Uppercase germandbls is coming to Unicode

[cuttlefish]

I think there is room for both the "Monumental" and "Leipziger" forms of the character, as one shape or the other will be more readable and graphically complementary within a given font, but it seems this is for the Germans to decide which one, if not both, is acceptable. (How, I don't know.)

What would a blackletter cap ß look like?

[hrant]

> letters must be simple.

It's very easy to read that in a detrimental way. I think the ideal is that letters are no simpler than readers can handle (which too often they are, like in the lc "el") otherwise you're reducing differentiation for no reason. Sure, maybe Chinese is too complex, but to me Latin is clearly too simple, and when given an opportunity to add a character to it we should diversify, not solidify.

As for what structure I prefer, I do tend to agree that your structure is better than that of Seidel & Stötzner, but Jason is right that the actual design on hand will always have a lot to do with it.

hhp

[andreas]

Just to stay on the right terms like Mr. Stötzner has "invented" it.
Leipziger shape = S / Dresdner shape = Z (З).

To the "problem" of the upper left corner, (make it true uppercase like a Γ or more round) is not a significant issue for me. Both approaches can work!
To me, Adams favorite shape is a derivate of the Dresdner shape like we have done it for our GTF typefaces.
I like the Leipziger shape too. It depends on the nature of the project which one I would prefer. The Leipziger shape is more complicated and usefull for some delicate titlings or for more calligraphic styles. The Dresdner shape I prefer for everyday typefaces.

--astype.de--

[poms]

>What would a blackletter cap ß look like?
Blackletter don't work* in All Caps and there is no beginning "ß" same as in Serif and Sansfaces. That's why, there is/was no Versal-ß for blackletter "possible".

*and never intended to work

[hrant]

There is no there is no.
Just picture a document that explains this cap-eszet business, set in blackletter.

hhp

[andreas]

Poms is right, since the ONLY word what has a tradition in gothic or fraktur typefaces (in German languages) to set in uppercase is the word for god/lord "GOTT" and "HERR". But if you need the latest Hells Angels golf club tattoo, then of course, you need an uppercase blackletter sharp s. :-)

--astype.de--

[poms]

@hrant and cuttlefish
I try to show you the "why it (especially in Frakur) was not intended" later, in another thread, cause i don't want to hijack this one.

Edit:
Not another thread :-)
You see why it was not intended. Even in Schwabacher or Neu-Gotische Schriften made in the beginning of the 1930s it doesn't work.

[hrant]

I love intent. But as a means, not a goal.

Saying a blackletter cap-eszet should't be
made does not befit a good type designer.

hhp

[cerulean]

This has certainly given me a lot more insight on how to treat the character in my unicase face.

[cuttlefish]

I am well aware that it is not right to set blackletter (and especially Fraktur) in all caps, and that no word in German begins with ß, but people will try to do either anyway. What is the harm in acommodating them? Let's leave that conflict of capitalization to the copy editors and tattooists. If they want to use the letters we design, let them do what they wish.

[Andreas Stötzner]

CORRECTION

The glyph prefered by the germantype-colleagues and me is the *Dresden glyph*, not the "Leipziger form". The Dresden glyph was first proposed by a man from Dresden in 1955.
It does not help to mix things up.

[hrant]

Actually having a descender on it (at least in fonts that have a descending
"J" and/or "Q") seems quite smart, as it helps differentiate it from the "B".

hhp

[ebensorkin]

Although the tail makes it easier to cofuse with Beta. That said; the tail appeals to my eye in big way and (obviously subjectively) helps me resolve what I am seeing as an Eszett much faster.

For more on my beta comment see here

http://en.wikipedia.org/wiki/Eszett

[Ralf H.]

Although the tail makes it easier to cofuse with Beta.

You would expect a greek lowercase letter in an uppercase German proper name!? ;-)

[cuttlefish]

Folks have used ß as a substitute for beta for as long as computers have had the character in their fonts, for such things as marking pre-release version numbers and such. Why anyone would want to uppercase their version number marker, I don't know, but if it extends utility, I don't have a problem with maintaining a passing resemblance.

[dan_reynolds]

>You would expect a greek lowercase letter in an uppercase German proper name!? ;-)

Why not? Most readers on the planet know neither German nor Greek. Yet we expect them to know the difference between betas and Eszetts. When I was doing corporate design work in Washington, DC, we had to disguise a lot of the companies who we researched. So instead of publishing a report on Acme, Inc., we called them Beta Corporation, or something like that. Readers of our reports knew who we were talking about, but we didn't name names without permission.

Anyway, the authors I worked with always thought that it would be clever for the first letter of Beta Corporation's name to be an actual beta, rather than a b. But their fonts did not have betas in them, or they didn't know how to input a beta. So they just write ß (ßeta Corporation). Most of them thought that the ß actually was a beta anyway. Good luck trying to educate them to see around that one ;-)

[hrant]

As a teen, I myself used to think the eszet was a beta. But I think in context (not least the context that somebody reading German probably knows German...) it's not a problem.

hhp

[Ralf H.]

So they just write ßeta Corporation.

That's a completely different thing. Expectations play a big role in reading texts. So when you write ßeta you suggest this way of reading.
Of course people could be confused with ß vs. β when placed as a single letter. But in this discussion of a capital Eszett in proper names this really isn't an issue.

[ebensorkin]

I think Ralph's point about context is a good one. Also, the sharp angle of the 1st loop does a good job at quickly clueing you in. And having a descender is a nice way to keep distiction with the UC B.

My hat is off to Adam like eveeryone else for introducing the thread and his highly lucid text on the subject, but to my eye the samples he gave feel cyrillic rather than latin which Ralph & Andreas' do not. And I think German aught to feel mostly latin - even in a glyph such as this. I do like the fact that he showed some descenders though.

