It's the start of a new year, a time when executives, project and portfolio managers, consultants, and analysts are dusting off crystal balls to try to predict what 2013 will bring, how much it will cost, and how long it will take to support.
Marty McFly is due to come back in three years, and still I have no flying car.
It turns out that predicting the future is hard.
We can, however, look at trends and predict improvements -- and also look at the intersection of movements to see what they mean.
Which brings me to the latest Internet trends survey from Kleiner Perkins Caufield Byers (KPCB). The research shows continued growth in web adoption -- but it won't come from our neighbors. If yours are anything like mine, the teenager already has a smartphone, and so do the kids she babysits.
Where the growth will come from in 2013
The new spike in Internet traffic will come from global, mobile sources, not the United States. This makes sense; the emerging world is just getting access to the Internet. Meanwhile, mobile devices do not require the costly cable lines and infrastructure of the more traditional services.
This commonsense is also backed up by data. According to that KPCB research, global mobile traffic was 4 percent of Internet traffic in December 2010, and reached 10 percent in May 2012.
That's about a half a percentage point a month.
Mature web properties that want to improve profits can only do it one way -- by catching the wave of that growth. That means that our money-making processes, ads, and checkout need to run in a mobile environment and need international support.
I think it's time to test your website for internationalization.
Two simple internationalization (i18n) tests for any website
Here are two quick tests for internationalization: Paste 'Øredev Gärtner Malmö' (include the quotes) into any web form, click save, and look at the redisplay.
This is a real problem. My friend, Markus Gärtner, has the very description of his name incorrect on his own book page on Amazon.com.
What's Wrong With This Picture?
Look very carefully at the author of the book.
Not only did Amazon's systems have a translation error, the problem seems to be in half the system, forcing both a correct and an incorrect name to appear. This is typically how it works, by the way. It's correct 90 percent of the time, wrong in an annoying 10 percent of the site that uses an obscure code library.
Want another? Try cutting, pasting, and saving 你好，我的朋友. The letters are Chinese; if they look like a mess, your browser doesn't support east Asian characters. You are probably running Windows.
Even if it works, what's next
Even if your website supports global input, the question is how much global text to display. Some companies simply have a mirror site with a different translation (this is common in Europe), while many web applications support the idea of a "locale," of storing only symbols, then looking up the correct word to display in a symbol table for that language.
For today, it's probably enough to find out if you can support global input and that you have the capability of localization. If not, find out what it will take to implement a symbol table.
Remember: Mobile customers are coming -- and they're coming from anywhere on the globe.
"Sure, we can do that boss. It will only cost so much and take so much time" is a very different conversation than, "Oh, no, I didn't realize that. We'll find out. Better get a pot of coffee going, call the family -- I'll be working all weekend."
I've had both conversations.
The first one is better.
— Matt Heusser is principal consultant of Excelon Development.