[Pharo-dev] default monospaced code font

btc at openinworld.com btc at openinworld.com
Wed Oct 16 13:56:09 EDT 2013


Goubier Thierry wrote:
>
>
> Le 16/10/2013 11:50, Sven Van Caekenberghe a écrit :
>>
>> On 16 Oct 2013, at 10:20, Goubier Thierry <thierry.goubier at cea.fr> 
>> wrote:
>>
>>> Interesting display, Sven.
>>>
>>> My take on that:
>>>
>>> * Aesthetics: the system has two fonts, not one. -1 if I review a 
>>> document with more than one font.
>>
>> In all documents, you have at least two fonts: body and headings, 
>> often quotes, examples, listings, etc have an another font to make 
>> them stand out. In the new approach, the idea is that monospaced 
>> fonts indicate code (in browsers, debuggers, workspaces). It is a 
>> useful principle.
>
> You're right. But nobody would dare write headings in a monospaced 
> font :) unless for an art project.
>
>>> * Coherence / uniformity: A class name, a method selector has a 
>>> different shape in the GUI (proportional) than in the code 
>>> (monospaced). Are they different objects? Can I recognize my class 
>>> name in the code without reading it?
>>
>> Syntax highlighting should take care of that I guess.
>
> I don't think so. This is no by making the selector green that it will 
> look more like the proportional version in the pane above.
>
> Kind of disrupting the uniformity of the underlying model, when I'm 
> pushing for things like smart suggestions where the GUI understands 
> the objects written in the code.
>
>> I think that if the monospaced font is a point size smaller that the 
>> main sans font (e.g. 12 and 11) the excessive width problem or visual 
>> shock is much more manageable. In any case, I am giving it a try.
>
> Probably. But then individual characters may become harder to read and 
> distinguish... sort of compromising character readability to make 
> space for the added whitespace inherent to the monospaced font.
>
> I'd be more impressed if the argument was helping me distinguish 
> between | and l.

Yes. It is designed to do that. Some common failings of monospaced fonts 
are noted [1] and dealt with. There comment section is also interesting.

The attached PDFs are the result of getting the urge to compare a broad 
coverage of code examples (taken from "Terse Guide to Squeak") against 
three fonts:
* DejaVu Sans 9 point
* Source Code Pro [1] [2] 9 point
* Source Sans Pro [3] [4] 9 point & 10 point, since the width of 10 was 
the same as the others at 9.
Also attached is the source excel file.

cheers -ben

[1] http://blogs.adobe.com/typblography/2012/09/source-code-pro.html
[2] http://sourceforge.net/projects/sourcecodepro.adobe/files/
[3] http://blogs.adobe.com/typblography/2012/08/source-sans-pro.html
[4] http://sourceforge.net/projects/sourcesans.adobe/postdownload?source=dlp
>
> I'l let you try, then :)
>
> Thierry

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Fonts-comparison-DejaVuSans9-SourceCode9-SourceSans9.pdf
Type: application/pdf
Size: 57547 bytes
Desc: not available
URL: <http://lists.pharo.org/pipermail/pharo-dev_lists.pharo.org/attachments/20131017/c5f01459/attachment.pdf>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Fonts-comparison-DejaVuSans9-SourceCode9-SourceSans10.pdf
Type: application/pdf
Size: 60414 bytes
Desc: not available
URL: <http://lists.pharo.org/pipermail/pharo-dev_lists.pharo.org/attachments/20131017/c5f01459/attachment-0001.pdf>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Fonts-comparison.xlsx
Type: application/octet-stream
Size: 51024 bytes
Desc: not available
URL: <http://lists.pharo.org/pipermail/pharo-dev_lists.pharo.org/attachments/20131017/c5f01459/attachment.xlsx>


More information about the Pharo-dev mailing list