However positive a step, though, Google acknowledged on its security blog on Tuesday, it's not without its wrinkles. To understand why that is, it's important to know that back in late 2009, the Internet Corporation for Assigned Names and Numbers opened the door to non-Latin Web site addresses.
And between that move and Google's last week, that means that it's now possible to get in your Gmail inbox an e-mail from, say, 'Buddy@Yahဝဝ.cဝm.' You'd have to be particularly eagle-eyed to realize that the o's have been swapped out in favor of the Myanmar letter Wa. It opens up a whole new universe of spam possibilities.
So today Google announced that it has begun rejecting e-mail addresses that have strange letter combinations in them, using as a barometer of oddness the internationalization group the Unicode Consortium's specification for mixed-character sets that might well be spammy.
The substance of the change aside, it's a bit fascinating to watch the Internet's patterns and practices circulate these days. IETF wrote its standards for non-Latin-character e-mail back in 2012. But they haven't gone much of anywhere. Google points out for those addresses to be truly universal, every e-mail provider, not to mention every web form, has to be willing to accept them.
"That's obviously a tough hill to climb," the company notes. "The technology is there, but someone has to take the first step."
More than that, increasingly the changes making their way around are recognitions that the Internet is an ever-more global medium, like re-orienting the Facebook globe icon to where a user is in the world. The Gmail character rollout is one example of the sort of change-adjustment cycle of decision-making that is churning along every day, determining along the way how the global Internet of the future will work.