Typing is tricky
With the spread of Unicode, it’s now possible to create texts in most of the languages of the world directly on the web. No more weird, inaccessible images of text.
So, everyone gets to write in their own language now, right?
Well, not so fast, unfortunately. There’s a problem that many folks seem to forget: Typing is tricky.
Think about trying to teach someone to type.
“This’ll be easy,” you think, “To explain how to type «a» I just point at the key with the «a» on it, and say, ‘Pound on that thingie right there!’”
Except, what do you see on the key you use to type «a»?
You see «A». That’s right kids, hit «A», get «a».
What if you actually want an «A»? Press the mysterious “shift” key. Well, explain that there are two, but they’re really the same (usually). And each has a redundant “arrow pointing up” icon (the «A» on the key is already “up”!). Got that? Good. Now, hold it down while you type «A».
That’s the thought process that new typists face… to type «A».
Now, multiply this challenge by the number of keys on the keyboard.
And multiply that challenge by the number of physical keyboard layouts in the world.
And that’s where we begin to see the problems that a computer newbie who bravely buys an hour’s worth of time in cybercafe faces.
I actually have a broader reason for getting into this topic, and there will be some more posts coming along, but for now…
Seriously, just think about this for a second:
7 comments.
Technorati tags: accessibility, global voices, input, keyboards, Language and the Web, language rights, typing, unicode
Someone who uses non-Qwerty characters in their native language probably types most of the time using a (hopefully) standardised keyboard layout that accommodates what they want, in an order that they learn. This is, I think, more an issue for those people who are typing characters in languages they don’t usually use. In that case there is likely to be another problem, ie. how to be sure which of several similar-looking characters they actually want to type. When I designed my Unicode Pickers I tried to arrange the characters in a way that helped with that, or provide a JavaScript lookup that highlights similar characters.
I mentioned this in passing when I spoke at Reboot this year: I can’t believe we’re still using *typewriters* as our interface.
Yes yes, Dragon Dictate, etc. etc. Roll on tangible interfaces/ cranial plugins/ and so on.
Future generations will find this hilarious.
Korea went full-tilt with a QWERTY based keyboard – albeit with a twist, in the sense that the characters on each key have nothing to do with the latin character they share space with.
In Chinese, I guess only the gwailos use pinyin / jyutping whatever other phonetic transcription system to input Chinese [that includes yours truly]. I tried bopomofo - blech - and the four corners or whatever, but let’s just say it wasn’t a frank success. Although watching HK kids typing sms, they look so freaking slow compared to, say, Koreans. Typing all-hanzi messages with just a few keys is dreadful. I bet if they bothered learning pinyin — after all pinyin is used in ChiCom to romanize Cantonese — they’d be faster…
As for Japanese, they have like Korean a qwerty to kana keyboard layout, which I never mastered, but it is quite fast, except for kana-kanji henkan. Realize that today the challenge is not so much inputing, say, Burmese on a computer keyboard, but on a mobile phone’s. And no cheating! Blackberrys stay home! ;-)
I agree, computer keyboards are usually not a problem anymore these days, it’s mobile phones that are.
My mobile phone is the only place left where I still leave out accents on Latin characters (a instead of á and so on) and so does everybody I know.
I’ve absolutely no idea how non-Latin alphabet folks type text messages on their phones but I guess it must be incredibly awkward.
I should make some some movies of hangul/kana~kanji/chinese sms. That would be interesting I guess…
I have to say that I disagree. I think there are still a LOT of challenges with traditional keyboards. I’m also willing to bet that they’ll be around for a few more years. The alternatives have shortcomings of their own: typing is still faster than dictating to a voice rec system, and dictating also causes significant voice strain.
As for inputting on cellphones, I agree that it’s a fascinating challenge and will assume more and more importance as time goes on. (Here’s a nice post with a good overview of interesting developments in Amharic by Ethan Zuckerman: …My heart’s in Accra » SMS in Amharic… and near-independent of Ethiopia Telecom.)
But I see SMS input as a separate problem, with distinctions solutions. For instance, in the case of the complex Amharic script, the SMS system mentioned in Ethan’s post actually eliminates certain homophonous characters from the repertoire. This is great for communicating quickly and easily on a cell phone, it’s just plain unacceptable for a novelist. (And, I might add, for speakers of the many languages which use other variants of the Ethiopic script.)
When thinking of keyboarding problems, I try to keep in mind what I see as the prototypical keyboarding problem: someone walks into a cybercafe in Phnom Penh, Kinshasa, or São Paulo (heh, right up the street from me…), purchases an hour’s worth of time, and wants to email a job application in Khmer, Tshiluba, or Portuguese.
Asciified text such as is often used in IM and SMS simply won’t do in such a situation. They need all the អ’s and ù’s and ã’s in their right place, end of story.
Right now, the situation is such that the average, informed user does not assume that this is possible when they sit down in that cybercafe. And by and large, they’re right. That’s a big problem, and needs to be fixed…
I remember having a booger of a time with text on my first blog on blogger. I wrote my first post in all caps and people though I was yelling at them.