[Trisquel-devel] GNUkCaH || gnuHaCk

[Aitor Ruano Miralles]

El ds 08 de 08 de 2009 a les 14:48 -0700, en/na william Herrera va
escriure:
> Hi,
> 
> I'll try to be as clear as possible ;) with this FaQ.
> 
> Well in first place thanks to Aitor, Germán and Ricardo to take into
> account this project, Trisquel GNU/* is a vibrant multi-language free
> software community and hope it remains this way. Also it is a 100%
> free libre distro :) heh
> 
> Dear hackers, certainly this is a set of growing ideas, so I'm still
> trying to order them. Initially when I commented the GNUkCaH ||
> gnuHaCk, named projects like GAP and Etoile cause I read about them,
> each one with different goals.
> 
> Is this an official desktop for Trisquel?
> Not now, this is just an idea. Rubén said we have the distro support
> to experiment. So maybe when it is ready, might get an official state
> for Trisquel and GNU.
> 
> Why this is neither GAP nor Etoile?
> I see GAP as project which is trying to develop as much applications
> as possible written in GNUstep platform in order to complete a
> “Desktop Environment” experience. Etoile is trying to “?” provide an
> user environment with project and document oriented in mind and more
> very nice concepts but with this “open-source (MIT/BSD licensed)” so
> obviously it out of focus.
> 
> Why GNUstep?
> GNUstep is one of the good options talking about software development
> platforms with a great performance, OO, high engineering specification
> with a fast development and deployment time, also features a dynamic
> and simple Objective-C language, so after a long research process
> this(GS) is the best option.
> 
> Are there clear objectives?
> Some are clearer than others, but sure there are. Let's name a few:
> * Low libs/apps dependency.- for ease of maintenance and distro
> packaging.
> * Completely FREE/libre GNU/fsf community project.- including
> copyrights, licenses,  documentation, standards, philosophy. I call it
> “GNUprestige” initiative.
> * Modern, functional and intelligent.- Emacs style showed to be
> intelligent, GNOME/KDE showed to be productive also GAP is trying to,
> Etoile is adding new concepts(modern).
> So? Well, I call it “Embedded I-Functionality” through emacs like*
> modes integrated natively; featuring a very good level of production
> experience through dedicated applications while it remains a modern
> environment.
> 
> Is “Embedded I-Functionality” possible and how it would be?
> Yes, It's been possible before, but this time with a modern approach.
> It would be a single decentralized dynamic piece of software,
> executing what the user really really needs.
> * MINI-applications.- dedicated workers doing what they need to do and
> just that, targeted to their domains, no extra features. Example: A
> music reproducer only plays music, an image and document viewer only
> shows them, and so with groupware, etc...
> 
> * MAJOR-applications.- bigger instances of user environment, I mean to
> all those tasks developed in scenarios where a considerable amount of
> functions are provided. Example: Emacs, Terminal/Console, Web,
> User/File Space. This can be seen as “Worlds” representing current
> activities in real computing, so the user might say: “Now I'm browsing
> the web using a set of web collaborative tools” and all around you is
> web, no menus no app's, just a real web instance and not a simple
> browser; “I'm passing commands” and you get a modern console with a
> toolbox of commands, a deeper design about what is a X-terminal, “I'm
> in emacs now” and you really are inside emacs. This way the desktop
> turns into anything you want it to be. If you want it to be a “point
> of sale” it will be; an “internet box” for netbooks, it will be; a
> “secretary workstation” it will be. Everything should be possible
> through mini-apps and its mini-jobs.
> So we can say we have a Web instance, a Terminal instance, a User/file
> instance, Emacs instance, and the whole desktop is what the instance
> is, like different “views”.
> 
> * I-applications.- these may work as common collaborative functions,
> here is where the intelligence come from, the “hard work”, “the
> brain”. Simply provide extra functionality  robbed from original
> applications to share with the desktop between mini-applications or
> major-applications(if needed), like Robin Hood .  Example: Organize
> playlist files, manage image galleries, document libraries, folders,
> calendars, to-do lists, processes, workflows, etc. Just like a super
> intelligent file manager and more.
> 
> Isn't it too complex?
> I don't think so, if you realize is a factoring procedure like in
> algebra, taking a big redundant desktop, minimizing it. Thus, instead
> of provide a set of full featured applications on the top of an even
> more full featured environment; You just get a set of non featured
> applications on the top of  mid featured environment(major apps),
> living in harmony through intelligent applications(i-apps) serving
> cool modern extra functionality.
