+1

The issue is dealing with double byte languages such as Japanese Korean and Chinese. Most applications are written for single byte languages.

To test if an application a piece of crap here is the test for Japanese

Copy and paste in a textbox the "string of death" 代表取締役 (it means Managing Director) as many times as possible. Then save the text. When the text is displayed, if you see a question mark and the last kanji is incorrect. You have an application that doesn't support Japanese.

Man did they ever get things wrong. I've seen misleading titles before but that was one of the first. The title and the start of the article makes it sound like they came up with a method of speeding up embedded code but reading through the article it appears that what they've come up with is just a means of speeding up the translation of text messages to new languages. That's a nice thing to have, but not at all what the title suggested.

"Working with the eking University in China" [sic]

EKE: To increase; to add to; to augment now commonly used with out, the notion conveyed being to add to, or piece out by a laborious, inferior, or scanty addition; as, to eke out a scanty supply of one kind with some other. "To eke my pain."--Spenser. [1913 Webster]

Sigh, this is what happens when you let lazy drongos "write" stories about things that they don't understand. Another gitwizard article which is a total waste of space, as it is completely incoherent and contains no useful factual content.

It wouldn't have been easier to just use gettext?