discuss-gnustep
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Cacao Linux - GNUstep based Linux distribution


From: Gregory Casamento
Subject: Re: Cacao Linux - GNUstep based Linux distribution
Date: Mon, 22 Oct 2018 19:44:14 -0400

Hey,

All of these are valid points.  The reason I rushed it a bit was because I wanted to spark the discussion and not have it held up on the list until some unclear decision was made. 

On Mon, Oct 22, 2018 at 12:06 Riccardo Mottola <riccardo.mottola@libero.it> wrote:
Hi,

Gregory Casamento wrote:
>
> I am launching a kickstarter project to help to create a Linux
> distribution based on GNUstep.  Please take a look, feel free to make
> suggestions and comment.  The link is here:
>
> https://www.kickstarter.com/projects/203272607/the-cacoa-linux-project-a-gnustep-based-distro

We had the idea of a "reference" distribution many times and I think it
is a good way to spread the GNUstep word. To be honest, the announce is
a bit unexpected, I would have liked to discuss the main topics and
goals, if not public, at least with some of the core members.
As of now, I can only share some doubts and/or things I would have
ordered differently

I read the announce after a couple of days again, just to have things
settle down.

The way the description is stated and the first goal of 12K implies that
for it you will get a "complete" distro and 10.12 APIs.
I think first, that it is not very realistic, second that it is also not
necessary to get all 10.12 APIs to get a full working distribution with
desktop!

I’ve taken down that claim.  Actually I did so days ago. 

A subset of these APIs is needed - after all ever since 10.4 one can say
that the Cocoa APIs are very complete for a good desktop!

Yes indeed, but 10.4 is not the pinnacle of Mac / cocoa / next evolution as some might believe it to be.   I think more realistically is to bring some of these up to par with 10.12 not all. 

Less exciting, but for the first step cleaning up what we have is
perhaps enough, while the rest is the work for actually fitting all
things together for a distribution.


WebKit port and WebCore are of course very much needed  - as well as a
set of supporting other Frameworks, at least partially implemented, like
CoreFoundation and CoreGraphics and some basic Framework for device
management.

I will at least have a working webview based on cef3. 



Instead of mentioning Darling, I would have mentioned a precise set of
Applications and Utilities native for GNUstep which are going to compose
this "distro".
After all, a Distro is a sum of Apps and Frameworks on top of a Linux
foundation.

Good idea. 

Having super Cocoa API, but nothing to show it with, is of little
interest for a distribution, in my opinion.

Agreed. 

I can write it the other way around: the "project" seems to sum up
different ideas living in GNUstep: desktop within a distribution,
porting (Mac compatibility) and porting up to the extreme of Darling.

A Distribution instead for me is pointing to the end-users, perhaps more
a developer or more desktop. I would have thus put these steps for a
"Distribution". Just as an idea, nothing definitive:

1) packaging and preparation of the distribution foundation (e.g. Devuan
+ all relevant packages we want in), build architecture, etc. The
"bases". Possibly a LiveCD that can be installed. Questions about
servers and repositories need to be addressed doo
2) Enhancements and fixes to
     a) Foundation and AppKit: e.g. completion of printing, fixes for X
bugs (etc etc)
     b) completion of certain core apps that want to be included (e.g.
Debugger in ProjectCenter, completion of Bean or any other thing which
could be relevant)
3) New Frameworks, e.g. WebKit/WebCore
4) Ported apps from Mac (which? Terminal, graphics utility ..... XXXX)
5) New Apps like Installer and Login
6) APIs just enough to support 3, 4 and 5

All of these are good points.  I think we need to focus on 2,3,5,6. 


Perhaps less shiny than the promise of Darling and 10.12 api, but more
"distribution" oriented. There are a lot of hairy things.

Nothing prevents us of a separate funding for other specific things.

True



Riccardo

--
Gregory Casamento
GNUstep Lead Developer / OLC, Principal Consultant
http://www.gnustep.org - http://heronsperch.blogspot.com
http://ind.ie/phoenix/

reply via email to

[Prev in Thread] Current Thread [Next in Thread]