Comma spacing in math mode

Comma spacing in math mode

[S Barmeier]

[-- Attachment #1.1: Type: text/plain, Size: 1818 bytes --]

I don't know if anyone feels the same, but I'm not happy with the
spacing of commas in math mode (pictures attached). For instance,
typesetting H^1(X,F), the comma seems to be closer to the F than to the
X. Equal spacing on both sides or setting the comma closer to the X are
I guess a matter of taste (inserting a \neghairspace after the X
achieves the latter). In any case, I would kind of prefer equal spacing.

Also, it seems odd that adding a negative space \neghairspace does not
change the length of the total formula - I had expected the modified one
to be slightly shorter...?

On a related issue, the spacing for lists, e.g. (0,0,...,0), is not
satisfactory. LaTeX provides a command \dotsc for dots between commas
(and other commands like \dotsm, \dotsi, \dotsb). I don't know if all of
these are necessary, but \dotsc and its cousins are not defined in
ConTeXt and both using \ldots or just ... give unsatisfactory results.
(It is also curious that the spacing after the first comma and the
spacing after the last comma in (0,0,...,0) is evidently different...)
Personally, I think I favour something approximating (0, 0, ..., 0) and
I find \ldots is spaced too widely, ... too closely, and in any case,
the spacing of the dots doesn't match with the spacing of the commas.

Any second opinions?

Best wishes,
Severin

P.S.: I'm also confused about \colon and : - the latter looks better to
me in $i: A \to B$. The spacing for projective coordinates [1:0: ... :0]
is wide but still acceptable. The index of a subgroup |G:H| is far too
widely spaced, but \colon only gives asymmetrical spacing.
The semicolon seems to undergo the same spacing as the comma, with the
same problems mentioned above.

P.P.S.: \hairspace does not seem to have any effect in math mode, while
\neghairspace does. Bug or feature?


[-- Attachment #1.2: Type: text/html, Size: 2182 bytes --]

[-- Attachment #2: commasmathmode.jpg --]
[-- Type: image/jpeg, Size: 24386 bytes --]

[-- Attachment #3: Type: text/plain, Size: 486 bytes --]

___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Comma spacing in math mode

[Aditya Mahajan]

On Sun, 27 Feb 2011, S Barmeier wrote:

> I don't know if anyone feels the same, but I'm not happy with the
> spacing of commas in math mode (pictures attached). For instance,
> typesetting H^1(X,F), the comma seems to be closer to the F than to the
> X. Equal spacing on both sides or setting the comma closer to the X are
> I guess a matter of taste (inserting a \neghairspace after the X
> achieves the latter). In any case, I would kind of prefer equal spacing.

\setupmathematics[autopunctuation=no]

@Hans: I would suggest that this should be made the default.

> Also, it seems odd that adding a negative space \neghairspace does not
> change the length of the total formula - I had expected the modified one
> to be slightly shorter...?

I don't know about this.

> On a related issue, the spacing for lists, e.g. (0,0,...,0), is not
> satisfactory. LaTeX provides a command \dotsc for dots between commas
> (and other commands like \dotsm, \dotsi, \dotsb). I don't know if all of
> these are necessary, but \dotsc and its cousins are not defined in
> ConTeXt and both using \ldots or just ... give unsatisfactory results.
> (It is also curious that the spacing after the first comma and the
> spacing after the last comma in (0,0,...,0) is evidently different...)
> Personally, I think I favour something approximating (0, 0, ..., 0) and
> I find \ldots is spaced too widely, ... too closely, and in any case,
> the spacing of the dots doesn't match with the spacing of the commas.
>
> Any second opinions?

Adding the \dots(m|i|b|) etc from amsmath is a good idea. I will look into 
that.

> P.S.: I'm also confused about \colon and : - the latter looks better to
> me in $i: A \to B$. The spacing for projective coordinates [1:0: ... :0]
> is wide but still acceptable. The index of a subgroup |G:H| is far too
> widely spaced, but \colon only gives asymmetrical spacing.
> The semicolon seems to undergo the same spacing as the comma, with the
> same problems mentioned above.

I think that the spacing of $i \colon A \to B$ is correct. (Compare from 
plain TeX output). I don't know what do you expect from |G:H|. : is 
defined as a relation and \colon is defined as a punctuation. Does |G{:}H| 
look better? (which forces : to be a math ord).

> P.P.S.: \hairspace does not seem to have any effect in math mode, while
> \neghairspace does. Bug or feature?

Don't know.

