Project status and proposal

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Project status and proposal

Gonzalo Aguilar Delgado

Hello all,

I was reading lasts emails from archive. Since I cannot respond the Alexandre e-mail I decided to open a new thread to not mix things.

It seems you are in the road to rewritting this wonderful piece of software that so many happines gave me in the past. So I can see if I can do something for the tool. Since the Gantt chart seems to be a problem I propose:

  1. Do it in Gdk - Cairo like baobab is doing. This should gave us threaded animations (nice feature) on gantt chart for free.
    1. I saw how to draw gantt charts by hacking data on a barchart. So if a library for barcharts is available I can hack a demo very fast...
  2. Since clutter will be mainstreamed and rebranded as GSK. I can also try to do something directly in GSK. That looks nice for the purpose (if ready).
    1. Other mean is to try with clutter then port to GSK.

What do you think?

I also moved to meson all my projects for building. The difference is GREAT. Compilation times and manageability are exceptional. And you don't have to wait 20 minutes for a build... I think it would be a win. Everyone is also moving.

About the use of Vala... will you reconsider it? Lot of wonderful projects are written in vala (gnome-builder for instance) and it's working nice. And no I don't think it will dissapear. Specially if we use it.

I did some hacking on baobab to add support for ceph and it was nice. Really. It saves a lot of time. Specially when you have to define the class init and object init functions, callbacks and so on. It has a lot of advantages.

About new features... I think the same than Andrew Miloradovsky. Maybe is not time for it. But giving a chance for new features it can give a boost to the project (but also kill it :-( ). Maybe new developers join... So we can propose, then it should be implemented... if possible.

So I propose the following:

  1. Add a Kanban style chart for the tasks. Grouped on milestones and with status depending if it has a resource or not, but also if someone has advanced it based on progress.
  2. Add Scrum backlog based on the tasks. And maybe grouped on sprints.
  3. Some progress reporting reports would be nice to have.

The agile development methodologies are advancing but the tools to manage them are not. Also agile methodologies have a problem when planning longterm. Thats why I use planner for the longterm management, but then I pass the tasks and milestones to a Scrum based tool where I get day to day (sprint to sprint) backlog and progress.

This could be a great win for the project and merging the best of both worlds can be a win. Cause the tools I see for Agile are specific for it. Not using a merge of Agile and traditional. And can give the project new visibility.

That's it. I know it's too much for first email hope I don't disturb much with it.


Best regards,




_______________________________________________
planner-dev-list mailing list
[hidden email]
https://mail.gnome.org/mailman/listinfo/planner-dev-list
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Project status and proposal

Alexandre Franke-2
On Thu, Aug 3, 2017 at 7:25 PM, Gonzalo Aguilar Delgado
<[hidden email]> wrote:
> Hello all,

Hi,

> It seems you are in the road to rewritting this wonderful piece of software
> that so many happines gave me in the past. So I can see if I can do
> something for the tool. Since the Gantt chart seems to be a problem I
> propose:
[…]
> What do you think?

As I suggested earlier, the way forward is to make it a bunch of GTK+
widgets (i.e. the whole chart is a widget containing widgets that each
represtent tasks, milestones…).

> I also moved to meson all my projects for building. The difference is GREAT.
> Compilation times and manageability are exceptional. And you don't have to
> wait 20 minutes for a build... I think it would be a win. Everyone is also
> moving.

Yes, meson is the hip thing and could be the way forward. However it
does not play nice for now with Python+C (+FlatPak) projects and there
is no rush to do that.

> About the use of Vala... will you reconsider it?

No.

> Lot of wonderful projects
> are written in vala (gnome-builder for instance) and it's working nice.

Builder is not written in Vala. You need to do better research.

> And no I don't think it will dissapear. Specially if we use it.

I couldn’t care less if it disappeared. I cannot recommend that anyone uses it.

