the space between the words. From a strictly optical point of view you might say that that was OK. But from a reader's point of view it's illegible.

This problem affects Unicode fonts in particular because the space glyph is normally made to look correct in roman, so it is too narrow for Greek. Something that could be corrected in a language-sensitive OpenType font, of course.

In a normal working classical Greek font (the kind described as 'dumb' by Adobe) the capital vowels with diacritics at the left are made with the left margin of the glyph level with the left edge of the diacritics. Better *optical* spacing would require the left margin to be moved rightwards a little, but only a little. However, this can cause difficulties when, for example, a capital with diacritics is preceded by a left paraenthesis. This is easily sorted with kerning, but since in the real world not all applications support kerning, it is advisable to make fonts in such a way as to work without it.

It's also worth noting that in classical Greek, sentences don't start with capital letters. So nearly all the capital vowels with diacritics occur after a single space, rather than punctuation and space.

Ralph Hancock

[David-Ar...@adobeforums.com]

Hello Ramon,

In the Quark side, I would not expect the regular demo version of QPxp
to work at all for Greek. Maybe the Quark Passport version?

I've installed an evaluation version of the Quark XPress Passport version indeed, though I do not think this should be necessary with version 7.

As soon as a program supports Unicode and OpenType, it can't really stop working when one throws at it characters outside of the Latin blocks.

And in fact Quark XPress 7 works fine with all the polytonic Greek documents I've imported into it as long as they do not use Garamond Premier Pro.

The really interesting thing in the Quark XPress 7 test is the way that application treats GPP's accented Greek capital letters: if they're imported or typed, it only displays the unaccented versions of the glyphs, just like every other "dumb" application. If the properly accented character is inserted via the Glyphs palettes, Quark XPress 7 does insert the character, but as two glyphs, one for the diacritical marks, one for the capital letter. Whereas in this case InDesign treats the inserted character as one unique "precomposed" glyph.

Moreover, as I've explained, whereas InDesign considers the default unaccented version shown in every "dumb" application and the properly accented "variant" accessible only in InDesign as the same Unicode codepoint, Quark XPress 7 only sees the default unaccented glyph as the correct Unicode codepoint. The properly accented variant, which seems to result from the addition of a combining diacritical mark via an advanced OpenType technique, is interpreted as a wrong and undefined codepoint.

Last, when typing those letters, the OS, by means of a keyboard layout, asks the applications, including InDesign, to insert the default Unicode codepoint. And in that case the "smart" InDesign is just as dumb as any other application and insert the default unaccented version of the glyph, not the properly accented one.

At least that's what I think is happening.

Garamond Premier Pro is a very strange beast indeed.

[David-Ar...@adobeforums.com]

Hello Neil,

I'm a graphic designer by trade, and do not know Greek. However, from
strictly a visual standpoint (correctly or not) the words you cite appear
to have word spacing that is too open when corrected as you suggest. Could
it be that Adobe was using optical spacing as the criterion when designing
the font, and not correct Greek usage?

Ralph Hancock has provided a very detailed answer, so I won't repeat his points.

What I've done is make available on the <http://dadaix.free.fr/garamond/> web page an InDesign document containing the Herodotus' text I've used as an example. That way you can download it if you want and "play" with the kerning settings in InDesign.

However, one thing is certain: when dealing with polytonic Greek, the default settings programmed into Minion Pro and Garamond Premier Pro are not satisfactory. In the case of Garamond Premier Pro, the problem is compounded by the very weird way in which the font deals with accented capital letters. But I'd say it's a general problem of the Adobe Pro fonts I've tested.

I've added a Minion Pro example on the web page which shows even more clearly the kerning issue.

So, indeed, I think you're right: those fonts have probably been designed and implemented without taking into account the peculiar typographical requirements of the classical Greek language.

I really love Garamond Premier Pro and I hope it will be fixed to satisfy the needs of Greek scholars (and typographers, who, IMHO, can't be satisfied either with the current situation).

[Neil_...@adobeforums.com]

David-Artur and Ralph.

Thanks again for continuing to take the time to explain what you see when attempting to use this Garamond.

I'm sure that Adobe is still reading this topic to understand the issue details, and what the choices are to fix them should they elect to do so (hopefully, that is the case).

Please continue to post any further observations, and links to examples. Ralph, you can use the free resource, pixentral.com to post a .jpg that links back here, if you wish.

Thanks!

Neil

[David-Ar...@adobeforums.com]

Hello Ralph,

It's also worth noting that in classical Greek, sentences don't start
with capital letters. So nearly all the capital vowels with diacritics
occur after a single space, rather than punctuation and space.

That is quite true, even though in the French editorial/typographical tradition -- as opposed to the English or German ones, where only the first letter of a chapter/section is a capital one --, all Greek sentences start with a capital letter, which, in the end, only adds more capital letters to the mix.

The Herodotus' text I've posted follows the French habit and so offers a lot fo capital letters to look at.

Thank you again for helping me clarify the issues at hand.

[Ramón_G_Castañeda@adobeforums.com]

Thank you, David-Artur.

[David-Ar...@adobeforums.com]

Hello Neil,

Please continue to post any further observations, and links to examples.
Ralph, you can use the free resource, pixentral.com to post a .jpg that
links back here, if you wish.

I've added several samples of the Herodotus' text using different fonts to the <http://dadaix.free.fr/garamond/> web page so that you can "see" how different fonts deal with the kerning issues.

Best regards.

[Ralph_...@adobeforums.com]

Many thanks to David-Artur for putting up these pictures, which illustrate the point nicely. A slight mental adjustment needed to allow for the fact that the lines are justified both sides, which exaggerates the word spacing.

