[Indicfonts] Samyak successfully harmonizes different Indic scripts

[rahul bhalerao]

Primoz PETERLIN <primoz.peterlin at biofiz.mf.uni-lj.si> wrote:
  >I believe Samyak has so far come closest to the ideal of being a true 
  >pan-Indic font, i.e. one that successfully harmonizes different Indic 
  >scripts, making them of proportional height and weight.
  
  
  First of all, THANKS for that appriciation. We actually started with  the same intention to harmonize these scripts. And its great to hear  that we have done it
  successfully.
  
  >I understand that while Samyak Sans is a compendium of other free OpenType 
  >Indic fonts, Samyak ("Serif", i.e. with modulated stroke) is your own 
  >creation. Am I right?
  
  Yes. Sans are based on other free fonts(did't want to waste time in something
  available already). But Samyak i.e. serif is completely our own creation.
  
  > I suppose another requirement is that the Shirorekha aligns in  Devanagari, >Bengali and Gurmukhi, and there are probabyl many more  such rules.
  
  Certainly we acknowledge that there is still lot to go ahead. Requirements are 
  huge and we shall make sure to come closer to satisfy them.
  
  >I noticed that the Samyak SFD files use quadratic splines rather than
  >cubic ones.  Does that mean that you indeed design glyphs that way, i.e.
  >with quadratic splines? I prefer to work with cubic splines - I find it
  >much easier to have a better control over a fewer control points.
  
  Even I find it easier to work with cubic splines and less number of control points.
  And we indeed started working with cubic splines. Most of our work was done
  with cubics only. But we faced some problems with them. While generating .ttf
  from them fontforge intrpolates and convert them to quadratic splines and while 
  doing this the control points are increased automatically. Again cubic  splines are still not supported on many machines and give problems  while installation.
  Also while drawing with cubic splines curve points occassionally get converted to
  corner pionts(when you try to stretch the curves). The shape may appear smooth
  in one view but sometimes after scaling, these curves don't appear as smooth as
  intended. As a developer we stressed on using as less control points as
  possible. But the font generation algorithm of fontforge sometimes add extra
  points. We couldn't find a good solution for this. If you know the best method,
  please let us know. 
  
  
  >Thanks also for the technical details about merging fonts with  anchor classes in  >FontForge(http://www.gnowledge.org/Gnoware/localization/merge.htm).  I've had 
  >problems with these, since I was not aware that the anchor class *had* to 
  >be defined in advance in the target font. I would like to mention though 
  >that point 6 (Copy-Paste) can be replaced with the function Element -> 
  >Merge Fonts.
  
  You may be right on using  Element -> Merge Fonts function. But still this point
  has some relation with point 7. Just to make sure nothing is lost we preferred this
  way. Anyway you just have to select the entire block of glyphs from one file and
  paste into the target file. All the typographic rules associated with them also get
  copied. So it works!
  
  >Maybe I should start with introducing myself. Some years ago, I  have started >the Free UCS outline font project, . The project aim  is to provide a set of free >OpenType fonts, which should eventually  span throughout the most used parts >of Unicode.
  
  >I any case, as a member of the Free UCS outline fonts project, I am
  >looking forward to borrowing from your work, and of course hope you will 
  >do the same if needed.
  
  It's really great  to get such response from someone like you. 
  Thanks once more!
  
  Regards,
  Rahul Bhalerao,
  [Indic Fonts Team,
   HBCSE,TIFR.]
  
  


		
---------------------------------
 Yahoo! DSL Something to write home about. Just $16.99/mo. or less
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://gnowledge.org/pipermail/indicfonts/attachments/20051225/7fcabbf0/attachment.htm