> About new features... I think the same than Andrew Miloradovsky. Maybe is
> not time for it. But giving a chance for new features it can give a boost to
> the project (but also kill it :-( ). Maybe new developers join... So we can
> propose, then it should be implemented... if possible.

Patching new features into legacy code is bad practice, a waste of
time and energy, and just increases the amount of work.

> Add a Kanban style chart for the tasks. Grouped on milestones and with
> status depending if it has a resource or not, but also if someone has
> advanced it based on progress.

You’ll have to explain how to map kanban swimlanes and columns map to
Gantt concepts. I think there is a clash between the methodologies
those tools support and I am not convinced a single tool should
contain both. I do use both, but in different contexts.

> Add Scrum backlog based on the tasks. And maybe grouped on sprints.
> Some progress reporting reports would be nice to have.

I haven’t used Scrum backlogs, but they seem similar to kanban boards
so the same concerns apply.

> This could be a great win for the project and merging the best of both
> worlds can be a win. Cause the tools I see for Agile are specific for it.
> Not using a merge of Agile and traditional. And can give the project new
> visibility.

How about we make it great again for traditional planning first, then
see where we can go from there?

Cheers,

--
Alexandre Franke
GNOME Hacker & Foundation Director
_______________________________________________
planner-dev-list mailing list
[hidden email]
https://mail.gnome.org/mailman/listinfo/planner-dev-list
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Project status and proposal

Gonzalo Aguilar Delgado

Hi Alexandre,

I cannot be more aligned with you except for the vala stuff.

For the gnome-builder you were also true. I mistakenly confused myself because was investigating the vala-pack plugin that's written in vala, but that's just one plugin. baobab is written in vala and working well, but anyway. 

Once we do the migration stuff of the critical stuff. We can start a new thread for the kanban and scrum.

But first, let's clean the code out if you agree. I will start trying to make a gantt widget with new components and avoid legacy code as much as possible.

Any guidelines, or points to examples of similar widgets will be welcome.

I have to write some widgets for a system I'm developing so this will be treated as a conceptual UI reasearch under GTK. That way I can work on it while get paid.

Let me start. I will open a bug report (if it's not already open) and do a branch for the update.

Do you agree?

Best regards,


El 03/08/17 a las 20:19, Alexandre Franke escribió:
On Thu, Aug 3, 2017 at 7:25 PM, Gonzalo Aguilar Delgado
[hidden email] wrote:
Hello all,
Hi,

It seems you are in the road to rewritting this wonderful piece of software
that so many happines gave me in the past. So I can see if I can do
something for the tool. Since the Gantt chart seems to be a problem I
propose:
[…]
What do you think?
As I suggested earlier, the way forward is to make it a bunch of GTK+
widgets (i.e. the whole chart is a widget containing widgets that each
represtent tasks, milestones…).

I also moved to meson all my projects for building. The difference is GREAT.
Compilation times and manageability are exceptional. And you don't have to
wait 20 minutes for a build... I think it would be a win. Everyone is also
moving.
Yes, meson is the hip thing and could be the way forward. However it
does not play nice for now with Python+C (+FlatPak) projects and there
is no rush to do that.

About the use of Vala... will you reconsider it?
No.

Lot of wonderful projects
are written in vala (gnome-builder for instance) and it's working nice.
Builder is not written in Vala. You need to do better research.

And no I don't think it will dissapear. Specially if we use it.
I couldn’t care less if it disappeared. I cannot recommend that anyone uses it.

About new features... I think the same than Andrew Miloradovsky. Maybe is
not time for it. But giving a chance for new features it can give a boost to
the project (but also kill it :-( ). Maybe new developers join... So we can
propose, then it should be implemented... if possible.
Patching new features into legacy code is bad practice, a waste of
time and energy, and just increases the amount of work.