(And I had never seen Vusillus subjected to Mac font smoothing before. Oh dear, it does the same as Windows 'Clear Type', which smears outlines more horizontally than it does vertically, causing uneven weight in an irrationally shaded font like this. But in Windows 'Grayscale'is the default, and with this or no smoothing at all, it looks quite even.)

I'd like to add one small example, of the delta apostrophe problem mentioned in my earlier posting, because it shows that Greek spacing is not all that easy:
<http://www.pixentral.com/show.php?picture=1BT6z9DA1172WPCqRRGPOxHYYnt>

This sample is taken from the kerning display of FontLab, and you can see that delta/apostrophe has positive kerning, and apostrophe/space negative kerning to prevent too much white space; but that even so, a lot of space is required to keep the apostrophe and lenis apart. I deliberately put them at different heights to distinguish them.

Ralph Hancock

[Neil_...@adobeforums.com]

David-Artur,

I replaced your inline images with their links -- they're large enough to blow out the width of this Web page. That, and the fact that some folks still depend upon per-minute dial-up accounts are reasons why anything but very small images are discouraged here, and why I recommend the use of a link to an off-site server.

Thanks for taking the time to put together the presentation and for understanding.

Neil

[Ramón_G_Castañeda@adobeforums.com]

David-Artur,

Under no circumstances should you take Neil's replacing your images with links personally. He's uniformly ruthless when it comes to that. :)

[David-Ar...@adobeforums.com]

Hello Neil,

I replaced your inline images with their links

That's perfectly fine.

I've updated the web page at <http://dadaix.free.fr/garamond/> and summed up everything I've discovered about the issues I've encountered.

Including the difference between InDesign and Quark XPress when looking inside the font or inserting characters using the Glyphs palettes; the technical explanations offered by Ralph Hancock about the "spacing" issue; what I think is happening when one tries to type accented Greek capital letters using Garamond Premier Pro; what the deal is with precombined glyphs vs. combining diacritical marks etc.

I'm going to leave that page up and running so that people can refer to it.

And I really hope Adobe decides to fix the font -- and Minion Pro as well. There can't be that many documents using those fonts to publish classical Greek texts. So compatibility shouldn't be a serious issue when updating the fonts to solve problems with the Greek Extended character set.

Best wishes.

[Neil_...@adobeforums.com]

David-Artur.

I've updated the web page...I'm going to leave that page up and running

so that people can refer to it.

Thank you! Your time and energy have produced an excellent resource on this particular, if somewhat arcane, problem. It can only help Adobe more fully understand its extent -- and hopefully lead to a correction.

And, yes, Ramón has me pegged when it comes to image replacement! <vbg>

Neil

[David-Ar...@adobeforums.com]

While searching the web for information about Garamond Premier Pro, I've sound this interesting entry on Thomas Phinney's blog:

<http://blogs.adobe.com/typblography/2005/12/quality_in_type.html>

Here is a particularly relevant quote:

There's also font quality. By this I mean the aspects of quality that
are specific to the digital format. For example, does the font technically
conform to the specifications for the font format it is made in, and any?
Achieving outstanding font quality requires a combination of considerable
expert human attention and examination with extensive automated tests.

Font quality is an area that Adobe tends to excel in. As far as I know,
only the operating system vendors, Apple and Microsoft, have the same
history of testing their fonts so extensively.

I suppose that Garamond Premier Pro has been extensively tested. The font has been several years in the making and Adobe's procedures appear to be excellent.

So I really would like to know why in this case "additional limitations imposed by the operating system, or most applications" have not been taken into account, since only InDesign CS2 knows how to handle the font, and even then imperfectly (using it, one can't _type_ classical Greek after all).

Please understand: I do not wish to stir up any controversy here. I'm genuinely curious.

It seems to me that there has indeed been a testing failure in the case of this font's Greek Extended character set support. Simply grabbing a page of Demosthenes or Thucydides and setting it up to use Garamond Premier Pro should have made the problems I've described perfectly clear. So I'm wondering why nobody has pointed them out before, when the font was being tested.

Best wishes.

[Neil_...@adobeforums.com]

David-Artur,

Please understand: I do not wish to stir up any controversy here. I'm
genuinely curious

I don't see this as a controversy at all. I think that all sides of this discussion are in accord.

I'm wondering why nobody has pointed them out before, when the font was
being tested.

Good question!

Neil

[David-Ar...@adobeforums.com]

A quick follow-up about something Thomas Phinney wrote earlier in the discussion:

We're still discussing this issue internally. I made a proposal for how
to change our code that would achieve the same effect in InDesign, but
would have the default forms for "dumb" applications be the accented forms,
instead.

I have forgotten to ask at the time, but what exactly is "the effect" that Adobe is trying to achieve by having the font behave the way it does?

Are the spacing issues Ralph Hancock and I have pointed out the desired results from the designers' standpoint? That seems hard to believe.

Is the desired "effect" something else entirely?

I'm still trying to figure out what's going on "behind the scene".

Best regards.

[David-Ar...@adobeforums.com]

Hello Neil.

Good question!

Thank you.

I believe it's all the more relevant as Minion Pro, which has been around much longer than Garamond Premier Pro and was one the first Unicode-savvy font available, while much more "compatible" in its treatment of accented Greek capital letters, displays just as severe spacing/kerning issues.

Could it be simply that the fonts have never been tested by scholars who actually need to _read_ Greek texts? Only by typographers and designers whose only concerns are technical and aesthetic?

