Comments on: The Software Localization Paradox

By: Always define the language and the direction of your HTML documents, part 02: Backwards English | Aharoni in Unicode, ya mama

Sat, 18 May 2013 12:14:32 +0000

[…] bug is also an example of the Software Localization Paradox: It manifests itself when Accept-Language is not English, but most developers install English […]

By: The Software Localization Paradox | Dario Solera

The Software Localization Paradox | Dario Solera — Thu, 31 May 2012 12:56:40 +0000

[...] (funny how I don’t remember how – was it Twitter? was it someone else’s blog?) an interesting blog postabout software localization, which is incidentally my company’s business. This is the most [...]

By: Wikimedia blog » Blog Archive » Interview with Wikimedia’s Amir Aharoni

Sat, 22 Oct 2011 09:46:27 +0000

[...] One other and somewhat more personal thing that i hope to achieve through my work in this team is spreading the word about the Software Localization Paradox. [...]

By: Dario Solera

Dario Solera — Thu, 01 Sep 2011 09:47:46 +0000

You made a nice point actually. My company works in the software localization industry, I am Italian and know English quite well. Funnily, I hate software in Italian and I use everything in English, starting from the operating system. Even my phone is configured with an English UI.

So I guess it’s not only users that in the end prefer English, it’s even people involved in the localization process…

By: Konst. Stampoulis (@geraki)

Konst. Stampoulis (@geraki) — Mon, 29 Aug 2011 09:24:56 +0000

I’ve published a post agreeing and commenting your article and I’ve added that many translators are not users of the software they are translating so errors come also from that way. Translating a string that you don’t know what is does and on what occasion appears at the interface, can have many funny results. So we need translation from people who know English and are also regular users of the software.

By: Geraki's blog » Μεταφράσεις λογισμικού

Geraki's blog » Μεταφράσεις λογισμικού — Fri, 26 Aug 2011 08:50:19 +0000

[...] ένα άρθρο του Amir Aharoni (βικιπαιδιστής από το Ισραήλ) σχετικά με [...]

By: cjl

cjl — Thu, 25 Aug 2011 12:39:43 +0000

India is a particular case with regards to English given their former colonial status. My speculation is that people prefer what they were exposed to when they learned something, in the case of software, all too often, it arrived with only English as an option or was learned in an educational system that promoted English language skills as a path to greater employment opportunity.

My answer to your question of what can be done to improve this situation is to create greater opportunity for learning (and computing) in native languages within the early educational system. At Sugar Labs we are working on localizing the Sugar Learning Platform into about 130 languages / dialects (at the moment). Sugar can be experienced as packages from many Linux distros, Live CD, Live USB (SOAS, or Sugar on a Stick), virtual machine images and, of course, on one of several million OLPC XO laptops currently in the hands of children.

http://translate.sugarlabs.org/
(Shameless plea for more localizers)

The image on an XO laptop is basically a Fedora spin that is dual boot in Sugar and Gnome UIs, so we need the upstream localized too.

The Gnome packages we pull are listed here:
http://l10n.gnome.org/releases/olpc/

Links to additional upstream packages (Fedora, Translate Project) can be found in the comments of these “tracking ticket” strings:

http://translate.sugarlabs.org/projects/upstream_l10n/

Catch the kids early and they will demand (or perform) localization of their tools later in life. I don’t disagree with any of your other suggestions, but I think that even more than localization testing, we just need more localization first.

By: Mel Chua

Mel Chua — Thu, 25 Aug 2011 02:54:01 +0000

People tend to do what’s easiest for them to do. I wonder how people would react to a screen at installation/upgrade (when they need to select the primary language they’ll be using the program in) that would let them opt-in “to improve the translation of this software into different languages.”

If you clicked “yes,” it would ask you to list the other languages you knew. Then the program would install and run as normal, in English (or whatever other primary language you picked). Once in a while, though, it would pop up and ask “would you like to help improve this program’s translation by using this application in Spanish/Mandarin/Tamil (or whatever other languages you listed) for the next hour?” If the user clicked “yes,” they’d get a thank-you message saying “remember, if you find mistranslations, you can report/fix them at this link,” and if they did report something, they’d get at least a thank-you email.

Basically, take the behavior you want to see and make it as easy as possible for people to do, then thank them profusely for doing it. I’m sure this would be far simpler to implement in some programs than others, but in the case of mature web applications or desktop environments and applications with stable translation/language-selection features, it could probably be a plugin that would toggle the startup language if installed.

By: HenkvD

HenkvD — Wed, 24 Aug 2011 18:41:02 +0000

Maybe a campaign with a sitenotice for a week would help. Ask people to try a different localization language and report and/or fix problems.

As for Indians and their preference of English. This is a known fact. Most education is done in English, as are the books. It will not necessarily apply to other native languages. It would apply for ‘dialects’

By: aharoni

aharoni — Wed, 24 Aug 2011 14:23:08 +0000

You should probably use it in Italian. (And if you know Milanese and Venetian well, then in them!) That’s because most likely there are many more people who can improve the English version.

Are you speaking about DownThemAll, by any chance?