The razor-toothed piranhas of the genera Serrasalmus and Pygocentrus are the most ferocious freshwater fish in the world. In reality, they seldom attack a human.

The only reason I mention this at the beginning of a technology column is because they're the two sentences used by Guinness World Records when testing contenders for the title of fastest texter. The record is currently held by Marcel Fernandes Filho, a 17-year-old Brazilian, who a few months ago completed the challenge in 17 seconds flat. I just tried it and it took me 58 – mainly because of repeated problems with the word "piranhas".

My texting speed has been dreadfully slow of late – in fact, ever since I upgraded my phone to a larger model. My fingers prod in the vague proximity of the correct key, but I'm sufficiently off target for my phone not to know what the hell I'm doing. "For crying out loud," I wail to no one in particular, "there must be an easier way".

Ever since virtual keyboards first appeared on touchscreen devices, developers have searched for ways of making typing more accurate and less frustrating. The most effective strategy has been to use autocorrect and prediction algorithms, making best guesses at what we're trying to type and mopping up our mistakes as we go. Fleksy, a third-party app for iOS and Android, prides itself on its superior ability to guess what we might be about to type, and Filho used that system when he broke the texting record in November.

But there's something about a computer thinking that it's one step ahead of you that feels mildly insulting; if it suggests "call" after I've typed "please give me a", I want to erase it and type "ring" just to assert my own independence. It's the same with emoji suggestions from keyboard apps like SwiftKey; it knows that the words "I love you" are often followed by a heart or a kiss, but the idea of being that predictable is galling. It makes me want to stick an octopus or an ambulance in there instead.

The other strategy, which is more idealistic but harder to foist upon the public, is to rejig the keyboard layout itself. It's well-known that the QWERTY keyboard was originally designed to slow up typing speeds in the age of mechanical typewriters, but alternative layouts boasting superior efficiency such as Dvorak (which begins PYFGCRL) never really caught on.

And it's the same with smartphones; alternative layouts such as KALQ and Minuum would undoubtedly improve the speed and accuracy of our typing if only we could be bothered to sit down and learn them. This week, another contender, Wrio, has thrown its hat into the ring via Kickstarter; it keeps vaguely the same layout as QWERTY but juggles it to create a keyboard of interlocking hexagons. It has two space "bars" (one for each thumb) and also claims to learn your slang and your typing behaviour – the best of both worlds, coming together with the hope of revolutionising touchscreen typing.

But QWERTY is ingrained in our culture. Any alternatives, no matter how clever they are, will ultimately come up against a brick wall of laziness. Speech recognition is becoming so good that it seems to offer the most attractive way out of onscreen keyboard hell – but it requires an internet connection, and we rarely want to be seen in public barking instructions at gadgets. How long until our phones are able to lip read?

Comments