Where do I put the font EULA?

[Jellby]

Originally Posted by Psymon

<p class="olde"><span class="dropcap drop1"><img alt="I" src="../Images/init-t.png" /></span>hou <span style="white-space: nowrap">appere¥</span> in my thouwts <span style="white-space: nowrap">ƒo</span> <span style="white-space: nowrap">o¤en</span> of lait and I can nowth healpe bwt thynke hauw vniqwe thou hauef becom, hauw <i>eccepcionall</i> ...

I'm sorry, but that code is wrong. You are using a font with non-standard mapping, and then some symbol codes for different characters. This is disallowed by the ePub spec:

"Fonts must not provide mappings for Unicode characters that would change the semantics of the text (e.g. mapping the letter "A" to a biohazard symbol)"

The long-s has its own Unicode spot, and ligatures should be dealt with by the font, not by using specific characters in the input. You should write:

Thou appereſt in my thouwts ſo often...

and use a font with ſ glyph and proper ſt and ft ligatures. Then you won't need those "nowrap" spans (as long as the reader understand that ſ is a letter)

if the word happens to be at the end of the line and can "break", it will, and it won't do so with a hyphen, so there's no indication to the reader

which, of course, was common in 16th century typesetting

[Psymon]

Originally Posted by Jellby

I'm sorry, but that code is wrong. You are using a font with non-standard mapping, and then some symbol codes for different characters. This is disallowed by the ePub spec:

"Fonts must not provide mappings for Unicode characters that would change the semantics of the text (e.g. mapping the letter "A" to a biohazard symbol)"

Well, that's depressing -- very seriously so. My ebook works just fine (in iBooks), and everything validates just fine, so who checks these things and, indeed, then disallows them? Apple states, for example, that any book uploaded for sale on their store "must" validate, and yet they have an ebook up for sale there -- about how to design ebooks, no less! -- that's chock full of errors and invalidations.

The long-s has its own Unicode spot, and ligatures should be dealt with by the font, not by using specific characters in the input. You should write:

Thou appereſt in my thouwts ſo often...

and use a font with ſ glyph and proper ſt and ft ligatures. Then you won't need those "nowrap" spans (as long as the reader understand that ſ is a letter)

Well, the thing is that in order to do this with an embedded font, the font has to be freeware, of course -- this JSL font set is the only one that I know of like that.

Apart from that, though, if what you're saying above is true, then wouldn't that mean that one can't use any "alt" or "supp" or "exp" fonts embedded in one's epub? I'm thinking, for example, of Adobe Caslon (among countless, countless others). If you wanted to use the "ct" ligature, found in Adobe Caslon Alt Regular, you specify that font and use the "c" character.

And re words breaking at the end of lines...

which, of course, was common in 16th century typesetting

Yes, but they would have a hyphen then -- when my "olde" words break because of those special characters, the two halves are treated as separate words, and no hyphen is inserted. :/

[Jellby]

[QUOTE=Psymon;2667963]Well, that's depressing -- very seriously so. My ebook works just fine (in iBooks), and everything validates just fine, so who checks these things and, indeed, then disallows them?[quote]

There's no way to automatically check this.

Well, the thing is that in order to do this with an embedded font, the font has to be freeware, of course -- this JSL font set is the only one that I know of like that.

I'm sure there are a few freely embeddable fonts with long-s and ligatures.

Apart from that, though, if what you're saying above is true, then wouldn't that mean that one can't use any "alt" or "supp" or "exp" fonts embedded in one's epub? I'm thinking, for example, of Adobe Caslon (among countless, countless others). If you wanted to use the "ct" ligature, found in Adobe Caslon Alt Regular, you specify that font and use the "c" character.

It depends on what you mean with "can't". It certainly means that it's against the specification, and for a very good reason: the text is garbled without the embedded font. And you just cannot assume that the embedded font will work at all, because:

- A user may disable them.
- A reader may not support them (or the specific format/version you are using).
- A medium may not use fonts at all (think text-to-speech)

and it would break searching too, I'd have to search "acion" if I want to find "action".

But there's probably an OpenType Caslon font with a proper "ct" ligature that appears (maybe optionally) whenever there is a c + t in the text... that's what one should use.