Aditya
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Comma spacing in math mode

[Stefan Müller]

Hi!

On 27.02.2011 20:58, Aditya Mahajan wrote:
> On Sun, 27 Feb 2011, S Barmeier wrote:
>
>> I don't know if anyone feels the same, but I'm not happy with the
>> spacing of commas in math mode (pictures attached). For instance,
>> typesetting H^1(X,F), the comma seems to be closer to the F than to the
>> X. Equal spacing on both sides or setting the comma closer to the X are
>> I guess a matter of taste (inserting a \neghairspace after the X
>> achieves the latter). In any case, I would kind of prefer equal spacing.
>
> \setupmathematics[autopunctuation=no]
>
> @Hans: I would suggest that this should be made the default.

IMO this does only partially help, consider the following example:

\starttext
$(B, Y, R, X)$
\setupmathematics[autopunctuation=no]
$(B, Y, R, X)$
\stoptext

In both lines the spacing after "Y" is way to big. I don't know if 
that's a bug or just ugly. (version 2011.02.15 16:11)

>> Also, it seems odd that adding a negative space \neghairspace does not
>> change the length of the total formula - I had expected the modified one
>> to be slightly shorter...?
>
> I don't know about this.
>
>> On a related issue, the spacing for lists, e.g. (0,0,...,0), is not
>> satisfactory. LaTeX provides a command \dotsc for dots between commas
>> (and other commands like \dotsm, \dotsi, \dotsb). I don't know if all of
>> these are necessary, but \dotsc and its cousins are not defined in
>> ConTeXt and both using \ldots or just ... give unsatisfactory results.
>> (It is also curious that the spacing after the first comma and the
>> spacing after the last comma in (0,0,...,0) is evidently different...)
>> Personally, I think I favour something approximating (0, 0, ..., 0) and
>> I find \ldots is spaced too widely, ... too closely, and in any case,
>> the spacing of the dots doesn't match with the spacing of the commas.
>>
>> Any second opinions?
>
> Adding the \dots(m|i|b|) etc from amsmath is a good idea. I will look
> into that.

+1. I normally use "$x, \dots\, ,x$" because then the spacing between 
the periods looks the same as between the commas and the periods, which 
I prefer. I personally appreciate the wider spacing in \dots.

Stefan
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Comma spacing in math mode

[Mojca Miklavec]

On Sun, Feb 27, 2011 at 23:49, Stefan Müller <warrence.stm@gmx.de> wrote:
> Hi!
>
> On 27.02.2011 20:58, Aditya Mahajan wrote:
>>
>> On Sun, 27 Feb 2011, S Barmeier wrote:
>>
>>> I don't know if anyone feels the same, but I'm not happy with the
>>> spacing of commas in math mode (pictures attached). For instance,
>>> typesetting H^1(X,F), the comma seems to be closer to the F than to the
>>> X. Equal spacing on both sides or setting the comma closer to the X are
>>> I guess a matter of taste (inserting a \neghairspace after the X
>>> achieves the latter). In any case, I would kind of prefer equal spacing.
>>
>> \setupmathematics[autopunctuation=no]
>>
>> @Hans: I would suggest that this should be made the default.
>
> IMO this does only partially help, consider the following example:
>
> \starttext
> $(B, Y, R, X)$
> \setupmathematics[autopunctuation=no]
> $(B, Y, R, X)$
> \stoptext
>
> In both lines the spacing after "Y" is way to big. I don't know if that's a
> bug or just ugly. (version 2011.02.15 16:11)

In text parts this is known as kerning. In OpenType math there are
complex bounding boxes.

Mojca
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Comma spacing in math mode

[Mojca Miklavec]

On Mon, Feb 28, 2011 at 00:10, Mojca Miklavec wrote:
> On Sun, Feb 27, 2011 at 23:49, Stefan Müller wrote:
>>
>> \starttext
>> $(B, Y, R, X)$
>> \setupmathematics[autopunctuation=no]
>> $(B, Y, R, X)$
>> \stoptext
>>
>> In both lines the spacing after "Y" is way to big. I don't know if that's a
>> bug or just ugly. (version 2011.02.15 16:11)
>
> In text parts this is known as kerning. In OpenType math there are
> complex bounding boxes.

(I accidentally sent the email too early.)

I wanted to add that unless/until somebody implements complex bounding
boxes (OpenType Math font for whatever font you are using), you cannot
avoid the problem. TeX has no idea how big Y is, it only knows its
rectangular bounding box. Cases like this one have to be manually
tuned.