Add a Kanban style chart for the tasks. Grouped on milestones and with
status depending if it has a resource or not, but also if someone has
advanced it based on progress.
You’ll have to explain how to map kanban swimlanes and columns map to
Gantt concepts. I think there is a clash between the methodologies
those tools support and I am not convinced a single tool should
contain both. I do use both, but in different contexts.

Add Scrum backlog based on the tasks. And maybe grouped on sprints.
Some progress reporting reports would be nice to have.
I haven’t used Scrum backlogs, but they seem similar to kanban boards
so the same concerns apply.

This could be a great win for the project and merging the best of both
worlds can be a win. Cause the tools I see for Agile are specific for it.
Not using a merge of Agile and traditional. And can give the project new
visibility.
How about we make it great again for traditional planning first, then
see where we can go from there?

Cheers,



_______________________________________________
planner-dev-list mailing list
[hidden email]
https://mail.gnome.org/mailman/listinfo/planner-dev-list
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Project status and proposal

Gonzalo Aguilar Delgado
In reply to this post by Alexandre Franke-2

Hello Alexandre,

I was taking a look to goocanvas that seems to fit in your design. It's out of development. No new features to be added but still supported.

Instead of hacking it for a new gantt widget I'm seeing the possibility to do an intermediate step in porting the current widget to this goocanvas. Latter we can create a specific gantt widget.

This allows us to setup everything, be able to step up gtk3 and continue porting meanwhile we evaluate if a full gantt widget does applies...

What do you think?

Best regards,


El 03/08/17 a las 20:19, Alexandre Franke escribió:
On Thu, Aug 3, 2017 at 7:25 PM, Gonzalo Aguilar Delgado
[hidden email] wrote:
Hello all,
Hi,

It seems you are in the road to rewritting this wonderful piece of software
that so many happines gave me in the past. So I can see if I can do
something for the tool. Since the Gantt chart seems to be a problem I
propose:
[…]
What do you think?
As I suggested earlier, the way forward is to make it a bunch of GTK+
widgets (i.e. the whole chart is a widget containing widgets that each
represtent tasks, milestones…).

I also moved to meson all my projects for building. The difference is GREAT.
Compilation times and manageability are exceptional. And you don't have to
wait 20 minutes for a build... I think it would be a win. Everyone is also
moving.
Yes, meson is the hip thing and could be the way forward. However it
does not play nice for now with Python+C (+FlatPak) projects and there
is no rush to do that.

About the use of Vala... will you reconsider it?
No.

Lot of wonderful projects
are written in vala (gnome-builder for instance) and it's working nice.
Builder is not written in Vala. You need to do better research.

And no I don't think it will dissapear. Specially if we use it.
I couldn’t care less if it disappeared. I cannot recommend that anyone uses it.

About new features... I think the same than Andrew Miloradovsky. Maybe is
not time for it. But giving a chance for new features it can give a boost to
the project (but also kill it :-( ). Maybe new developers join... So we can
propose, then it should be implemented... if possible.
Patching new features into legacy code is bad practice, a waste of
time and energy, and just increases the amount of work.

Add a Kanban style chart for the tasks. Grouped on milestones and with
status depending if it has a resource or not, but also if someone has
advanced it based on progress.
You’ll have to explain how to map kanban swimlanes and columns map to
Gantt concepts. I think there is a clash between the methodologies
those tools support and I am not convinced a single tool should
contain both. I do use both, but in different contexts.

Add Scrum backlog based on the tasks. And maybe grouped on sprints.
Some progress reporting reports would be nice to have.
I haven’t used Scrum backlogs, but they seem similar to kanban boards
so the same concerns apply.

This could be a great win for the project and merging the best of both
worlds can be a win. Cause the tools I see for Agile are specific for it.
Not using a merge of Agile and traditional. And can give the project new
visibility.
How about we make it great again for traditional planning first, then
see where we can go from there?

Cheers,



_______________________________________________
planner-dev-list mailing list
[hidden email]
https://mail.gnome.org/mailman/listinfo/planner-dev-list
Loading...