I will try to decide some goals for the project:
be simplistic and failrly transparent where possible;
be usable for completing various "high level" tasks;
an acceptably competent design;
work competently as a scripting language (possibly beyond my efforts);
allow a system to exist as a distributed mesh;
software on one node should be roughly similar to as if it were on another,
and, inversly, software on another should seem as if it were local;
the creation of a large shared world (possibly a game and a number of shared
" workspaces", ie: a desktop would just be a workspace being used only by
a single person);
a rough elimination of "applications" and "file formats", data should carry
it's own operation. manipulation should be directed at the data, and not
similarly, extending the data should be transparent, nearly anything that
exists within the system should be usable as data;
it should not be necessary to deal with having to load/close applications,
or to explicitly preserve the state of things;
the system should preserve it's own state, and the state of all the data
I do not agree with the view that things should be simpler than what is
needed, but I don't believe the system should be more complicated either.
the system needs adherence to a core design, but should allow what could be
useful (including allowing the potentially impure or kludgy). impurities and
kludges often hint at weaknesses in the system.
the system should be fairly easily altered, what power the user wants should
be present, the user should not be treated like a young child.
I believe that improvement in systems is based largely on competition,
and also on the innovations of the individuals. no one should control the
system, and there should be no single system (though what ones exist should
immitation/modification/alternate views/... are encouraged, but who am
I to argue this?...
I will not claim to reach these goals, only that they are goals and I will
do what I can. nearly all that exists is replaced eventually in the change
of the systems, and my existance/efforts are no different...
2004-01-28: Software Mesh