Mojca
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Comma spacing in math mode

[Aditya Mahajan]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 1312 bytes --]

On Mon, 28 Feb 2011, Mojca Miklavec wrote:

> On Sun, Feb 27, 2011 at 23:49, Stefan Müller <warrence.stm@gmx.de> wrote:
>> Hi!
>>
>> On 27.02.2011 20:58, Aditya Mahajan wrote:
>>>
>>> On Sun, 27 Feb 2011, S Barmeier wrote:
>>>
>>>> I don't know if anyone feels the same, but I'm not happy with the
>>>> spacing of commas in math mode (pictures attached). For instance,
>>>> typesetting H^1(X,F), the comma seems to be closer to the F than to the
>>>> X. Equal spacing on both sides or setting the comma closer to the X are
>>>> I guess a matter of taste (inserting a \neghairspace after the X
>>>> achieves the latter). In any case, I would kind of prefer equal spacing.
>>>
>>> \setupmathematics[autopunctuation=no]
>>>
>>> @Hans: I would suggest that this should be made the default.
>>
>> IMO this does only partially help, consider the following example:
>>
>> \starttext
>> $(B, Y, R, X)$
>> \setupmathematics[autopunctuation=no]
>> $(B, Y, R, X)$
>> \stoptext
>>
>> In both lines the spacing after "Y" is way to big. I don't know if that's a
>> bug or just ugly. (version 2011.02.15 16:11)
>
> In text parts this is known as kerning. In OpenType math there are
> complex bounding boxes.

I would call it a bug. The spacing is OK with cambria, but not with xits.

Aditya

[-- Attachment #2: Type: text/plain, Size: 486 bytes --]

___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Comma spacing in math mode

[Khaled Hosny]

On Sun, Feb 27, 2011 at 06:14:36PM -0500, Aditya Mahajan wrote:
> On Mon, 28 Feb 2011, Mojca Miklavec wrote:
> >>IMO this does only partially help, consider the following example:
> >>
> >>\starttext
> >>$(B, Y, R, X)$
> >>\setupmathematics[autopunctuation=no]
> >>$(B, Y, R, X)$
> >>\stoptext
> >>
> >>In both lines the spacing after "Y" is way to big. I don't know if that's a
> >>bug or just ugly. (version 2011.02.15 16:11)
> >
> >In text parts this is known as kerning. In OpenType math there are
> >complex bounding boxes.
> 
> I would call it a bug. The spacing is OK with cambria, but not with xits.

Because Cambria implements math kerning that Mojca is referring too,
XITS don't (and will not in the near future unless someone steps to
implement it; pretty tedious job).

Regards,
 Khaled

-- 
 Khaled Hosny
 Egyptian
 Arab
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Comma spacing in math mode

[Stefan Müller]

On 28.02.2011 00:20, Khaled Hosny wrote:
> On Sun, Feb 27, 2011 at 06:14:36PM -0500, Aditya Mahajan wrote:
>> On Mon, 28 Feb 2011, Mojca Miklavec wrote:
>>>> IMO this does only partially help, consider the following example:
>>>>
>>>> \starttext
>>>> $(B, Y, R, X)$
>>>> \setupmathematics[autopunctuation=no]
>>>> $(B, Y, R, X)$
>>>> \stoptext
>>>>
>>>> In both lines the spacing after "Y" is way to big. I don't know if that's a
>>>> bug or just ugly. (version 2011.02.15 16:11)
>>>
>>> In text parts this is known as kerning. In OpenType math there are
>>> complex bounding boxes.
>>
>> I would call it a bug. The spacing is OK with cambria, but not with xits.
>
> Because Cambria implements math kerning that Mojca is referring too,
> XITS don't (and will not in the near future unless someone steps to
> implement it; pretty tedious job).
>
> Regards,
>   Khaled

Ah, okay. Too bad. Some time ago I was wondering if there's something 
like "kerning" in math mode, too. Does that mean it would work if I 
switch to cambria math fonts?

Stefan
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Comma spacing in math mode

[Mojca Miklavec]

> Some time ago I was wondering if there's something like
> "kerning" in math mode, too. Does that mean it would work if I switch to
> cambria math fonts?

It would at least work better for cases such as different spacing
between letter F, Y or X and comma (where F and Y leave too much empty
space).

It would not effect \hairspace (this might be a bug, but I didn't
test) and it would not effect space *after* comma (unless maybe for
Y). Autospacing for commas is a different issue, but Cambria could
solve "bad kerning".