[hrant]

> I think German aught to feel mostly latin

Only after it feels fully German.

hhp

[ebensorkin]

Only after it feels fully German.

In the sense that a 'new' glyph should be 'on the team' and not stand out particularly much when used. Then yes. Entirely. In the sense that it should extend naturally from the historical German culture of letters. Again yes.

In the sense that it's 'Germaness' as an essence needs somehow to measured - then no.

Actually, I had been wondering if anybody would confuse my statement as being anti-blackletter. It wasn't. (There is a false but long held historical/cultural dichotomy of Roman/Latin vs. Blackletter/Fraktur) But actually what I meant by Latin was this visual/cultural continuity and the reality of intercultural pentration via loan words etc.

And in fact one great way to give consideration to a form might be to try to write it out in the various blackletter styles: Textura, Bastarda, Schwabacher, Fraktur, Burgundica, etc, to see what happens. BTW, I am also not suggesting that direct continuity of form from the Sans serif to the Blackletter would be desirable either. Looking at Fette Fraktur's C,L, and S certainly suggest the opposite might be true.

When I think about this letter it seems like the top 3 purposes would be :

- Monuments in a Trajanesque mould,
- Signs in a Sans Serif style,
- and things like Pickle Jars ( packaging )

... Admittedly tattoos and the back windows of pickup trucks might be candidates as well but I am less worried about those applications.

[John Hudson]

At Hrant's request, here's the text of a message regarding the uppercase eszett that I sent to the ATypI discussion list. I should note that I originally opposed the encoding of the new character in Unicode, arguing that it could be implemented as a ligature of SS, with a plain text distinction made using the Combing Grapheme Joiner control character. I still think that good arguments can be made in favour of such an encoding -- for instance, it enables documents to be backwards compatible with older fonts that do not support the uppercase eszett glyph --, but on the whole I've accepted Andreas Stötzner and, especially, Asmus Freytag's reasoning in favour of the proposed encoding.

Anyway, Hrant thought the following was a good synopsis of the matter, offered here as a contribution to the understanding of how we came to be where we are: wondering what this thing should look like.

A few comments about the proposed uppercase eszett character:

This has been proposed to ISO 10646 (and hence Unicode) by DIN, so the likelihood of it being accepted for encoding is very high. Since it is a German character, with no implications for other languages, no one feels particularly inclined to vote against a DIN proposal. If the Germans determine that they want this character, they can have it.

So why do they want it? Basically, they want it because at least some people are using it and it is considered better to have an unambiguous encoding distinct from SS. The uppercase eszett is not a new invention: it has been around for at least a hundred years. It has been used on-and-off by some publishers -- notably on the cover of the DDR edition of Duden -- and although it has never been enshrined in the official norms of German orthography it persists. As Asmus Freytag wrote on the Unicode discussion list: it is remarkable that something without official sanction should have persisted so long in such a rule-bound culture.

There can be little doubt that at least some of the users of the uppercase eszett have employed it in the interests of alphabet reform, i.e. they want to see it become standard. Others may have used it only as a display solution for a particular piece of typography. For whatever reasons, some Germans have found a use or a need for it, in preference to SS.

And let's be clear that the casing rule ß = SS is really messy for a nominally bicameral alphabet. Although ß began life as a ligature, it has long since ceased to function as one: it has distinct semantics from the sequence of two letters that it used to represent. It is a distinct character. The idea that it can be a distinct character in lowercase but not in uppercase is perverse, and causes major headaches in German text processing. The simple fact that case conversions in German are not roundtripable without dictionary support is crazy. Even if it were decided that the uppercase form of ß should always look like SS, a good argument could be made for encoding this as a distinct uppercase character. However...

...that's not how German is encoded, and the DIN proposal for uppercase ß is explicit that the new character should not interfere with existing standards for German text. So this means that ß still = SS, and the Unicode casing rules for German remain in place. The new uppercase ß character will exist alongside these rules. What this means is that any software performing case mapping between the new uppercase ß and the lowercase ß characters must do so at a level independent of the Unicode character properties. This is quite possible, and if the uppercase ß character starts to become widely used I think we can expect to see it handled in such ways.

The uppercase ß encoding is an interesting instance of a character being made available outside of the norms of encoding for the language in which it might be used. This might sound silly, but it allows this persistent heterographic letter to function in modern text encoding and display environments, and in so doing puts a lot of control in the hands of users to determine its future. It may fall by the wayside -- one more little used Unicode codepoint -- or it may, in time, become formally recognised and made part of the standard orthography.

At first glance, it may seem contrary that, on the one hand, the Germans have regularised and simplified the use of ß -- suggesting, at least to Bruno, that it is on the way out --, but on the other hand providing at least some recognition for an uppercase ß. But it is easy to see how these may be complementary moves: providing a distinct uppercase equivalent to a lowercase character regularises an anomaly in a bicameral alphabet and simplifies case conversion.

[John Hudson]

Monuments in a Trajanesque mould...

http://www.tiro.com/John/TrajanEszett.gif

This is an illustration I made during discussions of this topic on the Unicode list. I decided to put the uppercase eszett to the Trajan Test. In this instance, I was also interested to see how convincingly a form could be created that explicitly referenced the uppercase S for the right side. I'm not completely satisfied with the results, and in general prefer the models I have seen with the diagonal at the top right, but some people liked it.

[hrant]

Thanks John.

For the Trajan cap eszett, I think it's decent, but what I might try is
adding an explicit serif/terminal of some sort at the top-right corner.

hhp

[hrant]

Thanks John.

For the Trajan cap eszett, I think it's decent, but what I might try is
adding an explicit serif/terminal of some sort at the top-right corner.

hhp