Overview of Tunes

Brian Rice water@tunes.org
Wed Mar 28 14:50:03 2001

>I cannot seem to find simplified, summarized explanation for Tunes' concept.
>From reading the archive, I think I understand the grand scope of Tunes but
>I'm not clear as to the BIG CONCEPT, the specs or the goals are.
>Can anyone direct me to a URL which is a bit more illuminating and
>elaborative than the tunes.org website?  Or does anyone care to enlighten me
>with a quick briefing?
>I would be appreciative.
>Chad Prucha

This honestly depends on your background and education level.
I'll start with the RTFM-style URLs:

Disclaimer: I am in no way suggesting that the way TUNES' ideas have 
been published and expressed is in any way helpful. It's just that 
the nature and state of the project don't attract people with the 
time to devote to improving it (yet).

Now, with that out of way, you first need to know that this project 
has an extremely broad scope in the end. At the beginning, it just 
defines a class of programming language environments which are 
believed to lead to fulfilling that scope. Many people wonder why we 
must go out of our way to try and define new languages, concepts, 
etc. to achieve this goal.

This is where the education level comes into play. There are various 
logical limits to the way we are doing computing today that affect 
whether TUNES' end scope can be realized. There is *also* a strong 
suggestion in the patterns of computer science research that there 
can be made a way to build the right kind of system, and that we 
don't know how to do it yet. I should also stress that there is 
nothing stating that the goals of TUNES are impossible. To be 
involved with TUNES development right now means you have to be 
familiar enough with the research to define TUNES ideas in terms of 
them, and not rely on vague explanations and false promises.

Specifically, TUNES has a few levels that can be talked about. The 
first, (the one I think is most essential to it) is the acronym.: 
Useful Not Expedient Systems. This defines a really broad class of 
systems, but none has been realized which encompasses every possible 
implementation of that. A lot of people argue that some systems that 
we have today are enough to be called TUNES. I suggest that their 
horizon of acknowledging the nature of different concepts is not 
broad enough.

At the more specific levels, TUNES is also co-opted by Fare, since he 
pretty much wrote most of the site's conceptual material, and is also 
one of the few founding members who is still (even only slightly) 
vocal about anything in the project. Jecel is even more outspoken 
that he, although he has a different vision. Because of this, some of 
the TUNES vision may have been lost along the way. It's hard to say 
for me. I followed this project continuously from just a couple of 
months after their website was up until the present, but I didn't 
subscribe or join until 2 and a half years ago.

In terms of people and materials, there are a few people like myself 
doing really hard research and experimental coding in various 
directions. There are a few low-level coders who are waiting for the 
chance to write (OS-style or even language-style) bootstrapping code, 
and possibly don't care for the philosophy per se. And then there are 
several dozen people who subscribe to this mailing list who have 
either asked one question or never said anything. As for resources, 
we have Tril who spends a little of his scarce time maintaining 
Bespin as the server. A few others have mirror servers around the 
world. At least a dozen more promise such time once we have a 
concrete plan to implement. But basically only a very few make this 
their own work, or spend more than a tiny fraction of their time even 
thinking about TUNES.

I hope that helps.