Re: Command interfaces like the Lisp Machine

On Tue, 11 Mar 2014 22:22:49 +0700
Bruce Mitchener <[email protected]> wrote:

TL;DR: Dylan needs to re-evaluate its own identity to be straight
about the segment it wants to attract.

> I've been using it in a new build tool that I call "deft" or "Dylan
> Environment For Tools". Deft provides things like a package
> description and will know how to do builds with just "deft build" and
> tests with "deft test" and so on.  This should be a big
> simplification for people learning or using Dylan.

This is very nice and it help new users to embrace Dylan!

> Thoughts?

As you once sain on IRC: Dylan desperately needs new/more users and
that's why I believe it would be good to state more explicitly what it
is good at or why should one use it?

There is fierce competition on the programming language's market and
it is smart to estimate the strength/weakness of competition. Here,
based on my own experience, Julia is very strong competitor in the
nich of tge lanugage trying to build a bridge between static & dynamic
worlds.

While I was following Julia's mailing list in the short span of time
there were several requests/queries to make Julia more 'general
purpose' programming language.

Here, imho, lies the great opportunity for Dylan - people (still)
wants 'general purpose language' to provide Python-like ease of
development, decent performance, non-Lispy syntax and to be suitable
as 'general purpose' language.

Many new languages (e.g. D, Rust, Nimrod...) are trying to replace
C(++) and serve as 'system languages', but they are not best options
ofor general desktop/gui apps.

In the ML-camp, langs like Haskell, OCaml are also not the simplest
option for such purpose and/or there are not many examples pf apps
written in them in the above mentioned niche.

If however, the intention of Dylan hackers is to server some other
more specific niche, it would be good to mention it as well.

So, the present attempt to get "deft" in, I'm sure, the step in the
right direction. However, when it is going to be followed by
availability of REPL (it's hard to find serious language without it
today out of C(++) ) and GUI bindings (improving GTK, Qt or some
declarative DUIM-based model based on those), there is good chance to
attract new flock of users who can turn into devs later.

Some of the things which Bruce presented in his email, are falling,
imho, into the category of "it would be nice to have" but also might
be YAGNI for many of potential new Dylan users.


Sincerely,
Gour
 

-- 
When your intelligence has passed out of the dense forest 
of delusion, you shall become indifferent to all that has 
been heard and all that is to be heard.

http://www.atmarama.net | Hlapicina (Croatia) | GPG: 52B5C810