Though the spacing does look wrong as soon as diacritical marks are thrown into the mix, it's actually perfect, as you've yourself pointed out, when they disappear. So it's as though those fonts had been designed _without_ those marks in mind. It's as though they're an afterthought (after all, they disappear in almost every application known to man ;-) ). Yet that can't be the case, as they can't be ignored. And I'm sure Adobe has taken them into account. I just don't understand why they treat those letters that way.

And so we're back to the other question I've just brought up: what exactly is the effect that Adobe is trying to achieve by having the fonts behave that way when dealing accented Greek capital letters?

The spacing issue affects both Minion Pro and Garamond Premier Pro. The unaccented default glyphs vs. accented variants problem only the latter.

But both seem associated to a problem with the way the Adobe-designed "Pro" fonts deal with the Greek Extended character set (or at least with the capital letters in that block).

Best wishes.

[Neil_...@adobeforums.com]

David-Artur,

Are the spacing issues Ralph Hancock and I have pointed out the desired
results from the designers' standpoint? That seems hard to believe.

Could it be simply that the fonts have never been tested by scholars who

actually need to _read_ Greek texts? Only by typographers and designers
whose only concerns are technical and aesthetic?

To my "designer's" eyes, that is what their goal appears to have been. As the lines set with Adobe defaults, they look good. It is quite possible that a classical Greek scholar was never consulted.

Oddly, Greek uppercase and lowercase letters do not seem to work together very well (again from a designer's standpoint). And I'm a bit surprised that no attempt was made to modify the design of the capital letters to make them flow better with the lowercase, and perhaps, as a result, read better. Unless that is taboo in Greek.

Next stop: Cyrillic! <vbg>

Neil

[Ralph_...@adobeforums.com]

Neil Keller wrote:

Oddly, Greek uppercase and lowercase letters do not seem to work together
very well (again from a designer's standpoint). And I'm a bit surprised
that no attempt was made to modify the design of the capital letters to
make them flow better with the lowercase, and perhaps, as a result, read
better. Unless that is taboo in Greek.

This is getting off the main topic, so I'll be as brief as I can.

Greek capitals are based on the forms used in classical inscriptions, filtered through Roman and Renaissance styles, so that they have grown conventional serifs.

Greek lower case is based on 15th century Greek handwriting, which was in a cursive style and generally without capitals. Both upright and slanted forms existed at this time.

So a complete Greek font uses a mixture of styles; much more than roman, whose lower case comes from the formal upright style of late medieval roundhand.

Attempts to create a serifed Greek lower case style with the predominantly vertical shading of roman lower case (such as Century Schoolbook Greek; the metal font, not the WordPerfect one) have had an unhappy result. It just doesn't look Greek any more. Better to tolerate the uc/lc mismatch.

The worst mismatch of all must be that of Didot Greek, whose beautiful and irrationally shaded lower case is badly at odds with the thumping great modern-style capitals.

In a sans serif font, there is far less of a problem. Helvetica has a particularly good integration.

Ralph Hancock

[Ramón_G_Castañeda@adobeforums.com]

Neil,

Next stop: Cyrillic! <vbg>

Actually, Cyrillic support for Russian is pretty decent nowadays. I can't speak for those that need Cyrillic for other languages, such as Ossetian, etc., though

[Neil_...@adobeforums.com]

Ralph,

Thanks for the background! Much appreciated.

===

Ramón,

Good to hear that something works! <vbg>

Neil

[David-Ar...@adobeforums.com]

I've updated the web page at <http://dadaix.free.fr/garamond/> to incorporate the various "updates" I had added after running some tests with FontLab.

I was probably wrong about the font using combining glyphs to deal with capital letters, despite the strange way Quark 7 treats those characters. All the Greek letters seem to be precombined. And the font actually doesn't seem to contain all the needed combining glyphs.

In the end I do not understand at all why this font works that way with polytonic Greek capital letters. Why are the "default" glyphs incomplete? And why base the spacing of the font on those incomplete glyphs? Why didn't the beta test bring those problems to light?

I hope Thomas Phinney will find the time to shed some light on the subject.

Best wishes.

[Thomas_...@adobeforums.com]

I think we're clear that there are two things we need to change in new fonts, and almost certainly fix in existing fonts:

1) Better default spacing with the polytonic Greek caps. Increase sidebearings appropriately.

2) Have default polytonic accented caps keep their accents.

One thing that I wand to be clear on: are you saying that the behavior in InDesign is okay, as far as the first cap keeps the accents and the remainder (in an all-caps word) do not? Or is this also a problem?

With regards to Quark's behavior, it seems that their glyph palette is not quite as clever as InDesign's with regards to preferring to insert a base character plus OpenType layout, rather than a Unicode Private Use Area codepoint. Not too surprising, seeing as this is the first version of QXP to do either Unicode or OpenType layout.

As for our external beta program, although we have lots of people, it seems that we have a hole in that we don't have any people doing classical Greek typesetting in our beta program (or if we do, they are not active, or are not sufficiently demanding). I don't suppose you feel like volunteering?

Now I must run. My two-year-old daughter really wants me to read her a bedtime story, and I've been promising I'd be there "in a minute" for five minutes now....

T

[David-Ar...@adobeforums.com]

Hello Thomas,

And thanks a lot for taking the time to answer my questions. Especially if it steals you away from your daughter.

I think we're clear that there are two things we need to change in new
fonts, and almost certainly fix in existing fonts: 1) Better default spacing
with the polytonic Greek caps. Increase sidebearings appropriately.

Yes indeed.

2) Have default polytonic accented caps keep their accents.

Yes again.

One thing that I wand to be clear on: are you saying that the behavior
in InDesign is okay, as far as the first cap keeps the accents and the
remainder (in an all-caps word) do not? Or is this also a problem?

All right! Now I think I understand what this is all about!

