Is there ink-recognition relevance to the rsereach at an Elingsh uinervtisy, taht seowhd it deosn’t mttaer in waht oredr the ltteers in a wrod are? Should word recognizers try to exploit a similar strategy, concentrating on recognizing the initial and ending strokes of an ink group, the length of the group for a rough letter count, and just sort of muddling along in the middle?
Hey, speaking of ink-recognition, someone from Microsoft told me that the Tablet PC ink recognizer is derived from the Newton’s (in)famous recognizer via Pen & Internet’s Calligrapher. Just to make things confusing, Pen & Internet ships a “next generation” handwriting recognition tool for the Tablet called RitePen. 30-day trial available.
Your tax dollars at work: the latest PC Magazine reports “Researchers at the U.S. Department of Energy’s Idaho National Engineering and Environmental Laboratory have developed software that can detect extremely tiny differences — smaller than a fraction of a pixel — between two digital images….”
Update: PC Mag’s description is wildly inaccurate. The INEEL “breakthrough” is a program that rapidly aligns two digital images and then alternates them on-screen, which causes non-matched pixels to flicker, at which point the human observer notes the differences rapidly. The real impressive part is that “The alignment compensates for differences in camera angle, height, zoom or other distractions that previously confounded flip-flop comparisons.”
All the Tablet bloggers are pitching in ideas for Microsoft’s internal Power Toys competition (it’s so easy to think of software). Loren’s suggested a “snap on dwell” tool and Peter a “pen scrolling tool.” So my suggestion is that if you can’t pull off a fully customizable and skinnable tablet input panel (which, by the way, ought to have pen-friendly keyshapes):
…and you hate radial context menus…
… then can’t you at least see that we need an IE toolbar with an ink-enabled address combobox?