Fonts versus handwriting is going to be a difficult problem to solve
This bug report was converted into a question: question #274524: Fonts versus handwriting is going to be a difficult problem to solve.
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| Advanced Comic Book Format |
Invalid
|
Undecided
|
Unassigned | ||
Bug Description
I love what you've already accomplished with the Advanced Comic Book Format, and the reader, so thanks, and keep up the good work!
One problem that I think will be important to solve before ACBF can really take off is how to consistently render the text. The comic could perhaps embed TTF and OTF files, but that won't help with hand-written text. The only solution I can think of for that would be that the book's artist or one of the translators would have to rewrite the text for the different translation layers. For handwriting or for particularly complicated text positions, or letters with textural effects, it might be easier to render raster images instead of actual text.
Maybe you've already thought about this, but it's something that occurred to me when testing the ACBF Viewer and the Pepper & Carrot test comic.
| Robert Kubík (Pastierovič) (just-me) wrote | #1 |
| Changed in acbf: | |
| status: | New → Invalid |
Hi Andrew,
thanks for your comment. The text layer functionality cannot fully replace properly rendered image and it's not supposed to. ACBF format supports several semantic tags which can render text with different font, but that's as far as it goes, keeping it simple. It's supposed to be used as a translation layer for underlying original image, not as complete separation of text and image. I could make something like an extended SVG format, which could in addition support specific comic book features, but that would be very complicated to support in reader application.
I've already made quite a lot of text layers for different comics and I only came across minor issues, still improving the format and I think it works good enough even in version 1.0. If the translator wants to put more effort into making it as close to original as it gets, he/she should rather use some image editor and export the results into JPGs, making a new CBZ file.
Fonts embedding is planned for next specs version. That will add more options for translators, if they will want to use it. Currently fonts for semantic tags can be setup only on Viewer applicaton level by user.
Robo