Glyph scaling -- good or bad?

[Gene Trujillo]

There's a good thread on that here:

<http://www.adobeforums.com/cgi-bin/webx?128@@.1de6732b>

Basically, it's a matter of preference and moderation.

Gene

[michelle jean rice]

I am working on hnj settings for newspaper columns. A friend at Adobe suggested setting glyph scaling at 98,100,102. To my eye, that improves type color but does not distort the typeface. Others, however, are adamant that glyph scaling should be 100,100,100. What do you folks think and what kind of settings are you using? We are using a custom serif font at 8.4 on 9. Column width is 9p5.
Thanks,
Michelle

[Dave Saunders]

Gene,

Thanks for pointing out that topic in the Mac Archive. I'd forgotten about it. I just marked it to sort to the top of the archive and to not expire automatically.

Dave

[Dave Saunders]

98,100,102 looks good to me. I've seen 97,100,103 advocated too.

People who think that glyph scaling is a no-no should be shown a page from the Gutenberg Bible. That's when glyph scaling was first used, nearly six centuries ago.

Dave

[George Bilalis]

Michelle,

this advice resolves to +/-2%. This is about the maximum that goes unnoticed by any means. As far as I am concerned this is also true for point size and kerning.

Talking about narrow columns (newspaper work, we do here as well) this is something we normaly apply, and depending on the space band and hyphenation adjustments works pretty well. It is specially meaningfull where there are restrictions on the other parameters and you still have to justify properly.

What space band adjustments do you use?

regards
George

[michelle jean rice]

Hi George,
I've got word spacing at 98 100 210 and letter spacing at 0 0 8. Started by taking one of our Quark pages and opening it in InD, checking those settings and then adjusting for a zillion variations until I got one I thought stood out.
And thanks Gene, I will check out that thread. I did a search before posting, but it didn't turn up anything.

[Gene Trujillo]

I just marked it to sort to the top of the archive and to not expire automatically.

In that case, can we change the topic title to something a little less, um, contentious and more informative? "Glyph Scaling" would work, if you can do it.

I named it that with my tongue firmly implanted in my cheek because of a huge flame war at the time, but now I find it a bit embarrassing...

[Dave Saunders]

Michelle,

Without trying it, perhaps I shouldn't speak up, but those settings look horrendous (unless you're using them for ragged text, in which case only the middle numbers are significant).

Your settings favor spreading over squeezing by such a large amount that I can't imagine the result being pleasing for justified text.

Dave

[michelle jean rice]

Dave,
It seemed fairly outrageous to me, too. But I've set the same stories with those settings and the InD defaults (and lots of other variations). InD default has rivers and the 98 100 210 looks good. But, being more of a journalist than a typographer, I am certainly open to suggestion on other settings to try.

[Dave Saunders]

Done. I left the original name in parentheses to provide some continuity.

Dave

[Dave Saunders]

It was more the 0,0,8 that shocked me. I guess that in narrow columns you have to have a lot of slack, and better it be between the words than within them.

How does:

Word Spacing: 90 100,210
Letter Spacing: -2,0,5
Glyph Scaling: 98,100,102

compare to what you've been getting? You might even try:

W: 90 100 180
L: -2 0 3
G: 97 100 103

Dave

[George Bilalis]

Dave,

your

W: 90 100 180
L: -2 0 3
G: 97 100 103

is a good proposition for Michelle.
I have a feeling Michelle is setting in french. In this case (considering the average word length in french being longer (more hyphenation breaks per word) than english and with a good french hyphenation algorithm (I believe the one used in ID is good), there should be no great pain to justify his text.

George

[pleader]

Glyph scaling -- good or bad?

It depends. Experiment. Observe. Implement.

Regards,
Victor Lee.

[Dominic Hurley]

Glyph scaling, good or bad - bad I say, but then I'm a purist. I would never use glyph scaling (as implemented in ID) or letterspacing when justifying text.

And I can't let the mention of Gutenberg's Bible pass without comment. Sure, Gutenberg used glyphs of different widths. But I think he would be a little miffed if it were suggested that these were no more than mathematically scaled versions. They were created with the same care as the standard glyphs, just as all metal type used to be optimised for its size or width, not merely scaled. Herman Zapf's HZ program (which provided the basis for ID's h&j routines) allowed for glyph scaling, but, and this is important, the counters were to be adjusted, not the strokes. The difference is the same as between using a true condensed font and a compressed font. A full implementation of glyph scaling should have a result akin to using multiple masters, but Adobe has acknowledged that resource considerations were one of the reasons that this was not done in ID.

It seems to me that many people these days avoid hyphenating words at all, and if they do hypenate them, they do it no more than twice in a row, and that this has led to the prevalence of letterspacing and other undesirable practices (I'm sure we've all seen a single word spread out to occupy the full width of a narrow column). Indeed, in a recent design magazine, a so-called type "expert" wrote "hyphens are fine for compound words, but they don't belong at the end of a line of type". Sorry? Since when? She then went on to advocate adjusting horizontal spacing, which I take to mean letterspacing, instead of using hyphens. It used to be that letterspacing was considered a cardinal sin and that hyphenating was far preferable to uneven and wide word spacing. Now hyphens are out or limited to two in a row (by contrast, if we turn again to Gutenberg, we often see four, five, or six hyphens in a row in the 42-line Bible). I say it's time to strike a blow for hyphenation and reintroduce the concept of even tight spacing!

Dominic

[pleader]

The trick, of course, is to balance the numbers and obtain the best results (unobtrusively).

I, for one, am comfortable with glyph scaling of 2 or 3 percent, plus/minus. Not many people with bionic eyes out there.

There is no law that says any particular set of numbers can work across all situations.

Regards,
Victor Lee.

[Dave Saunders]

I'm inclined to think that had Gutenberg access to mathematically scaled type, he'd have saved a whole lot of bother by allowing +/- 2%.

I guess the issue for self proclaimed purists is: what exactly are you being pure about?

Dave

[Guy Smiley]

I'm sure the scribes considered type itself to be "impure".
--Mike Nitabach

[Thomas Phinney]

Dave:

The purists want the stroke thickness to be consistent from one line to the next.

Personally, I avoid glyph scaling except in relatively extreme situations. However, I see newspaper columns, with their completely unreasonable narrow width, to be such a situation.

Although most people won't *notice* differences of +/- 2%, the difference may still have an effect on the viewer. But I agree that it's pretty small.

