> Date: 2005-02-18 00:51 > From: Martin Duerst <duerst@xxxxxx> > At 11:02 05/02/12, Bruce Lilly wrote: > > Â>While I do not dispute that some mobile devices might use some subset > Â>of some version of Unicode for text in some languages, my point was, > Â>in response to John Klensin's "Until and unless every one of us has a > Â>keyboard that permits easy input of every Unicode character", that not > Â>only do I not expect to have a keyboard permitting *easy* entry (no, > Â>that doesn't mean "Grafiti" or "Decuma") of *every* Unicode character > Â>any time soon, I don't expect it *ever*, because the Unicode code space > Â>is expanding (in contradiction to the original Unicode Design Principles) > Â>faster than the available memory space on low-power, compact, mobile > Â>devices. > > Bruce - Would you care to supply some sources for the argument above? > Computer power, including mobile devices, is still pretty much > increasing exponentially. Unicode space allocation has never > increased exponentially, and is slowing down. The low end of computing devices -- what one should take into account per RFC 1958 section 3.1 -- has remained fairly constant in terms of available memory and processor power (price, power consumption, physical size, etc. have declined). Also, bear in mind that these devices may have a useful life of a decade or more, and that (software/firmware/hardware) upgrades for some older devices still in use might not be available, or might not be applied by some users. Regarding keyboard input, again the capabilities have remained fairly static at the low end, with input limited to about a dozen keys with multiple characters overloaded onto each key, and limited character selection. Unicode code size increased overnight by more than 4 orders of magnitude (a factor of 65536) when it went from 16 bits 65536 code points) to 32 bits (over 4 billion code points) at the same time that it incorporated musical notation etc. in contradiction to the Unicode Design Principles. So Unicode *everywhere* isn't likely to happen because there are and will continue to be low end devices that simply don't have the capacity to support normalization tables etc. for billions of code points -- now perhaps some of those low-end devices might be able to support Unicode version 2.1, but then there are the version incompatibility issues that I mentioned. _______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf