From the Beeb:

…the local authority e-mailed its in-house translation service for the Welsh version of: “No entry for heavy goods vehicles. Residential site only”.

The reply duly came back and officials set the wheels in motion to create the large sign in both languages.

One problem, the translator was out of office.

What do you get when you send an email to an out office responder?

Yep, the Welsh version of “sorry, not here…blah blah blah”. Unfortunately nobody could evaluate the problem until the signs were printed AND installed on the road.

Nobody could read or proof the Welsh. nobody knew what it said, but darnit, it looked about right…

So what failed here?

1. Lack of correspondence: meaning things matching things. Even in the most basic translations you have a relationship–1:1. What transpired was a transmission of pure data and a receipt without the necessary “envelope” of data around it. Usually in translation circles, we BILL people for X number of words translated. That would be a sign that something was done. Nothing is free
2. No review/edit/proof. If you are going to print a sign, it should at the very least be reviewed by a native/local language speaker for “do you understand this one?” in ANY language. if they bother to have it in Welsh they should bother to review it.
3. The Welsh-language community is abandoned.  If it was engaged in some meaningful way it would be a part of the operational framework. Some bureaucratic or functional body would exist to approve signage–this is highly inconvenient, but the fact is that somebody has to approve things–it’s like the complaint on review edit proof, but with a name attached. We ask “who cares?” if the sign is right–the people who answer should start work immediately!

Just for fun, here is a Welsh Language magazine just in case you English speakers have never seen it!

And I agree!! it’s legally libelous for you to have a Taiwanese flag in China and it’s not nice to have it elsewhere if  it’s a political situation. [will there be software updates given any global political situation?]

Flags are political colors–who cares if the main language group a political region is X if another similar political group shares the language or worse–HATES that political affiliation?  Simplified Chinese and Traditional Chinese are the most important and typical examples for a very good reason. As recently as the Olympics in China the Taiwan flag was replaced with the Myanmar [Burmese for the Americans!] because of political reasons…

If you are found to use the “wrong flag” in your software for localization [or other] purposes you can be fined, restricted from trade and otherwise “screwed” by the political apparatus of a given country for no good reason.  If you think that ZN-CH or ZN-TW or ZN-HK is really hard to visualize, then you don’t really care about localization.

There are arguments within the Simplified Chinese linguistic community as to what should be what, but adding a cultural bias to it [Taiwan] creates an unnecessary level of confusion, conflict and potential legal problems if someone doesn’t approve. Why risk it? Pretty colors that get confused when displayed in TINY pictures you can barely read?

No, let’s click one and then figure out that the language is not what I wanted, yes, that’s twice as bad. Now I can’t navigate out of the menus to change the language back…

Maybe we should focus on usability in a universal sense and localization in a “local” sense? I think that would mostly work.

First post will be about the localization life-cycle

second: what factors determine success

third and Nth: not there yet….