Sharp S can appear at start of word

Florian Pircher

John Savard said:

And, in addition, the lower-case eszett is no longer all that likely to be mistaken for f.

I think the proposed small ß looks more like an f, not less.

Franz Gratzer

John Savard said:

would be any better?

I don't really get what you are trying to do here. You want to "repair" the long established lower case \ß and in the process propose an other \ẞ too?

Well, it actually is a vertically mirrored 2, with something attached.

Unfortunately I get always confused how I should name mirroring: Am I supposed to call it after the axis along it is mirrored or in which direction the reflection is drawn? You seem to be certain that it needs to be the second case.

In any way: Why did you chose the \2 to construct the new glyph from?

Peter Baker

One of the subjects I particularly enjoyed covering when teaching History of the English language in years past was attempts at spelling reform, which date back to Orm in the twelfth century and gained speed in the sixteenth century, when some gentleman would develop an idea for how to fix English spelling (which has been, as everyone knows, completely broken pretty much forever), would spend obscene amounts of money getting a font made, full of new characters and weird squiggles, e.g.

(John Hart, On Orthography, 1569), and would proudly pay some printer to produce X copies.

And then—

Nothing would happen. The gentleman's wonderful book would be largely ignored, because it turns out that what readers want in a spelling system is not a closer phonetic representation of the spoken language, or a more consistent or efficient system, or references to the history of spelling and historic letter-forms, but familiarity, because reading is already difficult enough, and people are highly invested in both their written and their spoken language.

A partial success that proves the rule is Noah Webster, whose first attempt at an American dictionary included a pretty radical attempt at spelling reform, which he thought might catch on because it (cleverly) demanded that no new letters be cut. But he had to retreat in subsequent editions, except for relatively few innovations of the kind that didn't really rock the boat. These account for most of the differences between British and American spelling.

In sum: I doubt that German readers will thank or reward type designers for a redesign of the ß, though they might be receptive to the idea of dropping it altogether, since they already know the convention that ss can be substituted for it.

Christian Thalmann

The «authentic» eszett looks like this: ß. 

What is the purpose of trying to predict the evolution of a letter when it has actually evolved and we know the result? What makes your theorycrafting more relevant than reality? Writing, like language, exists only by convention, and there is no absolute truth to be found. Usage is king. Otherwise you’d have to go all the way back to Phoenician, or Linear A. 

In any case, pretty much all contemporary designs of ß can still be understood as «long s plus something», so they’re not even inconsistent with its origin. The fact that the something relaxed itself into a familiar shape (s, z) just comes from the need of a sound construction, which the above is still not by a large margin. 

John Savard

Franz Gratzer said:

In any way: Why did you chose the \2 to construct the new glyph from?

That is a question I can answer.

This diagram

shows the shapes previously illustrated, plus two versions of an additional shape.

The first shape is a lower-case eszett based on the earliest forms of the eszett in handwriting; added to a long s is the single stroke, without the addition on the bottom that led it to be taken as a z, to remove the "false etymology" of the eszett being an sz-ligature.

All the remaining shapes are proposals for a related upper-case eszett.

So the second glyph is a capital S, with the same addition as the long s.

But the lower-case eszett involves an addition to a long s, not the usual lower-case eszett. So possibly this wouldn't look right. So what would the upper-case version of a long s look like? (Never mind the fact that the distinction between the long s and short s only exists in lower-case. An eszett is supposed to have a long s in it!)

Well, since the upper-case of f is F, it seemed to me that the upper-case of a long s would look like a Greek gamma!

When I proposed this in another thread, though, there were howls of outrage. And, indeed, the Greek gamma doesn't make people think of an s.

So the shape involving the 2 was an attempt to split the difference - gamma on the top, but with a bowl at the bottom like that of an S. On further reflection, perhaps I needed to do a bit more work to make a hybrid S/gamma, and so the last two shapes illustrate another possibility.

Peter Baker said:

One of the subjects I particularly enjoyed covering when teaching History of the English language in years past was attempts at spelling reform, which date back to Orm in the twelfth century and gained speed in the sixteenth century, when some gentleman would develop an idea for how to fix English spelling (which has been, as everyone knows, completely broken pretty much forever), would spend obscene amounts of money getting a font made, full of new characters and weird squiggles, e.g.

(John Hart, On Orthography, 1569), and would proudly pay some printer to produce X copies.

And then—

Nothing would happen.

As it happens, the high school I attended had a copy of the Penguin edition of Androcles and the Lion that was published in order to illustrate the Shaw Alphabet, perhaps the last gasp of this long tradition.

Well, my plans for German are hardly so ambitious. It had been pointed out that certain forms of the eszett look like sz or ss ligatures, but this is misleading, and so I presented a form that would be true to its origin.

Franz Gratzer

John Savard said:

since the upper-case of f is F, it seemed to me that the upper-case of a long s would look like a Greek gamma!

Allright. Now I understand how you arrived there. Unfortunately, it still doesn't feel remotely right. I can't imagine it would replace the already established form. Even if it is misleading because it seems to be a combination of known letters despite that not being the case.

Christian Thalmann

There is no harm whatsoever in ß looking like a ſz or ſs ligature. Folk etymology or not, that's what it looks like nowadays, that's what its name says, that's what people often (and legally) replace it with in all caps, and that's how most people think of it in their minds. I doubt any significant fraction of the 130'000'000 speakers of German in the world would be willing to submit to your re-education on the matter, especially for a «benefit» that only a dozen or so people in the world even know and care about.

It also doesn't help that none of your designs even remotely look like a Latin letter, even after all these iterations.

Christian Thalmann

Well, since the upper-case of f is F, it seemed to me that the upper-case of a long s would look like a Greek gamma!