Are the default glyphs unaccented in order to accommodate switching a word to all-caps or small-caps and not having diacritical marks interfere with the spacing of the letters within the word?

If that's the case then let me explain why that should never happen when dealing with classical Greek. And it it does, then it should be treated as the exception, not the rule (i.e. the _default_ glyphs should not be without diacritics).

Accents and diacritical marks in classical Greek are a late editorial addition meant to help read the texts:

<http://en.wikipedia.org/wiki/Diacritics_(Greek_alphabet)>

Originally all the letters were capital letters only and carried no diacritical marks at all, nor were the words separated in any way. The "spiritus asper" was the first diacritic "invented" by the Alexandrian scholars, who actually used the letter "H" to indicate its presence.

In other words, _if_ one were to write an ancient Greek word in all-caps, one would only use regular _unaccented_ capital letters (the ones in the basic Greek block 0370-03FF), simply leaving out the diacritics. One would/should _never_ apply the all-caps or small-caps _style_ to a word written in minuscules. That would not make sense at all.

Just grab any current edition of Sophocles or Euripides in Greek and you'll see that all the characters' names for instance are spelled using _basic_ capital letters _without_ any diacritics.

In a nutshell: in classical Greek, all caps means no diacritics at all, i.e. the _basic_ capital letters contained in the basic Greek block (from Alpha 0391 to Omega 03A9).

So, from a technical standpoint, what should happen is this: if a Greek word written in small letters and so displaying diacritics should _by_mistake_ be switched to all-caps or small-caps (remember, that should never happen as one should use regular capital letters in such a case, not accented ones: but I realize that we're talking typography here, so it's easy to image a typographer "playing" with styles when designing a page and forgetting or ignoring the rules that apply to ancient Greek), _then_ maybe you can have OpenType instructions within the font to use the _unaccented_ versions of the accented Greek capital letters so that the text will look right in all-caps/small_caps while the operation remains reversible (i.e. remove the all-caps style and the small letters are back _with_ their accents).

Which is obviously what you intended to achieve with your original design except you made it the _default_behavior.

However, as you can see from my explanation, that substitution should be treated as an exception, not as the _default_ rule as it is now. It's a "worse case" scenario and should not be common at all.

I've examined the sample you provide on your store:

<http://store.adobe.com/type/browser/pdfs/GaramondPremierPro.pdf>

The Greek section now makes sense! Indeed if one copies the text from the .pdf and pastes it in TextEdit, the characters' names, which appear in the traditional all-caps style in the sample (KRATOS and HPHAISTOS), are now, in TextEdit, "regular" proper nouns, that is, they use a capital kappa and small letters, including an alpha with oxia for Kratos, and an initial _accented_ capital letter for Hephaistos followed by small Greek letters.

In other words, instead of being written using only basic capital letters, without any diacritics, as they actually _should_ be according to the ancient Greek editorial tradition, they're perfectly correct "minuscule" words switched to use the all-caps style.

Which in turn made you design the font to use _unaccented_ Greek capital letters as your default glyphs!

Indeed "ancient Greek is too serious a matter to leave to typographers"! ;-)

So to fix that issue you just need to reverse the "default" behavior to use _accented_ capital letters and to substitute _unaccented_ variant only if a word written in small letters and starting with a capital letter is switched to all-caps or small-caps.

I hope I'm being clear enough.

As for our external beta program, although we have lots of people, it
seems that we have a hole in that we don't have any people doing classical
Greek typesetting in our beta program (or if we do, they are not active,
or are not sufficiently demanding). I don't suppose you feel like volunteering?

Actually I'd be very happy to help you test your fonts that support polytonic Greek. You just need to ask. Making sure ancient Greek is easy to compose, edit, display and print is very important to me. In fact my job depends on it. I'll append my signature to this message so that you have all my contact info.

Now I must run. My two-year-old daughter really wants me to read her a
bedtime story, and I've been promising I'd be there "in a minute" for
five minutes now....

It's a feeling I know all to well (I have three little children myself).

Thank you again for taking the time to answer my question. I think we've clearly identified the problem now, which should make it easier for you to fix it both in your existing fonts (Minion Pro and Garamond Premier Pro) and in your future creations.

Best wishes.

--
David-Artur Daix
Centre d'Études Anciennes
Département des Sciences de l'Antiquité
École Normale Supérieure
45 rue d'Ulm, 75230 Paris Cedex 05
<mailto:da...@ens.fr>
<http://www.antiquite.ens.fr/index.html>
<http://www.greektranscoder.org>

[Neil_...@adobeforums.com]

David-Artur,

Thank you for your persistence, your patience, and for the time you've taken to provide a detailed explanation of the polytonic Greek problems.

I'll append my signature to this message so that you have all my contact
info.

Thomas' email info is available if you click on his name above his posts. If you wish to contact him directly that way, upon your request I can remove your personal contact information from your posts. Normally, we delete such info for privacy reasons.

Neil

[David-Ar...@adobeforums.com]

Hello Neil,

Thomas' email info is available if you click on his name above his posts.
If you wish to contact him directly that way, upon your request I can
remove your personal contact information from your posts. Normally, we
delete such info for privacy reasons.

Thank you. I'm going to send an e-mail to Thomas then.

But in fact, only the e-mail address needs to be removed from the previous post, as the other links and contact info are "public".

Thanks again for your patience.

I've updated the page at <http://dadaix.free.fr/garamond/> to include a copy (slightly edited for clarity -- I wasn't quick enough to catch all the typos and such before the post became permanent) of the above message so that it can serve as a reference.

Best wishes.

[Neil_...@adobeforums.com]

David-Artur,

I deleted your email address -- prevents some spam, if nothing else. Thanks, again, for the latest updates to your reference.

