SIGCONT

['Chris Harris' B. Harris]

Fare --

Welcome back!  Hope your exams haven't killed you....  =)


As for what you should be working on, I'm not really sure.  It seems that 
TUNES, as an entire project, really hasn't made that many decisions in 
some pretty basic areas.  (ie I still can't bring some of our high-level 
ideas (such as our definition of object) into anything that sounds 
implimentable)  So I think what you work on is really up to you.

LLL hacking might have the advantage in that we could have something to point 
to and say, "Okay, world, this is TUNES, 0.0.0.0.1a -- a small sample of 
the power TUNES will eventually have, but if it was done "wrong" or 
prematurely, you/we might end up re-writing the whole thing.  But that, I 
suppose, is inevitable -- for almost all subprojects!

A HLL definition would be helpful, since we'd be able to say to the 
world, "Hey -- look at this way-cool language we have, the basis of our 
up-and-coming OS!".  But has enough HLL research been done to be able to 
fully develop a HLL?  Behind-the-scenes code, such as a LLL, can be 
re-written without trashing all of the apps.  But is the same true for HL 
stuff?  I guess, having no applications, it wouldn't really matter.  But 
I dunno....


As for me, I'm _still_ not understanding what my Interfaces project is 
about 100%, and I'm getting the idea that everyone's view of the project 
is radically different.  My basic definition is, "the project that 
specifies how anything and everything interacts with anything and 
everything else".  Unfortunately, that sounds rather like a project 
responsible for the entirety of TUNES!

So what is it about?  Well, Fare keeps bringing up designing the HLL 
syntax.  Might work, but might it not also make sense for the HLL project 
to define the HLL syntax?  And are the semantics done?  They don't seem 
all that concrete to me, and how can you design syntax without 
semantics.  Also, who in the Interfaces project feels qualified to 
develop a syntax?  I certainly don't -- the only languages I know well 
are C/C++.  Trying, here, to pick up Scheme too.  But even if I got a 
sufficient grip on that, how does knowing 2.4 syntaxes qualify me to 
write one?  I don't think it does?

The seeming area of the Interfaces project I feel I could pull of is 
relating to UI stuff -- making worlds, or whatever they end up getting 
called, work.  I'll admit I haven't been doing the best job making 
progress on that front, and that won't be able to change until after this 
weekend (we US people like to go off and take long weekends around the 
4th of July, y'know).  But, in the long run, I think I could actually 
start "doing something".

Really, all areas of TUNES are interesting, but I just don't have the 
time/qualifications to do a lot of this stuff.  And getting more 
qualified takes more time, leaving less for TUNES, etc..  Lots of vicious 
cycles hanging around here.  Sort of like ones I see with this here 
project -- the project doesn't really move until the specs are more 
complete, but the specs don't get much more complete until people are 
moving, testing, refining (which comes about faster when the specs are 
more concrete).  So how to get out of these traps....


Well, off to eat dinner I am.  Hope all of this babble hasn't been 
totally useless.  =)

-Chris

--------------------------------------------------------------------------
"Be careful lest in casting out your devil you cast out the best thing 
that's in you." --Nietchie %% Me = Chris Harris (chharris@u.washington.edu)
Check out my web page: http://weber.u.washington.edu/chharris/

On Thu, 29 Jun 1995, Francois-Rene Rideau wrote:

> Dear joyous tunesers,
>    I've finished my exams now, so I'm back to work on TUNES !
> So, what's the next step to you ? Shall I hack on the LLL stuff,
> or try to define HLL syntax, or something else ?
> 
>    Basically, who is ready to do what on what bases ?
>    I'd like to you people to tell me more about who you are, what you
> intend to do, what you expect from others (including me), etc...