|
The problem is, achieving anything better than we have now would be expensive in terms of:
- human effort up-front --- requires better mark-up, esp. coding discretionary hyphens for words which are hyphenated different depending on which form of speech they are (present (to give) vice present (a gift))
- processor effort on the machine, increased requirements of testing for Q&A --- while a multi-gigahertz machine can typeset a TeX document all but instantly, back in the day of 28MHz 68040s my NeXT Cube could take _minutes_ to typeset a complex document --- a more complex algorithm will be more likely to have odd effects in edge cases
- there aren't any good h&j algorithms which will eliminate stacks (multiple instances of a word appearing at the beginning or end of a line) --- every effort to code one which I've seen has been unable to avoid getting stuck in an endless loop
Devices w/ larger screens are available --- the problem is they're _expensive_ (and are starting to go away --- Fujitsu seems to've discontinued their Stylistic ST-6012 slate computer)
William
|