RE: lines to vectors, vectors to GIF, [was RE: Back to svg-developers

Hi Domenico,

Yes, I do remember those caves, though I thought it was Lascaux. The sushi was excellent! 

 

I loved seeing your horses gallop again. I reposted the link on Ello and mentioned your work there. It was discussion there that prompted my initial questions here.

 

I think you are probably right (sigh) about the need for human intervention. Coming up with good vectors is just one of the problems with Escheresque drawings (see for example http://srufaculty.sru.edu/david.dailey/stillart.htm ). The next is to recognize the paths for filling and then recognizing how discontiguous paths might also be connected as they pass under and over one another.  Constructing paths from segments is addressed to some extent in http://cs.sru.edu/~ddailey/TGW2014/RectilinearMold2.html . Some of the formative ideas for this (particularly the part about making glyphs in the browser without relying on rectangular grids) was prompted by your workshop on font making at SVG Open 2010 in Paris. Perhaps you’ll remember that one of the folks from Institute Paris Telecom (whose speciality in mathematics was Bezier curves) and I were curious about why font design packages use rectangles for the placement of control points.

 

Cheers

David

 

From: [email protected] [mailto:[email protected]] 
Sent: Tuesday, November 11, 2014 5:50 AM
To: [email protected]
Subject: Re: lines to vectors, vectors to GIF, [was RE: [svg-developers] Back to svg-developers

 

  

Hi David,

 

> I have a bunch of line drawings done decades ago with India ink.

 

Remember when you and I and others were in the caves drawing SVG on walls with Africa charcoal?

 

OK, back to business. At this stage I don’t think you can get much better results with Illustrator or Inkscape, or other software; you do have to tweak manually. 

 

We’re talking discernment algorithms. The choices you make as an artist imply a huge mass of calculations (the results for some of which being already in the collection of “experience”), providing you with the necessary level of “anticipation”, the sine qua non that allows you to free the form that is captured in your mind and in the medium. “Mind” is the keyword that says all the difference between what you can do and what the software can. Think also about the words “touch”, “retouch”, “final touch”, and all those other concepts that the software cannot possibly grasp yet. 


When I made the http://svg.dotuscomus.com/demos/gallop.svg I remember starting off by doing that work on eight shots of a running horse, but very quickly I decided to do blueprint drawing instead; it would take only a fraction of the time needed to fix the software generated paths. But when you have to suggest convolution volumes internal to the shape you’re probably better off working with the generated paths. I generally don’t mind doing that kind of work when it comes about and, you know what? Even if the software yielded satisfying results right off, I would want to improve the work anyway!

Domenico





[Non-text portions of this message have been removed]



------------------------------------
Posted by: "David Dailey" <[email protected]>
------------------------------------

-----
To unsubscribe send a message to: [email protected]
-or-
visit http://groups.yahoo.com/group/svg-developers and click "edit my membership"
----