Yes, but they would have a hyphen then -- when my "olde" words break because of those special characters, the two halves are treated as separate words, and no hyphen is inserted. :/

Actually, I meant that old books used to break words without a hyphen or any other sign.

[JSWolf]

ADE 1.8 (I think) and later (2.0 does for sure) use ligature. When you have the letter combination such as fl you do get the ligature for fl if it is in the font. But older versions of ADE (1.72 and earlier) do not use ligatures and if you do put in the ligatures by hand, you do break searching.

So, the solution is to just put int he words as they should be without the ligatures and let the reading software deal with the ligatures or not. And to be honest, most people would know really care (or even notice) about there not being ligatures.

[Psymon]

Originally Posted by Jellby

'm sure there are a few freely embeddable fonts with long-s and ligatures.

I haven't found one -- not free, anyway, and not with all those ligs. If they have the ligs, they're not free, and either way when they have them they're often in a separate "alt" font. And this JSL set is not only free, but it's beautifully designed by an expert type designer who knows what he's doing (which many free font designers don't, and their fonts have horrendous kerning, etc.).

I gotta say, you guys almost had me giving up and resorting to graphics instead! But then I thought about your points, and realized that even your responses (I'm just focussing on yours here, Jellby to reply to) had issues, too.

I think I should point out, first of all, that this ebook of mine is in two parts/halves. The first half is my original text, which was written in plain ol' modern English, and that entire section makes use of nothing funny or unusual in the coding -- I don't even use an embedded font at all (other than for the title page and chapter headings). So that entire first half of my book is perfectly readable, perfectly searchable, and certainly text-to-speech readers would have no problem with it. So there's no issues at all with the first half of my book.

The second half of my book is the exact same text, but transliterated into late-Middle English, and then typeset (so to speak) using the typographic rules of that period (in a period font, of course).

So as far as anyone being able to search my book, it's not like the entire book is "lost" -- the first half (which is exactly the same text, really, as the second half) is perfectly normal English, in a plain ol' standard default font, and perfectly searchable.

As for the second half, well, I don't know too many people that have a clue about late-Middle English spellings -- indeed, many of the words are virtually unrecognizable (for example, "nevertheless" gets transliterated to "natheles"). This has nothing to do with the searchability of the font used -- who would be searching for late-Middle English words and spellings, let alone when typeset by authentic period rules?

And this latter brings up another issue with trying to do what I'm doing. Typesetting in that period goes far beyond the use of ligatures. For example, the use of the letter "u" versus "v" had nothing to do with their sound, but rather their placement within the word (similarly for the long-ess).

- words beginning with "u" or "v" would be typeset starting with "v", eg. "vse", "vp", "vnion", etc.

- "u" was used everywhere else in the word, except the beginning, hence we end up with "vniuerse" (universe), etc.

There's no way that you can typeset that correctly, by the rules of typography of that time -- even if you don't use any ligatures or special characters of any sort -- whereby someone would search for a word by the usual, modern spelling and then find the word.

But then, as I said, the first half of my book is in plain ol' modern English (with the exact same text).

As for some of your other comments...

It depends on what you mean with "can't". It certainly means that it's against the specification, and for a very good reason: the text is garbled without the embedded font. And you just cannot assume that the embedded font will work at all, because:

- A user may disable them.

That's certainly true, but I put a note at the beginning of the text of that second half of my book to the effect that readers should be sure to enable the original publisher's font, otherwise the text may not display correctly.

- A reader may not support them (or the specific format/version you are using).

I know -- that's why I'm only designing this for iBooks (unless I can find a way to do it for other readers). It works perfectly fine -- if not beautifully -- in iBooks.

- A medium may not use fonts at all (think text-to-speech)

I know, but even if I typeset that "olde" half of my book in a regular font, with no funny business (no ligs, etc.), I seriously doubt that a text-to-speech reader would be able to manage with the late-Middle English.

and it would break searching too, I'd have to search "acion" if I want to find "action".

As I said, there's still the first half of my book, in "regular" English.

Actually, I meant that old books used to break words without a hyphen or any other sign.

No they didn't -- and I've been studying the history of print and collecting early printed books for nearly four decades now, and have even had a website on the subject for almost two decades. If ever they broke words at the end of a line and didn't use a hyphen, it was an anomaly, and certainly not standard typography for the time. Trust me on that one.

