(2023-08-17) The future of nntrac
---------------------------------
The week has passed too quickly. But anyway, the first more-or-less working
nntrac version is on my SourceHut ([1]). I won't go over all its features as
they are described in detail in the repo's README file. But I promised to
tell you what I have been planning to do with it.
You see, I loved the idea of FirefoxOS/KaiOS: all phone userspace is
scriptable in a high-level language and non-obfuscated. The implementation
though... you know how it went. This is why I have an idea of creating my
own phone runtime (the target doesn't matter at this design stage, be it
SC6531E, MT6261 or anything more performant) that would be much better than
J2ME as it would be based upon a very lightweight and open scripting
language with a very optimized runtime. TRAC turned out to be the ideal
candidate for such a language. Especially with an implementation like mine,
which allows to easily embed new primitives into the language and the
language itself into C-based applications. And, as with any other homoiconic
language extended to the point of raw file loading, nntrac code can easily
be arranged into a set of modules responsible for various functionality. In
other words, I plan on creating a framework. The first TRAC-based framework
for featurephone applications.
I also cannot emphasize strong enough how far this language has been ahead of
its time. Especially with its "everything is a string" approach and, of
course, its homoiconicity. Too bad its history is more associated with
copyright freaks than with a technological genius of the 1960s. As long as
I'm alive, I promise to try and change that.
Stay tuned.
--- Luxferre ---
[1]:
https://git.sr.ht/~luxferre/nntrac
