Too many ideas for one message...

[Francois-Rene Rideau]

> You've probably already explored many of the same ideas I have
> (I haven't had time to read everything in your Tunes project),
> but I have been trying to carefully think out an operating
> system that seems to be the same concept yours has.. to free
> people from the tyranny of 'computer programmers' so they can
> program the damned things themselves.
>
I'm glad to know I'm not the only one...
If we can unite, that's all the better;
if not, well, good luck to you.


> Basically, I imagined (at first) taking Smalltalk and giving
> it a set of primatives that accessed the hardware directly..
> essentially turning Smalltalk into an operating system in its
> own right.
>
Smalltalk has may good ideas,
but it lacks good modularity to be a resource-multiplexing OS
in its own right.


[Graphical language]
It's been a long time since I've began to think about a graphical language;
surely, existing development environment are lame, and I hate text-based
stuff as a universal representation.
But the problem is deeper than that: graphic or text is just the interface;
what is wrong is the design, and just changing the interface won't help.


> After more thought, I realized this pseudolanguage could be
> a kind of low-level, almost CPU-like language that builds on
> itself.
>
The idea of a virtual machine sure is very useful.
Beware, of course, that a virtual machines introduces some overhead;
interpreting VM code can be dog slow, as compared to optimized assembly code;
however, compiling/assembling the virtual-machine code
instead of interpreting it (and thus having a code adapted to that)
allows to reduce the overhead to reasonable proportions
(and fat binaries could be used for performance-critical sections).
TAOS (and surely other systems) successfully uses such techniques for
interoperable distributed programming on heterogeneous platforms.
   However, such a VM surely doesn't solve any problem
for high-level programming and using of computers,
who don't get to see it anyway.
What it does provide them is interoperability of different hardware,
which could be done using higher-level techniques.
I like the idea of VMs, but not the idea of a priviledged low-level VM,
which would be statically decided once and for all for the whole system.


> I have some other ideas as well, but if this sounds too
> corny, or you're just not interested, I don't want to bore
> you.
You're not boring me.
I've considered using a VM in the past, and after pondering it a lot,
I concluded that I didn't want a one standard VM;
of course multiple substandard VMs could exist...

> I'm simply interested in seeing something come about
> that will make object development easier, free people from
> computer tyranny, and offer perhaps a serious alternative to
> past operating systems of lesser virtue.
In my opinion, people currently lack a good high-level language.
Once the language is here, it can still have a lot of implementations,
free or not.

> I'll make money some other way.
I hope you're successful...

> Just email me if you'd like to hear more stuff.  I'll try to
> read more of the information later.. there's quite a bit,
> and I've got much to do today.
>
You might want to propose things directly on the tunes mailing list
(see http://www.eleves.ens.fr:8080/home/rideau/Tunes/mailing-list.html),
instead of just e-mailing to me...

> Take care!
Thanks a lot.

== Fare' -- rideau@ens.fr -- Franc,ois-Rene' Rideau -- DDa(.ng-Vu~ Ba^n ==
Join the TUNES project for a computing system based on computing freedom !
                TUNES is a Useful, Not Expedient System
URL: "http://www.eleves.ens.fr:8080/home/rideau/Tunes/"