And yes, the scribes of Gutenberg's era considered typesetting to be an obviously inferior substitute. Lots of people thought it would never take over from hand-written books....

Cheers,

T

[Dominic Hurley]

Yep, as Thomas says, and as a self-proclaimed purist, I would like to see stroke thickness consistent across lines. That is, I don't want to distort the type forms just because I can't immediately see a better way to break the lines or don't want to take the time to do it. I also want to use optical masters, not simple scaled typefaces, though in many cases that's just not possible. My aim is to try to produce type as wonderful as the handset hot metal type of old. And that's not meant as a criticism of those who do use glyph scaling - it's just my personal answer to the question posed by Michelle.

But I do disagree that Gutenberg would have used scaling - after all, if he'd wanted to do things the easy way and save himself "a whole lot of bother", he'd never have hung the punctuation or created the wide range of different width sorts and ligatures that he did. That's why his Bible is held up as a classic of typography - he didn't take the easy way.

Dominic

PS. For those who doubt that minor changes can be noticed, try setting three identical lines of type, each fully justified, in ID. For the first, use glyph scaling of 100, 100, 100; for the second, 102, 102, 102; and for the third, 98, 98, 98, and then compare them, especially the last two. If you have minimum and maximum settings of 98 and 102 respectively, you could end up with a difference between two adjacent lines of 4 per cent, and I think even non-type people subconsciously (if not consciously) notice this difference.

[pleader]

Dominic,

That's a before and after comparison. Placed side by side, there would be an obvious difference for sure. But even in your test, if the results 100-100-100 and 98-100-102 were submitted to the typographic police, I think they would be hard put to say which is "better", although they might lean towards the latter because of the more even spacing.

Try it and see.

Regards,
Victor Lee.

[Dominic Hurley]

Well, I disagree with that - as I've said, I don't think distorting the letterforms qualifies in any way as better and I prefer to achieve even spacing by careful line breaks and hyphenation. But that's just my opinion and I know you and others feel differently. But I think you've missed my point, which is that, although everyone here has been talking as though max and min settings of only a couple of points either side of 100 won't be noticed, the difference is actually quite clear. I agree that you won't see two lines of type side by side like this, but you will see two "the"s or two "and"s in adjacent lines and these may vary from each other by up to 4 per cent (given the settings mentioned earlier). And I believe readers can subconsciously (or consciously, if they're so informed) pick up on these differences in the glyph strokes.

[Anders Svedin]

I'd like to add a side-note to this, which has to do with the (obvious) fact that ID won't always succeed in setting the lines according to your preferences. When playing around with glyph scaling I noticed that if you set it to anything other than 100-100-100, ID might in problematic cases scale specific lines way below or above your settings. Glyph scaling settings of 99-100-101 would be acceptable for me, but I won't touch them until I know that no line of my text will be stretched more than that. Way too huge word-spacing is equally ugly of course, but directly visible on screen - unlike incorrect glyph scaling that might look OK at a quick glance, but come out a major embarrasment in the printed product.

Cheers,
Anders

[pleader]

Dominic,

Have you done a "blind" test? I think that experiment would be very telling.

Anders,

I've never noticed InDesign breaking the specified glyph scaling limits. If you are right then it would be a serious bug in my opinion.

Regards,
Victor Lee.

[Anders Svedin]

Victor,

Unless it's been fixed in version 2.0 (I only tried it in 1.5): make a new paragraph style with word spacing set very tight (90-100-110 or even less) and glyph scaling 98-100-102, allow max 1 hyphenation in a row and ofcourse no letterspacing. Then create a couple of fairly small columns and fill them with placeholder text, perhaps deleting a few spaces as well so as to make longer words. Under these extreme circumstances I'll think you'll see it happen.

Otherwise you can always force an override by using em spaces instead of normal ones. In this case though I guess it's an understandable last resort and perhaps better than breaking the line or forcing it out of the frame as overset text, although I would frankly prefer the later.

Agreed, these are very unlikely situations in actual typesetting. The main reason I distrust ID's glyph scaling is that the text in the proofs I made with it set at 98-100-102 seemed very disharmonious and uneasy, and many lines made me suspect overrides. Either that or I can actually see the shifts in glyph scaling when I read - in which case I would definitely advice against using it.

Cheers,
Anders

[Dominic Hurley]

Hi Victor

I'm not quite sure what I would be testing, but if you're asking whether I've tested to see whether an average reader could pick up ID's glyph scaling, then, no, I haven't done it. But, even if a suitable test could be designed and run, and if the results were to show that people didn't notice or didn't mind glyph scaling, then it still would make no difference to me. As I said, I am a purist in this regard, and so I am coming from a philosophical point of view - namely that I don't believe in distorting the letterforms (however slightly) to make it easier to fit the text in. (Similarly, I don't believe in editing text to make it easier to fit in, but that's another story.) I think that careful hyphenation and line breaking should be used instead of scaling, but I also accept that still sometimes one's desired word spacing parameters may not be met. However, I don't see that as sufficient justification for using scaling and I do think that people do notice such distortions (witness Anders' "disharmonious and uneasy" comment). But, even if I'm proved completely wrong in holding that last belief, it's not going to change my mind on the appropriateness of scaling.

So, I'm not going to run a test. I'm not looking for validation for my position or an excuse to change it, and I'm not looking for "evidence" to convince others to change theirs. Even if I'm the only one on this forum (or even in the world) who doesn't believe in scaling glyphs, then that's okay with me.

Dominic

[Richard Rönnbäck]

Very interesting subject this one!

There is at least one more thing to mention about the effects glyph scaling
might have:

If you don't use it then obviously something else has got to give. Usually
this would either mean a greater variation in word spacing or the
introduction of more hyphens.

Likewise obviously, the number of hyphens is effected by the length of the
words used in the paragraphs.

If you typeset in English, with it's relatively short words (as compared to
many other languages) you may achieve an even spacing without having to
resort to glyph scaling, and without introducing a large numbers of hyphens.

However, the equation is simply not the same if you set type in other
languages. Languages such as Italian or French has got longer words than
English, and if you go any German language (such as Dutch, Danish, Norwegian
or of course German) the average word length is far longer than in English.
(And don't even get me started on Finnish <g>)

In addition to that, many of these languages, due to their nature, combine
shorter words into longer ones. This means that in similarly constructed
languages the number of hyphens will typically be far greater than in
English.

