View Single Post
Karel Jansens's Avatar
Posts: 3,220 | Thanked: 326 times | Joined on Oct 2005 @ "Almost there!" (Monte Christo, Count of)
#15
Originally Posted by benny1967 View Post
While I'm of course always excited to learn about new technologies, I am not at all happy to learn that this is apparently planned for the N800.

Why? It's closed source, proprietary code again. When the original 770 came out and included some proprietary parts, my expectation was that they were used only for a quick start to fill in the gaps. I hoped that Nokia would either actively support free projects to supply alternatives or develop those alternatives and make them available as parts of the free platform.

With OS2007, it seems we're moving to the other side very quickly: First they announced Skype where they should have improved GTalk and SIP-handling. Now they include patented XT9 technology instead of trying to imprvove existing GPLed projects. What will be next?

Again, I'm not trying to say XT9 isn't a good thing. It certainly is. Such as MS Windows is a good operating system. I only wonder if the N800 maybe not only introduces technical changes, but also a change of concept: will the N900 be a closed, non-hackable box such as any other device?.
I feel your Open Source pain.

However, do note that there exist exactly zero -- as in: none -- Open Source handwriting recognition engines for Linux -- or for any other platform that I know of (warez don't count). The best ever made for Linux was Xstroke, a copy of Graffiti2, and that one's essentially dead.[*]

If I have to choose between no Open Source solution and an excellent (from what I've been told) Closed Source one... Well, let's just say I won't be in dilemma for more than a couple of microseconds.
____
[*] It's a pity, really. I'm on the Xstroke mailing list, and a while back someone (Dr. Limburg IIRC) proposed he'd merge his own HWR neural net algorythms into Xstroke, thereby transforming it into a real HWR solution. I haven't heard from him since that one message.