A while ago I was moaning about a sad state of javascript i18n tools. When I finally started working on it, I faced a dilemma.

Should I wait until it’s finished and then reveal it like Moses to chosen people? Look guys what I etched with a toothpick during my lunch time break! You know, I was waiting for my Crème Brûlée and had to kill those few pesky minutes.

Or should I choose a more honest approach, showing steps and missteps between embarrassing first version and hopefully usable last?

Well, there’s nothing better than public humiliation to instill some humility so let’s see what is behind the door number two.

The goal is still the same. Develop a short and relatively fast library that allows you to translate javascript code and HTML content and to do so with as little fuss as possible and with a clear separation between HTML, javascript and translations, but without breaking pages in case translation doesn’t exist yet.

And here’s my humble start together with a simple demo . Three functions, two of them a support needed later on. stripStr and stripStrML just strip whitespace at the start and end of string with stripStrML doing it on every line of a given string. We’ll need this later on so we can find strings to translate even if they are enclosed in whitespace to make source display more pretty.

Then there’s a function called _, which takes a string and translates it if javascript object i18nDict exists and has this string as an attribute with a non empty value. Otherwise it just returns the same string, so it doesn’t break page even if translation doesn’t exist in translation object . It’s what is usually called a proof of concept prototype.

That’s it for today. I’ll try to post updates every couple of days or so, if progress warrants it.

Update: Follow up here and here .