I would like to give you just one example from my native tongue, Swedish to
clarify: The English words "Steam boat captain" would translate to
"ångbåtskapten" in Swedish. As you immediately notice, instead of three
short words we would use one long. (and no, it's not all bad, because it
means that sentences are usually clearer and that we can use fewer commas,
brackets etc.)

Like Dominic, I would be prepared to throw in a word or two for increased
use of hyphenation. BUT, (and here is my actual point) Glyph scaling affects
InDesign's choice of hyphenation point, if you use glyph scaling, you will
often notice that a "good" hyphenation is chosen over a "bad" one.

Since bad hyphenation points has very negative impact on readability, glyph scaling
simply can't be dismissed ONLY because it distorts the letter form (which
also has a negative impact) - you simply have to make a balanced judgment.

With all that said, I personally start by trying other means of getting even
textures and "good" hyphens, and only if the overall result in the end is
better, I resort to glyph scaling.

It would be very interesting to see some of you examples and if I get around
to it I will post a few of mine some day.

Thanks

Richard

[George Bilalis]

You are welcommed Michelle.

George

[michelle jean rice]

Thanks all.
Dave and George, I did some tests and think we'll go with 90,100,210 and -2,0,5. I think the columns look better with glyph scaling of 98,100,102 but I'm not prepared to fight that battle right now, so we'll stick with 100,100,100.
Dominic, you're right about the aversion to hyphens. The reasoning is that type with slight squeeze or expansion is more readable than multiple hyphenated lines. (We'll be doing a lot of work customising the dictionary to make sure the hyphens we do get are in the right place.)
Thanks again,
Michelle

[Dominic Hurley]

"Type with slight squeeze or expansion is more readable than multiple hyphenated lines" - another statement I completely disagree with, but I won't go there. When people can unashamedly write that hyphens don't belong at the ends of lines of type (as did the "expert" I mentioned), I think that particular type refinement (ie, hyphenation) has been well and truly lost to the masses.

In reply to Richard, yes, I agree, language will play a part, as will copy. If you're setting scientific or technical copy in English, you could well expect to run into long words. As with copy concerning the internet, which may have lots of non-breakable web addresses. That's why you've got to have this in mind when you choose your measure, typeface, layout, etc. For this particular example, where we were talking newspaper columns, I did a quick calulation. Assuming a column width of 14 picas (quite a wide column), a 9pt font with a cpp of 3 (which is what you'd expect for a newspaper font), then, if you had a full line of type with no spaces at all (which you wouldn't), the maximum compression or extension you'd get with settings of 98, 100, 102 glyph scaling would be plus or minus 3.36 pt. Because the cpp is 3, that may well not be enough to drag a single character back from the next line and thus will not likely affect hyphenation points much. (I'm assuming here that most use of glyph scaling will be in loose lines.)

Dominic

[Richard Rönnbäck]

Dominic,

While I respect your standpoint and appreciate you calculation I know from
experience that under many real world conditions glyph scaling is sufficient
to play a part in hyphenation. That in itself is not of course an argument
to prefer it or choose to use it - but it will play a part.

Naturally the genre of the text will impact the frequency of long words. My
point however was that to combine shorter word into longer is essential, and
very frequent, in many other languages than English - regardless of genre!
You of course do the same in English sometimes (sic!) but nowhere near as
often as in (particularly) German languages.

Fact is, I got the inspiration to a script that will count the frequency of
long words in a text. If all goes well I will publish it on the scripting
forum in a few days. It will be apple script only as I don't know visual
basic.

However, if I'm right, this thread will get the attention of Olav Kvern
shortly and if so he will probably beat me too it. <g>

Regardless of our opinion on glyph scaling I firmly believe that discussions
such as this one will improve our judgment - I know it does improve mine,
and so has a value in itself, even if you stick to your original opinion.

Thanks

Richard

[George Bilalis]

Hi Dominic,

Doing this thing for many years for newspaper work in greek language (average word length 6.8 characters), the most pronounced effect was always with space band adjustments along with carefull hyphenation (minimum hyphenable word - characters before, characters after).

Using the proper fonts and an efficient hyphenation algorithm with properly adjusted space bands, no glyph scaling was necessary for good justification (or was available for justification purposes in the old qxp 3.X or PM6.5).

George

[George Bilalis]

Richard,

correct me if I am wrong, but I know of no layout program using glyph scaling as a parameter (before ID) for automatic justification purposes.
You could do it manually, but automatic (the way ID does it)?

George

[Gene Trujillo]

Manually scaling glyphs has been possible for a long time -- I've certainly seen enough grotesque examples! What's new is building it into part of the justification engine. So far as I know, ID is a first in this.

I would say so unconditionally, but I don't know for sure about Tex, not having used it. I don't ~think~ it can, but it's so flexible I won't discount the possibility without hearing otherwise.

Gene

[George Bilalis]

Dominic, you wrote:

Because the cpp is 3, that may well not be enough to drag a single character
back from the next line and thus will not likely affect hyphenation points
much.

This would be true, not taking in consideration the paragraph composer, but using the single-line composer.
The paragraph composer being recursive in calculating can have a play with glyph scaling on. Without a paragraph composer, glyph scaling would not matter much (as you correctly state) in auto justification of the paragraph.

George

[pleader]

Dominic,

My suggestion of a "blind" test was not so much to demonstrate the fact that an ordinary reader would scarcely notice a difference, but to possibly show that even one so highly attuned to type as YOURSELF would have difficulty sifting the "wheat" from the "chaff" so to speak!

;^)

Regards,
Victor Lee.

[George Bilalis]

Hi Dave,

this is exactly what I mean is the way it goes. As for who's counting, usually this is the publisher <grin>.
The point being whether or not to scale type (by any percent) I agree with the purists (if I can avoid it) but then we have to make a living and scaling type is one way to do it.

Happy New Year Dave
George

P.S. Well if it goes well over 10% I usually suggest the author re-editing

[Richard Rönnbäck]

Thanks George

If I wouldn't ridicule myself I would say evtikismenos y kenorgios kronos to
you (please tell me how it's properly spelled, my Greek friends just say it
to me) to you too <g> but instead I just whish you the same

On 03-01-06 18.55, in article 1de73...@WebX.la2eafNXanI, "George Bilalis"
<g...@ipirotiki.gr> wrote:

> Hi Victor,
>
> this more generic approach of character scaling (or not), is more specifically
> examined in situations like having a book of say 880 pages that has to be
> "shrinked" to 800 pages.
> Other things (body text, character point size - leading) being equal (not
> negotiable) the only option would be to scale type by 10% down.
> This being a real situation that one has to compromise - besides convincing
> the author to cut 80 pages of text, (or the publisher to pay 10% more paper),
> I believe settles the matter the way it usually has to.
>
> Happy New Year to all
> George

[George Bilalis]

[Dave Saunders]

George,

9.2% would probably hack it, but who's counting?

