Flexpansion Ltd

Flexpansion Ltd was founded in 2008 by Dr Tim Willis, during his PhD, in which he researched an advanced form of word prediction called ‘flexible abbreviation expansion’. Tim named his version ‘Flexible Text Expansion’, abbreviated :-) to Flexpansion.

This lets users create emails, text messages, documents etc. using far fewer letters, inventing abbreviations as they go along, without needing to memorise any set shorthands. For instance it will offer to replace input of any of ‘exprn’, ‘xprc’, or ‘xpnc’ with the full word ‘experience’ (as well as other words). Phonetic shorthands like ’4tnl’ for ‘fortunately’ and ‘fon’ for ‘phone’ also work. We call it ‘word prediction for the texting generation’.

An alternative is ‘fixed abbreviation expansion’, which offers greater savings in keystrokes and time, at the expense of the user needing to memorise preset shortcuts. The Flexpansion system is actually a hybrid of the best of both techniques, since its personal dictionary allows the addition of such predefined correspondences by the user. As an example, it can easily be programmed to insert their entire address and contact details when they type ‘zz1′, and their work details for ‘zz2′ – or whatever other abbreviation or full equivalent they want to use.

An early prototype was developed as an add-in to Microsoft Word 2007 as a stepping-stone to mobile. At that time, very few phones were capable of storing or processing the amount of data required to achieve Flexpansion’s functionality. With the widespread recent adoption of smartphones, this barrier has now fallen, and the company chose Goole’s Android platform for their first commercial release (market share: 48%, Oct 2011 : http://www.bbc.co.uk/news/technology-15400984 ).

A free product came out in August 2011, followed a couple of months later by a paid-for ‘Pro’ version with extra features. All early adopters who had already downloaded were upgraded for free to the Pro version, and further features continue to be added to differentiate it more fully from the free version. These include voice recognition, keyboard visual & auditory themes such as an old-fashioned typewriter complete with the sound of letters hitting paper and a whirrrr-zing! when you press Return, and the facility to back up your personal dictionary in case of loss or moving to a new device.

Founder – Dr Tim Willis

Dr Tim Willis is surely one of the foremost technical experts in this area in Europe.

Former CTO, O2 Telefonica

Tim has spent much of the past two decades in teaching, study and research in Computational Linguistics, interspersed with some sales and marketing. His doctoral research at the world-famous School of Informatics at Edinburgh University was on the methods people employed in abbreviation, in order to inform a system for automatic abbreviation expansion. This was initially aimed at computer users with motor disabilities which slow their typing, to enable them to type more text with less effort, and express themselves more fully. He then realised that the technology also had a much wider application in the massively growing market for mobile devices – since everyone’s input is hindered by their small keyboards.

He holds a BA in Linguistics & Phonetics (Leeds University), plus an MSc in Cognitive Science & Natural Language, and PhD in Informatics, both at Edinburgh. He was a Research Assistant at Leeds, and Lancaster University, building and annotating large databases (‘corpora’) of English language in high detail, including semantic, grammatical and case structural information.

Tim has experience in project and team management, and has worked in sales for a web design company in Cumbria. He also dealt with the correspondence and merchandise for the band Utah Saints in the 1990′s (still going strong :-), which produced some great anecdotes.

Tim enjoys white-water kayaking (‘terrifying but exhilarating’), climbing and snowsports. He has also done a couple of stand-up comedy routines in front of audiences of a hundred (‘just terrifying’).

Publications include:

Willis, T.A. (1992), “Parsing Natural Language”, Chapter in Roach, P. (ed.), Computing in Language & Linguistics, Academic Press.

Willis, T.A., (1995), “Using the TOSCA analysis scheme for ICE tags, to tag and parse the Lancaster-IBM Treebank.”, Proceedings of ICAME (International Computer Archive of Modern English), Toronto.

Willis, T.A., Bateman, J., & Forrest, J., (1997), “The use of syntactic annotation tools: partial and full parsing”, Chapter 11, in Garside, R., Leech, G. & McEnery, A. (eds.), Corpus Annotation: Linguistic Information from Computer Text Corpora, Longman, London.

Willis, T., Pain, H., Trewin, S., & Clark, S., (2002) “Informing Flexible Abbreviation Expansion for Users with Motor Disabilities”, ICCHP, Linz, Austria. This paper also presented at CLUK 2003, Edinburgh University.

Willis, T., Pain, H., Trewin, S., & Clark, S., (2002), “Informg Flexibl Abreviatn Expansn 4 Usrs wth Motor Disablitis”, ASSETS 2002, Edinburgh.

Willis, T., Pain, H., Trewin, S., & Clark, S., (2005), “Probabilistic Flexible Abbreviation Expansion for User with Motor Disabilities”, Accessible Design in the Digital World conference, Dundee, Scotland.