On 8 July 2012 21:14, la .lindar.
*Regardless* of the arguments, I'm staunchly against it precisely for the fact that people will *use* it. Regardless of whether or not *I* have to use it, *other* people will, which means that *I* have to learn to read it. Therefore, it *does* affect me, even if I *do* choose to use the original sumti connectives. Therefore, I will always veto this idea every time forever.
That being said, I will gladly bring it to Robin's attention, who I presume will laugh and say, "Fuck no.", but very well may not.
I'm not proposing a change in the language, not yet, anyway. I'm using this as a case study to find out the *purely technical* problems this kind and magnitude of change would cause, especially at the level of my present parser framework.
Using a parser as the basis for various linguistic tools can simplify many tasks, e.g,, syntax sensitive re-coding. It turned out to be a trivial task to add an official-to-mad coder process to my parser, and adding a mad-to-official coder is just mechanical work now that the basis for re-coding is there. After a two-way re-coding mechanism is available, the parser can be used to ascertain that a supposedly one-to-one change doesn't break anything. The re-coding process doesn't appreciably affect the parsing time, even for Alice the increase is of the order of one second.
If we ever get so far that we are seriously considering moving on from the baseline, we need tools like this to study the effect of the proposed changes. I also think that before that stage it would be quite useful and wise to do various kinds of hypothetical case studies, which would help to chart the territory. It is well nigh impossible to say anything definite on the basis of a few hand-crafted examples. A proposed change may look good in theory but turn out to be a flop in practice, and vice versa. Pushing sections of the existing corpus through an automatic coder and looking at the result will tell a lot more about various aspects (intelligibility, learnability, aesthetics &c.) and perhaps help to overcome prejudices concerning a proposed change, both pro and con, that is.
veion