Dave <grin>

[Richard Rönnbäck]

Thanks George,

Now I will surprise my Greek friends next year!

Actually, I quoted your message intentionally, so that anyone not familiar
with the orthodox new year would know what this message was about

[George Bilalis]

Hi Richard,

Happy New Year in greek goes:

"Eftichismenos o kenourgios chronos", but then we have a shortcut:
"Kali chronia", with the same meaning

regards
George

P.S. please adjust your setup not to repeat my posts. These are already there.

[George Bilalis]

Richard,

no need to wait next year. You can accomplish this by :

1. KALI SARAKOSTI for Mardi Gras
2. KALO PASCHA for happy Easter
e.tc.
We have all a full hand of wishes (as well as you) for any good occasion around the year. Actually what's intriguing is if anybody has time to work in a year full of holidays. Today for one, is a national holiday in Greece, where nobody works (I do - grin)

regards
George

[Dominic Hurley]

Gosh, 13 new posts on this thread since I last dropped in! It's a popular discussion.

Anyway, in brief: Victor, to be a truly blind test, someone else would have had to test me on this without my knowing the purpose of the test (and possibly without the person administering the text knowing the purpose), and nobody's done that. Besides which, it would be very hard to devise a test which allowed for a true comparison between glyph scaled text and text with no scaling but better hyphenation. So many variables to consider! But in a straight comparison between scaled and non-scaled text, I would certainly hope that I would notice a 4 per cent difference. (When I've been reading newspapers, I've certainly noticed fonts being compressed, so that's sort of a blind test.) I may not consciously notice a one per cent difference, but that doesn't mean the subconscious effect I was talking of doesn't come into play. (As an example of this, while working as a proofreader, I would sometimes know a word somewhere in the paragraph/sentence I was reading was wrong before I actually identified the mistake. Sort of like a proofreader's sixth sense! (Cue "X Files" music!) As all proofreaders know, the mind tries to correct errors when reading and will automatically make allowances, and that's why some proofreaders will read copy backwards. I attribute my "sixth sense" to picking up on the shape of the word being wrong, even while another part of my mind was trying to correct the mistake for me to allow me to carry on reading. I think a similar process operates for typography. (And, if anyone wants to give me tens of thousands of dollars to study this, I'll enroll you all in a trial programme.)

Hi Richard and George. Yes, I'm sure you're right - in the real world glyph scaling will affect hyphenation points, and I was dealing only with the single-line composer in my example. My simple example was just intended to illustrate that, at shorter measures (as in newspapers) and with glyph scaling of only a couple of points, the impact on hyphenation might be a lot less than some expect. It will be more pronounced at longer measures, but then they're usually easier to satisfactorily space without resorting to scaling anyway.

And, as you point out, my discussion has been based solely on English text. Maybe I would have a different view if I were setting German. (Though I don't think I'd like the look of all the capitals in German - I'm a lowercase kind of guy!) As for other programs using glyph scaling, Zapf's HZ program uses it, and with strokes intact as I recall. But I don't know if it's actually been implemented in any DTP program. Possibly TeX uses a variant and maybe there's a plug-in for XPress, but I don't know. Then, of course, there are the big proprietary typesetting sytems. It wouldn't surprise me if some of them used scaling. Anyone have any experience with Atex? Many newspapers here are/were set on Atex and I've seen lines of compressed and expanded text in them (though it could well have been done manually).

Finally, if I had to reduce a book by 10 per cent of its length, I wouldn't go for glyph scaling to do it. If I had to leave everything else exactly the same, I would choose a condensed version of the font (ie, a properly designed condensed font). If there were none and I couldn't commission one, I'd just say the reduction couldn't be done with those constraints (or not by me). I just wouldn't want to put out an 800-page book under my name with the type compressed by 10 percent. That's getting far too much like WordPerfect's "make it fit" feature for my liking! But again, that's just me.

Apologies for the ultra-long post.

Dominic

PS. Where is Olav Kvern? I'm sure he's another stickler for not scaling!

[pleader]

If I recall correctly, Ole is in favour of small amounts of glyph scaling, like the rest of the world!!!

Regards,
Victor Lee.

[michelle jean rice]

Dominic,
Atex composition coding includes ways to squeeze text, but I know of nothing that specifies glyph scaling. Our subs routinely squeeze one or two paragraphs to eliminate widows and it looks pretty awful. (Although I'm not sure the average newspaper reader would notice.) I can only imagine what they'll want to do when they switch to InDesign. If you thought my 98,100,102 glyph setting was a bad idea.....

[George Bilalis]

Michelle,

you are right about Atex. Then there comes a point in every automatic hyphenation/justification algorithm (procedure) where human intervention is required. We have all been through it, and we all know of tricks to accomplish the feat - rearranging things - to make things proper.

George

[Dominic Hurley]

Olav, where are you? I seem to recall from an earlier discussion of glyph scaling (early May of 2002) that you said you were opposed to it too, but those messages are not in the archive or on google (only the last half of the thread got preserved).

Michelle, George, when you say Atex squeezes text, are you saying it removes letterspacing as opposed to compressing the letterforms? Of course, I can't find any examples in my local paper now when I go to look for them. At least they seem to have realised that justifying a single word across a column is a bad idea! And maybe you can tell me whether a bug I've noticed in the papers here is a recognised Atex thing - when a picture is inserted and the columns wrap around it, the text line immediately below the caption in the columns to the left and right of the picture has a point or two less leading than the rest. It's weird, and it's been a consistent bug in the papers here for years now.

[michelle jean rice]

Hmm, I know of a "change set size" and a "change set width" command but I thought that applied to letter and word spacing. Now I am second-guessing myself.
I'm not sure about the text wrap. The paper you read might not be produced with Atex. Not that many are anymore.

[Scott McCullough]

As others have noted in this curiously active thread, some publishers are strongly averse to excessive hyphenation, particularly when targeting non-native readers.