You should be able to see Thomas Phinney's public email address when you click on his name in blue (above any of his posts). I don't want to give it here in case I'm seeing it only because I'm a forum host.

Neil

[David-Ar...@adobeforums.com]

Hello Neil

You should be able to see Thomas Phinney's public email address when you
click on his name in blue (above any of his posts). I don't want to give
it here in case I'm seeing it only because I'm a forum host.

I think that's what's happening: I can't see the email address myself, just a link to Adobe's "Type" web page.

But that's OK. Thomas Phinney should be able to contact me easily if he wants to, and I'll be very happy to help him and his team test the polytonic Greek support in the current as well as new Adobe "Pro" fonts should the need arise.

I'm very happy with the way this entire discussion has turned out, as I feel it's been really useful. Which is not so common an occurrence on public fora.

Thanks again.

Best regards.

[Thomas_...@adobeforums.com]

Hi, David-Artur,

The catch with the formatting-as-upper-case example is that such changes are done by the application, not the font. That is, the font does not link lower-case characters to the glyphs for the upper-case characters. In an OpenType architecture, case conversion is something that is left to applications and appropriate support infrastructure (for example, a library such as IBM's open source ICU for Unicode case conversion tables).

I'll be writing you off-forum about the beta program. Your expertise will be very useful to us, as both of the typefaces we have next in our development pipeline right now are slated to support polytonic Greek. One of them is even my own design. Additionally, we are planning revisions to our type library, when we could make corrections to Minion and Garamond Premier.

Regards,

T

[David-Ar...@adobeforums.com]

Hello Thomas,

The catch with the formatting-as-upper-case example is that such changes
are done by the application, not the font. That is, the font does not
link lower-case characters to the glyphs for the upper-case characters.
In an OpenType architecture, case conversion is something that is left
to applications and appropriate support infrastructure (for example, a
library such as IBM's open source ICU for Unicode case conversion tables).

In that case, the best way to handle the situation, I think, would be to always use accented Greek capital letters as the default glyphs.

In "smart" applications such as InDesign, you should be able to implement a way to use unaccented "variants" when one applies the all-caps or small-caps styles to a word written in minuscules. Just like now InDesign is generally able to use the correct accented "variants" instead of the unaccented default glyphs when importing texts -- only in reverse.

"Dumb" applications will probably always use the accented letters (unless Mac OS X could be made to handle the OpenType instructions correctly), and the result will be awful when applying the all-caps/small-caps styles. But that's perfectly all right since one should definitely _not_ do that. To repeat, if one wants an all-caps Greek word, one should use basic capital letters, not apply a style to a word written in minuscules.

I'll be writing you off-forum about the beta program. Your expertise will
be very useful to us, as both of the typefaces we have next in our development
pipeline right now are slated to support polytonic Greek. One of them
is even my own design. Additionally, we are planning revisions to our
type library, when we could make corrections to Minion and Garamond Premier.

That is great news. I'll be happy to help you in any way I can.

Thank you.

Best wishes.

[Neil_...@adobeforums.com]

David-Artur.

It looks like this has been a very productive thread! I'm anxious to see how this turns out.

Again, many thanks for sharing your knowledge and for your guidance.

Neil

[David-Ar...@adobeforums.com]

Hello Neil,

It looks like this has been a very productive thread!

Indeed.

I'm anxious to see how this turns out.

Me too. The sooner I can use Garamond Premier Pro as my default font in my word processor, the happier I'll be. It's really a beautiful font.

I learned a lot via this thread -- particularly that certain assumptions
about the construction of current-vintage OpenType fonts were incorrect.

The spacing issue was very easy to identify, all the more since both GPP and Minion Pro are affected.

But it took me a long time to understand what "effect" the Adobe team was trying to achieve with the current "buggy" scheme used to deal with accented Greek capital letters in GPP.

I think that from Thomas Phinney's standpoint the need to deal with the "all-caps" and "small-caps" styles was so obvious that he didn't feel the need to "spell it out". But from my POV it was so "far-fetched", for all the reasons I've detailed in my previous posts, that I would never normally have thought that could be the reason why the font was using unaccented glyphs by default to display accented capital letters.

Now it all makes sense and I hope the fix will be easy to implement by simply reversing the current instructions (i.e. always use accented capital letters by default and, in "smart" applications, call upon unaccented variants when applying the all-caps/small-caps style to a text written in minuscules).

BTW I realize I haven't answered Thomas as precisely as I should have.

He wrote:

One thing that I wand to be clear on: are you saying that the behavior
in InDesign is okay, as far as the first cap keeps the accents and the
remainder (in an all-caps word) do not? Or is this also a problem?

To be perfectly clear, in ancient Greek:

-- In an all-caps or small-caps word, no letter, not even the first, should carry any diacritic. None. (KRATOS, HPHAISTOS etc.) No diacritical marks at all.

-- In a word written in _minuscules_, on the contrary, if the first letter is an accented capital letter, it should _always_ carry the needed diacritic(s). No exception. Failing to display the diacritics in that case results in "barbarismes" of the worst kind.

Again, many thanks for sharing your knowledge and for your guidance.

You're most welcome.

Thank you for moderating this discussion and asking Thomas Phinney for his help in resolving the issue.

Best wishes.

--
David-Artur Daix
Centre d'Études Anciennes
Département des Sciences de l'Antiquité
École Normale Supérieure
45 rue d'Ulm, 75230 Paris Cedex 05

<http://www.antiquite.ens.fr/index.html>
<http://www.greektranscoder.org>

[Neil_...@adobeforums.com]

David-Artur.

