Proposal for Outreachy (May-August 2020)

Proposal for Outreachy (May-August 2020)

[Raghav Gururajan]

[-- Attachment #1: Type: text/plain, Size: 1794 bytes --]

Hello Guix Folks! I would like to contribute to Guix via Outreachy program. Based on domains listed in http://guix.gnu.org/contribute (http://guix.gnu.org/contribute/), I would like to work on "Packages". After brief discussion with Tobias (nckx) on IRC, I came up with following ideas. 1) Improve GNOME Software Stack and GNOME Desktop Environment's user-experience in Guix. (a) Revise Package(s): Add missing packages, remove deprecated packages and fix erroneous packages. (b) Revise Meta-Package(s): Clean-up gnome and create gnome-minimal. (c) Revise Package Module(s): Rebase gnome.scm and gnome-xyz-.scm into gnome-core.scm, gnome-deps.scm, gnome-sdk.scm and gnome-extras.scm. (d) Revise Service(s): Add missing services, remove deprecated services and fix erroneous services. (e) Revise Service Module(s): Rebase gnome-dektop-service-type and %desktop-services into gnome-desktop-service-type and gnome-minimal-desktop-service-type. 2) Improve Xfce Software Stack and Xfce Desktop Environment's user-experience in Guix. (a) Revise Package(s): Add missing packages, remove deprecated packages and fix erroneous packages. (b) Revise Meta-Package(s): Clean-up xfce and create xfce-minimal. (c) Revise Package Module(s): Rebase xfce.scm into xfce-core.scm, xfce-apps.scm, xfce-plugins.scm and xfce-bindings.scm. (d) Revise Service(s): Add missing services, remove deprecated services and fix erroneous services. (e) Revise Service Module(s): Rebase xfce-dektop-service-type and %desktop-services into xfce-desktop-service-type and xfce-minimal-desktop-service-type. We can discuss more in detail on IRC to see the gravity of these ideas and then it can be decided whether to go with (1) or (2) or both. Please let me know about your thoughts on this. Thank you! Regards, Raghav Gururajan.

[-- Attachment #2: Type: text/html, Size: 2087 bytes --]

Re: Proposal for Outreachy (May-August 2020)

[Raghav Gururajan]

Hello Guix Folks!

I would like to contribute to Guix via Outreachy program. Based on domains listed in http://guix.gnu.org/contribute, I would like to work on "Packages".

After brief discussion with Tobias (nckx) on IRC, I came up with following ideas.

1) Improve GNOME Software Stack and GNOME Desktop Environment's user-experience in Guix.

(a) Revise Package(s): Add missing packages, remove deprecated packages and fix erroneous packages.
(b) Revise Meta-Package(s): Clean-up gnome and create gnome-minimal.
(c) Revise Package Module(s): Rebase gnome.scm and gnome-xyz-.scm into gnome-core.scm, gnome-deps.scm, gnome-sdk.scm and gnome-extras.scm.
(d) Revise Service(s): Add missing services, remove deprecated services and fix erroneous services.
(e) Revise Service Module(s): Rebase gnome-dektop-service-type and %desktop-services into gnome-desktop-service-type and gnome-minimal-desktop-service-type.

2) Improve Xfce Software Stack and Xfce Desktop Environment's user-experience in Guix.

(a) Revise Package(s): Add missing packages, remove deprecated packages and fix erroneous packages.
(b) Revise Meta-Package(s): Clean-up xfce and create xfce-minimal.
(c) Revise Package Module(s): Rebase xfce.scm into xfce-core.scm, xfce-apps.scm, xfce-plugins.scm and xfce-bindings.scm.
(d) Revise Service(s): Add missing services, remove deprecated services and fix erroneous services.
(e) Revise Service Module(s): Rebase xfce-dektop-service-type and %desktop-services into xfce-desktop-service-type and xfce-minimal-desktop-service-type.

We can discuss more in detail on IRC to see the gravity of these ideas and then it can be decided whether to go with (1) or (2) or both.

Please let me know about your thoughts on this.

Thank you!

Regards,
Raghav Gururajan.

Re: Proposal for Outreachy (May-August 2020)

[Jan]

Hi!