Poppycock! The uppercase of ſ looks like this: S. Remember that all of ſ, s, ß are derived from the vernacular form of the original Latin S, not the other way round. Talk about historical authenticity...

John Hudson

At this point, it’s hard not to suspect that JS is trolling y’all.

John Savard

Christian Thalmann said:

I doubt any significant fraction of the 130'000'000 speakers of German in the world would be willing to submit to your re-education on the matter, especially for a «benefit» that only a dozen or so people in the world even know and care about.

I don't claim to have "a dog in this fight" myself; until very recently, when one of those dozen or so people educated me about the matter, I was perfectly happy to think of the eszett as being a long-s/s ligature.

I have only been so bold as to chime in on the subject because it is inherently fascinating.

If it is viewed as a problem that a number of typefaces, by having forms for the eszett suggestive of an ss-ligature, decieve unwary foreigners into regarding the eszett as a ligature - apparently, ordinary German speakers are not decieved, and regard the eszett as a full-fledged letter without the assistance of a modified eszett such as I have proposed - then I offered, though without too much seriousness, a form of the eszett which would prevent this confusion, and bring those who are not German speakers to a better understanding of the character.

On further reflection, though, while I can say that my design proposals are not an attempt to impose anything on the German-language community, but are merely in the spirit of a gift that they can use if they wish... in the German language, the word "Gift" means poison!

Franz Gratzer

Thank you John. I really can follow your thought. But I seriously would rather drop the "ẞ/ß" than replacing it with an other glyph. But it's not up to me anyway.

Christian Thalmann

- apparently, ordinary German speakers are not decieved, and regard the eszett as a full-fledged letter without the assistance of a modified eszett such as I have proposed -

That's really not surprising — after all, most English speakers uuould think of W as a full-fledged letter of the alphabet rather than as a u_u ligature.

Come to think of it, the appearance of W as a V_V ligature is misleading. To right this outrageous injustice, I propose using this new and improved form of W for English orthography:

Ɯɯ

You're ɯelcome.

John Savard

Christian Thalmann said:

Come to think of it, the appearance of W as a V_V ligature is misleading. To right this outrageous injustice, I propose using this new and improved form of W for English orthography:

Ɯɯ

You're ɯelcome.

Touché!

However, while this would be more in accord with what we call the letter, it is potentially confusing in another way.

After all, English spelling includes a number of digraphs for sounds, such as "th", "ch", and "sh". These make the orthography more complicated than it needs to be. Also, the families of consonants such as (p,b), (f,v), (g,k) and so on only have two members each, which means that without diacritical marks, it isn't really possible to adequately transcribe the sounds of either the Wu dialects of Chinese (such as Shanghainese) or Eastern Armenian into English texts.

So clearly the Latin alphabet should have the letters Θ, Ч, Ш, and Ж, along with П, Б, and Γ added to it (along with something like ŋ for the "ng" sound). So that form of W would be confused with "sh".

Come to think of it, though, neither uu nor vv really represents the completely separate and independent consonant sound of "w" adequately. So that letter should perhaps by entirely scrapped, and replaced by something based on the "wynn" rune.

In fact, of course, English once had a letter "wynn", plus the "thorn", an alternative of its own to the Greek theta.

Craig Eliason

Christian Thalmann said:

Come to think of it, the appearance of W as a V_V ligature is misleading. To right this outrageous injustice, I propose using this new and improved form of W for English orthography:

Ɯɯ

You're ɯelcome.

Okay, but you'll also need a /w.loclFRA and /w.loclESP that look like this: w

RichardW

Using Ɯɯ would be confusing for IPA users, as it's already the high back unrounded vowel, and easier to write by handed than the horned u (ư) of Vietnamese.  Handwritten 'w' can already look like a small omega - that would be a better basis - except that some African language already uses omega as a letter in its Latin script alphabet.  (The capital is the small omega write large, not the ohm symbol.)

Denis Moyogo Jacquerye

Fun fact about Ɯɯ, the capital looks like a cyrillic capital letter sha, Ш, in Zhuang and Bouyei documents from when the letter was used. The uppercase is in Unicode for those languages and the lowercase for various uses. Of course the uppercase doesn’t have that shape in Unicode and pretty much every digital font, and, on top of it, a couple languages started using the letter with the current shape.

John Savard

Of course, though, it isn't really fair to do a serious critique of Ɯ and ɯ, as they were only mentioned as a new shape for W and w to illustrate how terrible my suggestions for changing the shape of ß were - Ɯ and ɯ weren't being seriously suggested in their own right.

george_thompson

The following list of words comes from:

https://www.idiotikon.ch/Texte/Landolt/Jiddisch_Landolt_EuroComGerm_Bd2

Christoph Landolt 

Jiddisch. 

In: Janet Duke (Hrsg.): EuroComGerm. Germanische Sprachen lesen lernen. Band 2: Seltener gelernte germanische Sprachen. Afrikaans, Färöisch, Friesisch, Jenisch, Jiddisch, Limburgisch, Luxemburgisch, Niederdeutsch, Nynorsk. Shaker, Düren 2019, ISBN 978-3-8440- 6412-4, S. 127–160 und 298. 

a ßach – viel(e) 

ßejder – Ordnung, Reihenfolge 

ßmach – Grundlage

ßojcher – Händler 

ßchojre – Ware 

ßakóne – Gefahr 

ßibe – Grund, Ursache 

ßod – Geheimnis 

ßojne – Feind 

ßam – Gift

ßimche – Freude, Fest 

ßof – Ende 

ßof-kl-ßóf – letztendlich 

ßtam – schlecht- hin, einfach 

ßejchl – Verstand

ßimen – Zeichen, Merkmal 

ßtire – Widerspruch 

ßofek – Zweifel 

ßwiwe – Umgebung, Milieu