In any case, for all of the above reasons, no matter how I might try to represent this period text (16th century late-Middle English) so that it displays as it would have in that time, it's not going to be searchable -- and sure, I could do it all up as graphics so that then I could easily make it available on all platforms, no problem, but graphics wouldn't be searchable either!

And like I said, my book is in two parts/halves. The first part is really the "readable" text, in plain English -- no funny font business, and perfectly readable/searchable. The second half is, well, really meant to be typographic "art".

If you still think that I'm going to have issues with trying to publish this book (only on iBooks, unless I can figure out a way to make it work on other platforms), then I'm truly all ears! Seriously, if you guys or anyone else foresees any issues I might have with this -- in light of my explanation(s) above -- then I really do want to hear.

This work is very near and dear to me, and it's very important to me that I get this right, and do this right.

[Tex2002ans]

Originally Posted by Psymon

Here, I'll include a screen shot -- see the attachment here -- of one of my pages (actually two pages, of course), as it looks in iBooks. I'm using a digitized period font, of course, to pull this off, embedded in the epub, that has tons of ligatures and stuff, both in roman and italic.

Oh yeah, the instant I saw your image and read your explanations, I remember you posting about it in another topic (I have your sample EPUB sitting right on my MobileRead folder in a "Psymon" subfolder). Isn't that the same exact one Hitch warned about marketshare?

Originally Posted by Psymon

And just to show you how screwy the coding is for this, here's the HTML for the selection of text that you see in that screenshot...

Any reason you are not using CSS for those nowraps?

[...] <span class="nowrap">appere¥</span> [...]

.nowrap {
	white-space: nowrap;
}

Originally Posted by Psymon

And this latter brings up another issue with trying to do what I'm doing. Typesetting in that period goes far beyond the use of ligatures. For example, the use of the letter "u" versus "v" had nothing to do with their sound, but rather their placement within the word (similarly for the long-ess).

- words beginning with "u" or "v" would be typeset starting with "v", eg. "vse", "vp", "vnion", etc.

- "u" was used everywhere else in the word, except the beginning, hence we end up with "vniuerse" (universe), etc.

Ugh, this brings back bad reminders of a book I finished converting a month ago. The author was quoting a lot of books from the late 1500s-1600s:

<p class="negindenttwo">[<i>a</i>] euerie man is desirous to make ouer money by exchange, and that money which should be employed vpon the commodities of the realme, is deliuered by exchange to the great hindrance of the vent and aduancement of our home commodities: and yet the forrein commodities not any way therefore sold the better cheap<a href="#fn51" id="ft51">[51]</a>; [and]</p>

I mostly updated the 'i' -> 'j' and the 'v' -> 'u' just to make it more readable, as those were quite jarring:

vse -> use
iust -> just
deliuer -> deliver
aduance -> advance
vp -> up
ouer -> over
aboue -> above
loue -> low
obiection -> objection
vs -> us
giuen -> given
obserued -> observed
euerie -> every

[Psymon]

Originally Posted by Tex2002ans

Oh yeah, the instant I saw your image and read your explanations, I remember you posting about it in another topic (I have your sample EPUB sitting right on my MobileRead folder in a "Psymon" subfolder). Isn't that the same exact one Hitch warned about marketshare?

Yup! It's my first and only ebook (so far), so that has to be it.

I'm flattered you actually have a folder for me -- don't expect it to get filled up too soon! I've been putting a great deal of effort into this first ebook of mine (not to mention that with it being my first, it's been quite a learning curve for me), and although I do have another one planned for after this one is actually up on for sale, after that I have nothing else currently in mind to do. Who knows what the future holds, though???

Any reason you are not using CSS for those nowraps?

Ha! Because I'm an idiot! In all honesty, I really don't know why, but it simply never occurred to me. Thanks! I'll definitely change that first thing next time I dive into my code!

Ugh, this brings back bad reminders of a book I finished converting a month ago. The author was quoting a lot of books from the late 1500s-1600s:

I mostly updated the 'i' -> 'j' and the 'v' -> 'u' just to make it more readable, as those were quite jarring

Oh, I can totally relate, I've done that, too -- taking "olde" texts and updating them to modern typographic rules -- but that's just it, in my case I'm not updating something, but backdating something! This isn't actually an old text, I wrote this myself -- the original work is a literary piece that was written in plain, modern English, and I took that and transliterated it into late-Middle English, and then endeavoured to set it in early 16th century typographic style. The whole point of this work is to have it looking genuinely -- and typographically -- "olde".

