Monday Sep 19, 2011

Squirting is a Software Experience? Mind Your Language, Please

The language used in an application's user interface (UI) is a critical aspect of the user experience (UX), bit one often overlook. Des Traynor (@destraynor) brought this importance artfully to life at Refresh Dublin in his presentation on the Language of Interfaces. Well worth checking out, Des emphasized how language choice determines user action and engagement, with the simple choice of text for a button label or placeholder for status update making all the difference.

In Oracle Fusion Applications, for example, there's a big difference between the button labels Save, Submit, or Done, and the action that they imply to take on a page. Save implies an intermediate state during data object or process creation that the user will return to later before the task can be finalized. Submit is a final action, committing an object to the database or handing off a process, thus ending the task. Done is generally used to conclude the user review of a read-only page, closing it.

Save and Submit buttons together on a page


Google Wave's choice of Done however (as pointed out by Des) didn't help much with the puzzling concept of what anyone was expected to do with a wave to begin with. Language alone isn't going to save a rubbish UX.

Google Wave UI Done button


Des used some great examples from social media to as examples. Compare the language and action implied of the Facebook friend with the LinkedIn contact or the contact categorizations of Google+'s circles. Determining the action should shift from a third-person to first person paradigm led Facebook to change its status update text to What's on your mind? Twitter switched from What are you doing? to What's happening?

US English Twitter and Facebook status placeholder text

Not every natural language follows the English direction however. What's up with that? And, what about the challenges offered by crowdsourced language (as in the Dutch version of Twitter)? Facebook's community translation feature, as I pointed out before, is as much a user engagement strategy as a way of obtaining translated UIs (but not help) very quickly for the local market.

French and Dutch Twitter status placeholder text

French and German Facebook Status Placeholder Text

This choice of evolving or action-intended words can be a challenge for controlling the action globally. My old friend Frank Dietz in Multilingual magazine tells of the challenge of finding German translations for gaming concepts (buff, debuff, kiting, toon hop, and so on) for example, having to rely on transcreation, Denglisch, or the English term itself.


What the presentation didn't cover was how the language in the UI drives the creation of language around the intended action within the user community too. Unfriend, for example, appeared nowhere in the Facebook UI, but is a well-established word now. ReTweeting (or RTing) was a term and concept that came from the Twitter community, before it was codified. Personalization features that allow users to control the language or add their own are critical UX features too, particularly in the mobile space.

Apple iOS5 shortcut personalization feature

As for the choice of squirting to convey the sharing of music in Microsoft Zune (see Des's presentation), well, nobody over the age of five should be squirting anything at anybody, should they? What were they thinking? And yet,they're back with internet charms...


Find those comments...

Wednesday Jan 19, 2011

Language and Usability: It Doesn't Have To Be That Way. And It Shouldn't

I enjoyed this very telling comic strip about how some usability professionals might regard language when it comes to designing the user experience:

"...as for the text... just copy something from Wikipedia and polish it up a bit."

However, it's unfunny because very often it's true. Still don't get it? Read the comments of some of our sales folks, partners, and customers (names have been removed to protect the innocent):

  • "When the language is IT-speak and not normal speak, it is part of the usability problem." 
  • "When users know what things mean, they're halfway there. When options are 'simple' English, it's obvious where to look and what to do... the language for users must be clear and helpful." 
  • "The quality of language in the user interface (UI) is a deal breaker."
  • "The competitive playing field has changed since the old days. Customers even do their own usability testing and include language heuristics now. It's a competitive issue."

So, in Oracle Fusion Applications we've spent a great deal of time getting our language (terminology, style and grammar) right for our users. Developers don't create the language you see in the UI or user assistance. There is a comprehensive terminology creation and management process and a multi-stakeholder review process for all UI strings, messages, and help content. Language is a user experience issue and that's why we continually invest in our information quality efforts. We have to. It's a strategic competitive issue for sales and as a productivity issue for users when they work. Watch out for more information about information quality in Oracle Fusion Applications on this blog in 2011.

Thursday Sep 30, 2010

No More Fart Apps. Would We Ever?

I loved Lucy Kellaway's (she of the Financial Times) article "Words to describe the glory of Apple" comparing the language style used by Apple and Microsoft and wondering if language style impacts the bottom line. If you don't want to register for the article, then the podcast version is here.

iFart Screen

Ms Kellaway tells us that Apple's language makes for content that is "fun to read", "elegant", and "makes you laugh". It has a tone that is "direct', 'comic", and "elegantly threatening". She contrasts this with the Microsoft language used for the new version of Internet Explorer, "architected to run HTML5, the beta enables developers to utililize standardized markup language across multiple browsers" and the rest. This is "standard stuff" from Microsoft, and Luce is "irritated", "bored", "alienated", and "restless" with it.

Actually, I think the article is unfair to Microsoft, who do care about language and the example used is not representative of language used in other Microsoft products - games, for example. Plus, the audience for their words is different to Apple's; basically it's a marketing pitch to Microsoft's partners, encouraging uptake of a beta release.

The Apple language that Ms Kellaway admires is taken from the App Store Review Guidelines (full PDF version); a set of guidelines more likely to be read by geek and hobbyist developers working from home than the corporate equivalents. Such language is not used in other Apple products themselves. In fact, language quality is not an acceptance criterion for App Store submissions at all. That of course, is telling in itself. Why not let the market, the users, decide on the language? In line with this, the Human Interface Guidelines from Apple for the iPhone has a common sense approach to language style. For example:

In all your text-based communication with users, be sure to use user-centric terminology; in particular, avoid technical jargon in the user interface. Use what you know about your users to determine whether the words and phrases you plan to use are appropriate.

