What is this?

Web previewer for XKCD Time (explanation) translations by Russian What If? and XKCD translators team. The main goal is to provide a combined view of a given original image and original text with a corresponding translated image and translated text. Generated web pages also include links to corresponding frames on geekwagon and links to translators’ discussion resources.

Project also contains translated images.

That’s just for translation process support and not for comfortable reading.

Installation and page generation

First install llgal, version 0.13.15. The version is significant because .llgal/llgal perl script is a patched version of the original script from the llgal-0.13.15 package.

Clone this repo, then run ./regen.sh.

If everything’s alright, the script will generate lots of *.html files in the repo directory and lots of png thumbnails in .llgal subdirectory.

Tell your web-server about this directory, and you’re done.

Add new section

1. Fill metadata for new section in .llgal/metadata-sXpY.txt directory, where X and Y is section and part numbers.
2. Add metainfo_parse metadata-sXpY.txt statement into file .llgal/llgal.
3. Download source images to orig/Scene X Part Y directory from URLs like http://geekwagon.net/projects/xkcd1190/data/frames/${N}.png
4. Put translated images to Scene X path Y directory.
5. Regenerate captions with ./regen.sh gc command.

TODO: Exclude orig from regenerated captions.

Bugs around images

In my Firefox browser (version 29, Linux x86_64) on my notebook indexed png images viewed uncorrectly. Fix that works for me: change gfx.color_management.mode option in about:config from 2 to 0. See https://bugzilla.mozilla.org/show_bug.cgi?id=621474 .

License

Perl code distributed under the same license as well as llgal: GNU General Public License V2 or later.

Images and related metainfo (metainfo-*.txt files) distributed under CC BY-NC 2.5 as well as sources from xkcd.com.

CSS styles, HTML templates and other stuff in this repository is in public domain.