I see you were replying to an earlier post of mine -- if you go back to my last/previous reply here, I explained what this work is actually about a little more (with it being in two parts, the original "modern" version, plus this "olde" version of the exact same text -- both of which I wrote myself).

[Jellby]

Originally Posted by Psymon

I haven't found one -- not free, anyway, and not with all those ligs. If they have the ligs, they're not free, and either way when they have them they're often in a separate "alt" font. And this JSL set is not only free, but it's beautifully designed by an expert type designer who knows what he's doing (which many free font designers don't, and their fonts have horrendous kerning, etc.).

With FontForge it is relatively easy (if anything can be said to be "easy" with FontForge) to take the ligatures from the "alt" font, insert them in some free private slot of the regular font, and add the ligature definition when needed. Or move the glyphs around. With free fonts, that's all you need. (That, and change the kernings of the moved glyphs, I guess.)

If the fonts are designed by someone knowledgeable, you could contact the author and ask whether he would mind doing that himself, or if he'd oppose your doing that for your specific needs.

I don't have time for a full search now, but:
Heuristica has long-s, and a number of f* and ſ* ligatures (but no ct, ſt)
Espinosa Nova apparently has long-s and ligatures (the Regular font is free, but I don't know if embedding is allowed)

No they didn't -- and I've been studying the history of print and collecting early printed books for nearly four decades now, and have even had a website on the subject for almost two decades. If ever they broke words at the end of a line and didn't use a hyphen, it was an anomaly, and certainly not standard typography for the time. Trust me on that one.

OK, then it might be in even older times, or in a different language, but I'm pretty sure I have seen words broken across lines without hyphens as a normal case. I've just checked the Gutenberg Bible now, and it has "="-like symbols for broken words, so it might have been somewhere else...

[Psymon]

Originally Posted by Jellby

With FontForge it is relatively easy (if anything can be said to be "easy" with FontForge) to take the ligatures from the "alt" font, insert them in some free private slot of the regular font, and add the ligature definition when needed. Or move the glyphs around. With free fonts, that's all you need. (That, and change the kernings of the moved glyphs, I guess.)

But that's what the font(s) that I'm using already do, and about that you yourself wrote in a previous message:

I'm sorry, but that code is wrong. You are using a font with non-standard mapping, and then some symbol codes for different characters. This is disallowed by the ePub spec:

"Fonts must not provide mappings for Unicode characters that would change the semantics of the text (e.g. mapping the letter "A" to a biohazard symbol)"

So I may as well just stick with what I've got! After all, it works great, looks great, not to mention there was an enormous amount of work/time that went into creating it -- if I was to change the font, I'd pretty much have to start all over, from scratch.

If the fonts are designed by someone knowledgeable, you could contact the author and ask whether he would mind doing that himself, or if he'd oppose your doing that for your specific needs.

But I already have a beautiful free font.

I don't have time for a full search now, but:
Heuristica has long-s, and a number of f* and ſ* ligatures (but no ct, ſt)

Cool, thanks! I've downloaded it (haven't unzipped and looked at it yet, though), but I'd really miss those latter two ligs. :/

Espinosa Nova apparently has long-s and ligatures (the Regular font is free, but I don't know if embedding is allowed)

Cool again! But I need italics, too. That regular does look like it has a whole ton of really nice ligs, though...

http://www.myfonts.com/fonts/estudio...ar/glyphs.html

...but like I said, I need the italics, too. :/ Looks like a very, very nice font set, though! What I prefer about the JSL font over this one, though, is the somewhat "distressed" look of it (the former) -- it actually looks like it was made with metal type, with all its imperfections. The Espinoza Nova one looks a bit too smooth and "perfect" -- it looks digital.

OK, then it might be in even older times, or in a different language, but I'm pretty sure I have seen words broken across lines without hyphens as a normal case. I've just checked the Gutenberg Bible now, and it has "="-like symbols for broken words, so it might have been somewhere else...

Yeah, it's possible that some printers did that (break words at the end of a line but not insert a hyphen), but off-hand I don't recall seeing it before -- it certainly wasn't the norm, that's for sure.