For example, when I worked as an editor/designer for a university with many international campuses, even two consecutive hyphens was considered quite undesirable. The reasoning was that non-native readers have enough trouble understanding the information without breaking it into pieces. The obvious solution, of course, is to write as clearly and concisely as possible, and to set all copy ragged right. But even then it is sometimes impossible to eliminate all hyphenation without creating absurdly short lines.

Two other considerations are inappropriate hyphenation and compound hyphenation. At the university I mentioned earlier, the greatest sin was to hyphenate cam-pus (for obvious reasons!). In other situations, such as technical publishing, hyphenating compound terms is a real no-no (i.e., sub-molecu-lar).

Scott

[Guy Smiley]

"The obvious solution, of course, is to write as clearly and concisely as possible, and to set all copy ragged right."

Is anyone aware of a narrow-column publication that actually does this?

--Mike Nitabach

[Scott McCullough]

>Is anyone aware of a narrow-column publication that actually does this?

Does which--writing clearly and concisely or setting all copy ragged right? I think we're more likely to see ragged right. 8^)

Scott

[Dominic Hurley]

I'm pretty sure the daily here still uses Atex (they always seem to advertise for Atex familiar staff), but whatever they use I'm pretty sure it has a bug!

And the papers here often use ragged right columns and I've seen ragged right columns also in glossy magazines.

[Olav Kvern]

Gang--

Okay, I'll bite.<g>

I have never been, and will never be in favor of any sort of glyph scaling, anytime, ever.

I am also against allowing variation in letter spacing in body copy.

I allow word spacing to vary, and I don't object to hyphenation (though I do try to avoid "ladders" of hyphens). When word spacing reaches 150% or so, I rewrite. When word spacing *frequently* reaches 150% for a given text, I redesign.

I do this because I believe that the reader's eye is more sensitive to changes in letter spacing and character shape from line to line than it is to word spacing variation.

I also believe that variation in character shape is not something that a reader perceives consciously--it's something that affects the subconscious perception of the text. The question "can you see a difference" therefore misses the point. Note, too, that this means that the effect on the *untrained* eye is greater than on the eye of a typesetting professional--we've elevated more of our perceptions of type the the level of consciousness.

All of that said, I understand that not everyone has the kind of control I have over the texts I lay out (I can always rewrite to get around composition problems), and that sometimes one has to bend the rules a bit to get a job done. This doesn't mean, however, that it's a good thing.

Thanks,

Ole

[Dave Saunders]

Ole,

And this qualifies you to write books in the "Real World" series?

ROFLMAO!!!!

Dave

[Guy Smiley]

I think Dave was just kidding around, alluding to the fact that in the "Real World" typographers do not generally have the freedom to rewrite copy or redesign publications.

--Mike Nitabach

[Gene Trujillo]

Nothing stopping you from writing a book with your own opinions, Dave. In fact, I encourage you to do so; I think you're a good writer with a lot of knowledge of putting together documents. I'd buy it.

I feel much the same about Ole, and I can't wait for my copy of RWID2.

Gene

[Dave Saunders]

Gene,

Of course, I'm only kidding, but in this case, Ole's opinion seems more idealistic/ivory tower than "real world."

But then, I was ever the pragmatist. And, in spite of my current way of making a living, I never was known for having much of an eye.

Dave

[Dave Saunders]

Pragmatist that I am, I've done both, frequently.

Dave

[Gene Trujillo]

It is rather ironic that he decries glyph scaling and espouses editing for copy fitting purposes in the same post.

I think we've entered the realm of religion.

[Olav Kvern]

Gene, Dave--

In the "real world," for most of the "real world" publications I worked on, an *editor* rewrote copy to fit. I can't afford an editor now, so I have to do it myself.<g>

What is "ironic" about rewriting copy to fit? I don't get it. If there were something bad about rewriting copy to fit, it would make sense. But in most of the magazines/newspapers I've worked on, rewriting to fit generally improved the content of the text.

My point, however, is that using glyph scaling to fit text is not, in an ideal sense, a Good Thing. It's something one might turn to as a last resort. I don't have to, so I can stick with my ideals.

I've noticed that there's often a relationship between the professionalism of a writer and their acceptance of editing: James Joyce rewrote an excerpt from "Portrait of the Artist as a Young Man" to fit; the editor of a local Coast Guard Reserve newsletter I've worked on insists that not a word of his text be altered in any way.

Thanks,

Ole

[Olav Kvern]

I specifically wanted to comment on: "And this qualifies you to write books in the "Real World" series?"

Is it possible that I worked on other publications before I started writing "Real World" books?

When I had a job producing more than 50,000 unique typeset pages per year, I cut corners on typesetting. Do I have to be proud of or recommend those techniques today? No. I'm ashamed that I couldn't take the time/trouble to do it right.

But feel free to doubt my qualifications, anytime.

Thanks,

Ole

[Peter Kahrel]

Ole,

I think Dave was kidding, as others have suggested, though maybe he did not make a good job of it. I'm sure he knows that you are perfectly qualified to write books in the Real-world series.

I disagree with Dave's position that the situation of a writer/typesetter (Ole's, in this case) is idealistic/ivory tower. Of course, if you're the writer and the typesetter, that makes copy-fitting easy. But in my experience, publishers welcome comments from typesetters about bad pages and suggestions for rewrites.

Peter

[Dave Saunders]

Ole,

Of course you're qualified. Sure everybody knows what "ROFLMAO" means. Maybe not: "Rolling on (the) floor laughing my ass off."

I wasn't commenting on your qualifications but on your extreme position on the real world issue of pumping good looking type out the door as quickly as possible.

Some have the luxury of taking all the time in the world to achieve perfection with the presentation of their own words. Most of us have hard deadlines for presenting other people's words.

So I made a little joke.

Sorry if you thought I was attacking you. I was just having what was supposed to be a little bit of fun.

Dave

[Gene Trujillo]

Ole,

What is "ironic" about rewriting copy to fit? I don't get it. If there
were something bad about rewriting copy to fit, it would make sense.

My point was only that some people are just as adamantly against rewriting to fit as you are against glyph scaling.

Interesting anecdote regarding Joyce.

Gene

[Olav Kvern]

Gang--

No, I meant, seriously, question authority (one of which I've somehow become).<g>

As to people being against rewriting--it all depends on the text and the publication. I can't see rewriting, say, Jane Austen. But, on the other hand, "The New Yorker" rewrites fiction to fit. I've laid out plenty of books where I couldn't touch the text--in most cases, I was able to find a way to get them to look pretty good without letter spacing or glyph scaling. But not always. Even some of my own books--where I *could* rewrite--are marred by hideous text spacing.