On Tue, 07 Jan 2020 11:32:10 +0000
"Raghav Gururajan" <raghavgururajan@disroot.org> wrote:

> Hello Guix Folks!
> 
> 1) Improve GNOME Software Stack and GNOME Desktop Environment's
> user-experience in Guix.
> 
In my opinion we should stay away from GNOME, because it is a Red Hat
project, rather than GNU now. More than this, GNOME makes some bad for
user freedom decisions, for example they promote nonfree software in
their "Software" application. They also promote nonfree services such as
Facebook, Google mail, etc. in their "Online accounts" application.
They also make some poor design decisions such as adding support for
flatpak, which is going to have support for "paid", which most
probably means nonfree applications and they add more and more
systemd-only features (systemd's problem is feature creep, software
bloat, uncooperative developers and merging projects into systemd,
without any reason).
That's why I would like to see XFCE being improved rather than
GNOME.
Ideally both GNOME and XFCE should work well on Guix, but if we have to
choose between those two, I vote for XFCE.

But that's my private opinion, it'd be better, if Guix maintainers told
what they think about it.



Jan Wielkiewicz

Re: Proposal for Outreachy (May-August 2020)

[Raghav Gururajan]

Hello Jan!

> In my opinion we should stay away from GNOME, because it is a Red Hat
> project, rather than GNU now. More than this, GNOME makes some bad for
> user freedom decisions, for example they promote nonfree software in
> their "Software" application. They also promote nonfree services such as
> Facebook, Google mail, etc. in their "Online accounts" application.
> They also make some poor design decisions such as adding support for
> flatpak, which is going to have support for "paid", which most
> probably means nonfree applications and they add more and more
> systemd-only features (systemd's problem is feature creep, software
> bloat, uncooperative developers and merging projects into systemd,
> without any reason).

Yeah, I am too disappointed by some or half of the things they incorporated into the project. The thing is, I want to bring more newcomers of GNU+Linux System into Guix. I would like to see Guix being deployed as mainstream system for domestic and commercial purposes. For domestic users, there is a high requirement for graphical interfaces. Unfortunately, currently we have only GNOME and KDE as full-blown desktop environments. They both have similar issues regarding software freedom and software creep.  GNOME Software, QtWebKIT etc. Also. they both are tightly integrated redhat creeps such as network-manager, pulseaudio, dbus and systemd. The only reasons for me to lean towards GNOME over KDE, is that 1) Being GNU Project (hope it stays) 2) Being GTK based (shared dependency with most other free softwares).

I also see another light. Enlightenment. The EFL library is independent of gnome, kde and redhat related stuffs. It has systemd only as optional dependency. Also uses connman over network-manager. May there is some hope here. Once Enlightenment grows into drop-in replacement of GNOME or KDE. I will definitely be shifting my focus from GNOME to Enlightenment.

> That's why I would like to see XFCE being improved rather than
> GNOME.

I love the design principles of Xfce. Modularity and re-usability. The only downside I face with Xfce are 1) Depends on GNOME stuff, dbus etc. 2) Not as feature-rich as GNOME or KDE. Anyway, Xfce is good light-weight DE fpr new-comers. That's why Xfce is still on my hope list.

> Ideally both GNOME and XFCE should work well on Guix, but if we have to
> choose between those two, I vote for XFCE.
> 
> But that's my private opinion, it'd be better, if Guix maintainers told
> what they think about it.

Let's not constrain ourselves by stating something as just opinion. :-) The matters you mentioned are real issues and factually causing trouble in software freedom. :-)

Regards,
RG.

Re: Proposal for Outreachy (May-August 2020)

[Raghav Gururajan]

Oh, I forgot mention something in my previous reply.

One more reason I lean towards GNOME over KDE is because of GNOME Human Interface Guidelines.

1) Cognitive Ergonomics (https://en.wikipedia.org/wiki/Cognitive_ergonomics)
2) Human-Computer Interaction (https://en.wikipedia.org/wiki/Human%E2%80%93computer_interaction)
3) Human Interface Guidelines (https://en.wikipedia.org/wiki/Human_interface_guidelines)
4) GNOME Human Interface Guidelines (https://developer.gnome.org/hig/stable/)

Re: Proposal for Outreachy (May-August 2020)

[Efraim Flashner]

[-- Attachment #1: Type: text/plain, Size: 1932 bytes --]

