Kerning Tables in InDesign?

rosaiani's picture

Hi everyone,
Quark has this feature where it lets you adjust kerning parameters for an entire font. I wonder if InDesign has something similar to it or the only way is to change the tables in the font itself.

Thanks!

as8's picture

The limitation on Cool Kerning is, that the demo version only
will apply a limited amount of kerning values.
http://www.knowbody.dk/productbasement/index.asp?language=2&PBSelProd=I000001&PBSelMenu=Download

Please remember to watch our video under the Manual section
http://www.knowbody.dk/productbasement/index.asp?language=2&PBSelProd=I000001&PBSelMenu=Manual

That is what wrote S

pattyfab's picture

I'm reviving this thread because it's three years and two upgrades later... but I can't seem to find a way to change the kerning tables permanently for a font in InDesign. It's quite easy in Quark but none of the help topics or the InDesign book I own seem to explain how to do this in InD. I'm using Perpetua and for some reason the italics are kerned way too tight around quote marks. I'd like to adjust this globally for the font rather than on a case by case basis. Does InDesign allow you to do this? If so, how?

blank's picture

I looked into this recently and it appears that there is still no way to apply global kerning tweaks. And I was really hoping it would turn up in CS3—now that Indesign has table and cell styles, it would be nice to globally kern all of the numbers in those cells without editing my fonts.

Miguel Sousa's picture

I wouldn't hold my breath for such feature. I understand that this was an important add-on back in the day when fonts were constrained to a limited number of kerning pairs, but nowadays?!...
If the font is lacking an obvious kern pair, or is kerned too tight, you can almost certainly call it a bug. And the job of layout applications is to provide the fonts "as is"; layout applications are not the right tools to fix fonts.

pattyfab's picture

Miguel - I don't agree that we should be constrained by the font "as is" or be forced to kern each pair manually. Even some beautiful fonts have kerning problems. When I have "metric" kerning turned on in InD, I can see the kerning pairs but can't change them globally. I would think that an application created to help designers and typesetters do their job as best they can wouldn't just dismiss a tool that was so useful in Quark.

Nobody is asking the application to fix the fonts, but typesetting is an exacting discipline (if it's to be done well) and why not provide a valuable tool to help the designers do that? It would only improve the software. I'm actually considering starting this job over again in Quark in order to finesse the kerning.

blank's picture

I understand that this was an important add-on back in the day when fonts were constrained to a limited number of kerning pairs, but nowadays?!…

Just because a font can have a lot of kerning pairs does not mean that they will be applied well, nor does it mean that the pairs will be appropriate for all, or even most, situations.

If the font is lacking an obvious kern pair, or is kerned too tight, you can almost certainly call it a bug.

Call it a bug to whom? The font designers? The myriad of fonts with crap kerning pairs for numbers shows just how many font designers don’t really pay too much attention to the users on this issue.

And the job of layout applications is to provide the fonts “as is”; layout applications are not the right tools to fix fonts.

Now you are just being absurd. Typesetting needs change dramatically for a number of reasons and layout software should offer users the flexibility to cope with those changes. Kerning is not fixing a font, it is a method of adapting spacing to a situation. Being able to do this globally would be a boon to Indesign users—especially Indesign users who have purchased font licenses that do not allow modifications.

charles ellertson's picture

if the font is lacking an obvious kern pair, or is kerned too tight, you can almost certainly call it a bug.

Then (pardon my vent), every font Adobe has issued is buggy. Try setting "quotedblleft" plus "quoteleft" -- and of course, "quoteright" "quotedblright".

Or footnote calls (superiors) following a quote(dbl)right -- or following a period -- or comma -- or an f or g or or r, t, v, w (sometimes x) or y -- etc. How about fractions, particularly the one.numr sort -- following any number? The reason for having arbitrary fractions is so if/when you have to set 11/32, you can. That stops any strategy of using an unnaturally large left sidebearing on the numerator to get the needed space.

Etc. In any event, a bunch of kern pairs are omitted; others are too loose or too tight bye my eye.