But I don't want to start a project saying "It's OK to letterspace/glyph scale/use range kerning to get text to fit." It might be something that's forced on one by circumstances, but I don't like having it as part of the overall plan.

Thanks,

Ole

[Dominic Hurley]

I was pondering this issue today over a cup of tea and a newspaper. Some of this discussion has concerned newspapers, both because of their narrow columns and because of their highly deadline-driven nature. These are the justifications often given for glyph scaling (excuse the pun). But, what has not yet been touched on is the question "What of the days before digital type?" The days when hot metal meant that there could be no glyph scaling. Was newspaper typography really that much worse when the typesetters and comps couldn't scale the glyphs? Was that any less real world typesetting? And were the daily deadlines any less tight then? If not, why is it suddenly so necessary to employ glyph scaling today?

While I haven't studied old newspapers, I have examined books set with hot metal, and I can say that the standard of typography of many general issue paperbacks exceeds that of mass market paperbacks of today (Tschichold's and Schmoller's Penguin books anyone?). So, I just don't see why the rejection of scaling is regarded as such a non-practical or extreme approach. Just because technology now means we can do something doesn't mean we should.

As for editing to fit, as I said earlier, I am also opposed to it. But I'm obviously not talking here about a newspaper situation, where an editor wants 500 words to fill a space, or where stories are so constructed as to allow cuts to be easily made. I was talking about situations where, say, a novel is delivered to a typesetter and the typesetter wants to change a couple of words to avoid a bad break. If the author is okay with that, that's fine, but I think those sort of decisions should rest with the author, not the typesetter. Where the typesetter is the editor and the editor has the authority to make such changes (as often is the situation I find myself in), the issue gets less clear-cut. I find it a good discipline not to edit the text to meet a design problem, preferring to find a design solution. But, as Olav says, editing can often improve the text, so in that case it might be appropriate to edit the text. I would certainly never edit the text to make it fit if that left the text reading the worse for it.

Anyway, my apologies to Olav for trying to force his hand and drag him into the discussion, perhaps against his will!

[michelle jean rice]

I don't go back to hot metal days, but I've seen some pretty awful things done to type with an exacto knife in the name of filling out a newspaper column.
I've also rarely seen a newspaper story that couldn't be cut.
Now we have a much more sophisticated tool. Our challenge is to not wield it like a knife.
But there certainly doesn't seem to be any consensus about what the rules should be.
I know of one design director who won't allow vertical justification. If a story is a line or so short, paragraph returns or words are added to fill it out.
In a newspaper, I guess it comes down to using InDesign to create standard settings that make columns of type fit as efficiently and attractively as possible to begin with -- and then making sure everyone knows where management stands in the "form vs content" debate.

[pleader]

Olav,

I apologise if I misrepresented your position in any way, I had recalled (quite wrongly) that you agreed that glyph scaling was fine as a matter of unobtrusive compromise.

I am firmly in the camp of typesetting as a balance between the pragmatic and the aesthetic. No buts about that for me. So minor, unobservable glyph scaling is de rigeur as far as I am concerned.

I also think that it is a relatively rare case where the typesetter assumes the simultaneous role of an editor, especially in large multi-party jobs. It could happen more often in small one-outfit jobs where the firm handles design and editorial services.

Regards,
Victor Lee.

[Thomas Phinney]

Just to add another data point, in the font I've been designing recently, the lowercase was too wide. So I narrowed most of the glyphs by an average of 6% (optically, keeping stem weights constant).

I found this to be a very significant, quite visible difference, even though I didn't alter the stroke weights (which glyph scaling does).

Similarly, I find allowing glyph scaling of +/- 2% (possible 4% variation) can be visible.

T

[pleader]

It depends on the fonts and other factors really.

--Victor.

[Dominic Hurley]

Thomas, are you going to release your font commercially (Adobe Phinney, perhaps) or is it for personal use?

[Guy Smiley]

And what style of typeface is it?
--Mike Nitabach

[George Bilalis]

Right you are Victor saying:

It depends on the fonts and other factors really.

Glyph scaling (or any other thing) can be perceived in relative. I bet no one can spot a difference per se (without having something to compare to). So -/+2% has to be evaluated as perceptible (or not) to the limits of discrimination by the human eye.

regards
George

[Dominic Hurley]

"I bet no one can spot a difference ... without having something to compare to". I'd agree with that. If you had a font that was wholly scaled 10 per cent, you wouldn't know whether it was just an awfully designed font or whether it was a scaled version of a good quality font. But that comparison doesn't have to be on the page in front of you - it could be in your memory. I think many people would recognise a scaled version of Garamond, for example.

But you would imagine anyway that glyph scaling wouldn't be used consistently across a whole text. If it were, then it obviously wouldn't be being used to solve justification problems - a text all scaled 2 per cent is just the same as having a text not scaled at all in terms of aiding in justifying text. So I'd say there would always be some unscaled text there to compare the scaled text to.

Cheers, Dominic

[George Bilalis]

Dominic,

I still recall the first time I 've seen one working with quark 3.0 (many years ago) that he proudly displayed his ability to drag the side of a headtitle text box, holding down command key and streching interactivelly the title.
He was so fascinated by that feat, and very pleased to have the type disfigured.

This practice has been known to be applied by every newcome graphic designer of that time (especially in newspaper layout - to fit the headline). I recall my disapproval of his deforming the type (my background was Ventura Publisher of GEM environment) only to collect a comment about me being "old fashioned".

Happy New Year Dominic
George

[Dominic Hurley]

Yes, I've also seen some hideously disfigured type done in the name of "modernity"!

Happy New Year to you, too, George.

[pleader]

We've all been hit before by 3D extrusions and page curls and "clever" photo-montages, haven't we?

Off-topic....

Regards,
Victor Lee.

[George Bilalis]

Yes we have been and then we all rediscover decent typography - where the requirement is to read the text and get the context, not to hit the reader squarely between the eyes, dead.

Off topic... continued. One thing that beats me is the "alternative" kind of font design, of distorted type (erratic - broken_up, jagged outline). I cann't understand who needs to buy distorted misformed type where a simple out of order offset press will efficiently create such type.

Happy New Year Victor
George

[pleader]

Don't get me started on the aesthetics (or lack thereof) of "grunge" type, George, pleeeeeeeze!

Charis kai eirene to you, George.