I admit it did take me a couple of posts to realize that you were not posting the same problem as often described by others. But your needs, while certainly not pertaining to a more common use of the font, are legitimate, and certainly important enough to be properly and consistently executed by it.

I hope that, with your help, Thomas and his team can correct this flaw in an otherwise superbly designed font.

I wonder how well other extended OpenType fonts, from other foundries, handle this...

Neil

[David-Ar...@adobeforums.com]

Hello Neil,

I wonder how well other extended OpenType fonts, from other foundries,
handle this...

As far as I know, apart from Minion Pro and Garamond Premier Pro, the only other _commercial_ OpenType font with polytonic Greek support is Palatino Linotype. I haven’t tested the version currently for sale by Linotype and converted to the .otf format, so I can’t tell you if it includes new OpenType instructions compared to the version of the font that comes with Windows.

_That_ version displays no spacing issues, as shown in the screen capture that uses it on the web page I’ve set up, and only includes “regular“ accented Greek capital letters.

So when you switch some text in Palatino Linotype to use the all-caps or small-caps style, the result is ugly as hell. But it’s perfectly expected and normal.

And the rest of the time one gets the desired capital letters with diacritics. Which is a _good_ thing ;-)

The Adobe fonts that support polytonic Greek, Minion Pro and Garamond Premier Pro, are quite exceptional in that regard. The spacing issues only affects them: no other Greek font I’ve tried works that way, except in some older encodings such as Sgreek or SPIonic where all the diacritics are combining glyphs and so can end up over the space between two words if one does not type _two_ spaces before a capital letter (one for the actual space and one meant to receive the combining diacritical mark(s)).

Using such fonts/encodings usually does not produce good-looking results. Which is why fonts/encodings that work with precombined glyphs are much more popular. And why when typing in Unicode the precombined glyphs usually give better-looking results than using combining diacritics (it’s difficult for an application or an OS to position several combining diacritics correctly over a letter, especially since Unicode only supports “simple” diacritics: for instance one needs to combine both an acute tone mark and a comma above with a vowel to get the “lenis + oxia“ diacritical combination, which in turn means that the OS/app must combine three or more glyphs adroitly to produce one good-looking character).

Which is also why at first I thought Garamond Premier Pro (and also Minion Pro in that case) was using such combining marks. It was the first explanation that came to my mind when confronted with those ugly spacing/kerning issues.

The all-caps/small caps issue is, again, in a league of its own: Garamond Premier Pro is the only font I’ve ever tested that tries to deal with that issue. The whole idea is so “new” that I simply could not imagine at fist that _that_ was the “effect” the Adobe team was trying to achieve (to repeat, when typing an all-caps word in ancient Greek, one normally would use basic unaccented capital letter, not apply a style to a word in minuscules).

Though the current solution offered by the font is exactly the opposite of what it should be, I believe it’s actually a good idea to deal with the problem if it can be done right, as people are quite fond of applying styles to their text and text designers in particular, especially those with no particular knowledge of the Greek language, might perform such operations without realizing the issues they may raise.

So if a font can handle the issue for them automatically, that will be nice.

To conclude, I think that Adobe is experimenting with the features that can be implemented in OpenType fonts and indeed is “pushing the envelope”. Which is great when it works correctly. No so great when it results in a “buggy” implementation. But once the kinks are worked out -- and Thomas Phinney and his team seem quite willing to correct the fonts to make sure they work as they should, both Minion Pro and Garamond Premier Pro: I’m crossing my fingers --, a font like Garamond Premier Pro will be a real gem.

No only in InDesign, but in every application, “smart“ and “dumb“ alike. It will simply shine even more brightly when the application or the OS (if Mac OS X gets smarter) knows how to use its OpenType features to their full extent.

Best regards.

[Neil_...@adobeforums.com]

Thanks, again, David-Artur.

Neil

[Stan_C...@adobeforums.com]

Well, now, after all these scholarly remarks about Greek diacritics, I hesitate to ask such a mundane question.

Garamond Premier Pro Display includes the variations Light, Medium, Semibold, and Bold, along with their Italic partners. There's one more. Would it be fair and appropriate to call it Book?

Thank you.

Stan Coutant

[Don_Mc...@adobeforums.com]

Stan

As far as I know there is no Book weight in Garamond Premier Pro.

[Neil_...@adobeforums.com]

In the Adobe Store, under Garamond Premier Pro, I see the following five basic weights offered, each with a number of variants:
light, regular, medium, semibold and bold.

Neil

[He...@adobeforums.com]

I believe he's talking about the Garamond Premier Pro Optical series,
which includes these weights of Garamand Premier Pro Display:

Garamond Premier Pro Display
Garamond Premier Pro Light Display
Garamond Premier Pro Medium Display
Garamond Premier Pro Semibold Display
Garamond Premier Pro Bold Display

And he's asking whether the "Garamond Premier Pro Display" itself
could be considered as a "book" weight.

The answer is ... maybe. Adobe doesn't identify it as "book", but as
"normal" or "regular'. In FontLab, though, if you select "Book" as
the weight, it assigns a weight of 400, which is the same as Adobe's
"normal weight.

[Note: I don't have the GPP opticals, and have based this on other
Adobe font families]

- Herb

[Stan_C...@adobeforums.com]

YES! This is precisely what I meant. Thank you, Herb. Having just acquired the "optical size variants" edition, I had not yet realized that there is a difference between the optical series and the original GPP group.

Thanks again.

Stan

[Dominic...@adobeforums.com]

There's info here <http://studio.adobe.com/us/type/main.jsp?xhtml=topics/opticalsize> on optical masters. Display is for setting type above 24pt, subhead for 14-24pt, regular for 9-14, and caption for smaller than 9pt.