I sort of agree though, that kerning should be in the font. The user should be able to fix the kerning, IN THE FONT. To it's credit, Adobe's EULA permits this. But man, is it hard taking apart the kerning of an OT font & re-doing it. It can be so frustrating that I often go back to our Type 1 fonts & make OT versions of them. Today was spent on Adobe Caslon (Type 1 to OT). Of course I have Adobe's Adobe Caslon Pro -- it came with some software -- but for good kerning & character naming, it is easier to start over, even if I miss kerning Tcommaaccent with everything else. I, anyway, set a lot fewer Tcommaaccents than I do footnote calls, fractions, and quotes within quotes.

Charles Ellertson

Miguel Sousa's picture

> and why not provide a valuable tool to help the designers do that?

There's one plug-in available to do that, isn't there? I doubt that such functionality will ever be implemented natively in InDesign, unless thousands of users ask for it...

> nor does it mean that the pairs will be appropriate for all, or even most, situations.

That sounds like it would be better to have fonts with no kern at all; this way the user would kern as needed. As a font developer, I'd certainly wouldn't mind that... at all!

> shows just how many font designers don’t really pay too much attention to the users on this issue.

I have kerned fonts, and I certainly tried to cover all the pairs that I could think of, and which made sense, so I know that it's impossible to predict all the situations where and how the fonts will be used.

> Kerning is not fixing a font, it is a method of adapting spacing to a situation.

Right, and a good font should have kerning for 97% of those situations. The other 3% can be done manually, I think.

> Try setting “quotedblleft” plus “quoteleft” — and of course, “quoteright” “quotedblright”.

What kind of kerning do those need?

> others are too loose or too tight bye my eye.

I hear you; we all have different perceptions of how much close/far glyphs should be from each other, but for the most part what's in the font ought to work for most people, no?

pattyfab's picture

I don't get it at all - why NOT provide designers with a valuable tool to help us control the appearance of type.

That sounds like it would be better to have fonts with no kern at all; this way the user would kern as needed. As a font developer, I’d certainly wouldn’t mind that… at all!

No offense but this is patently absurd. Then why not go for monospace? I would think font designers would want to create the best possible appearance for their font. But kerning is subjective, and just like there is good design and... not so good design, the same holds true for kerning tables. I'm not asking InDesign to make these decisions for me (altho it seems to want to) but to provide me with the tool I NEED to make these decisions for myself.

or the most part what’s in the font ought to work for most people, no?

and comic sans works for a lot of people too. Does that make it good? Are you trying to appeal to designers or the lowest common denominator?

It's a great application, but this is a glaring omission. Your attitude, again no offense, seems pretty shortsighted. You really need to wait for "thousands of users" before you add a feature? If you're gonna ignore the needs of designers who care about kerning then I'm gonna stick with Quark.

(edited for tone)

Miguel Sousa's picture

Let's come down, alright? You have your views, I have mine, and there's absolutely no reason for getting down to that tone.

I may work for Adobe but I'm not in charge of InDesign's development. If you think that's such an important feature, use the best channel to communicate it to the right people:
http://www.adobe.com/cfusion/mmform/index.cfm?name=wishform

Have a good day.

timd's picture

Since the ‘bug’ existed before InDesign it certainly seems an omission (and more important to me than many other features that are available). Even the best crafted typefaces cannot meet the requirements of every project and users shouldn’t expect them to, it is shortsighted to write off a feature that would enable designers to better set type and compile artwork, which is the prime purpose of InDesign surely.

Tim

k.l.'s picture

A few thoughts on this:

Spacing/kerning should be as perfect as possible in a font. However, there are limits to this requirement:
First, spacing/kerning always expresses the 'philosophy' of the type designer, while the typographer's spacing/kerning 'philosophy' may be quite different.
Second, it is impossible to care for all possible glyph combinations. The larger fonts get, the harder this is to achieve. (In old 256 glyph fonts, it was a possible maximum of ca 65000 pairs to kern, or check at least; in recent 1000-2000 glyph fonts, this is 1-4 mio. pairs to kern, or check.)
With every font I kern I wonder what to kern. I mean: beyond the usual stuff. Letters with numerals? 'T4' or 'A4' can both be pretty ugly combinations. Quotation marks with all kinds of math symbols or miscellaneous glyphs? One may refer to them in a text and surround them by quotes ...
E.g. in some fonts I noticed kerning pairs for .’ and ,’ (in English), but not for ’. and ’, (in German). Or fonts that only kern uppercase with following punctuation marks, but not the other way round -- which may occur in email addresses. Not very useful, from a typographer's point of view.

