jBrowse

If you want to read Japanese web pages, but your Japanese isn't perfect, you might be interested in jBrowse.

It's a plugin for Microsoft Internet Explorer that adds furigana and word definitions to Japanese web pages.

It's also a Japanese dictionary and kanji tool.

It's also free.

What does jBrowse do?

jBrowse's feature set is as simple and elegant as the tea ceremony, yet as fascinating and powerful as Godzilla.

1. Inject Definitions into a page in your browser
2. Add Furigana (explanatory kana) into a document that has hard kanji
3. Lookup Words -- it's also a decent two-way dictionary
4. Find Kanji by radical, stroke count, pronunciation, meaning, jouyou level, JIS number...

Look at the screenshots for more details.

Did I hear you say it was free?

Yes, and it is in fact free. You can register if you want to support jBrowse, but there is nothing forcing you to do that.

What do I need?

You need Microsoft Internet Explorer 5.5 or above, with a Japanese font available. You also need to be running Windows NT/2000/XP (any language version). If you are running one of the 'home' Microsoft operating systems, you will need to have the Japanese version.

Where do I get it?

Click the 'download' link on the left and get jBrowse. It's a convenient self-installing package that can also be uninstalled with a single click.

How does it all work?

Glad you asked. There are three parts to jBrowse:

• The dictionary, which is a front-end to Jim Breen's famed EDICT and KANJIC data files.
• The browser plugin, which can reach into a page in your browser and change things around.
• The Japanese parser, which can recognise which characters in a page belong to which actual Japanese word.

If you would like more information on EDICT and related projects, please go to the links page.

The most difficult part of jBrowse is the browser plugin part. It would be far easier to write a separate standalone program or a web service rather than adjust the document while it's in the browser. But that wouldn't give the fast response and smooth ergonomics that we want.

The most interesting part of jBrowse is the Japanese parser. It is not at all trivial to figure out where words begin and end in Japanese text, and jBrowse is not yet perfect, although it gets it right almost all the time and it's usually pretty obvious when it gets it wrong. If people used kanji all the time the task wouldn't be so hard, but when a few kana words, particles and the like come along in a row, it gets just a tiny bit complicated.

If you are interested in this kind of thing, please contact us because there is much thinking still to be done in this area.