Quote:
Originally Posted by Psymon
I was going to use that project to try to get it out there beyond just iBooks, and learn whatever quirks and idiosyncrasies of Kindle, etc. -- and once I've got that under my belt, and a little more experience, then maybe I'll go back and try to come up with a "second edition" of my first book.
|
Sounds like a good idea.
Quote:
Originally Posted by Psymon
Yeah, I appreciate that -- I appreciate the sentiment of "getting it right" and being done with it, and also appreciate all the help everyone has given me! Thanks to you, and to everyone else!  |
You are welcome.
Quote:
Originally Posted by Psymon
That was an interesting thread, actually! I'm not quite sure what you meant here, though, by "ligatures -> images -> shrink/expand along with the digital text"?
|
So a 'character' in a font is really just a mathematical collection of lines/shapes. The computer can then generate those lines+shapes at any size it needs (this is what allows you to change font sizes in any of your programs/devices).
Let us take a very simple character, like a 'T'. The font tells the computer, "when someone wants to draw a 'T', draw a line straight across X units, then draw a line from the middle Y units down."
This is then where font-size comes into account. The computer can multiply those X and Y coordinates by any number.
This means I can make the 'T' a "12-pt font", "16-pt font", "72-point font", "1000-pt font", and the 'T' will not lose any clarity and look just as crisp. When making it a 72-point font, the computer just reads the instructions for a 'T' shape, and says "draw a line straight across 72 times further, draw a straight line down in the middle 72 times further":
T
T T T T T T
Now, with images, there is two types: "raster" and "vector". Raster is formats like JPG, PNG, BMP. Vector is formats like SVG, AI.
Raster is your typical image of a photograph you see. It tells the computer "stick this shade of red at this exact position on the screen, stick this shade of blue at this exact position on the screen, stick yellow at this exact position on the screen, stick this shade of black at this exact position on the screen...."
Side Note: Raster does not scale well at all. If you make the image larger, the computer has to "guess" which colors go where, and the larger you go, the more pixelated the image becomes. See Wikipedia:
https://en.wikipedia.org/wiki/Raster_image
Vector formats are a lot like the fonts, they are all math based and tell the computer what shapes to draw. So you can scale to any size with zero loss. For example, in that same topic, with the help of RbnJrg, I was able to add an SVG apple right into the text:
https://www.mobileread.com/forums/sho...2&postcount=28
Side Note: A "font" is just a special way of packaging a lot of vector images together and storing them in a certain way. Everything is all math based, so you can scale this to any size and not lose any quality. See Wikipedia:
https://en.wikipedia.org/wiki/Vector_image
So what I meant by "ligatures -> images -> shrink/expand", is pulling out the "shapes" from the font into a format like SVG. And then scaling them like I did the apple. So you can take this "ff" ligature and stick it in between the text:
http://openclipart.org/people/gezegen/ligatures_ff.svg
Of course, this brings along its own maintainability problems. BUT, I would say SVGs of ligatures throughout the text are slightly less worse of a problem than non-standard mappings of characters in a font.
Best bet though is to just get a font that is built correctly.
Quote:
Originally Posted by JSWolf
Besides, if you ignore our advice a second time, we'll just not bother trying to help.
|
Meh.. I will continue to help anyone. It is not necessarily helping the original poster, but it may potentially still help someone who is lurking or who stumbles upon this topic in the future.
Perhaps there is a lurker out there who read all of our discussion on non-standard character mappings and how they are a bad idea in fonts... and now they will not use it in their books!