> MINI-applications(simple tasks) will be inside
> MAJOR-applications(environments) and I-applications will be where they
> are needed, MINI or MAJOR a sort of “agents”.
> 
> What would be the relation with GAP?
> Ricardo has wrote recently, and I'm happy for that, but as people can
> see this is a completely different thing. We will use some GAP bits
> and it might get some bits from here, do not really know, perhaps a
> fusion... But anyway Happy Hacking...
> 
> What is the problem with the “application” concept?
> None at all. All this is about “applications” but performing only its
> essence, leaving app-features for another component.
> 
> What is the problem with GNOME or KDE?
> There are no real problems, just a fact of personal dissatisfaction
> when I see official GNU projects that are far away from its standards
> and freedom positions. Maybe I see these projects in a comfortable
> area where innovation has been replaced by “simple mass industry
> products” for the user.
> 
> Note: This is for GNOME, KDE, GNUstep, etc.. It is important to take
> care of what libraries or applications are classified as officials.
> Sometimes it's hard look inside a deep sea of bits and check what is
> this or is not. In GNUstep case is very clear, that is a reason why we
> are using it, but still should be more precise.
> 
> Are we interesting in become GNUstep hackers?
> Yes, why not... since this is development platform, is important for
> all of us. But following and make others follow GNU standards. Simply
> because is an Official GNU project. I know GS has a statement that ask
> for copyright assignation to the FSF for large code.
> 
> Are non-GNU projects bad things?
> Absolutely NO.. they are good things.. We are just trying to promote
> the “GNUprestige” and Free Software movement. This is NOT about hate
> to other movements, they have their positions and that's it. Equality,
> peace and love among the differences.  :)  We're 'n You're frEE... as
> in frEEdom
> 
> And the look and feel?
> It is obvious that we are missing the “state of art” thing, We have
> not dig inside GNUstep yet, but certainly much work have to be done.
> 
> What about the Window Manger for the project?
> A new concept must be introduced here as well, I've said in the -devel
> list that we're not trying to clone NeXT environment, not porting
> GNOME/KDE to GNUstep, but creating from the ground.
> One point might be, an appIcon Manager + appAlien Manager +
> viewManager. I explain myself.
> According to the specification used by GNUstep, applications show an
> appIcon box when they are being executed, which is not really
> attractive. So a beautiful “state of art” appIcon Manager would be
> perfect. The alien manager would work the same way but for external
> applications. The view manager should provide a flexible “vista” for
> working instances or applications, allowing the user to call what he
> wants to see in the desired way, mixed, simple, etc.. This is even a
> more experimental concept.
> 
> Traditional applications will exist?
> Yes, there are thousands of applications and we should provide an
> app-launcher for alien applications and a better user experience.
> 
> Some Points
> * I tried to explain some technical and low-level concepts, also I
> will write other answer with some more graphical and functionality
> behavior.
> * If you guys have some ideas, they are welcome...
> * Aitor you really got the point....
> * I do not have GNUstep skills(neither Aitor, I guess), so this is a
> research stage.. we need to handle the platform first. And Germán
> documentation is great, the offical GS doc too.
> 
> Thanks,
> 
> William H.
> GNU – Resistencia Digital

You are welcom William :)

I think with this all the main ideas are very clear, great work
William :P

As you have said, maybe not everybody is get used to GNUstep, and we
need to do a little bit of research on it. For that I recommend this:
http://io.debian.net/~tar/gnustep/

It's an easy to use debian based live CD with the GNUstep framework
implemented on it, so you can try it without having to install anything
on your computer, it has by default all the GNustep stable (more or
less) applications, including WindowMaker of course, also the developers
ones, such as gorm, project manager, gWorkSpace, etc...
I haven't tried to install it permanently, but I supose it's possible,
so it's a nice platform to work on. The only problem I noticed is that
its kernel is a bit old, I don't remember the version at all but I think
it was 2.6.25, so there can be problems with hardware detection, in fact
I had :(
By the way, 100% free as in freedom :)

Now we have the ideas clear, we should start to write a list of tasks,
from more important to less and start hacking and documenting the
progressions. I have never developed a desktop environment so I have no
idea what we have to do first, I hope somebody knows :P


Thank you for understanding my horrible English! ;)
> 

-- 
Aitor Ruano Miralles <aitux.boot en gmail.com>