User:Max/rwm/bridge

Predictive Text

The first text message was sent in 1992 from Neil Papworth, a former developer at Sema Group Telecoms. Mobile phones did not have keyboards at the time, so Papworth had to type the message on a PC. Papworth's text — "Merry Christmas" — was successfully sent to Richard Jarvis at Vodafone. Most early GSM mobile phones did not support sending SMS (Short Message Service) text messages and only Nokia supported the sending of SMS text messages. Like often with new technology (Link to M McLuhan) the initial growth was slow.

0.4 SMS were sent per month per person / customer in 1995. In the year 2000 customers sent 35 SMS per month. Then in 2007 for the first time there were more text messages than phone calls per month.

Writing text messages on mobile phones was always hard. Because the available space on a mobile phone is very limited keyboards are not very convenient. A mobile phone in the early days had 9 number-keys, the first common method of commercial texting was "multi-tap", which means that each key displayed three to four letters. For example the key "3" holds "D", "E" and "F". To select the letter "E" you would have to tap two times the number-key "3". "Multi tap" was not very efficient.

In the 90s Tegic Communications developed a system/technology to simplify keying: T9. T9 stands for Text on 9 keys and was one of the first predictive text technologies. Instead of selecting each letter by tapping multiple times on the number keys with T9 the word gets formed by a single keypress for each letter. The groups of letters on each key are connected to a dictionary of words and the telephone looks up in this dictionary what words can be built out of the typed sequence. For example pressing "4663" will typically be interpreted as the word "good" (alternatives like "home", "hood" and "hoof" are also valid interpretations of this sequence). By offering words it speeds up the process of writing texts. It was also possible to extend the dictionary with individual entries and phones could learn to adapt to the user (= feedback loop, Norbert Wiener).

Full keyboards on mobile phones was first introduced in 1997 with the Nokia 9000i Communicator. It became a popular feature in the late '90s to early '00s.

In 2007 Apple introduced the iPhone which had a multi-touch interface with a virtual keyboard. Other companies followed soon. The virtual keyboards supported auto-correction, later also auto-completion. In 2014 with iOS 8 then Apple introduced the predictive text keyboard QuickType. It is a row with three words (word proposals) above the virtual keyboard. The more you use QuickType, the smarter it will get as it learns your style of writing and choice of words. Very similar predictive text keyboards exist on Android and Windows Phone devices.

My Current Work

In my work Love Letters (working title) I will take use of these current predictive text algorythms to generate text messages. In the performance there will be two phones lying side by side, one is mine the other one from my girlfriend. The disply will show a text message application. A primitive robotic arm will formulate a text message by using only the three word suggestions offered by the phone. Then it will be send to the second phone which then starts answering.

The work deals with the increasingly automatization of our communication and the the constructed reflection of our identity in our communication devices.

---

Norbert wiener

boston arm

hearing glove / seeing glove

Keywords:

hot cold

participation

redundancy

1500 words

read: Cybernetic Brain

1) Explain my work

2) compare it make bridges to

McLuhan > Media circulates and changes the society

Wiener > Feedback Loops, systems adapt and learn (iPhone dictionary)

Shannon > Redundancy / Messages