South by Southwest was an absolute blast. I could go on and on about the panels I attended or the people I met, but I'll just say they were all awesome. Instead this will be about what I saw as the overarching themes of the conference: accessibility and data ownership/transferability.
Big Point #1: Everyone is online. We're all aware of this, but too few of us produce work that is accessible to anyone besides english readers in standard web browsers. What about assistive technologies? What about non-english readers? How will we deal with localizing our work? There are some great projects to help with the creation and maintenance of content in multiple languages. More and more, we're going to have to make use of these things.
First, it's important to nail down the key terms. Internationalization does not equal translation. Internationalization, according to User Experience Designer Jon Wiley from Google, refers to the overall design of a product. That means getting down to architectural considerations, like whether or not the visual presentation of a site is flexible enough to allow for changes in character set height and width or direction that appear with localization of the actual content. According to him, English has relatively short words, which can expand greatly when translated. As a rule of thumb, he says that any layout should allow a +40% size tolerance to account for that expansion.
Beyond those architectural decisions, content must be localized. Wiley describes localization as the act of making content appropriate for a target audience. This goes beyond translating words according to culture to color considerations. For example, what does red mean to a target audience? It might mean urgency, or it might imply luck.
Google is available in 117 languages, and gracefully shifts from left to right text (ex: english) to right to left text (ex: hebrew). Google's visual layout is also fairly simple compared to a lot of our work. Luckily, they've produced some useful tools. CSS Janus does a fairly decent job of translating left to right-focused CSS to right to left. Google Translate is useful to check the fitness of a visual design for proper expansion for other languages (NOT as a translation service, of course).
So to summarize, everyone's online, and content needs to be accessible to more people. Localization is no small task. Following accessibility guidelines is not easy. So this point, machine translation projects cannot produce worthwhile text, so budgets will have to expand to work with translators. Still, it's crucial to get the full breadth of the audience on board with whatever it is a site is promoting, so it sounds necessary to me. Honestly, I'm excited about this stuff. I'm in the process of a full redesign of the website of a community bike shop I work with, which will be localized for both english and spanish-speaking audiences.
Big Point #2: We want control of our data. Again, we're all aware of this. Down in Austin, I felt a genuine air of paranoia about who owns data, and how protected and portable it is, and how we sort through it all. We've all dumped loads about ourselves on the web, and clearly business has an interest in using it to target marketing.
Users need choices and clear explanation about what is public and what is not. I'm probably not alone in being unsure as to which parts of a Facebook profile are public and which aren't. Things need to be simple and easy. Catering to the lowest common denominator is not always preferable, but with everyone using data harvesting services, maybe it's necessary. I can't imagine it's a bad thing that everyone knows exactly how Netflix works. They bop you over the head on every page as to what ratings do and how they produce better recommendations, so even my parents know how it works. That sort of transparency is good.
What about who owns the data? Facebook's culture is to just dump everything about yourself. Zuckerberg (Facebook's CEO) says that he's in this to change the way we communicate, and says relatively little about how that information will be used. Seemed like Beacon wasn't widely loved. What about our loved services? Flickr rules, right? Well, it's already owned by a fairly large company, which may soon be owned by an even larger and not always loved company. What about services like Twitter that exist with no revenue and tons of users. We love Twitter, but I don't understand how any company could exist with no revenue forever. How will these sorts of services use that information? Again, I love you, Twitter. This is not an attack, just a question.
How do we access all this stuff? There was a great panel that focused on OpenID and oAuth. Creating a new profile for every site we visit seems a little silly. Decentralized single sign on sounds great to me, but it's crucial that we trust those third party authentication services. I'm hoping to do more with OpenID in my projects at Echoditto.
In closing, the big points are things we've been aware of and thought about for a while now. The difference is that it's clear the rest of the web is thinking about these things too. We're in for some interesting times. Drupalers, time to look at the internationalization handbook. Maybe it's time to read about RDF, as it might blow our data concerns wide open. There's some neat stuff coming down the pike.