For example, the Wi-Fi Networks preferences screen uses clear, nontechnical language to describe how the device connects to networks. Very reasonable from a user experience perspective. But, then Lucy says: "You might think there was a clear commercial advantage to be had in writing clearly and stylishly. But you would be wrong."

Not quite. There is a relationship, though it may not be all that visible to key decision-makers. That's because the commercial advantage does not come from writing clearly or stylishly per se, but its application. It comes from writing content that users actually want, in a way that they can understand, using terms and language that suits them; and that facilitates easy search and retrieval. The result is a quicker transfer and comprehension of information leading to better productivity for users and less training and support costs. And that's a competitive advantage.

In the enterprise applications space the opportunities for a product language tone that is "fun", "comic", or "elegantly threatening" doesn't exist in the same way it does for the iPhone app development community (and let's face it - for all the BS about the iPhone - most apps are little better than free low-tech toys designed by rank amateurs).

But that's not the point. The point is the language should suit the audience for the information. We need to spend less time worrying about our internal language style and all its nuances and rules and concentrate more on how users - our customers - themselves want it to be, and actually use it. Bringing terminology in line with user expectations and concentrating on a few basic writing principles grounded in research on information search, retrieval, consumption, and problem-solving would yield far better bottom line results than fretting about a rigorous adherence to every single aspect of a style guide for no other reason than it's there.

Saturday Aug 14, 2010

Language Should Never be 'Plain'

I see a lot of UX professional debate on the subject of plain language. Many of these arguments are decontextualized. They often import personal frustrations from filling out government forms in the US or EU, anecdotal evidence about technical error messages, and so on. This is not very helpful for making a case about what language user assistance should use for the enterprise applications user experience.

Sure, we must communicate ideas clearly and succinctly, but we must also do so using the terminology of the target audience's roles,expertise, and how they actually work. Generalizations about plain language certainly seem very reasonable when we discount such variables, aren't they? Often we see recommendations made for one design context that simply don't apply in the other.

The most notorious one in the user assistance area is the conflation that all online users read publicly available web content the same way that they would read content in the enterprise space (for example the "golden triangle" or "F-shape" argument). These findings do not hold true in the enterprise applications world.

Same for language on the web. Instead of talking about some globally applicable 'plain language' being required by all users, the discussion needs to move in the direction of information quality, and away from the dominance of internal linguists, and towards the conversations in the user community.

Information we deliver in user assistance components should reflect the needs of the user and how they work. We must use terminology that our users recognize and use consistently when interacting with the UI, searching, reading online, and most importantly, when they seek help or help each other. Getting your terminology right is central to information quality.

Engaging the user community and their conversations is key to this. On one level, it's easy. Why call "breadcrumbs" something else if the term is widely accepted in general use? Or why say "Enter a valid password" when you can say "Enter the correct password"? And, then, of course, we have some language used without thought of use at all (like "OK" in dialogs instead of more meaningful button options).

But, in the enterprise space with such a broad range of domain expertise and many possible roles, involving specialized vertical, functional or technical expertise, additional layers of complexity are added to our choice of words. User experience is about knowing who our users are, so use that research.Terminology, and therefore language, can hardly be 'plain'. This is an area that interests me greatly.

So, stay tuned to this blog and the usableapps website  for news about UX developments concerning language and user conversations.

Monday Jul 05, 2010

Globish?

I've just been reading Robert McCrum's Globish: How the English Language Became the World's Language. Interesting story (for that's really what it is) about how the development of the English language combined with political, technological and globalization forces brought out a form of English called 'Globish' that's been accepted as a global cultural norm and certainly can lay claim to being the lingua franca of business worldwide.

Jean-Paul Nerrière coined the term 'Globish' in 1995. It was "a utilitarian vocabulary" of 1,500 words; a form of English intended for non-native speakers. We can see it in action when McCrum says:

"Globish is essential to India's globalizing ambitions... (it) is the emerging supranational lingua franca that enables a call centre in Bangalore to answer impossible queries, or sell new product, as far afield as Cheltenham in the UK, Cedar Rapids in the United States, or Co. Cork, Eire."

However, bear in mind other research that shows that consumers prefer language versions of websites when it comes to buying goods and services. And we know that products and services must be localized along with their delivery mechanisms for users gain international markets. Quality of language is a deal-breaker in the enterprise space too and it definitely impacts adoption and productivity too. Try selling apps in the public sector in the wrong language and see what happens!

So, before you think we'll all end up speaking English some day, think again. Language (translated or otherwise) is part of the UX. I couldn't agree more with Gerd Schrammen of Verein Deutsche Sprache (the German Language Association), when he says, "It is perfectly legitimate to expect one's own language to be used at home."

About

Oracle applications user experience (UX) assistance. UX and development outreach of all sorts to the apps community, helping to design and deliver usable apps.

Profile

Ultan Ó Broin. Director, Global Applications User Experience, Oracle Corporation. On Twitter: @ultan

See my other Oracle blog about product globalization too: Not Lost in Translation

Interests: User experience (UX), user centered design, design patterns, tailoring, BYOD, dev relations, language quality, mobile apps, Oracle FMW and ADF, and a lot more.

Search

Archives
« April 2014
SunMonTueWedThuFriSat
  
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
   
       
Today