Re: about founding z developement of zsh

[Sebastian Gniazdowski <sgniazdowski@gmail.com>]

On 19 June 2018 at 11:12, Marc Chantreux <eiro@phear.org> wrote:
> hello people,
>
>> The problem with that is none of the core zsh developers have
>> ever asked for contributions in a quarter century of its existence,
>> and all are unpaid volunteers.
>
> well ... i am a zsh user since circa 2001 and always felt i didn't gave
> zsh enough (same for perl, vim, linux, dwm, gnome and all those awesome
> software that gave me 20 years of fun using floss).
>
> i only have 1 patch in zsh, i don't really contribute to zsh but i try
> to be active in the ecosystem
>
> * support french users
> * promote zsh in technical mailing lists, local meetups and
>   other conferences
>   (http://act.perlconference.org/tpc-2018-glasgow/talk/7338)
> * start writing a zsh programming guide
>   (https://github.com/eiro/zsh-programming-guide)
> * start writing a plugin system with a very opiniated coding style
>   (as perlish as we can) (https://github.com/zsh-uze)
>
> basically, i try to spend some of the time i spared using zsh to promote
> it because ... gosh ... when i have to read some bash/ksh/sh/posix code,
> my eyes are bleeding and i really want to see more zsh outzere.
>
> i started a list of things i would like to see appearing in zsh here
>
>     https://github.com/eiro/zsh-programming-guide/blob/master/c/wishlist.md
>
> and i guess some topics means huge work so unpaid patches coming time to
> time are not enough.
>
> and i wonder if this is worth a founding campain... and to be honnest,
> sometimes i also wonder if a fresh start using a modern langage (like crystal or rust)
> should be an option. there is the elvish shell

I think that fresh start is a heavy thing, I recall Linus once said
"creating by destroying first". I'm not sure if the message that
"fresh start" conveys it's actually this, to turn away from current
Zsh, i.e. "destroy" (heavy word, but the discussion Linus had, maybe
even about "rewriting in C++", was little emotional). To then engage
into heritage-free coding in something that is placed as opposite to
C.

I think different approach is maybe possible. Rust allows gradual
rewrite, there's a project gradually rewriting Emacs (Wilfred/remacs).
If this would be started, then – one association comes to my mind: it
would be like in many Anime/Manga, where characters change form, but
keep their identity, and also essential traits (new form still
expresses character's main properties). ;)

Zsh would keep its identity, it would be just expressed in a fresh
form. I think this would be good to many people, to act again like
when e.g. new completion system was coded (I assume it took some time
and there was some mobilization to quickly have better product than
the previous completion system).

-- 
Sebastian Gniazdowski
News: https://twitter.com/ZdharmaI
IRC: https://kiwiirc.com/client/chat.freenode.net:+6697/#zplugin