tunes poll -- please answer
Tue, 10 Jan 95 18:44:41 +0100
I just joined this mailing-list yesterday and promptly got that poll
form. Since I only have a vague idea of what Tunes is about, I am a
bit unsure about what to write in the form. Some questions are very
specific and answering them properly requires knowledge of what has
already been discussed. How do I catch up with those 1.5M of logfiles?
Some kind of digestion mechanism might be useful to browse the
messages. But Emacs M-x undigestify-rmail-message says: Not a Digest.
Any hints on how to do this?
Oh, Farë just sent the url for the tunes home page, let me see ..
looks interesting ... still, a digest is not provided.
I think i'll just tell you what I can right now and maybe send an
PS: One more question: Iff one does academic work in a frame like
this, how can one make sure to get credited for it?
*********** Identity ***********
** E-Mail address:
email@example.com (until summer)
** Network capabilities:
any; fast to usa, slow to fr or uk; dunno why.
** Aux addresses:
firstname.lastname@example.org (probably more permanent)
** Physical address:
home: Stettiner Str.61, 37083 Goettingen, Germany
office: Drittes Physikalisches Institut
Buergerstr.42-44, 37073 Goettingen, Germany
** Phone Number:
++49+551 -7702884(home) -397731(office) -397720(fax)
** Birth day:
** Work/Study situation:
preparing masters thesis in physics on the topic of speech
recognition by artificial neural networks
*********** Computer Background ***********
** computer languages fluently spoken (in order of decreasing interest):
c/c++, pascal, html
x86 assembly, apple2 stuff (graforth, basic, 6502 assembly :)
** languages I'd like to know:
prolog, haskell, beta
** The language I'd like to use as Tunes' HLL:
i'd like the DREAM language, of course.
- based on ideas from lisp and self
- reflects as much as possible:
- syntax (parser&grammar a la mango represented in and for itself)
- semantics (accesible interpreter/compiler/control-structures)
- code represented as data, a la lisp
- allow modification of own implementation (for lisp this would
mean access to cells of symbols, for self access to maps etc)
- allow creation of new low-level types including explicit handling
of pointers (like in C). of course, these should be wrapped in such
a way that they are safe to use. an example would be to introduce
conses to self.
- allows use of multiple parsers/readers/printers
=> pick your favorite syntax style,
even switch it in a single 'source-file' (if one uses such)
- get rid of the concept of source files
if this should still not turn out to be the ultimate language, it
would at least make it easy to build it. THINK BIG!
** political *computer* opinions:
no patents on ideas (algorithms), only on implementations
put more work in deciding what to work on
prevent work from being done twice
** Architectures/OS combination I can work on:
sunos, solaris, ultrix
*********** Parts of the system I'm ready to develop: ***********
** New protocols I'd like to design:
high squared language
** Existing protocols I'd like to implement:
** Architectures I'd help write the low-level stuff for:
*********** Parts of the system I'm eager others to develop: ***********
** New protocols I'd like to see designed:
** Existing protocols I'd like to see implemented:
** Architectures I'd like the system to be ported to:
*********** Project Management ***********
** How much I am available:
not (for six months)
** How I think the project could be managed better:
- use www to coordinate.
- provide digests of discussed threads. once they subside,
boil them down to some sort of annotated document. use a
format that can be converted to html, of course.
(read `As We May Think' by Vannevar Bush. the url is
we should try to use the web like he would have meant it to be used.)
** Sub-project I'm ready to manage:
newbie => none yet
** What I think about the poll:
interesting to learn about you all