Mojca
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Comma spacing in math mode

[Andreas Harder]

Am 28.02.2011 um 00:14 schrieb Aditya Mahajan:

> On Mon, 28 Feb 2011, Mojca Miklavec wrote:
> 
>> On Sun, Feb 27, 2011 at 23:49, Stefan Müller <warrence.stm@gmx.de> wrote:
>>> Hi!
>>> 
>>> On 27.02.2011 20:58, Aditya Mahajan wrote:
>>>> 
>>>> On Sun, 27 Feb 2011, S Barmeier wrote:
>>>> 
>>>>> I don't know if anyone feels the same, but I'm not happy with the
>>>>> spacing of commas in math mode (pictures attached). For instance,
>>>>> typesetting H^1(X,F), the comma seems to be closer to the F than to the
>>>>> X. Equal spacing on both sides or setting the comma closer to the X are
>>>>> I guess a matter of taste (inserting a \neghairspace after the X
>>>>> achieves the latter). In any case, I would kind of prefer equal spacing.
>>>> 
>>>> \setupmathematics[autopunctuation=no]
>>>> 
>>>> @Hans: I would suggest that this should be made the default.
>>> 
>>> IMO this does only partially help, consider the following example:
>>> 
>>> \starttext
>>> $(B, Y, R, X)$
>>> \setupmathematics[autopunctuation=no]
>>> $(B, Y, R, X)$
>>> \stoptext
>>> 
>>> In both lines the spacing after "Y" is way to big. I don't know if that's a
>>> bug or just ugly. (version 2011.02.15 16:11)
>> 
>> In text parts this is known as kerning. In OpenType math there are
>> complex bounding boxes.
> 
> I would call it a bug. The spacing is OK with cambria, but not with xits.

As it comes to math, there is still this ugly rendering of $n\choose k$ … with all OpenType math fonts. 

Andreas
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________

Re: Comma spacing in math mode

[Mojca Miklavec]

On Mon, Feb 28, 2011 at 00:14, Aditya Mahajan <adityam@umich.edu> wrote:
> On Mon, 28 Feb 2011, Mojca Miklavec wrote:
>
>> On Sun, Feb 27, 2011 at 23:49, Stefan Müller <warrence.stm@gmx.de> wrote:
>>>
>>> Hi!
>>>
>>> On 27.02.2011 20:58, Aditya Mahajan wrote:
>>>>
>>>> On Sun, 27 Feb 2011, S Barmeier wrote:
>>>>
>>>>> I don't know if anyone feels the same, but I'm not happy with the
>>>>> spacing of commas in math mode (pictures attached). For instance,
>>>>> typesetting H^1(X,F), the comma seems to be closer to the F than to the
>>>>> X. Equal spacing on both sides or setting the comma closer to the X are
>>>>> I guess a matter of taste (inserting a \neghairspace after the X
>>>>> achieves the latter). In any case, I would kind of prefer equal
>>>>> spacing.
>>>>
>>>> \setupmathematics[autopunctuation=no]
>>>>
>>>> @Hans: I would suggest that this should be made the default.
>>>
>>> IMO this does only partially help, consider the following example:
>>>
>>> \starttext
>>> $(B, Y, R, X)$
>>> \setupmathematics[autopunctuation=no]
>>> $(B, Y, R, X)$
>>> \stoptext
>>>
>>> In both lines the spacing after "Y" is way to big. I don't know if that's
>>> a
>>> bug or just ugly. (version 2011.02.15 16:11)
>>
>> In text parts this is known as kerning. In OpenType math there are
>> complex bounding boxes.
>
> I would call it a bug. The spacing is OK with cambria, but not with xits.

But that is probably a "bug" in font, not in ConTeXt; and something
that cannot really be handled in LM or in MKII at all.

I already sent a similar report to Khaled (when I was testing XITS in
Word), but he was somehow reluctant to change metrics (it has to be
done carefully for the whole font, it needs a lot of time & work;
might become obsolete with next release of stix; and just fixing two
out of thousand such "bugs" doesn't really help).

Mojca (who didn't try out the examples)
___________________________________________________________________________________
If your question is of interest to others as well, please add an entry to the Wiki!

maillist : ntg-context@ntg.nl / http://www.ntg.nl/mailman/listinfo/ntg-context
webpage  : http://www.pragma-ade.nl / http://tex.aanhet.net
archive  : http://foundry.supelec.fr/projects/contextrev/
wiki     : http://contextgarden.net
___________________________________________________________________________________