[Project-ideas] GSoC idea for improving the aesthetic quality of Indian Scripts

[Mayank Jha]

Thanks for your generous feedbacks.

Improved Implementation details:

I will be implementing this application for the Devanagri script. To
implement the testing application what I propose to do is, first of all set
standards or rule-sets ie what to search for in any Devanagri script. There
exists a current application called fontQA but that runs only inside
FontLabs, which is not freely available, therefore it is very difficult to
try, modify and test it.

Therefore I plan to make a script to start with based on the Font Forge
framework. The idea of the script, would be extract the data from the font
in the following fashion:


 We would be analysing the font's shaping engine algorithm, as to how it
actually parses the input from the user into glyphs by using the GPOS &
GSUB methods, to achieve this we can use a standard text similar to “A
quick brown fox jumps over the lazy dog.” which uses the characteristic
properties unique to the type of font.

Standard properties of a font can't be altered by the developer, while the
dynamic properties are those which can be modified at developer's will.


 I think the best way to test fonts based on these properties in the script
would be from *classes*, one for standard properties and the other for
dynamic properties, with appropriate methods built inside to check for the
properties and their values.


 Retrieving static properties would be easy, since most of them would be
numerical values like font weight, point size etc.

To check for the dynamic properties, one needs to look at when and how the
actual shaping of the glyph takes place after the glyph buffer for the text
has been created.


 When a rendering engine handles a chunk of text (usually a syllable), the
processing is divided into a set of predefined “features”, which are
“applied” one after the other in a standardized way. What differentiates
one font from the other in the manner in which these “features” are
applied. We could implement this by making methods inside the class to
check for these. How one must go about making these methods depends largely
on the rendering engine used (I would be using Pango for my case) and the
parameters which exhibit each instance of applying these features.


 These “features” for Devanagri script are well described for the
Uniscribe, W3C specifications!


 The points kept in mind would be the OpenType specifications for the
script, and will have scope for addition. These are again specified, at the
unicode standards.


 Is the proposal draft, ok ? I have omitted mentioning the Unicode
standards and the OpenType specifications as they are pretty long. Should I
copy paste it into my proposal ? Would that make it more good ?

Cheers!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ankur.org.in/pipermail/project-ideas-ankur.org.in/attachments/20130424/bf30f7f7/attachment-0003.htm>