Since there is the Cool Kerning tool (at a reasonable price), need to implement this functionality into InDesign is not that pressing.

However, I think it could be a useful addition to InDesign.
Alongside the current kerning box (for kerning a particular glyph combination at a particular place) and the tracking box, I can imagine another kerning box for kerning of a particular glyph combination whereever it appears throughout a the document. Yet it's not that easy, because: does the value apply to this particular font only (i.e. style)? or to this font family? or to all fonts in a document? Moreover: only in this document or in others too?
I just watched the Cool Kerning demo. Its approach to create font groups illustrates that the issue is more complex than just additional box -- it requires some more options.
[Edit:
Now I think these user-defined kerning pairs are additions/corrections to the font, not to a single document alone. Since CS is a suite, the kerning pairs saved in one app could be available in the others too.
So, when entering a kerning pair in this additional box, this would be added to a special preference file read by all CS apps.]

As Mr Ellertson says, modifying metrics in OTFs is really tricky and may even break other feature functionality in fonts (FLS does not import all kinds of features when opening an OTF/TTF).
So an option to do this in the layout application maybe is not the worst scenario.

(I speak of 'fonts' here since I refer to the technical side.)

Karsten

pattyfab's picture

I did post the suggestion to the Adobe forum but thanks for supplying this form - it seems more likely that it will get read this way.

Also sorry if my tone was harsh but Miguel it felt to me like you were just dismissive of my suggestion as not gonna happen without really considering what value it is to designers to be able to control kerning. The choice between "no kern at all" or just using the fonts "as is" (both of which you suggested) is not much of a choice, is it?

I am relatively new to InD and was kind of stunned not to find this feature. As a result, when I have a choice, I think I'm more likely to continue using Quark. Most publishers in NY still use Quark and this may be part of the reason.

Miguel Sousa's picture

Don't get me wrong, I understand the value of such feature, specially for book designers. But consider this: When InDesign came along, Quark already had this feature for years; nonetheless, InDesign is already in version 5, and there's no sight of such feature (I have a vague idea that version 1.0 had it, but I may be wrong). The people in charge of InDesign are bright, committed, and surely listen to the users' needs. We all know that InDesign and Quark are running for the same piece of the cake, so, don't you think that if that feature was *that* important, it would had already been added?
Again, I see the value of being able to edit the kerning, but that feature may not be added in the near future, unless a lot of users demand for it, I think.

charles ellertson's picture

don’t you think that if that feature was *that* important, it would had already been added?

Is this a variant on "We get the politicians we ask for"?

John Nolan's picture

Adobe included KernEdit with PageMaker for awhile, so this aversion to allowing users to influence kern pairs is a new doctrine.

Linda Cunningham's picture

We all know that InDesign and Quark are running for the same piece of the cake, so, don’t you think that if that feature was *that* important, it would had already been added?

Heck no. Adobe doesn't think it's important, that doesn't mean it isn't to the rest of us.

charles ellertson's picture

There will always be some difference of opinion between those who design a product and those who use it. My complaint with Miguel isn't so much over InDesign as it is over the attitude that if a product doesn't offer something, then it generally isn't needed.

Many of you are type designers rather than "applications" designers. Would it shock you to know that as a user, I find OpenType sadly lacking? It lacks a feature I think very important to the user, namely, the ability to (fairly easily) modify it for a user's individual needs. Whatever arguments you want to advance as to why this is incorrect are also available to the "applications" designers about their one-size-fits-all products.

pattyfab's picture

It was that tone that got my dander up too. Obviously this is a useful feature or it wouldn't be in Quark.

I would think Adobe would be eager to make their product even better rather than to dismiss their customers suggestions. But they already have my money.

will powers's picture

Let me add one more voice to the chorus of those suggesting it is a good thing to allow typographers to adjust kern pairs within a page layout application as they are needed in particular settings. This feature alone makes Quark a more valuable tool for me.

Has there ever been offered an explanation about just why InDesign does not offer this feature? I don't know the first damn thing about building applications, but how tough could it have been to add kern pair adjustment?

powers

charles ellertson's picture

Will, it can be a problem. The way Quark worked was to read in those kern pairs already in the font, then let the user add to them -- or remove them, if that was your preference.