--Victor Lee.

[M Blackburn]

Long post already, but I'll bite, put my two cents in, and make it a bit longer.

Let's start with a comment from Dominic: "Was… typography really that much worse when the typesetters and comps couldn't scale the glyphs? … why is it suddenly so necessary to employ glyph scaling today?" It isn't necessary; it's an available option. (And PS, for anybody to say, purist or not, what Gutenburg or anyone else would have done had they had more options is pure hubris. <g>).

Ole has defended the paragraph composer as a step out of stone age of typography and here he eschews any use of glyph scaling. Sorry folks, the paragraph composer is only tolerably intelligent if one also takes advantage of other program features, including — glyph scaling. Even then it sometimes it simply has to be turned off to get what is desired. This talk of typographic purity and readability seems to me to have a bit of "can't see the forest for the trees" syndrome going on. I remember being taught to look at calligraphy upside down if you wanted to know if it was well done. The idea being that the overall look and feel is as important to readability as anything else. If it ain't pleasing to the eye from outside of reading range reading it may not even be attempted. It should also be remembered that the glyph scaling is only going to be used when the other methods of justification are pushed to the limit. I'm looking through a recent newsletter and I can't tell where glyph scaling was applied. Changes in letterspacing are much easier to detect and affect legibility more. I personally don't like negative letterspacing at all. (For that discussion "I bet no one can spot a difference ... without having something to compare to" — it seems both sides forget that there is something to compare. If you're using glyph scaling, theoretically it's totally possible to have one line using the maximum glyph scaling and the very next line using the minimum.)

There are specific cases in abundance to make any claim for "correct" typography. To ever say that any one thing should never be done is a recipe for mediocrity. But the one thing that comes close to never-never-land is the idea that good typography can be achieved by copy editing. That's backwards. Words don't need type, type needs words; typography serves words.

[Dominic Hurley]

You seem to have missed or misunderstood at least some of the points I was making. I mentioned hot-metal typesetting exactly to highlight that glyph scaling isn't necessary, because I perceived an attitude that to reject it was somehow to not be facing up to real world typography or to be taking an overly theoretical viewpoint.

And I've already mentioned that it is possible to have maximum and minimum scaling in adjacent lines; indeed, much of the discussion centred on this cumulative effect. Likewise, the issue of nothing to compare scaled type with has also been canvassed.

As for not noticing glyph scaling, I actully picked up a printout of an Adobe PDF on InCopy when I was tidying up this morning, and the glyph scaling stood out
a mile. In particular, two adjacent lines ended with the word "flow" and the difference was very clear (I would say to just about anyone). So it's not just an academic argument.

I don't agree with you that glpyh scaling or letterspacing is necessary for the paragraph composer to give good results, but I always check line breaks manually anyway, because I don't think it's as good as the human eye and mind.

As for your "forest for the trees" comment, I don't follow that at all. You seem to be saying that what is being missed is the readability of the text, whereas I thought the whole discussion was implicitly on the readability of the text.

I don't know that anyone here has ever said that "any one thing should never be done". While I personally don't like or use glyph scaling, I'm not advocating that no one else should ever do it. To each their own. You may call my viewpoint "a recipe for mediocrity", but that's fine with me. One man's meat and all.

Cheers, Dominic

[George Bilalis]

Hi again,

Glyph scaling, in perspective.
For one this procedure is a new one. We agreed (upwards) that before ID no other layout program ever used glyph scaling for H&J (and so goes with paragraph - or multiline, iterative composers). They used letter spacing and line composers.
As far as I am concerned, it's another tool towards the end (good typestting). We can use it or not, if we think so.

And then we have to pay respect to other things as well. Good typesetting depends on many important variables besides glyph scaling and needs a lot of understanding.

Cheers
George

[M Blackburn]

Dominic,
Only a couple of clarifications and then I think I'll duck out of this one.

Never did I say that glyph scaling couldn't be bad. I said that in my newsletter I couldn't notice any scaling though I did have it set for my body text.

Readability is not affected by small imperfections of letterform as much as it is by spacing (particularly letterspacing), line length, or any number of other things. In fact I've read of tests that indicate slicing off the bottom third of the letters has little affect on legibility.

Ole's comment about glyph scaling seemed to be rather 100% against it. But I didn't refer to your viewpoint (or Ole's either) as "a recipe for mediocrity" it was rather the idea that anything should be condemned on principle. Glyph scaling is just another tool, how good it is depends largely on how it's used.

[pleader]

I'd also like to chip in with a (hopefully) final word.

Even if glyph scaling is noticed (e.g. in the case where two consecutive lines end in the same word) the deformation to the stem width per se might not be perceptible, and we would still have a "Gutenberg" phenomenon (for want of a better term) to satisfy "purists", wouldn't we? Anyway, I would imagine that the document in question probably has a much wider range than plus/minus 2%.

Regards,
Victor Lee.

[Olav Kvern]

Gang--

To clarify: I am 100% against it. The topic title includes "Good or Bad"--which implies that we're talking about the feature in idealistic, black-and-white terms.

I have no problem at all saying that glyph scaling is Bad, in an ideal sense. Yes, it can and should be condemned, "on principle."

The world is not an ideal place, and contains far more shades of gray. Glyph scaling is one of those bad things that we sometimes have to use to get the job done. I end up doing many things I am 100% against.

My point was that you should not start a job saying, "I'm going to use glyph scaling! It's great! I want those mangled character shapes!" (unless, of course, you're going for a special effect). It's something that one should consider as a last resort.

I don't think anyone here has argued that it's a desirable effect in an ideal sense. Again, to me, that means that the answer to the question in the title is:

Bad.

Finally, let's throw out the "Gutenberg Bible" argument. It's not applicable. The altered glyphs were not mathematically scaled versions of an original character, but are entirely re-cut versions of the character. Which means they're more like alternate glyphs from an OpenType font than they are like text that's been squashed or stretched using glyph scaling.

Thanks,

Ole

[Scott McCullough]

What say we let Ole have the penultimate (and authoritative) word on the issue and put this poor, mangled thread out of its misery?

[Guy Smiley]

But we're getting close to 100 posts!
--Mike Nitabach

[Thomas Phinney]

Sounds good to me!

Off topic: The typeface is too early in design (heh!) for me to comment much on it.

T

[pleader]

Olav,

I would like to clarify that in my post #83 I was trying to point out that within acceptably narrow bounds there is no effectual difference between scaled glyphs and constant-stroke glyphs, so even phenomena like two consecutive lines ending in the same word are not anathema per se.

