Language development computer:
Computer-based method for aiding language development seems like an interesting idea, the trick with this would be in getting the computer to take on part of the role of the human in the checking process.

In all computer aided learning applications aimed at the very young, there is a danger that the computer may be seen as a replacement for a human instructor, however computer are notorious for not volunteering information or inability to deal with erratic behaviour. Getting this right would be a real challenge. As it seems now certain that inappropriate language teaching (forms of language teaching which are not in tune with the real as opposed to chronological development age of the child) leads to 'induced' (as opposed to real) dyslexia, finding the right targets for such systems will be a challenge in its own right.

With two such competing challenges, this project is really interesting.

Spell checking:
The first problem here is the fact that language is an abstract concept mainly built on rules that can be incorporated into logical software, but with some rules which are subjective and cannot be automated so easily.

The problems stem from what language actually means, what we are doing is producing a series of stylized sounds, which can be interpreted as language. You don't actually see language, it is an abstract concept built on rules. The only way you can know that it exists is through interpreting the results of its interactions with you. In the same way you can't smell intelligence!

Standard spell checking functionality tends to use a binary tree data structure to process a dictionary and a boolean operation to check the spelling of a word and returns true if the spelling matches an entry, false if not. The problems with this approach are compounded, especially with dyslexics, in an assumption that the initial letter, at least, is correct.

This project, on the other hand, starts from an identification of the meaning of the chosen word, and will return a series of correctly spelled results, none of which may match the target word. It is likely that the user will be requested to select one from a range of possible meanings, before being provided with the results.

The first problem here is related to how we get into the unconscious mind? The only way is to see what happens when people use words and see the effects of these things on that part of the world that you can perceive. However this is in effect arguing by analogy

Also there are some rules that are subjective and cannot be automated, the other half of them are not, where the program would consist of a database using some sort of parsing routine that detects the misuse of a word with respect to its contextual position.

Word Find:
Assuming it is possible to disregard syntactic information and ignore all syntax that functions as syntactic and yet still objective definition of the scaling constants.

This project could be very useful, a reverse thesaurus has many applications where it is helpful to find smaller words. For example situation where the uneducated are incapable of comprehension of polysyllabic words. The task here is to first create a way of breaking down works in a way that does not.

Stage one of this process is complementary to the Spell Checker discussed above.

The exciting part, from my perspective, is this. A slow/late learner, aged 18, is currently required to use books designed for primary school children, lacking the emotional maturity and depth of story to attract the older person.

A truly programmable word find could automate the 'translation' of the great (or exciting) works of literature into language accessible by persons with different levels of language development. Such persons could then develop the excitement in reading which drives reading development.

I am the website administrator of the Wandle industrial museum (http://www.wandle.org). Established in 1983 by local people determined to ensure that the history of the valley was no longer neglected but enhanced awareness its heritage for the use and benefits of the community.