GSOC Introduction

[John Casey]

Hi all,

So I've been crushed by mid-terms for the last three weeks, as a result, I'm only now finishing my proposals for GSOC 2017.

OSS has been an interest of mine for a while and GSOC has been the push I needed to actually get involved.

Looking at the issues list, much of it seems to be far more complex than could be satisfied over the summer. However, the issues raised in #1117 seem appropriate to me.

I like the idea of improving the usability of the project because a scary or non existence interface can often cause users move away.

Playing with GTK, I think many of these issues should be suitable for GSOC.

If anyone has any usability/GUI tickets that you think I should look at do let me know.

So, I'm gonna call this fashionably late.

John.

[Marek Marczykowski-Górecki]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On Sat, Apr 01, 2017 at 01:52:27PM -0700, John Casey wrote:
> Hi all,

Hi!

> So I've been crushed by mid-terms for the last three weeks, as a result,
> I'm only now finishing my proposals for GSOC 2017.
>
> OSS has been an interest of mine for a while and GSOC has been the push I
> needed to actually get involved.
>

> Looking at the issues list <https://github.com/QubesOS/qubes-issues>, much

> of it seems to be far more complex than could be satisfied over the summer.
> However, the issues raised in #1117 seem appropriate to me.

This list is very broad, I strongly suggest focusing on some subset of
it. Have you seen GSoC ideas list[1]? There are also GUI and/or UX
related ideas.

As for ideas not mentioned there, take a look issues with label "UX".
For example those:
https://github.com/QubesOS/qubes-issues/issues/2304
https://github.com/QubesOS/qubes-issues/issues/2329
https://github.com/QubesOS/qubes-issues/issues/2139
https://github.com/QubesOS/qubes-issues/issues/899
https://github.com/QubesOS/qubes-issues/issues/952

> I like the idea of improving the usability of the project because a scary
> or non existence interface can often cause users move away.
>
> Playing with GTK, I think many of these issues should be suitable for GSOC.

Oh, and there is one more thing - VM settings window. In Qubes 4.0 we're
going to decompose Qubes VM Manager into separate applications[2]. VM
status monitoring (moved to Xfce4 panel applet) is already handled by
Kalkin-, but it isn't all. Initially we've planned to reuse old Qt-based
VM settings window as standalone application and simply call it from the
menu and/or that applet. But if you'd like to rewrite it in GTK as GSoC
project, that would be great! That old Qt-based code isn't perfect piece
of software... There was an attempt to do that already and also some
code is available[3], but it's far from complete and finishing that work
(if any of that would be good enough to use...) would still be a perfect
3-month task.

> If anyone has any usability/GUI tickets that you think I should look at do
> let me know.
>
> So, I'm gonna call this fashionably late.
> John.
>

[1] https://www.qubes-os.org/gsoc/
[2] https://github.com/QubesOS/qubes-issues/issues/2132
[3] https://github.com/QubesOS/qubes-issues/issues/1870

- --
Best Regards,
Marek Marczykowski-Górecki
Invisible Things Lab
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBCAAGBQJY4ByQAAoJENuP0xzK19cs0dkH/1mRoFW2LSzzPyCwFOQn6+Cc
U7HTDi0vC6mQ5l8DZIkacTCl1mkrhdNQI1wKA2fRr2rJzhR4d6RqwHSr798vgd7J
GObKMuhErjSZaPtcYuy+JQYnf3x3aiwGowCn8ytCuhTzto8fG2eJ/nJJoqO8X6d/
VBbMDFNKea6cK36nQJ2fIgjP3wrvrlGXAQlidbWGIfZzAPWBPu+Op7NK8qoFW6jH
2yNq56+1ZuggtI9qknjbVell/F2PQ943XZ3m8jLRdgNXFF4/AVHu0jpPMFoApiY9
x26OvcfbqiNUexIp/Wfg/rOJT03L0mHGC6TOmdNxbI1J0VwvSgRAukgxDtc9nFg=
=6joP
-----END PGP SIGNATURE-----

[John Casey]

On Saturday, 1 April 2017 22:33:10 UTC+1, Marek Marczykowski-Górecki wrote:

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On Sat, Apr 01, 2017 at 01:52:27PM -0700, John Casey wrote:
> Hi all,

Hi!

> So I've been crushed by mid-terms for the last three weeks, as a result,
> I'm only now finishing my proposals for GSOC 2017.
>
> OSS has been an interest of mine for a while and GSOC has been the push I
> needed to actually get involved.
>
> Looking at the issues list <https://github.com/QubesOS/qubes-issues>, much
> of it seems to be far more complex than could be satisfied over the summer.
> However, the issues raised in #1117 seem appropriate to me.

This list is very broad, I strongly suggest focusing on some subset of
it. Have you seen GSoC ideas list[1]? There are also GUI and/or UX
related ideas.

As for ideas not mentioned there, take a look issues with label "UX".
For example those:
https://github.com/QubesOS/qubes-issues/issues/2304
https://github.com/QubesOS/qubes-issues/issues/2329
https://github.com/QubesOS/qubes-issues/issues/2139
https://github.com/QubesOS/qubes-issues/issues/899
https://github.com/QubesOS/qubes-issues/issues/952

Reading those, it strikes me that they could be a useful "Hello, World" for contributing to Qubes. Especially seeing as many of them involve interacting with different sub-systems within Qubes. 

> I like the idea of improving the usability of the project because a scary
> or non existence interface can often cause users move away.
>
> Playing with GTK, I think many of these issues should be suitable for GSOC.