And yeah, with fraktur-style ("blackletter") fonts, like in the Gutenberg Bible, the hyphen did generally look like an "="-sign, but with Roman and Italic the double-line was reduced to just a single "dash".

Thanks very much for your input! You haven't quite changed my mind on anything, but you've certainly gotten me thinking a lot! I do think that I should be okay, as I've done it -- all I can do is finish it up, as I've been doing it, wait for my application with iTunes Connect to get approved (still waiting since I applied last week), and then load it up and see what happens, I guess.

[Jellby]

Originally Posted by Psymon

But that's what the font(s) that I'm using already do

Erm... No, if I understood it correctly. Your font does not use the private Unicode slots, but it uses some already reserved slots, for different characters, and makes them look like something different.

And it doesn't have a proper ligature definition, instead it forces you to use specific and foreign characters in your text. Or, if it does, then you are not using it correctly. If you write "ct" in your text, does the font display the ct ligature? If it does, then that part is correct, and you don't need to write the ¢ or whatever it is (but it still should be placed in the private Unicode region).

So I may as well just stick with what I've got! After all, it works great, looks great, not to mention there was an enormous amount of work/time that went into creating it -- if I was to change the font, I'd pretty much have to start all over, from scratch.

As long as it's your personal work, do whatever you think is best for you. And about the last remark, that's precisely one of the things it would solve. If you leave all that ligature handling to the font, you could change fonts without having to change all the text. Suppose you now find a new font that has everything this one has, plus a nice "Th" ligature... you'd have to go back to your text and change all "Th" into "þ" or something like that. If you had it right from the beginning, and the fonts are "correct", all you have to do is change the font.

But I already have a beautiful free font.

It shouldn't take very long to "fix" it, as long as the license allows it (and often that's fine as long as you change the font name).

[Psymon]

Originally Posted by Jellby

Erm... No, if I understood it correctly. Your font does not use the private Unicode slots, but it uses some already reserved slots, for different characters, and makes them look like something different.

Just like the aforementioned Adobe Caslon Alt Regular font does -- it puts the "ct" lig in the "c" slot.

And it doesn't have a proper ligature definition, instead it forces you to use specific and foreign characters in your text. Or, if it does, then you are not using it correctly. If you write "ct" in your text, does the font display the ct ligature? If it does, then that part is correct, and you don't need to write the ¢ or whatever it is (but it still should be placed in the private Unicode region).

I just fired up my word processor and tried it, and it doesn't seem to (i.e. change "ct" to the lig, just by typing "ct"). However, that font has a lot more ligs than most fonts do: ct, st, sh, fi, ss, ffi and ff; and the italic has those plus is, sl, ll, as well as an alternate v.

As long as it's your personal work, do whatever you think is best for you. And about the last remark, that's precisely one of the things it would solve. If you leave all that ligature handling to the font, you could change fonts without having to change all the text. Suppose you now find a new font that has everything this one has, plus a nice "Th" ligature... you'd have to go back to your text and change all "Th" into "þ" or something like that. If you had it right from the beginning, and the fonts are "correct", all you have to do is change the font.

Well, I've looked and looked and looked, and although there are a couple of other fonts that look reasonably nice (that is, reasonably "authentic" for the period, and not overly-cleaned-up, digitally faked reproductions), I have yet to find any other font that is as nice and "complete" as this one -- and that's even with paid, non-embeddable versions, never mind freeware. In that regard, it's actually astonishing to me -- and certainly fortuitous for me -- that this JSL font set is free, and embeddable.

In any case, though, correct me if I'm wrong, but your only point in stressing that the font should use "legitimate" slots for where it puts the ligs is so that the text can be searchable and stuff, right? I mean, what other reason matters? It works just fine, and visually looks just fine -- the only issue might be readability and searchability. Well, in both those regards, that's virtually irrelevant: the text that it's displaying isn't modern English, but late-Middle English. Who would be doing a "search" within that (unless they're some prof or something who knows the language -- and all the insane variants of spelling, etc. -- from that time)?