[Neil_...@adobeforums.com]

Herb,

I oversimplified. But, yes, that is the direction I was going in.

Neil

[Don_Mc...@adobeforums.com]

Now I understand, Stan.

You should not assume that a regular weight is the same as a book weight. Many fonts will have both. What is frustrating is that there is no "rule" or even standard in type names. I have seen cases where heavy is darker than black, and the reverse. And the order of medium, book and regular can also be quite arbitrary.

(So far I haven't come across a case where light is darker than bold, but it may just be a matter of time. :) )

[Dave_S...@adobeforums.com]

To further Don's point, there are (were) fonts where the boldened Light weight maps on to the Black weight, thereby causing a bold-Light to be heavier than a bold-Regular.

The original Helvetica Light was packaged this way.

Dave

[Neil_...@adobeforums.com]

medium, book and regular

Back in the days of metal type, as I recall, I would expect medium, heavy, semibold, bold, extra bold, ultra bold, and black to be incrementally heavier than book; then regular, light, and thin would be the progressively anorexic versions of the face. Many typefaces only had regular and bold plus their representative italics in their families.

(And neither "Emaciated", "gaunt", nor "skeletal" were not further degrees of lightness.) But, then again, there was none of the computer-age silliness of where you wind up with "black" when you apply "bold" to a light face.

Neil

[Giorgio_D...@adobeforums.com]

In Mellel 2.1 is possible to get classical greek capitals with diacritics: in Mellel forum I have found this:

"beneficus
Got the styles thing figured out

Joined: 22 Oct 2005
Posts: 5

Posted: Sun Oct 22, 2006 11:46 am Post subject:
Mellel does not support glyph variants through the systems palette . If you want diacritics in your capitals (in Garamond Premier Pro), you have to choose your text and from the "Character Appearance" Palette, choose OpenType and check the "Contextual Alternates" feature. In this way every time you type a capital greek letter the accent will appear AS LONG AS you follow with small letters.
To be clear consider the pronoun "ostis" (=whoever)
the letter "O" alone (e.g. with a dasia and oxia) will NOT be displayed
if you type "O stis" you will NOT get diacritics
if you type "OSTIS" you will NOT get diacritcs
but
the word "Ostis" will be displayed with a dasia and an oxia provided you have selected the text and applied the Contextual Alternates feature.
Hope I helped... just tested it to make sure it works"

And indeed this works fine.

But perhaps it's not a problem of GPP, see also this from Mellel forum:

"Mart°n
Oversees development (just a rank...)

Joined: 21 Oct 2005
Posts: 321
Location: Germany
Posted: Sun Oct 22, 2006 11:34 pm Post subject:
gidelve wrote:
The problem is strictly related to Garamond Premier Pro, doesn't affect Minion Pro and doesn't depend from keyboard layout.

It’s not a problem of Garamond Premier Pro but a problem with Mellel. Mellel’s OpenType support doesn’t support all and every OpenType feature, as the mentioned Glyph variants. There are more fonts out there, that support multiple Glyphs per character. You could force Mellel to use most of those Glyphs with the contextual alternates (as beneficius has described above) or by selecting “stylistic alternates” (or the other options like Small Caps, Proportional numbers) but you don‘t have access to all Glyphs (the striked through zero, for examplevia). The OS X Glyph palette doesn’t work at all with those fonts which is another drawback.

I have posted a example some time ago made with Zapfino (an extreme example). You could see that there are some variants of the letter “p” available - but not in Mellel. It doesn’t matter if you try to use the OS X Zapfino (which is a dfont Font) or a OpenType version. Currently you couldn’t access those letters with Mellel and I hope a future version will change this."

[Giorgio_D...@adobeforums.com]

If the problem depends from GPP (but see my previous post), when Adobe will fix it, how can I know it?

Best regards,

Giorgio

[David-Ar...@adobeforums.com]

Hello Giorgio,

And thanks for chiming in with details about Mellel.

To answer your questions, the problem is first with the way Garamond Premier Pro is designed: it considers that by default accented capital letters should be unaccented and only uses the glyphs "variants" with accents in "smart" applications based indeed on the context, i.e. when a capital letter is immediately followed by small letters (as in the (/Ostis example you quote).

The reason behind that behavior is that the Adobe engineers wanted people to be able change the text case and get a proper result when using the all caps style, that is no accents at all.

But it doesn't work properly when implemented with unaccented default glyphs. Indeed, accented capitals only make sense in a mixed context (i.e. around minuscules), but they must _always_ have their diacritics _by_default_. On the other hand, a text in all caps should actually use the regular unaccented capital letters, not accented capitals deprived of their accents.

Here is an example that will show you why adding accents based on the context to unaccented default glyphs is bound to fail: sometimes a word is composed from only one unique capital letter (a relative pronoum at the beginning of a sentence for instance), in which case the font's contextual features do not kick in (no minuscule right after the capital letter) and one must insert the proper glyph variant by hand because otherwise the accents will _never_ appear. That's a "bug" for sure, and it can't be fixed as long as the current implementation (default unaccented glyphs) is in place.

What's needed is to reverse the current implementation.

By default, the Unicode codepoints corresponding to ancient Greek accented capitals should always be represented by _accented_ glyphs. And in smart applications, the context (a case transformation for instance) could then force of the use of unaccented variants so that a sentence turned to all caps would seem to lose its accents (the "right" way to do it would be to substitute regular unaccented capitals here -- which can happen in InDesign if one applies a "hard" case transformation, i.e. not reversible --, but then the operation wouldn't be "undoable", the diacritics being lost forever).

