Re: Lot's of new (incompatible) Lisp dialects

In article <87wsvtg0r2.fsf@xxxxxxxxxxxxxxxxxx>,
"Perry E. Metzger" <perry@xxxxxxxxxxxx> wrote:

Rainer Joswig <joswig@xxxxxxx> writes:
New Lisp dialects seem to be the new hobby. They are mostly
incompatible to existing and emerging dialects.

Proponent@xxxxxxx writes:
This makes one worry that no replacement for CL will ever become
sufficiently widespread as to be truly viable. Is it perhaps in the
nature of people who are attracted to Lisp that they tend to go off
each in his own direction?

New Lisp dialects are not a new hobby -- they are a very old
hobby. From 1960 to about 1985, new lisp dialects flourished, and with
them new ideas and new visions of what Lisp could do and what it
could become.

On a regular basis, new Lisps would appear, and with them, new
frontiers were crossed. Lexical scoping, object systems, even such
basic tools as strings as a first class type, were tried out in new
dialects, and ultimately spread and made the community stronger.

I doubt that you need a new Lisp dialect to try out
dot notation for access of array data. Not in 2007.
Not in 1980.

Since the development of Common Lisp and the AI winter, it has become
unfashionable to create new dialects, and a dim view has been taken of
experimentation.

The experimentation has moved to a different level. Once
you have a stable base language, you still can experiment with
libraries, interfacing, tools, ...

Some claim that new dialects divide the community,
and that there is nothing to learn from the new dialects.

I won't say that there is nothing to learn from new dialects.
The implementor of a new dialect will learn how difficult
it is to create a good language, a good implementation,
good documentation and good libraries. Maybe some person years (ten?) to
get basics going. I have high respect for that.

Orthodoxy
has set in, as well as stagnation, and the main dialect has remained
largely untouched in fifteen years, on a basis that is 25 years
old. The community has also, sadly, turned inwards and looked back to
the past instead of towards the future -- one hears constant laments
for Lisp Machines and other parts of the glorious golden age of yore,
instead of about what we can do next.

Personally I'd like to move back and forward at the same time.
There is the vision of a workstation that is built for
the (Lisp) programmer. From the ground. Then there is the
vision of a software environment that is powerful enough
to support programming for most current purposes. Then there
is the vision of being able to develop state-of-the art
software that integrates with current operating systems,
so that you get the power of interactive programming even
for those. These visions have some overlap. Some people
are trying to work on different versions of the vision.
For example to make Lisp implementations good citizens
in a GNU/Linux environment (there is a log activity in this
area). But these different visions can be moved
forward while using the same base language. It makes
the exchange of ideas and code much easier.

I'm personally totally uninterested in 'new' 'languages'
that focus on micro-advancing the state of syntactic
expressibility. Also, if a new language is developed without
software, it is not worth it. If it claims to be better,
more modern, whatever, it should be able to demonstrate that
and to compete with other Lisp systems. I'm somehow interested
in a Lisp implementation, but I'm MUCH more interested
in using/writing software applications written and extended in
(Common) Lisp.

That's also what has happened. Quite a few people have 're-'
discovered Lisp and tried to apply it to some (new) domains.
There are now several active implementations and many people
able to maintain them. There are now people experimenting
how to develop software, what the environment should look
like for that (given the current surroundings), what
tools one needs in the Internet age (project sites,
chat channels, web forums, bots, paste boards, ...).

The emergence of new dialects of Lisp is not a sign of trouble. It is
a sign that the community is finally regaining its health. Springtime
has come at last after a long ice age, new flowers are growing where
the glaciers have receded. I'm glad of it.

For me the sign of spring would be Lisp software, not Lisp
dialects.


Let a hundred flowers bloom: let a hundred schools of thought contend.


Perry
.

Relevant Pages

  • Re: threading and multicore on the "free" lisps -- any unified interface similar to CFFI for
    ... constant cheerleading for clojure actually makes clojure look lamer ... One of the key points that attracted me to the Lisp community is that they are typically very honest about strengths and weaknesses of the respective dialects, ... What we don't need are the cheerleaders and hooray-sayers you typically find more often in other language communities, especially of hyped languages. ...
    (comp.lang.lisp)
  • Re: Lots of new (incompatible) Lisp dialects
    ... New Lisp dialects are not a new hobby -- they are a very old ... CL, static compilation at the cost of language dynamism from C/C++, notation from APL, and type inferencing from some retarded family of languages I forget. ... "Algebra is the metaphysics of arithmetic." ...
    (comp.lang.lisp)
  • Re: An Arc Tangent
    ... you're basically changing the ... So I wouldn't agree that language experimentation has stopped ... fanfare and a new catchy name just to be considered new dialects. ... Lisp dialects are "silent". ...
    (comp.lang.lisp)
  • Re: Lambda lists.
    ... dialects already. ... possibility of runtime lamba lists manipulations, ... i want all the flexibility Common Lisp has down ... even all of that language, if that's your idea of a good time. ...
    (comp.lang.lisp)
  • Re: Is function overloading per valence good or bad thing in lisp
    ... been many Lisp dialects. ... I don't favor overloading as a language construct. ... Although a few details of the congruency rules seem wrong to me, ...
    (comp.lang.lisp)