The whole point of that second half of my book (with the first half being the same text in modern English) is more for the sake of "art" than anything else -- I suppose it's a study in language, too, of course, but even that is an artistic endeavour. Readability (that is, the easy understanding of the language), searchability and whatever else is secondary, really, than the visual impact of the work. And you might say that I'm making it inaccessible to visually impaired people, but, well, if I created an art book full of paintings, that's rather inaccessible in that way, too. I can't help that.

So while I do understand your point about this font set -- and how it would most definitely apply to, say, the first half of my book (in modern English) or any other "regular" book I might create -- I'm at a bit of a loss as to what exactly the issue might be in this particular case, with what I'm trying to accomplish here.

With that said, again I do genuinely appreciate your input, Jellby! This has certainly been a thought-provoking discussion, and these have all been very good things to think about -- alot of it I hadn't even considered, but even if I still think I should be okay in doing things the way I've been doing them (for lack of any other way to do them), at least I can now say that I'm doing them for well-thought-out, rational, mulled-over reasons.

[Jellby]

Originally Posted by Psymon

Just like the aforementioned Adobe Caslon Alt Regular font does -- it puts the "ct" lig in the "c" slot.

Which is exactly why it is not "correct" either

I just fired up my word processor and tried it, and it doesn't seem to (i.e. change "ct" to the lig, just by typing "ct"). However, that font has a lot more ligs than most fonts do: ct, st, sh, fi, ss, ffi and ff; and the italic has those plus is, sl, ll, as well as an alternate v.

It doesn't matter how many ligatures, all the font needs is some instruction saying "when these characters appear together, display this other glyph". For alternate forms and other fancy stuff, there are OpenType features... but those are usually rather poorly supported, and are more complex to use correctly.

In any case, though, correct me if I'm wrong, but your only point in stressing that the font should use "legitimate" slots for where it puts the ligs is so that the text can be searchable and stuff, right? I mean, what other reason matters?

Searchable and readable if the font is not used or available (for whatever reason). And all this began because you were complaining (among other things) of your words breaking between lines without a hyphen, which would not happen if the reading software knew that those ligatures are normal letters and not weird symbols.

And it would be easier for you to edit the book, and it would be possible to copy-paste, etc.

The problem is not the font per se, but the fact that, in order to use the font as you want, you have to alter the text in such as way that it is only readable with this particular font. If you are fine with that, go on, no policeman will be knocking at your door, but you should be aware of the possible problems.

It works just fine, and visually looks just fine -- the only issue might be readability and searchability. Well, in both those regards, that's virtually irrelevant: the text that it's displaying isn't modern English, but late-Middle English. Who would be doing a "search" within that (unless they're some prof or something who knows the language -- and all the insane variants of spelling, etc. -- from that time)?

Do not assume that "nobody will want to search", or "no one will ever copy paste that", or "no sane person would use a text-to-speech with this". The world is full of weird people

So while I do understand your point about this font set -- and how it would most definitely apply to, say, the first half of my book (in modern English) or any other "regular" book I might create -- I'm at a bit of a loss as to what exactly the issue might be in this particular case, with what I'm trying to accomplish here.

Not a real issue, I was simply calling your attention on this particular problem, it's up to you to decide whether it's important enough for you.

That, and the nitpicker in me wanting the world to be perfect

If the font license permits it, I'm willing to have a go at "fixing" it... I might even use it myself (check the Gulliver PDF if you don't believe it).

[Psymon]

Originally Posted by Jellby

Searchable and readable if the font is not used or available (for whatever reason). And all this began because you were complaining (among other things) of your words breaking between lines without a hyphen

Actually, not exactly -- my mentioning that was because I mentioned that my ebook will only work correctly in iBooks, and Tex2002ans then asked me "May I ask what you are using in your book that only works in iBooks?", and so that's how I then mentioned the font issue. I started this thread to inquire about where to put the font EULA (hence the subject here, of course), all this other stuff came up rather inadvertently.

And it would be easier for you to edit the book, and it would be possible to copy-paste, etc.

I don't need to edit it anymore, there's nothing that's going to be changing as far as what the text "says".

The problem is not the font per se, but the fact that, in order to use the font as you want, you have to alter the text in such as way that it is only readable with this particular font. If you are fine with that, go on, no policeman will be knocking at your door, but you should be aware of the possible problems.