The way we worked, on a PC with TeX, (where TeX made use of the AFM to make up a TFM for metrics), was simply to edit the AFM itself. Nothing was perfect, if you used Fontographer for your kerning, and if you had characters (as they were called in those days) above character number 256, Fontographer would mess up the character names in the kerning -- so you had to learn work-arounds.

But OpenType is different, with its class-based kerning, which I imagine is tables within tables. And just when kerning is applied could also get in the way -- if you are substituting glyphs with other features, kerning obviously has to wait until all such substitutions have been made. Since I'm not a programmer, this seems like a tougher nut to crack than the simple Quark approach to Type 1 fonts. I would also worry that Adobe might farm the programming of this one out, as it appears the exception dictionary was farmed out, with similar results.

The more I work with OpenType, the more I'm against class-based kerning, on the grounds that creates more problems than it solves.

Miguel Sousa's picture

What I find interesting about this thread is that it sounds like all of sudden people realized that they're using a tool (assuming that they're using InDesign) that does not have a kern table editing feature, when in fact it's been like that forever. I mean, I'm all in favor of giving more power to the users (see Who wants/needs a Glyph Palette in Photoshop?, for example), but it's the first time I hear so many voices complaining about the lack of such feature.

If nothing else, I truly wish that this thread helps you all get what you want. I'm a big fan of InDesign, and if it turns out that this feature has been on your wishlist for a long long time and hasn't been added yet, then speak up; our combined forces will hopefully make their way to the right people. (But again, what blows my mind is that from 2004, when this thread was started, until now, I don't recall anyone raising this issue. Therefore, thanks to Patricia for resurrecting this discussion.)

> I would think Adobe would be eager to make their product even better rather than to dismiss their customers suggestions.

AFAIKT Adobe *is* eager to make their product even better, and all input towards that is welcome. But, this thread didn't start as a customer suggestion, neither your revisitation of the topic sounded like one. But I agree that it is turning into a customer suggestion, for which I'm thankful that you brought it up.

My replies were simply my interpretation of why a kerning tables feature isn't available in InDesign, and certainly do not express the reasons behind what went into the decision of not adding it. (FWIW, InDesign has been around longer than I've been working for Adobe)

Having said that, I think it's time for all of us to steer this thread into the right direction (or start a new one), and contribute constructively for the discussion.

Thanks!

pattyfab's picture

Miguel - I've only been using InDesign for a few months or I'd no doubt have raised the flag earlier. Late to the game (as I mentioned, book publishing is still pretty Quark-centric). I did follow your suggestion and make a feature request, hope others did the same and we'll have a groundswell and either get a plugin or it will be implemented in CS 4.

I did download that Cool Kerning plug in but haven't a clue how to install it.

timd's picture

Like Patty I have only recently started using InDesign where the client has requested it. I use the kerning pairs feature regularly in Quark.

Tim

Miss Tiffany's picture

Patty, where did you find it? Is it really $99.00 US?

Bert Vanderveen's picture

I have always been under the impression that the Optical Spacing feature of Id was meant to take care of kerning issues. I even kind of remember that it was based on an algorithm developed by (or inspired by) David Kinderleys work in that field. I may have understood all of that, but I did a lot of work with Id over the years, and hardly ever found fault with kerning Apart from some of the usual situations that would turn up (mostly specific to the Dutch language and my personal preferences) — these I solved manually.

In my Quark days (well, actually 13 years) I have used the Kerning tool only for global kerning (larger sizes, less space, etcetera). I DID use Fontographer to mess with really awkward fonts, mostly MonoType ones. : )

But if building Kerning into Id will ease the work of type designers, I'm all for it...

William Berkson's picture

I believe that Adobe's Optical spacing is based on a program developed by URW inconjunction with Hermann Zapf, and bought by Adobe; I don't think Kindersley's work, which was different, had anything to do with it.

Optical Spacing can take care of some problems, but it won't do the kind of thing Charles wants, with kerning of quotations marks and footnotes etc. in exactly the way he, the compositor, specifies.

pattyfab's picture

Tiff - I followed the link above, in this thread. I have no idea if I downloaded the Demo or the real version but didn't pay for it, so it must be the demo. I also have no idea how to install it, or what they're talking about when they say "All plugins require the KLiC plugin in order to work correctly." So I haven't done anything with it. It's sitting on my desktop.