On Tue, Jan 07, 2020 at 07:06:58PM +0100, Jan wrote:
> Hi!
> 
> On Tue, 07 Jan 2020 11:32:10 +0000
> "Raghav Gururajan" <raghavgururajan@disroot.org> wrote:
> 
> > Hello Guix Folks!
> > 
> > 1) Improve GNOME Software Stack and GNOME Desktop Environment's
> > user-experience in Guix.
> > 
> In my opinion we should stay away from GNOME, because it is a Red Hat
> project, rather than GNU now. More than this, GNOME makes some bad for
> user freedom decisions, for example they promote nonfree software in
> their "Software" application. They also promote nonfree services such as
> Facebook, Google mail, etc. in their "Online accounts" application.
> They also make some poor design decisions such as adding support for
> flatpak, which is going to have support for "paid", which most
> probably means nonfree applications and they add more and more
> systemd-only features (systemd's problem is feature creep, software
> bloat, uncooperative developers and merging projects into systemd,
> without any reason).
> That's why I would like to see XFCE being improved rather than
> GNOME.
> Ideally both GNOME and XFCE should work well on Guix, but if we have to
> choose between those two, I vote for XFCE.
> 
> But that's my private opinion, it'd be better, if Guix maintainers told
> what they think about it.
> 

For better or worse Gnome is the dominant desktop in the GNU/Linux
market space and we do have it packaged and offer it as part of our
default desktop example. Furthermore Raghav has already been working on
making Gnome better in Guix.

As far as the privacy concerns, those can be addressed on a
package-by-package basis but shouldn't be cause to skip it entirely.

-- 
Efraim Flashner   <efraim@flashner.co.il>   אפרים פלשנר
GPG key = A28B F40C 3E55 1372 662D  14F7 41AA E7DC CA3D 8351
Confidentiality cannot be guaranteed on emails sent or received unencrypted

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: Proposal for Outreachy (May-August 2020)

[Jesse Gibbons]

On Tue, 2020-01-07 at 11:32 +0000, Raghav Gururajan wrote:
> Hello Guix Folks!
> 
> I would like to contribute to Guix via Outreachy program. Based on
> domains listed in http://guix.gnu.org/contribute, I would like to
> work on "Packages".
> 
> After brief discussion with Tobias (nckx) on IRC, I came up with
> following ideas.
> 
> 1) Improve GNOME Software Stack and GNOME Desktop Environment's user-
> experience in Guix.
...
> (b) Revise Meta-Package(s): Clean-up gnome and create gnome-minimal.
...
> 2) Improve Xfce Software Stack and Xfce Desktop Environment's user-
> experience in Guix.
...
> (b) Revise Meta-Package(s): Clean-up xfce and create xfce-minimal.
> 

Yes! I don't think I'm alone in frustration over unnecessary packages
installed systemwide. 

Re: Proposal for Outreachy (May-August 2020)

[Raghav Gururajan]

[-- Attachment #1: Type: text/plain, Size: 108 bytes --]


> Yes! I don't think I'm alone in frustration over unnecessary packages
> installed systemwide. 

:-)

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: Proposal for Outreachy (May-August 2020)

[Jan]

On Wed, 08 Jan 2020 07:06:26 +0000
"Raghav Gururajan" <raghavgururajan@disroot.org> wrote:

> Hello Jan!
> 
> Yeah, I am too disappointed by some or half of the things they
> incorporated into the project. The thing is, I want to bring more
> newcomers of GNU+Linux System into Guix. I would like to see Guix
> being deployed as mainstream system for domestic and commercial
> purposes. For domestic users, there is a high requirement for
> graphical interfaces. Unfortunately, currently we have only GNOME and
> KDE as full-blown desktop environments. They both have similar issues
> regarding software freedom and software creep.  GNOME Software,
> QtWebKIT etc. Also. they both are tightly integrated redhat creeps
> such as network-manager, pulseaudio, dbus and systemd. The only
> reasons for me to lean towards GNOME over KDE, is that 1) Being GNU
> Project (hope it stays) 2) Being GTK based (shared dependency with
> most other free softwares).
>
Okay then, if it is going to help promoting Guix, I'm okay with this.
 