Oh, and there is one more thing - VM settings window. In Qubes 4.0 we're
going to decompose Qubes VM Manager into separate applications[2]. VM
status monitoring (moved to Xfce4 panel applet) is already handled by
Kalkin-, but it isn't all. Initially we've planned to reuse old Qt-based
VM settings window as standalone application and simply call it from the
menu and/or that applet. But if you'd like to rewrite it in GTK as GSoC
project, that would be great! That old Qt-based code isn't perfect piece
of software... There was an attempt to do that already and also some
code is available[3], but it's far from complete and finishing that work
(if any of that would be good enough to use...) would still be a perfect
3-month task.

That sounds fantastic, as it would definitely improve my GTK knowledge. I see that there hasn't been a commit to [0] in almost a year. Has bnvk changed direction, or is it just waiting for someone to pick up the code and work on it?

Would rewriting it in GTK be busywork, or would it be more effective to learn and build it in Qt. Does Qubes plan to change to GTK entirely?

> If anyone has any usability/GUI tickets that you think I should look at do
> let me know.
>
> So, I'm gonna call this fashionably late.
> John.
>

[1] https://www.qubes-os.org/gsoc/
[2] https://github.com/QubesOS/qubes-issues/issues/2132
[3] https://github.com/QubesOS/qubes-issues/issues/1870

- --
Best Regards,
Marek Marczykowski-Górecki
Invisible Things Lab
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBCAAGBQJY4ByQAAoJENuP0xzK19cs0dkH/1mRoFW2LSzzPyCwFOQn6+Cc
U7HTDi0vC6mQ5l8DZIkacTCl1mkrhdNQI1wKA2fRr2rJzhR4d6RqwHSr798vgd7J
GObKMuhErjSZaPtcYuy+JQYnf3x3aiwGowCn8ytCuhTzto8fG2eJ/nJJoqO8X6d/
VBbMDFNKea6cK36nQJ2fIgjP3wrvrlGXAQlidbWGIfZzAPWBPu+Op7NK8qoFW6jH
2yNq56+1ZuggtI9qknjbVell/F2PQ943XZ3m8jLRdgNXFF4/AVHu0jpPMFoApiY9
x26OvcfbqiNUexIp/Wfg/rOJT03L0mHGC6TOmdNxbI1J0VwvSgRAukgxDtc9nFg=
=6joP
-----END PGP SIGNATURE-----

[0] https://github.com/bnvk/qubes-manager-new 

[Jean-Philippe Ouellet]

On Sun, Apr 2, 2017 at 1:50 PM, John Casey <john.c...@ucdconnect.ie> wrote:
> On Saturday, 1 April 2017 22:33:10 UTC+1, Marek Marczykowski-Górecki wrote:
>> On Sat, Apr 01, 2017 at 01:52:27PM -0700, John Casey wrote:
>> > I like the idea of improving the usability of the project because a
>> > scary
>> > or non existence interface can often cause users move away.
>> >
>> > Playing with GTK, I think many of these issues should be suitable for
>> > GSOC.
>>
>> Oh, and there is one more thing - VM settings window. In Qubes 4.0 we're
>> going to decompose Qubes VM Manager into separate applications[2]. VM
>> status monitoring (moved to Xfce4 panel applet) is already handled by
>> Kalkin-, but it isn't all. Initially we've planned to reuse old Qt-based
>> VM settings window as standalone application and simply call it from the
>> menu and/or that applet. But if you'd like to rewrite it in GTK as GSoC
>> project, that would be great! That old Qt-based code isn't perfect piece
>> of software... There was an attempt to do that already and also some
>> code is available[3], but it's far from complete and finishing that work
>> (if any of that would be good enough to use...) would still be a perfect
>> 3-month task.
>
> That sounds fantastic, as it would definitely improve my GTK knowledge. I

> see that there hasn't been a commit to in almost a year. Has bnvk

> changed direction, or is it just waiting for someone to pick up the code and
> work on it?

See [0], especially bnvk's note at the bottom.

[0]: https://github.com/QubesOS/qubes-issues/issues/1870#issuecomment-223055937

> Would rewriting it in GTK be busywork, or would it be more effective to
> learn and build it in Qt. Does Qubes plan to change to GTK entirely?

See discussion in
https://groups.google.com/d/topic/qubes-devel/2pWolRF3otY/discussion

AFAIK the current answer is "Yes, the plan is to change to GTK
entirely", but it's not clear to me whether or not this is the best
plan.

[John Casey]

Thanks, that explains rather bare-bones nature of the above repo. That comment also serves as good introduction to the requirements for any implementation independent of the GTK vs Qt discussion.

 

> Would rewriting it in GTK be busywork, or would it be more effective to
> learn and build it in Qt. Does Qubes plan to change to GTK entirely?

See discussion in
https://groups.google.com/d/topic/qubes-devel/2pWolRF3otY/discussion

AFAIK the current answer is "Yes, the plan is to change to GTK
entirely", but it's not clear to me whether or not this is the best
plan.

I find it very interesting the process for choosing a technology in open source projects. In addition, it seems that no matter which is chosen, there will always be those who would have preferred the other. An example being the XFCE vs KDE argument I've seen elsewhere in the Qubes community.

So the Gnome change has been pushed into the future, and as a result, a lot of the benefit of GTK as also been pushed to the future. Is that approximately correct?

Additionally, I haven't been able to find any current sources for a Gnome transition to Qt, and that discussion is from 2013.

Finally, I feel the need to underline by inexperience in these matters as until the last month my knowledge of OSS has been limited to download, build and run. So take my input with a pinch of salt.

Regards,

John.