I just don't have much choice! To create what I want to create, I need a font that LOOKS good (and looks like it's from that period in printing history), includes both roman and italic, and has a nice bunch of ligatures -- and, of course, is free and embeddable. This JSL font is the only one that I know of like that, that meets that criteria -- and I've looked around, a lot, for years and years.

Do not assume that "nobody will want to search",

I didn't quite say that -- or mean that, in the sense you mean. As I've said before, my book is in two parts/halves -- both the exact same text. One half is regular, modern English (with no weird "font stuff" happening), and the other half is the same thing transliterated into late-Modern English and typeset according to rule/practices of the time. No matter what font I might choose to do the latter in -- even if the font is all "correct" (as per your specs) -- then if a person was to search for, say, the word "universe", they would find it in the modern half of the book, but in the "olde" half of the book it would be typeset as "vniuerse".

I find it highly unlikely that someone would be doing a search for "vniuerse", let alone any of the other, far more odd spellings and variations of words. And if for some crazy reason they really wanted to find it, well, they could search for it in plain, modern English, find it in the modern half, and then easily go find the equivalent page in the "olde" half.

or "no one will ever copy paste that",

Even if they could, I'll repeat again that that part/half of my book is meant as a visual thing -- even if it was copy/pastable, no matter what font I set it in, unless they have that font on their system it's not going to come out the same in whatever other program they're pasting into.

or "no sane person would use a text-to-speech with this".

I seriously doubt that there's a text-to-speech reader that can "speak" in late-Middle English.

The world is full of weird people

Indeed -- and I guess I'm one of them, so why not cheer me on?

Not a real issue, I was simply calling your attention on this particular problem, it's up to you to decide whether it's important enough for you.

Like I said, there's no other way to do it -- that's the only font that does what I want to do (that I'm aware of, after years of looking for fonts like that).

That, and the nitpicker in me wanting the world to be perfect

Yes, but at least you're "self-aware."

If the font license permits it, I'm willing to have a go at "fixing" it... I might even use it myself

I'm not sure what it is that you're offering to "have a go at" or "fix", but in any case here's what the EULA for that font says in that regard...

"JSL Ancient and JSL Ancient Italic are copyright (c) 1997-2000 by
Jeffrey S. Lee. Permission is granted to freely distribute them,
provided that they are distributed unaltered, both the roman and italic
versions are distributed together, and they are accompanied by this text
file."

If you're curious to see/have this font, it's easy enough to find -- just search for "JSL Ancient" and it comes up right away.

(check the Gulliver PDF if you don't believe it).

Oh, I believe you, don't worry! I can see that you have a better handle on this stuff than most people do. Again, I appreciate the input -- but like I said, I don't really see much way around this, to do things differently than I have.

And as I pointed out at the beginning of my reply here, this isn't even why I started this whole thread -- all this font stuff just came up "by accident" rather inadvertently. It wasn't an "issue" for me anymore -- I've been quite happy with how things have turned out. It's just a shame that the JSL Ancient font would appear to be the only font that meets my needs and that my ebook (as it is now) will only work in iBooks. It could work in other platforms, I'm quite sure, if only they paid heed to that CSS "nowrap" styling. If you ask me, that's a far bigger issue that they (non-iBooks platforms) have -- in not allowing what is actually a fairly simple, straightforward and, indeed, rather important feature of CSS that undoubtedly many, many people might want to make use of -- than any font issue that I have.

[Jellby]

Originally Posted by Psymon

Well, I've looked and looked and looked, and although there are a couple of other fonts that look reasonably nice (that is, reasonably "authentic" for the period, and not overly-cleaned-up, digitally faked reproductions)

Linux Libertine has the long-s and ligatures (for "st" and "ct" you have to enable the "hlig" feature). It's probably too "clean" for you, but at least I found one

[Psymon]

Originally Posted by Jellby

Linux Libertine has the long-s and ligatures (for "st" and "ct" you have to enable the "hlig" feature). It's probably too "clean" for you, but at least I found one

Yes, it's too "clean", it doesn't look "olde" -- there are other fonts out there that do look "olde" (Caslon Antique, for example), but there's just no other font that I've found that works as well as these JSL fonts.

I appreciate your efforts to help look, but I suspect that you're invariably just going down the same paths that I've already explored -- and the thing is, I'm happy with what I've done, it's finished and over with, and I can't see the point in re-doing it all only to make it all look less nice. :/