In other words, those glyphs should not "gain" accents in some contexts (when followed directly by minuscules), but instead "lose" accents when switched to all caps.

The default glyphs should be accented and the variants unaccented.

So in that regard the second poster is not entirely correct: there _is_ a problem with Garamond Premier Pro. It suffers from an unfortunate design choice with nasty consequences for classicists.

It does seem that Mellel itself has some problems dealing with some advanced OpenType features, but seeing that Quark Express itself doesn't know how to deal properly with Garamond Premier Pro either, it's already quite a feat that Melel supports Garamond Premier Pro as well as it does, according to the first poster.

Last, about Adobe fixing the issue, I don't have an answer. But now that the issue is clearly defined, one can hope that it will be dealt with.

Best.

[Giorgio_D...@adobeforums.com]

Thank you so much. I hope that Adobe will fix this soon and then let know to the users of GPP how get the new font.

Best regards,

Giorgio

[Giorgio_D...@adobeforums.com]

Thank you so much.
Only another question, please: do you think that new (fixed) release of GPP will be free for the users?

Best regards.

Giorgio

[Don_Mc...@adobeforums.com]

do you think that new (fixed) release of GPP will be free for the users?

Thomas will have the definitive answer, but I suspect the answer will be no. Adobe does not register fonts, so there would be no identification trail for verifying upgrades.

[Neil_...@adobeforums.com]

Sidebar: Unfortunately, for those of us who bought many Adobe fonts on floppies many moons ago -- with registered serial numbers -- the same holds true.

The big problem now is transfering all those fonts to more stable (and usable) media -- and apparently under Mac OS X (at least) 800 K floppy disks are not readable on the few USB floppy drives I've tried. And there are no software patches that I've been able to find.

Neil

[Don_Mc...@adobeforums.com]

Neil

Can you not back up your fonts folder onto a CD-ROM to create a backup?

I remember when I migrated to PC from Mac ... I lost about 1000 fonts, as the Type 1 fonts were not cross platform. I kept that old Mac 20 meg hard disk around for ages in hopes I would come across a tool that would let me convert them. Finally trashed it.

[Neil_...@adobeforums.com]

Don,

If I'm gonna archive the fonts, I'd rather go back to the "pristine" source media, rather than risk copying some corrupted fonts.

Neil

[Don_Mc...@adobeforums.com]

Ah yes, you are using that OS that tends to corrupt fonts. I forgot about that.

:)

[Dominic...@adobeforums.com]

Only if Neil using an old version of that OS. It's not an issue with the later versions.

[Neil_...@adobeforums.com]

Don,

Mac OS X 10.4.8 has not corrupted any software for me. Still...

Just from using fonts and moving them from drive to drive can eventually corrupt them. I also wouldn't copy my Photoshop from an old machine to a new one. I always do a fresh, clean install, and then do any updates.

Neil

[Don_Mc...@adobeforums.com]

Just from using fonts and moving them from drive to drive can eventually
corrupt them

Never heard of this problem on the Windows platform. That is the little dig I was making. It is only a Mac problem, so far as I know. (And probably an obsolete one, if Dom is correct.)

[Neil_...@adobeforums.com]

Don,

I would hazard to say that making copies of copies of copies of copies of...[etc.] any software can result in file corruption, any platform.

Neil

[Don_Mc...@adobeforums.com]

I would hazard to say that making copies of copies of copies of copies
of...[etc.] any software can result in file corruption, any platform.

I can't see how, knowing how file copies are made. CRC checks make sure that each byte is correct, or your copy program will reject, as when you try to copy to a bad floppy. It is not like making a photocopy or printing where photography is involved, and eventually things get fuzzy. On a disk it is 1 or 0, and there is no blurring of those.

However disks can go bad over time ... that one I agree to.

[Thomas_...@adobeforums.com]

I've heard that the main problem with fonts on Classic Mac OS was that they were system resources, and the resource fork was kept as an open file by the OS, and was subject to corruption in a crash for that reason.

Unlike most things typographic, I can't personally vouch for the accuracy of this tale....

Cheers,

T

[Neil_...@adobeforums.com]

Don,

When files are copied from one drive to another, I don't believe there is the same kind of verification process as there is when burning a CD or DVD.

Neil

[Giorgio_D...@adobeforums.com]

In the meantime, has Adobe fixed the problem with GPP? Is there a new release of the font?

Thank you so much and regards

Giorgio

[Giorgio_D...@adobeforums.com]

Thank you so much, Thomas.
Please, could you say in which way Adobe will make know the update?

Thanks again.

Giorgio

[Thomas_...@adobeforums.com]

Well, we'll be updating all the fonts in our library at the same time, so we'll be talking about that fairly widely.

Regards,

T

[Neil_...@adobeforums.com]

Thomas,

Any advance clues to what kinds of global changes there will be? Paisley instead of solid color fills? Chase lights in the outlines? Animation? Speaking letters? Viewers want to know!

Neil

[Thomas_...@adobeforums.com]

Nothing so dramatic, I'm afraid. Mostly very minor technical issues that will be invisible to users. Probably the biggest change is a few more fonts supporting truly arbitrary fractions.

Cheers,

T

[Neil_...@adobeforums.com]

Thomas,

Thanks for bringing me back to Earth!

I know that when the info is available, it will be posted either here and/or on adobe.com. I just ask that info be included on any upgrade paths that might be available.

Neil

[Giorgio_D...@adobeforums.com]

Thomas,

will be fixed also the main question of this topic, i.e. diacritics with greek capitals?

Regards

Giorgio

[Thomas_...@adobeforums.com]

Yes - that's what I was trying to say a couple of days ago (msg #93). Sorry if I wasn't clear enough.

Cheers,

T