Optical spacing is definitely an improvement over Metric, at least for Perpetua Italic. But I am using this font in InDesign for a book and would really like to have been able to tweak it myself. If I were working in Quark I'd spend about an hour improving it before I start laying out the book. This client is the only one I have who won't take Quark. I have at least 3 or 4 who use Quark exclusively, and several who take both. Quark Passport is a valuable tool for publishers who print foreign co-editions. I don't know if InD has multi-language support.

hrant's picture

> if building Kerning into Id will ease the work of type designers

Huh, ease? You want less work?!

Actually I've personally become quite happy that InDesign doesn't support the addition of custom kerning, because I just got a contract to respace a font that's being used in an InDesign document. :->

hhp

John Nolan's picture

Patty:
You might do well just buying Typetool 3 for your kerning, remembering, of course, to rename your re-kerned fonts, and to check your EULAs!

Miss Tiffany's picture

Well, d'oh. I didn't see that before.

Patty, find your InDesign application folder. Then within that find the Plugin folder. You can just drop the two files in there.

pattyfab's picture

John,

I'm not gonna bother with that thanks. That creates all sorts of other problems, such as then how does the client license the re-kerned font? Fugeddaboutit. I'll just have to weigh the kerning options b4 deciding which app to use.

I'm not compulsive about kerning, but often enough I find a font that needs some work and am used to being able to do that in the design app.

hrant's picture

> how does the client license the re-kerned font?

The EULA will tell you. Usually if the EULA allows modification the client just needs to own a copy of the original. Sometimes the person modifying it needs to own a copy too, but not always, for example if the work is technically (certainly not necessarily physically) in-house, or if you remove the font from your own computer once the work is complete. Many licenses are transferable, in which case as long the font is only present at one "entity" at a time you're legally OK with just one copy.

hhp

John Nolan's picture

Patty: Out of the box, InD CS3 supports all of the languages that Quark Passport does, and then some (e.g: Bulgarian, Lithuanian, Latvian, Ukrainian, Romanian, Slovenian, and full support for Hungarian.) I can't tell you anything about how well the 2 apps do on these languages (how well the hyphenation works, etc.), but InD CS3 wins on the specs!

William Berkson's picture

>all of sudden people realized that they’re using a tool ...that does not have a kern table editing feature.

Miguel, I think it's interesting that both Charles and Patty are book designers, and both are recently switching to InDesign--one from TeX, one from Quark. To me that suggests a couple of things: first because of books' longevity that some people are willing to take more care in typographical detail. Second that maybe only recently are you getting book people on board with InDesign.

If these two very experienced designers are indicative, it would mean that the kerning feature is in fact important for people in book production, and having the feature might help win them over to InDesign.

charles ellertson's picture

Optical Spacing can take care of some problems, but it won’t do the kind of thing Charles wants, with kerning of quotations marks and footnotes etc. in exactly the way he, the compositor, specifies.

Actually, I believe the optical kerning routine won't do this at all. I set quoteright-quotedblright in InDesign, and switching on optical kerning. Nothing happened -- there were still three "raised commas" all together. No change in the note call spacing, either. Maybe I did something wrong. &BTW, the space between a quotedbl(left or right) and a quote(left or right) is about half a word space, optically.

I don't mind having to kern some things by hand, in the file. In fact, our rule-of-thumb is all display type has to be hand-kerned, because what's right depends on the setting size, how many words there are in the line, etc., etc. Text is another matter -- we had one comp that who had to spend 4 hours extra on a job hand-kerning all the note calls that followed a close quote, and finding all those bloody places where the open quotes were kerned back into a word space (this kerning was in the font), where something like an "f" preceded the word space. A number of times, the "f" was closer to the open quotes than the open quotes were to the letter that followed, even though technically a word space intervened. Customers would mark such proof "P.E.--add word space."

Anyway, four hours work & a grumpy comp.

Oh -- Bill, I don't really call myself a designer. I can design when needed, but I'm only a compositor -- what AIGA calls a "tradesman."

John Nolan's picture

Patty:
As Hrant says, the licensing needn't be any more complicated than usual: the client would need to have a license for the font anyway, right?

