Linked by Thom Holwerda on Sun 6th Aug 2017 20:52 UTC
OSNews, Generic OSes

Before I link you to the story this item is actually about, I want to tell you about one of my biggest frustrations with computer hardware and software. It's something that I have to work around every single day, and its consequences bother me almost every few minutes.

Hardware and software have no idea how to handle people who lead multilingual lives.

Like hundreds of millions of people, I speak and understand several languages, but on top of that, I use two languages every single day: Dutch and English. I switch between these two all the time, often even multiple times a minute when juggling multiple friends, clients, work-related material, entertainment, and so on. I might be writing an e-mail to a client in English, work on a translation in Dutch, WhatsApp with a friend in English, and write a Facebook post in Dutch - switching between all of these.

Software has no idea what to do with this. The most operating systems like Windows and OS X can do is offer a small icon somewhere tucked away to manually switch input languages, which is incredibly cumbersome and just wholly impractical to perform every time you have to switch languages. It gets even worse on mobile operating systems, which are heavy on the autocorrect (I cannot type on a touchscreen), so if my input method is still set to English while I'm typing something in Dutch, it gets autocorrected into meaningless garbage (it's only recently that both Android and iOS at least offer some form of true multilingual input).

It's even worse when it comes to these voice assistants the entire technology industry is trying to ram down our throats, like Google Assistant or Apple's Siri. Do you know what you need to do to switch voice assistant input language on an Apple Watch or Android Wear device? Are you ready for it?

You need to perform a full wipe and set up the device as new.

Since my use of Dutch and English is split about 50/50 - or maybe 60/40 - the end result is that for about 50% of the time, I cannot use any of these devices to reply to an e-mail or write a text message. While Android Wear 2.0 has a keyboard and handwriting recognition, I have no idea how to change the input language for those input methods. Even if I could by tapping around - the point of these things is that you can use them without having to look away from whatever you're doing (e.g. cycling).

And just in case you think this kind of multilingual use is rare or an edge case: just in the United States alone, dozens of millions of people speak both Spanish and English every single day. This is not an edge case. This is not a peculiarity. This is daily reality for possibly hundreds of millions of people all over the world.

There's countless other daily irritations that arise from this inability of software to deal with multilingual use (Win32 vs. Metro vs. Chrome vs. Office vs. etc., which all have their own input language switching mechanisms I manually have to keep track of), but the point I want to make is the following.

Because software has no idea how to deal with multilingual use, I know for a fact that very few of the engineers working on Windows or Office or iOS or WatchOS or Android or whatever lead multilingual lives, because any person who uses multiple languages every single day would be able to spot these problems within 15 minutes of use. If the manager responsible for WatchOS led a multilingual life, or had a bunch of people on his team that led multilingual lives, WatchOS would've never been released without the ability to easily switch Siri input language.

Despite what some low-level Googler claims in his rambling manifesto of idiocy, diversity matters. Or, as ex-Googler Yonatan Zunger puts it way more eloquently:

Engineering is not the art of building devices; it's the art of fixing problems. Devices are a means, not an end. Fixing problems means first of all understanding them - and since the whole purpose of the things we do is to fix problems in the outside world, problems involving people, that means that understanding people, and the ways in which they will interact with your system, is fundamental to every step of building a system.

If, at this point in time, you still don't understand the importance of diversity when developing products, you are beyond help, and have no place on any product development team.

Thread beginning with comment 647589
To read all comments associated with this story, please click here.
Umm ... what?
by kristoph on Mon 7th Aug 2017 02:40 UTC
kristoph
Member since:
2006-01-01

1) I have an English IME and Japanese one. I can switch between them by tapping one key on the IPhone keyboard. In the English IME I get English suggestions and in the Japanese one I get Japanese suggestions. English, as it happens, is the 5th language I learned to speak and one of a number I use regularly. I don't have any of the problems you describe.

2) I can change the language Siri speaks through Settings > Siri > Language. The reason why it can't speak two languages isn't some conspiracy against diversity, it's just that is way WAY harder to teach a neural network to recognize two distinct languages accurately ( one day, I have no doubt, Siri and Alexa and Hello Google, will be effortlessly polylingual ). In the mean time I don't see how Siri is any different then the billions of people who only speak one language.

K

Reply Score: 2

RE: Umm ... what?
by patrix on Mon 7th Aug 2017 09:57 in reply to "Umm ... what?"
patrix Member since:
2006-05-21

and yet Google Now can recognize at least 3 languages interchangeably without changing any setting, beyond the initial setup of choosing 3 languages. At least in my experience of choosing French, English and Japanese. Of course it doesn't mix-and-match within a single query, but it DOES work, at least it did when I tried it earlier this year.

Reply Parent Score: 1