Not quite sure if this is what you did, but you shouldn't use contextual substitutions in the salt feature. You should only use simple substitutions (sub x by x.alt ; or sub x from [x.alt1 x.alt2 x.alt3] ;, for example). No idea whether this is what's causing the problem.

Sorry Mark,
SALT was simple one to one, not contextual. I didn't actually write the code for it, but it seems fine and works properly on my computer, just not on their CS2 for some reason. I keep wondering if ID CS2 either handles type differently or if there is something different with preferences or such that is doing it to them. I think InDesign doesn't use SALT though anyway, but who knows what the glitch here is.
Stephen

I do recall reading about a change that occurred in CS3 where an old model for scripted substitutions was dumped in favor of another. Fonts made with the new version (5) of fontlab contain code that should work in both cs 2 & 3. The two places I recall bumping into something like this were in the fontlab 5 PDF guide and maybe also in a PDF available from Thomas Phinney's blog. I will see if I can dig these citations up. Hopefully they are relevant.

In the meantime why not post the code so we can see what you are really dealing with?

feature calt {
sub @upper @upper' by @calt1;
sub @lower @lower' by @calt2;
#
sub [@upper @lower] space @upper' by @calt1;
sub [@lower @upper] space @lower' by @calt2;
#

} calt;

* @calt1 is the uppercase alts @calt2 is lowercase alts

feature ss01 {
sub [A B C D E F G H I J K L M N O P Q R S T U V W X Y Z] by @calt1;
} ss01;
feature ss02 {
sub [a b c d e f g h i j k l m n o p q r s t u v w x y z] by @calt2;
} ss02;

feature salt {
sub [A B C D E F G H I J K L M N O P Q R S T U V W X Y Z] by @calt1;
sub [a b c d e f g h i j k l m n o p q r s t u v w x y z] by @calt2;
} salt;

feature c2sc {
sub [A B C D E F G H I J K L M N O P Q R S T U V W X Y Z] by [a.smcp b.smcp c.smcp d.smcp e.smcp f.smcp g.smcp h.smcp i.smcp j.smcp k.smcp l.smcp m.smcp n.smcp o.smcp p.smcp q.smcp r.smcp s.smcp t.smcp u.smcp v.smcp w.smcp x.smcp y.smcp z.smcp];
} c2sc;

Its their proprietary font and doesn't have diacritics and such.

Here is the text I was thinking of from the Fontlab User Guide:

Contextual substitutions in invalid legacy format

Due to a misunderstanding between Adobe and Microsoft, early versions of
Adobe InDesign (1.0, 1.5 and 2.0) contained a bug in the interpretation of
contextual substitutions (for features such as calt or clig) so only “invalid”
contextual substitutions work in these early versions . In Microsoft
applications, only “valid” contextual substitutions work. InDesign CS and
CS2 have a special routine so both “invalid” and “valid” contextual
substitutions work.

If you enable this option, FontLab Studio will generate a font with “invalid”
contextual features that will work correctly in InDesign 1.0, 1.5 and 2.0 as
well as InDesign CS and CS2, but these contextual features will not work in
OpenType-savvy applications from Microsoft or other vendors.

If disabled, “valid” contextual OpenType features will be generated that
will work in InDesign CS and CS2 and all other applications, but not in
InDesign 1.0, 1.5, 2.0. It is recommended to keep this setting disabled.

Does this help?

Eben, I don't think that has to do with the problem. Even if the font was coming out of FontLab with the “invalid” format, it would still work correctly in InDesign CS2, which is one of the apps mentioned by Stephen.

Not sure Eben.
She's using InDesign CS2, so sounds like it should work. I'm not exactly sure what they mean by "invalid", but it seems like the code is correct. It works in InDesign CS3 okay.
I looked for that option in preferences. If only I knew what I was doing???

Is that the order in which the features are in the font? I'd put 'calt' after all the other features, otherwise you might experience undesirable interaction between them.

feature order is:
aalt
smcp
salt
ss01
ss02
c2sc
calt

I looked for that option in preferences. If only I knew what I was doing???

The option which you marked red in your screenshot should be off, yes.

If I understand your feature order (post of 23.Jan.2008 7.40pm) and the feature code (post of 23.Jan.2008 5.56pm), then if a user switches on 'ss01'/'ss02', then all 'normal' glyphs will be replaced by alternates, since substitution in these features is non-contextual. And once that happened, your 'calt' will not show any effect, of course.
One possible solution is using contextual substitution in 'ss01' and 'ss02' too. Then, if both 'ss01'/'ss02' and 'calt' are on, there's nothing left for 'calt' since there's no match for the context any more.

You write that

The Stylistic sets (and SALT) are all set up the same as CALT feature.

but your code does not reflect this. 'ss01' is not contextual, 'calt' is. (I think that 'salt' doesn't matter here since it is not addressed by InDesign.)
Also you may check if the code

    sub @upper @upper’ by @calt1;

really does what you want. In your description you say

In Calt feature it basically substitutes an alternate every time 2 characters are typed in a row. So you'd get c o o.alt l for "cool".

but what the feature does is substitute every second uppercase or lowercase by its alternate form, so if you type "cool" what you get is c o.alt o l.alt.

Karsten

Your right Karsten.
I probably need to be a bit more accurate in my description here. Still the code seems to at least work on my computer and not on theirs. I wonder if their system software ( OSX 10.1.1 ) might be part of their problem. I'm thinking of taking out the stylistic sets feature and seeing if that might help. Its kind of an odd arrangement because I'm not the designer of this font but agreed to do fixes as a favor to a friend. I'm hoping I can get this resolved soon.

Stephen

Yes, giving them an alternate font without Stylistic Sets to see what happens may be the best thing. (I doubt that the OS is the reason since, AFAIK, Adobe apps manage/render fonts by themselves.)