Mike (Nitabach),

I also say let's go for a century if we are able to keep up an intelligent and cordial exchange!

Regards,
Victor Lee.

[Guy Smiley]

I am certainly learning a lot from following this expert discussion!

--Mike Nitabach

[Dominic Hurley]

I composed the following offline as a closing post, but now when I come to post it I see there are still some who are happy to continue. So, I'll post it anyway, but it may not be a last post after all!

Well, I've been greatly enjoying this discussion, and I hope no one feels they have to follow the thread and read every post, but we do seem to have reached a point where we're going over some of the same ground, so I'm happy to call it quits. Thanks to those who debated the issue with me; my workmates' eyes glaze over when I mention "glyphs", let alone "glyph scaling", so it's nice to talk over the subject with others with an interest (needless to say, my workmates are not publishing types - I'm the sole DTPer there).

In closing, I am also against glyph scaling in principle (unless you're after a special effect). But that principle was not born in a vacuum; it is the product of my experience, my reading, my aesthetic sense, and my general attitude. Like all of us here, I have to make aesthetic judgements in my work all the time; it's an undeniable fact that some of my layouts are just better than others - they're not all equal in quality. And just as I make aesthetic and value judgements on my work, so I make such judgements on the tools that I use (or choose not to use). But that's only my personal opinion for my personal use. While I have no problem identifying guidelines and absolutes for myself, I'm not one for imposing typographic rules on others (don't get me started on the oft-stated "rule" that states "Never use underlining - it's the mark of the amateur"). I original offered my thoughts here not as any kind of instruction to Michelle but as a viewpoint that she might find helpful to consider.

I do understand how others might see my position as imposing an arbitrary and unnecessary constraint upon my work. But I don't see it as a constraint, any more than I see having to work with a consistent point size and measure within a paragraph a constraint. I see it as a challenge and part of the fun of this art we call typography. (But then I speak as one who really enjoys taking the time to check the bad breaks and loose lines and figuring out how best to compose the text - "walking the lines" as Olav once memorably described it - which probably well qualifies me as an overly obsessive type.)

Cheers, Dominic

[Anders Svedin]

In aiding this thread towards the centennial post, I'll chip in with some last words as well.

I agree with Dominic on the enjoyment of "walking the lines" - it means taking responsibility for your designs and seeing to it than you can leave them feeling proud. Not that the enjoyment would be any bit lessened if there's nothing to fix, but in the end I don't think automatic settings will make line breaks as good as those decided by a typesetter for a long time yet.

As a result of this thread I've actually allowed a small amount of glyph scaling (99-100-100) in a new book project. The columns are sufficiently narrow to make it viable I think, and it does help in reducing the amount of bright yellow lines. I'm reluctant to go beyond this 1% though, and keeping a very sceptical eye on the outcome - but so far so good.

A point that hasn't been covered in this thread so far (I think) and that's seems important to me is whether or not glyph scaling also affects fixed width spaces? I would guess and hope not but I haven't worked with glyph scaling enough to be able to say for sure. Anyone knows?

Cheers,
Anders

[Dominic Hurley]

An interesting question. I did a test, and, yes, it does affect the width of fixed and non-breaking spaces.

[Anders Svedin]

That's not good. While helping out with some issues, glyph scaling evidently also introduces completely new problems!

[Dave Saunders]

But the glyph scaling we're talking about here applies only in justified text, so a couple of percentage points of variation in the "fixed" spaces (if you are even using them in such paragraphs) is probably no more an issue than the same couple of percentage points has on the glyph shapes.

Dave

[Anders Svedin]

Dave,

Not quite. I use them a lot in justified text for lining things up where tabs is not a first option.

Example 1: Numbered paragraphs. I use a fixed width space after the "#." to line the following passages up (using fixed width small cap numerals (is that the correct term?)). This could be done by tabs ofcourse, but then i'd have to use one style for "#." and one for "##." Easier for me with the spaces. But then again, the glyph scaling of the number might be enough to put the whole thing out of whack, so I'd guess it would be best to leave glyph scaling out anyway.

2. Further reading. In reference lists I use "[thin space]–[em space]" for repeated author names. This has the extra benefit over tabs of letting me manually kern the space before the first character to make the optical line-up as good as possible. But again, the glyph scaling of the "–" might be enough to ruin things here as well.

3. I want the fixed width spaces used in conjunction with abbreviations, paragraph signs, numbers over 3 digits (Swedish custom 1 000 rather than 1.000), etc. to remain absolutely constant throughout a text.

All in all, as I was saying, glyph scaling introduces new problems in my typesetting and new variables to account for. For many of my paragraph styles, I'd have to create duplicate sets, one with and one without glyph scaling. Not sure I think it's really worth it just to avoid going over a few lines manually.

Cheers,
Anders

[Dave Saunders]

Not because of glyph scaling but to avoid needing two styles for numbered lists, I use tab [#]#.tab for numbered list (where the first tab stop is a right tab). That way all my numbered lists align with each other whether or not they go past 9.

Also, I rarely, if ever set numbered lists other than left aligned (but that's because I'm usually working with number lists rather than numbered paragraphs -- I see a distinction between the two).

I don't see how a 2% variation in your examples 2 and 3 would really make any more difference than the same variation in the glyphs. If you find the one unacceptable, you'll find the other equally so, but if you're of a mind to accept the glyph scaling, the "fixed" space scaling won't be any more noticeable.

Dave

[Dominic Hurley]

I imagine Dave's right - those who don't mind glyph scaling probably aren't that worried about the fixed spaces changing. (Whereas, as I'm sure you've all guessed, I like my fixed spaces to be fixed!)

However, it does raise another interesting point, which is that, because glyph scaling is triggered when your maximum word space limit is exceeded, and seeing as glyph scaling also scales normal spaces, then I presume the glyph scaling further increases your already too wide word spaces!

Again, not likely to be of much concern to those who like glyph scaling to start with, but it does bring us closer to the 100-post mark!

Dominic

[pleader]

The behaviour described is not intuitive from the typographical point of view. I would consider it a bug to be rectified. A fix, plus a "look-across-space-characters-optical-kerning-engine" feature, would make me happy to upgrade.

Regards,
Victor Lee.

[Dave Saunders]

Guy,

Hey, knock it off!

Dave

PS: Sorry, I couldn't resist.