But, don't get me wrong, I strongly support having kerning addressed within InDesign.

pattyfab's picture

Easier just to use Quark, innit?

John Nolan's picture

I agree it's a pain. The only plus is that the re-kerned font would be available to you in other apps.

blank's picture

Easier just to use Quark, innit?

Only if you actually know Quark and own a copy.

k.l.'s picture

Charles Ellertson -- The more I work with OpenType, the more I’m against class-based kerning, on the grounds that creates more problems than it solves.

Somehow I like the idea that an application, or suite of applications, allows to do "font modifications" externally to a font. Whether by modifying an AFM file ("semi-external") or adding to a kerning preferences file. Doing these things externally, it does not matter at all whether a font's kerning is class based or not. Simply enumerating pairs is simple, but highly unproductive as regards font-internal kerning. External modifications can (should?) work without classes, irrespective of how the font is built internally.
The only issue I can imagine is when kerning gets contextualized.  ;-)

Other pluses of adjusting kerning externally to a font are (legal) avoiding EULA issues, and (usability, typographer's p.o.v.) avoiding different versions of a typeface on the HD.

hrant -- Huh, ease? You want less work?! Actually I’ve personally become quite happy that InDesign doesn’t support the addition of custom kerning, because I just got a contract to respace a font that’s being used in an InDesign document. :->

So here speaks the type designer, not the typographer?  :)

charles ellertson's picture

Karsten,

In theory, I've no complaints with an approach where kerning modification is external. In practice, it depends on how it works. I never used Quark, so what I don't know is if, once read in for a job, the kerning was always with a file, or you had to remember to read it in each time you opened a file. For example, books go through a cycle of proofs. You finish the first proof & ship it off to the customer. Six weeks or so later, proof is returned & you make corrections. If the first proof used the external kerning program but you forget to read it in for second proof, all your line ending will change -- or In InDesign, lines where the "wax" is disturbed. The problems due to the bugs with InDesign's exception dictionary that can be made a part of the file come to mind.

Whatever, it would be an improvement on the current situation.

As to class based kerning, yes it is an aid in dealing with the increased character set of OpenType, but in practice, it all to often seems to be a token effort. Much kerning involves vowels, which take diacritics. Frequently the class for a vowel simply includes the vowels with all the diacritics, and unfortunate kerning results. And since the original "kerner" tends to deal with shapes rather than meaning, it is wasteful in a way; "Gamma" will never (rarely?) occur with "e" in a word, the characters are from different alphabets.

Having said all that, I confess I do use a little class-based kerning, but always feel a little guilty, as the values entered are usually a compromise I think I can get away with.

k.l.'s picture

Hello Mr Ellertson, indeed class kerning's merits are at the same time its downsides. It is on the type designer to make good use of the possibilities which the technology offers, e.g. to care for diacritic. (This is best addressed in two ways: by defining exceptions to class kerning where necessary, and by designing diacritic marks such that they do not require too much space and so avoid collision at least in some cases.)
Personally, I would not dare touch an OpenType font's kerning information because I do not know the logic by which classes are made.

(As regards document reflow: I have spent some thoughts about how an InDesign implementation might look like. I can think of a solution which does not not change the kerning if the according user-defined kerning preference file is not present. And I think the way these kerning pairs are defined can be done more comfortable than in XPress' and Cool Kerning. The latter, however, does a very good job, because it only uses existing interface elements, and also avoids document reflow in case that the plugin is absent.)

hrant's picture

> So here speaks the type designer, not the typographer? :)

Remember, I just make fonts, I don't use them. :-)

hhp

k.l.'s picture

Aua-aua-aua!  :))

charles ellertson's picture

Karsten,

For goodness sake (unless it's too terribly forward of me), call me Charles. All this started because long ago I was on another forum, where your *name* field was limited to 15 characters -- my name doesn't fit. So I was "Charles," until, inevitably, there was another "Charles." I then became "charles_e". The downside is when I feel strongly enough about something to make a statement & don't want to do so anonymously, I have to put my full name at the bottom of the post.

When grandpa came over from what is now called Sogn og Fjordane -- he wasn't the oldest son & the farm was too small to split again -- we sort of lost what was left of any old world manners. If I'm being too forward, forgive me.

Syndicate content Syndicate content