> 
> I also see another light. Enlightenment. The EFL library is
> independent of gnome, kde and redhat related stuffs. It has systemd
> only as optional dependency. Also uses connman over network-manager.
> May there is some hope here. Once Enlightenment grows into drop-in
> replacement of GNOME or KDE. I will definitely be shifting my focus
> from GNOME to Enlightenment.
>
I found it a bit strange, guess I'll have to write my own DE anyway :)
> 
> I love the design principles of Xfce. Modularity and re-usability.
> The only downside I face with Xfce are 1) Depends on GNOME stuff,
> dbus etc. 2) Not as feature-rich as GNOME or KDE. Anyway, Xfce is
> good light-weight DE fpr new-comers. That's why Xfce is still on my
> hope list.
>
Understandable.
> 
> Let's not constrain ourselves by stating something as just
> opinion. :-) The matters you mentioned are real issues and factually
> causing trouble in software freedom. :-)
> 
What I meant by this is I can't really make decisions for the whole
community, because I'm just a package maintainer. But you're right
those are serious issues.
> Regards,
> RG.


Jan Wielkiewicz

Re: Proposal for Outreachy (May-August 2020)

[Jan]

On Wed, 8 Jan 2020 12:40:52 +0200
Efraim Flashner <efraim@flashner.co.il> wrote:

> 
> For better or worse Gnome is the dominant desktop in the GNU/Linux
> market space and we do have it packaged and offer it as part of our
> default desktop example. Furthermore Raghav has already been working
> on making Gnome better in Guix.
> 
> As far as the privacy concerns, those can be addressed on a
> package-by-package basis but shouldn't be cause to skip it entirely.
> 

I only hope fixing those issues won't take more effort than maintaining
our own thing/better alternative one day. It's fine when you can just
use configure flags to disable bad features during build time, but what
if they start adding proprietary parts to systemd and GNOME will use it
even harder? systemd is LGPL'd after all. 

Embrace, Extend, Extinguish.


Jan Wielkiewicz

Re: Proposal for Outreachy (May-August 2020)

[Jan]

Don't know if this is a good place, but gnome-color-manager in this
commit:
https://git.savannah.gnu.org/cgit/guix.git/commit/?id=29ae18db0578bbdbc0119703f7412fe91927e80e
has GPL2 license in the "license" field and the COPYING file also tells
the license is GPL2, but if you check a random source file, for example
https://gitlab.gnome.org/GNOME/gnome-color-manager/blob/master/src/gcm-brightness.c
Then files are licensed under GPL2 or any later.
> * This program is free software; you can redistribute it and/or modify
> * it under the terms of the GNU General Public License as published by
> * the Free Software Foundation; either version 2 of the License, or
> * (at your option) any later version.

I guess the license is GPLV2 then, but I'm not sure, can someone
confirm this?


Jan Wielkiewicz

Re: Proposal for Outreachy (May-August 2020)

[Ludovic Courtès]

Hi,

Jan <tona_kosmicznego_smiecia@interia.pl> skribis:

> Don't know if this is a good place, but gnome-color-manager in this
> commit:
> https://git.savannah.gnu.org/cgit/guix.git/commit/?id=29ae18db0578bbdbc0119703f7412fe91927e80e
> has GPL2 license in the "license" field and the COPYING file also tells
> the license is GPL2, but if you check a random source file, for example
> https://gitlab.gnome.org/GNOME/gnome-color-manager/blob/master/src/gcm-brightness.c
> Then files are licensed under GPL2 or any later.
>> * This program is free software; you can redistribute it and/or modify
>> * it under the terms of the GNU General Public License as published by
>> * the Free Software Foundation; either version 2 of the License, or
>> * (at your option) any later version.
>
> I guess the license is GPLV2 then, but I'm not sure, can someone
> confirm this?

The “or any later version” wording is present in source file headers, so
it’s definitely ‘gpl2+’, not ‘gpl2’.

I fixed this one and a couple of others in
7bd000d104f79f2b1c51c48c9d01f247189f9935.

Thanks for the heads-up!

Ludo’.

Re: Proposal for Outreachy (May-August 2020)

[Jan]

On Sun, 12 Jan 2020 00:39:55 +0100
Ludovic Courtès <ludo@gnu.org> wrote:

> Hi,
> 
> 
> The “or any later version” wording is present in source file headers,
> so it’s definitely ‘gpl2+’, not ‘gpl2’.
Right, that's what I was supposed to say, but my brain removed the "+".
> I fixed this one and a couple of others in
> 7bd000d104f79f2b1c51c48c9d01f247189f9935.
> 
> Thanks for the heads-up!
No problem.
> Ludo’.


Jan Wielkiewicz